Proefschrift Aarts

Kaft Floor - Psychological Attachment in Obesity Final.pdf 1 24-04-14 14:45 Uitnodiging Floor - Psychological Attachment in Obesity (drukbestand).pdf 1 18-04-14 15:36

Psychological Attachment in Obesity

Significance for Bariatric Surgery

Floor Aarts

Psychological Attachment in Obesity. The Significance for Bariatric Surgery

Academic thesis, University of Amsterdam, Amsterdam, The Netherlands

ISBN 978-94-6108-685-3

Author Floor Aarts

Coverdesign Coen Siebenheller, 7Pixels Media, Arnhem, The Netherlands

Layout and print Gildeprint, Enschede, The Netherlands

© 2014 Floor Aarts, Amsterdam, The Netherlands

All rights reserved. No part of this publication may be reproduced, stored, or transmitted in any

form or by any means, without written permission of the author.

Printing of this thesis was financially supported by: Stichting Klinisch Wetenschappelijk Onderzoek

Slotervaart Ziekenhuis, Academic Medical Center, Covidien, Novo Nordisk BV, Julius Clinical BV

te Zeist

Psychological Attachment in Obesity

Significance for Bariatric Surgery

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad van doctor

aan de Universiteit van Amsterdam

op gezag van de Rector Magnificus

prof. dr. D.C. van den Boom

ten overstaan van een door het college voor promoties

ingestelde commissie,

in het openbaar te verdedigen in de Agnietenkapel

op dinsdag 10 juni 2014, te 12:00 uur

door

Floortje Kara Aarts

geboren te Arnhem

Promotiecommissie

Promotores: Prof. dr. D.P.M. Brandjes

Prof. dr. R. Geenen

Co-promotores: Dr. S.C.H. Hinnen

Dr. V.E.A. Gerdes

Overige leden: Prof. dr. P.M.M. Bossuyt

Dr. M. de Brauw

Prof. dr. E. Fliers

Prof. dr. R. Sanderman

Prof. dr. S. Visser

Content

Chapter 1 General introduction and Outline of this thesis 7

Chapter 2 The significance of attachment representations for obesity: 23

a systematic review

Submitted for Publication

PART I ATTACHMENT REPRESENTATIONS, OBESITY AND PREOPERATIVE

ASSESSMENT

Chapter 3 Psychologists’ evaluation of bariatric surgery candidates influenced 41

by patients’ attachment representations and symptoms of depression

and anxiety

Journal of Clinical Psychology in Medical Settings, 2014; 21(1).

Chapter 4 Coping style as a mediator between attachment and mental and 57

physical health in patients suffering from morbid obesity

International Journal of Psychiatry in Medicine, 2014; 47(1).

Chapter 5 Mental health care utilization in patients seeking bariatric surgery: 75

the role of attachment behavior

Bariatric Surgical Practice and Patient Care, 2013; 8(4).

PART II POSTOPERATIVE: ATTACHMENT REPRESENTATIONS AND EFFECT OF

FAMILY MEMBERS

Chapter 6 Attachment anxiety predicts poor adherence to dietary recommendations: 89

an indirect effect on weight change one year after gastric bypass surgery


Chapter 7 The significance of attachment representations for quality of life one 105

year following gastric bypass surgery: a longitudinal analysis


Chapter 8 Gastric bypass may promote weight loss in overweight partners in the 119

first year after surgery


PART III SUMMARY AND APPENDICES

Summary 133

Samenvatting (Dutch Summary) 137

Dankwoord 143

PhD Portfolio 147

List of Publications 149

Curriculum Vitae 151

1General introduction and Outline of the thesis

Floor Aarts

R1R2R3R4R5R6R7R8R9

R10R11R12R13R14R15R16R17R18R19R20R21R22R23R24R25R26R27R28R29R30R31R32R33R34R35R36R37R38R39

8 | Chapter 1

R1R2R3R4R5R6R7R8R9R10R11R12R13R14R15R16R17R18R19R20R21R22R23R24R25R26R27R28R29R30R31R32R33R34R35R36R37R38R39

General introduction and Outline of the thesis | 9

1Morbid obesity: definition and treatment

Obesity is a growing health problem and can be described as having disproportionately more

body weight in relation to body height.1, 2 The most common used classification for obesity is

Body Mass Index (BMI), defined as weight in kilograms divided by height in squared meters.

A person with a BMI above 25 kg/m2 is considered overweight, with a BMI above 30 kg/m2

obese and with a BMI above 40 kg/m2 morbid obese.3

After an increase in the past decades, worldwide more than 20% of the adults are overweight

and approximately 10% are obese.4 In The Netherlands in 2012, 48% of the population were

overweight, and 12% were obese.5 Some other European countries and the US show even higher

rates. The prevalence of overweight in the US in 2007–2008 was 68%, and the prevalence of

obesity in 2012 was 34.9%.6, 7 Although obesity rates remain high, the prevalence of obesity

remained relative stable the last years.7

Obesity is seen as a chronic disease. It is associated with several diseases and conditions such as,

type 2 diabetes mellitus, hypertension, dyslipidemia, coronary heart diseases, obstructive sleep

apnoea syndrome (OSAS), cancer, psychopathology and increased mortality.8, 9 Since obesity is

often combined with somatic and psychological problems, the overall health care costs related to

obesity are higher than for non-obese subjects.10

Dietary and exercise regimens are used as primary treatment for obesity. However, patients with

morbid obesity seem to respond poorly to this traditional form of treatment and therefore turn

to bariatric surgery.11 Bariatric surgery, which consists of several surgical weight loss procedures

is currently the treatment of choice for patients with morbid obesity when conservative regimens

have failed.12-14

A common type of bariatric surgery is the gastric bypass operation. This procedure combines two

alterations: restriction of gastric volume (limitation of food intake) and diversion of the ingested

nutrients away from the proximal small intestine.15 The gastric bypass procedure creates a small

gastric pouch via stapling (10-30 ml), and a limb of the jejunum (small intestine) is attached directly

to the pouch, which results in ingested food bypassing 90% of the stomach, the duodenum, and

the upper portions of the small intestine (Figure 1).16

R1R2R3R4R5R6R7R8R9


10 | Chapter 1

Before surgery After surgery

Figure 1. Gastric bypass procedure

Gastric bypass outcome

Several studies have reported long-term follow-up results of weight loss and quality of life in

patients after gastric bypass surgery. The majority of the patients lose 25-35% of their initial

body weight with gastric bypass surgery within one year after surgery.13 Although the majority of

patients benefit from a gastric bypass operation, there is still a small but considerable portion of

patients who are unable to benefit optimally from a gastric bypass operation in terms of weight

loss and quality of life.15, 17, 18 The amount of weight loss after gastric bypass surgery will to a

large extent depend on the degree to which the patient succeeds in adopting healthy dietary

behavior.19 Both being successful in adopting healthy dietary recommendations and a person’s

ability to bring about enduring changes in quality of life will be determined by psychological

factors.

Psychological aspects

A standard component of the clinical evaluation of candidates applying for bariatric surgery is

a pre-surgical psychological assessment to identify possible indicators of suboptimal adherence

and outcomes.20-22 A history of psychiatric problems and current psychiatric comorbidity (e.g.,

anxiety and depression) are among the factors assessed.8, 20-23 The importance of these factors is



1supported by studies showing that psychiatric comorbidity was associated with less weight loss

after the initial year of the gastric bypass operation.24-26 This relationship may be explained by

difficulties with adherence to dietary and/or exercise recommendations.27

The focus of this thesis is on attachment representations, habitual states of mind with respect

to interpersonal relations. It is expected that –in addition to and related to current and past

psychological problems– patients’ attachment representations will influence adherence to dietary

recommendations. Moreover, attachment theory is expected to be a relevant determinant of

preoperative and postoperative quality of life in the group of patients with morbid obesity.28, 29

Attachment theory

According to attachment theory, people internalize early childhood experiences that centre

around the interaction with primary caregivers resulting in enduring beliefs and expectations

(i.e., internal working models or schemes) about the self (e.g., as worthy of love and care) and

about others (e.g., as trustworthy and caring).30-33 These enduring expectations are referred to

as attachment representations and in adulthood have been conceptualized as a set of mental

states concerning anxiety about rejection and abandonment, and avoidance of intimacy and

interdependence.30, 31, 34 Attachment representations impact among other things the way people

regulate emotions and deal with stress.35, 36

Description of attachment representations

Figure 2 presents a two dimensional, four categorical model of adult attachment. Attachment

representations have been characterized by their position on two dimensions reflecting anxiety

and avoidance.37

Persons who are securely attached (i.e., those low on attachment anxiety and low on attachment

avoidance) have a positive view of the self and a positive view of others, are self-confident,

explorative (e.g. curious, problem solving) and are comfortable in seeking support when needed.38

They have a sense of social resiliency, that is, they dispose over psychosocial skills (e.g. social and

communicative competences) and are capable to use a broad range of coping strategies (e.g.

social support, active problem solving) in times of stress.39

Persons who are anxiously attached (i.e., those high on attachment anxiety) have a negative

view of the self and a positive view of others, feel fragile, unlovable and unworthy of care and

are hypervigilant for rejection or abandonment. Their sense of vulnerability and hypervigilance

R1R2R3R4R5R6R7R8R9


12 | Chapter 1

for threats results in high levels of perceived stress and distress.38 They have been found to make

stronger attempts to seek proximity in order to try and elicit increased attention and support from

others often to the point of being ‘clingy’ in order to regulate their emotions.40, 41 Despite their

strong desire for closeness and reassurance, research shows that support is hardly effective in

reducing distress in these people.42

Low Avoidance

High Avoidance

Low Anxiety High Anxiety

AnxiousSecure

Avoidant Disorganized

Figure 2. Two dimensional model of adult attachment related to the four attachment representations31

Persons who are avoidantly attached (i.e., those high on attachment avoidance), have a positive

view of the self and a negative view of others, perceive others as unavailable and unable to

provide adequate support when needed, and therefore value independency and self-reliance.43,

44 As a way to reinforce their self-sufficiency and to avoid relationships with others, they tend to

dismiss symptoms of distress and vulnerability.38 They deal with stressors by distancing, avoiding

and repressing negative emotions.45-48

Persons who are disorganized attached (i.e. those high on attachment anxiety and high on

attachment avoidance) are a mixture of both characteristics, the avoidant and anxious attachment

pattern.37 They have a negative view of both self and others. They are cautious, avoidant, and

distrustful and expect others to be harsh or rejecting and experience difficulties with assertiveness

(shy) and social inhibition (timid).49, 50 Although they may experience intense negative affect, they

rather suffer than seek help.51, 52



1Attachment as predictor of dietary adherence

Both attachment anxiety and attachment avoidance have been found to be related to poorer

adherence to medical regiments in chronically ill patients.53, 54

More anxiously attached patients have been consistently shown to be more prone to distress

when confronted with stressors.28 In stressful situations people high on attachment anxiety may

view themselves unable to deal with the stressors and they may rely on smoking, alcohol and

high caloric food to regulate their emotions.39, 55, 56 In accordance, attachment anxiety has been

found to be associated with obesity in both children and adults.56, 57 Due to their high levels of

distress and their tendency to rely on external and behavioural modulators of affect such as high

caloric food, more anxiously attached patients can be expected to find it more difficult to adhere

to dietary recommendations after bariatric surgery.

More avoidantly attached patients, on the other hand, stress the importance of independence

and self-reliance, are reluctant to seek support and feel uncomfortable trusting others, including

health care providers.43, 44 Due to their high level of self-reliance and low collaboration with health

care providers, it can be expected that they will be less adherent to dietary recommendations

after bariatric surgery as well.

Attachment as predictor for quality of life

The improvement in quality of life after bariatric surgery will in addition to the amount of

weight loss depend on individual characteristics58 such as one’s attachment representations.

Both attachment anxiety and attachment avoidance have been uniformly found to be associated

with impaired mental and physical functioning in healthy people,28, 29 chronically ill patients59

and morbidly obese patients.60 In a cross-sectional study in morbidly obese bariatric surgery

candidates an association between attachment avoidance and poor mental health quality of

life was observed,60 but it is as yet unknown whether attachment representations impact the

postoperative course of quality of life.

The effect of gastric bypass on family members

The development of obesity is multifactorial with a sedentary lifestyle and a hypercaloric

diet playing important roles.61 Parental weight has proven to be one of the most important

independent predictors of childhood obesity, and consequently of obesity in adulthood.62, 63 While

parents and children share both genetic and environmental factors, if one’s partner becomes

obese, the likelihood that the other partner will become obese is increased by 37%.64 Following

R1R2R3R4R5R6R7R8R9


14 | Chapter 1

gastric bypass, patients are instructed to implement diet and lifestyle changes which may lead to

partners and children mimicking the altered behaviours of the patients undergoing gastric bypass

surgery.65

Outline of this thesis

This thesis examines social and emotional aspects of bariatric surgery and obesity with a focus on

attachment representations. In this thesis we approached this subject on two levels: (1) the aim

of the first part is to examine the role of patients’ attachment representations in obesity and the

assessment before bariatric surgery (2) the second part focuses on the postoperative situation by

examining attachment representations as a predictor of the treatment outcome of gastric bypass

surgery for morbid obesity and the effect of gastric bypass surgery on weight and eating behavior

of family members. Chapter 2 presents a systematic review of the main topic of this thesis. Next,

the three preoperative cross-sectional studies are described in chapters 3, 4 and 5 and the three

postoperative longitudinal studies in chapters 6, 7 and 8 (Figure 3).

Part I: Attachment representations, obesity and preoperative assessment

It is now clear that the aetiology of many chronic diseases including obesity concerns not only

genetic and current environmental factors, but also the way in which early repeated interactions

with significant others results in enduring ways of reacting to stress and dealing with negative

affectivity. In chapter 2 we systematically evaluate the existing evidence on attachment

representations in relation to obesity.

The main focus of the next three chapters lies on the role of attachment on mental well being and

functioning in the group of patients referred for bariatric surgery.

Attachment may influence many aspects psychologists are likely to incorporate into their

evaluation, including the anxious and depressive symptoms patients’ experience. The aim of

Chapter 3 was to examine whether patients’ self-reported attachment representations and

levels of depression and anxiety were associated with psychologists’ evaluations of patients with

morbid obesity applying for bariatric surgery.

Chapter 4 examines in patients applying for bariatric surgery the association of attachment

representations and coping styles on the one hand with mental health and physical functioning

on the other. Less securely attached patients (those high on attachment anxiety or attachment



1avoidance) may use more ineffective coping strategies which in turn may increase the risk at

mental problems and limitations in physical functioning in patients suffering from morbid obesity.

Obesity may be a factor contributing to mental health problems in patients seeking bariatric

surgery. Whether or not a person uses mental health care for one’s psychological problems may

have its roots in attachment behaviour. In Chapter 5 attachment behavior is hypothesized to be

associated with mental health care utilization in morbidly obese patients seeking bariatric surgery.

Part II: Postoperative: attachment representations and effect on family members

There is a small but considerable proportion of patients who are unable to benefit optimally from a

gastric bypass operation in terms of weight loss. The final outcome of the operation will to a large

extent depend on the degree to which to which the patient succeeds in adopting healthy dietary

recommendations. Current and past psychological problems and attachment representations are

expected to be determinants of adherence to dietary recommendations.

In Chapter 6 we aimed to examine the mediating role of adherence to dietary recommendations

between on the one hand, current and past psychological problems, attachment anxiety and

attachment avoidance, and on the other hand, weight reduction one year after gastric bypass

surgery.

The main aim of Chapter 7 was to examine whether attachment anxiety and attachment

avoidance, independent of body mass index (BMI), predict the level and course of physical

functioning and mental well-being after gastric bypass surgery. The improvement in quality of

life after bariatric surgery will in addition to the amount of weight loss depend on individual

characteristics such as one’s attachment representations.

Chapter 8 describes a 1-year longitudinal study examining weight and eating behavior changes

in cohabitating family members of patients after gastric bypass surgery. Obesity is increasingly

recognized as a family trait, with family members imitating each other’s lifestyle. Following

bariatric surgery, patients are assumed to implement diet and lifestyle changes which are expected

to have a positive effect on the body weight of family members.

R1R2R3R4R5R6R7R8R9


16 | Chapter 1

Attachment representations

Gastric bypass

Weight & Quality of life

Family members

Chapter 3 & 4

Chapter 6 & 7 Chapter 8

Obesity

Chapter 2 & 5



1Reference List

1. Gortmaker SL, Swinburn BA, Levy D et al. Changing the future of obesity: science, policy, and action. Lancet 2011;378(9793):838-847.

2. Swinburn BA, Sacks G, Hall KD et al. The global obesity pandemic: shaped by global drivers and local environments. Lancet 2011;378(9793):804-814.

3. WMO expert consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet 2004;363(9403):157-163.

4. Finucane MM, Stevens GA, Cowan MJ et al. National, regional, and global trends in body-mass index since 1980: systematic analysis of health examination surveys and epidemiological studies with 960 country-years and 9.1 million participants. Lancet 2011;377(9765):557-567.

5. Bakel AM, Zantinge EM. Neemt het aantal mensen met overgewicht toe of af? In: RIVM, editor. Volksgezondheid Toekomst Verkenning, Nationaal Kompas Volksgezondheid. Bilthoven: 2013.

6. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of obesity in the United States, 2009-2010. NCHS Data Brief 2012;(82):1-8.

7. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of childhood and adult obesity in the United States, 2011-2012. JAMA 2014;311(8):806-814.

8. Lawrence VJ, Kopelman PG. Medical consequences of obesity. Clin Dermatol 2004;22(4):296-302.

9. Visscher TL, Seidell JC. The public health impact of obesity. Annu Rev Public Health 2001;22:355-375.

10. Quesenberry CP, Jr., Caan B, Jacobson A. Obesity, health services use, and health care costs among members of a health maintenance organization. Arch Intern Med 1998;158(5):466-472.

11. Colquitt J, Clegg A, Loveman E, Royle P, Sidhu MK. Surgery for morbid obesity. Cochrane Database Syst Rev 2005;(4):CD003641.

12. Adams TD, Davidson LE, Litwin SE et al. Health benefits of gastric bypass surgery after 6 years. JAMA 2012;308(11):1122-1131.

13. Buchwald H, Avidor Y, Braunwald E et al. Bariatric surgery: a systematic review and meta-analysis. JAMA 2004;292(14):1724-1737.

14. Carlsson LM, Peltonen M, Ahlin S et al. Bariatric surgery and prevention of type 2 diabetes in Swedish obese subjects. N Engl J Med 2012;367(8):695-704.

15. Suter M, Donadini A, Romy S, Demartines N, Giusti V. Laparoscopic Roux-en-Y gastric bypass: significant long-term weight loss, improvement of obesity-related comorbidities and quality of life. Ann Surg 2011;254(2):267-273.

16. Kral JG, Naslund E. Surgical treatment of obesity. Nat Clin Pract Endocrinol Metab 2007;3(8):574-583.

17. Karlsson J, Taft C, Ryden A, Sjostrom L, Sullivan M. Ten-year trends in health-related quality of life after surgical and conventional treatment for severe obesity: the SOS intervention study. Int J Obes (Lond) 2007;31(8):1248-1261.

18. Aftab H, Risstad H, Sovik TT et al. Five-year outcome after gastric bypass for morbid obesity in a Norwegian cohort. Surg Obes Relat Dis 2013.

19. Karmali S, Brar B, Shi X, Sharma AM, de Gara C, Birch DW. Weight recidivism post-bariatric surgery: a systematic review. Obes Surg 2013;23(11):1922-1933.

20. Snyder AG. Psychological assessment of the patient undergoing bariatric surgery. Ochsner J 2009;9(3):144-148.

21. Wadden TA, Sarwer DB. Behavioral assessment of candidates for bariatric surgery: a patient-oriented approach. Surg Obes Relat Dis 2006;2(2):171-179.

22. Ritz SJ. The bariatric psychological evaluation: a heuristic for determining the suitability of the morbidly obese patient for weight loss surgery. Bariatr Surg Pract Patient Care 2006; 1: 97-105

R1R2R3R4R5R6R7R8R9


18 | Chapter 1

23. van Hout GC, Vreeswijk CM, van Heck GL. Bariatric surgery and bariatric psychology: evolution of the Dutch approach. Obes Surg 2008;18(3):321-325.

24. Averbukh Y, Heshka S, El Shoreya H et al. Depression score predicts weight loss following Roux-en-Y gastric bypass. Obes Surg 2003;13(6):833-836.

25. Rutledge T, Groesz LM, Savu M. Psychiatric factors and weight loss patterns following gastric bypass surgery in a veteran population. Obes Surg 2011;21(1):29-35.

26. Kinzl JF, Schrattenecker M, Traweger C, Mattesich M, Fiala M, Biebl W. Psychosocial predictors of weight loss after bariatric surgery. Obes Surg 2006;16(12):1609-1614.

27. Gorin AA, Raftopoulos I. Effect of mood and eating disorders on the short-term outcome of laparoscopic Roux-en-Y gastric bypass. Obes Surg 2009;19(12):1685-1690.

28. Hunter JJ, Maunder RG. Using attachment theory to understand illness behavior. Gen Hosp Psychiatry 2001;23(4):177-182.

29. Kotler T, Buzwell S, Romeo Y, Bowland J. Avoidant attachment as a risk factor for health. Br J Med Psychol 1994;67 ( Pt 3):237-245.

30. Crowell J, Fraley R, Shaver P. Measurement of individual differences in adolescent and adult. In: J. Cassidy & P. R. Shaver, eds. Handbook of attachment: Theory, research, and clinical applications: 434-465. New York: Guilford Press 1999.

31. Bartholomew K, Horowitz LM. Attachment styles among young adults: a test of a four-category model. J Pers Soc Psychol 1991;61(2):226-244.

32. Bowlby J. Attachment and loss: retrospect and prospect. Am J Orthopsychiatry 1982;52(4):664-678.

33. Griffin D BK. Models of the self and other: Fundamental dimensions underlying measures of adult attachment. J Pers Soc Psychol 1994; 67(3): 430-445.

34. Brennan KA, Clark CL, Shaver PR. Self-report measurement of adult attachment: An integrative overview. In: J. A. Simpson & W. S. Rholes, eds. Attachment theory and close relationships: 46-76. New York: Guilford Press 1998.

35. Allen JG, Stein H, Fonagy P, Fultz J, Target M. Rethinking adult attachment: a study of expert consensus. Bull Menninger Clin 2005;69(1):59-80.

36. Mikulincer M, Shaver P. Attachment processes and emotion regulation. Attachment in adulthood: Structure dynamics and change. New York: Guilford Press; 2007:188-217.


38. Maunder RG, Hunter JJ. Assessing patterns of adult attachment in medical patients. Gen Hosp Psychiatry 2009;31(2):123-130.

39. Maunder RG, Hunter JJ. Attachment and psychosomatic medicine: developmental contributions to stress and disease. Psychosom Med 2001;63(4):556-567.

40. Brennan KA, Wu S, Loev J. Adult Romantic attachment and Individual Differences in Attidudes Toward Physical contact in the Context of Adult Romantic Relationships. In: Simpson JA, Rholes WS, editors. Attachment theory and close relationships. New York: Guilford Press; 1998:394-428.

41. Kobak RR, Sceery A. Attachment in late adolescence: working models, affect regulation, and representations of self and others. Child Dev 1988;59(1):135-146.

42. George C, West M. The development and preliminary validation of a new measure of adult attachment: the adult attachment projective. Attach Hum Dev 2001;3(1):30-61.

43. Florian V, Mikulincer M, Bucholtz I. Effects of adult attachment style on the perception and search for social support. The Journal of Psychology 1995;129(6):665-676.

44. Priel B, Shamai D. Attachment style and perceived social support: effects on affect regulation. Personality and Individual Differences 1995;19(2):235-241.

45. Mikulincer M, Orbach J. Attachment style and repressive defensiveness: The accessibility and architecture of affective memories. Journal of Personality and Social Psychology 1995;68(5):917-925.



1

46. Turan B, Osar Z, Turan JM, Ilkova H, Damci T. Dismissing attachment and outcome in diabetes: The mediating role of coping. Journal of Social and Clinical Psychology 2003;22(6):607-626.

47. Fraley RC, Shaver PR. Adult attachment and the suppression of unwanted thoughts. Journal of Personality and Social Psychology 1997;73(5):1080-1091.

48. Vetere A, Myers LB. Repressive coping style and adult romantic attachment style: is there a relationship. Personality and Individual Differences 2002;32:799-807.

49. Ciechanowski P, Russo J, Katon W et al. Where is the patient? The association of psychosocial factors and missed primary care appointments in patients with diabetes. Gen Hosp Psychiatry 2006;28(1):9-17.

50. Klohnen EC, John OP. Working models of attachment: a theory-based approach. In: Simpson JA, Rholes WS, editors. Attachment theory and close relationships. New York: Guilford; 1998:115-140.

51. Ciechanowski PS, Walker EA, Katon WJ, Russo JE. Attachment theory: a model for health care utilization and somatization. Psychosom Med 2002;64(4):660-667.

52. Ciechanowski PS, Katon WJ, Russo JE, Dwight-Johnson MM. Association of attachment style to lifetime medically unexplained symptoms in patients with hepatitis C. Psychosomatics 2002;43(3):206-212.

53. Bennett JK, Fuertes JN, Keitel M, Phillips R. The role of patient attachment and working alliance on patient adherence, satisfaction, and health-related quality of life in lupus treatment. Patient Educ Couns 2011;85(1):53-59.

54. Ciechanowski PS, Katon WJ, Russo JE, Walker EA. The patient-provider relationship: attachment theory and adherence to treatment in diabetes. Am J Psychiatry 2001;158(1):29-35.

55. Maunder RG, Hunter JJ. Attachment relationships as determinants of physical health. J Am Acad Psychoanal Dyn Psychiatry 2008;36(1):11-32.

56. Wilkinson LL, Rowe AC, Bishop RJ, Brunstrom JM. Attachment anxiety, disinhibited eating, and body mass index in adulthood. Int J Obes (Lond) 2010;34(9):1442-1445.

57. Anderson SE, Whitaker RC. Attachment security and obesity in US preschool-aged children. Arch Pediatr Adolesc Med 2011;165(3):235-242.

58. Loving TJ, Smets EM. Romantic relationships and health. In: Simpson JA, Campbell L, editors. The Oxford handbook of close relationships. New York: Oxford Univeristy Press: 2013:617-637.

59. Martin LA, Vosvick M, Riggs SA. Attachment, forgiveness, and physical health quality of life in HIV + adults. AIDS Care 2012;24(11):1333-1340.

60. Sockalingam S, Wnuk S, Strimas R, Hawa R, Okrainec A. The association between attachment avoidance and quality of life in bariatric surgery candidates. Obes Facts 2011;4(6):456-460.

61. Simopoulos AP. Characteristics of obesity: an overview. Ann N Y Acad Sci 1987;499:4-13.

62. Agras WS, Mascola AJ. Risk factors for childhood overweight. Curr Opin Pediatr 2005;17(5):648-652.

63. Keane E, Layte R, Harrington J, Kearney PM, Perry IJ. Measured parental weight status and familial socio-economic status correlates with childhood overweight and obesity at age 9. PLoS One 2012;7(8):e43503.

64. Christakis NA, Fowler JH. The spread of obesity in a large social network over 32 years. N Engl J Med 2007;357(4):370-379.

65. Woodard GA, Encarnacion B, Peraza J, Hernandez-Boussard T, Morton J. Halo effect for bariatric surgery: collateral weight loss in patients’ family members. Arch Surg 2011;146(10):1185-1190.

R1R2R3R4R5R6R7R8R9


PART IAttachment representations, obesity and

preoperative assessment

2The significance of attachment representations

for obesity: a systematic review

Floor Aarts, Rinie Geenen, Victor E.A. Gerdes, Dees P.M. Brandjes, Chris Hinnen

Submitted for publication

R1R2R3R4R5R6R7R8R9


24 | Chapter 2

Abstract

Theoretical considerations and empirical results suggest that interpersonal patterns known as

attachment representations are of relevance to obesity. This paper systematically examines the

peer-reviewed evidence regarding the relationship between attachment representations and

obesity. Peer-reviewed literature published between 1990 and 2013 was derived from PubMed,

PsycINFO and reference lists of included papers. Ten studies met the selection criteria. Overall the

studies suggest a relationship between attachment insecurity and obesity. Particularly attachment

anxiety, (i.e. the anxiety about rejection and abandonment by others) was associated with current

and future obesity. Possible explanations for an impact of attachment insecurity on obesity can

be found in heightened physiological responses to stressful situations and the underdevelopment

of emotion-regulation, which is an issue for future inquiry. Despite the early stage of theory and

research in the field of obesity, there is potential in considering attachment representations in

obesity care.

Keywords: obesity; attachment; body mass index; child; adult; systematic review


Significance of attachment for obesity | 25

2

Introduction

Obesity is a complex global health problem, which has proven difficult to prevent and to treat1

and is also affecting children.2 It is now clear that the aetiology of many chronic diseases including

obesity concerns not only genetic and current environmental factors, but also the way in which

early repeated interactions with significant others results in enduring ways of reacting to stress

and dealing with negative affectivity.3, 4 Problems with stress management and affect regulation

have been repeatedly linked to obesity in both children and adults.5, 6 One specific theoretical

framework that describes individual differences in dealing with stress and affect regulation based

on early childhood experiences is attachment theory.7

According to attachment theory, individuals internalize early childhood interactions with primary

caregivers in enduring beliefs and expectations about how others behave towards oneself and how

oneself behaves towards others.8-13 These enduring expectations are referred to as attachment

representations or internal working models and are thought to be the mechanisms by which

the influence of childhood experiences are sustained into adulthood.13 In adulthood, internal

working models of attachment are generally conceptualized as sets of global beliefs about the

self (e.g., as worthy of care and lovable) and about others (e.g., as trustworthy and caring).10, 14 In

terms of their affective –motivational characteristics, these global beliefs are referred to as anxiety

about rejection and abandonment and avoidance of intimacy and interdependence.13 These

two dimensions can be combined into four attachment styles– one secure and three insecure

subtypes: preoccupied, dismissing and fearful.10, 14

Individuals low on attachment anxiety and low on attachment avoidance (i.e., secure) have a sense

of social resiliency. That is, they dispose over psychosocial skills (e.g. social and communicative

competences) and are capable to use a broad range of coping strategies (e.g. social support, active

problem solving) in times of stress.4 Individuals high on attachment anxiety (i.e., preoccupied)

have a sense of vulnerability and hypervigilance for threats, resulting in high levels of perceived

stress and distress.15 They have been found to make stronger attempts to seek proximity in order

to try and elicit increased attention and support from others often to the point of being ‘clingy’

in order to regulate their emotions.16, 17 Despite their strong desire for closeness and reassurance,

research shows that social and emotional support is hardly effective in reducing distress in

these people.18 In contrast, individuals high on attachment avoidance (i.e., dismissing) tend to

dismiss symptoms of distress and vulnerability.19 They deal with stressors by distancing, avoiding

and repressing negative emotions.20-23 Consequently, more avoidantly attached patients may

R1R2R3R4R5R6R7R8R9


26 | Chapter 2

experience and report to be non-distressed while showing considerable biological distress (e.g.,

increased blood pressure, heart rate variability).24, 25 Individuals high on both attachment anxiety

and attachment avoidance (i.e., fearful) show a mixture of both preoccupied and dismissing

attachment patterns.14 Although they may experience intense negative affect, they rather suffer

than seek help.26, 27

Due to their prototypical ways of dealing with stress and affect regulation, eating has been

suggested to be a regulatory mechanism for more insecurely attached individuals to deal with

stressors.28 Therefore, several investigators expected higher levels of attachment anxiety and

attachment avoidance to be associated with obesity.

Insight into the relationship between attachment representations and obesity is important as

it may help to determine who is at risk of obesity as well as to develop person-customized

prevention and intervention programs. Therefore, in the present study we systematically reviewed

the literature on the association between attachment and obesity.

Materials and Methods

Document eligibility

We aimed to identify articles which covered any aspect of the relationship between attachment

representations -in both adults as children- and obesity, published between 1990 and 2013.

Dissertations were excluded.29 The study design and document type was unrestricted.

Search strategy and document selection

A systematic search was implemented in the following two bibliographic databases: PubMed

and PsycINFO. The search strategy included the following combination of key words/MeSH

terms: ‘attachment’ OR ‘relationship style’ AND ‘obesity’ OR ‘overweight’ OR ‘body weight’ OR

‘body mass index’ OR ‘waist-to-hip ratio’ OR ‘BMI’. In line with the document eligibility criteria,

publication date and human studies limits were applied. The search strategy is shown in Table 1.

First, titles and abstracts and, second, full-text version of records identified by the search strategies

were assessed by two authors (FA and CH) against document eligibility returned 350 articles

in PubMed and 86 articles in PsycINFO of which 421 articles remained after the removal of

duplicates.



2

Table 1. Search strategy

Database KeywordsPubMed (attachment[tiab] OR “relationship style”[tiab]) AND ((“obesity”[MeSH Terms] OR

“obesity”[tiab]) OR (“overweight”[MeSH Terms] OR “overweight”[tiab]) OR “body weight”[tiab] OR “body mass index”[tiab] OR BMI[tiab] OR “waist-hip ratio”[tiab] OR “waist-to-hip ratio”[tiab])

PsycINFO (exp Attachment behavior/ or relationship style.ti,ab. or emotional attachment.ti,ab) AND (exp Overweight/ or exp Obesity/ or exp Body Weight/ or exp Body Mass Index/ or obesity.ti,ab. or overweight.ti,ab. or body weight.ti,ab. or exp body weight/ or body mass index.ti,ab. or exp body mass/ or waist-hip ratio.ti,ab. or waist-to-hip ratio.ti,ab.)

Quality assessment

The Newcastle-Ottawa Scale (NOS) for assessing the quality of non-randomised studies in meta-

analyses was used as a guide to assess the quality of the observational studies.30 This scale assesses

three broad areas: (i) selection; (ii) comparability; (iii) outcome or exposure. Quality of the included

studies was assessed independently by the same two reviewers (FA and CH). No attempts to mask

for authorship, journal name or institution were made. Appendix 1 details the quality assessment

and scoring system.

Results

Characteristics of studies

A total of 10 articles met the inclusion criteria and were identified in this review. Table 2 depicts the

study characteristics and main results. Six studies were cross-sectional and four were longitudinal.

Six studies investigated an adult population whereas four studies investigated children. The

majority of the studies were performed in the general population with a prevalence of obesity

between 15%-20%.

Obesity and attachment measures

As a definition of obesity, BMI was used in the majority of studies; however, also waist-to-hip

ratio (WHR) was used in two studies.31, 32 In children an adjusted BMI score for youngsters was

used, or categories relative to the 2000 US growth reference were used.12, 33 For the evaluation

of attachment representations, nine methods were used comprising both categorical (e.g., RQ,

AHQ, AAPR) and dimensional measures of attachment (e.g., AQS, ECR-R, MAQ, IPPA-R, SC).

R1R2R3R4R5R6R7R8R9


28 | Chapter 2

Tab

le 2

. Ove

rvie

w o

f st

udie

s on

the

rel

atio

nshi

p be

twee

n at

tach

men

t an

d ob

esity

Stu

dy

Pop

ula

tio

nD

esig

nA

ttac

hm

ent

mea

sure

Ob

esit

y ch

arac

teri

zati

on

Ou

tco

me

Qu

alit

ySi

gn

ifica

nt

Coo

per

et

al. 44

145

youn

g w

omen

Cro

ss-

sect

iona

l A

HQ

-qu

estio

nnai

re17

.2 %

BM

I >25

Th

e su

bsca

le n

egat

ive

pare

ntal

dis

cipl

ine

was

ass

ocia

ted

with

BM

I Bet

a=.0

17 p

=

.02,

but

not

the

sub

scal

es s

ecur

e ba

se,

thre

ats

of s

epar

atio

n or

pee

r af

fect

iona

l su

ppor

t.

Low

Yes

for

pare

ntal

di

scip

line,

no

for

secu

re b

ase/

thre

ats

of

sepa

ratio

n/ a

ffec

tiona

l su

ppor

t

Wilk

inso

n et

al

. 46

200

stud

ents

C

ross

-se

ctio

nal

ECR-

R-

ques

tionn

aire

Mea

n BM

I 23.

0, r

ange

fro

m

17.4

- 4

1.1

(sd

= 3

.2)

Att

achm

ent

anxi

ety

and

BMI r

= .1

5 (p

<

.05)

; att

achm

ent

avoi

danc

e an

d BM

I w

ere

not

sign

ifica

ntly

cor

rela

ted.

Low

Yes

D’A

rgen

io e

t al

. 45

50 n

on-o

bese

150

obes

eC

ross

-se

ctio

nal

RQ-

ques

tionn

aire

Mea

n BM

I non

-obe

se h

ealth

y pa

rtic

ipan

ts 2

3.38

±2.

85M

ean

BMI o

bese

pa

rtic

ipan

ts=

41.3

3 (s

d=6.

80)

Mea

n BM

I obe

se p

artic

ipan

ts

with

cur

rent

psy

chia

tric

di

agno

sis=

38.2

7 (s

d=6.

69)

The

odds

for

obe

sity

was

1.2

3 (9

5% C

I, 1.

08-1

.41,

p =

.002

) hig

her

for

anxi

ous

vs. s

ecur

e at

tach

men

t

Med

ium

Yes

Kie

sew

ette

r at

al.

63, 6

9

44 o

bese

pa

tient

sC

ross

-se

ctio

nal

AA

PR-

inte

rvie

wM

ean

BMI=

37.3

(sd=

7.4)

Mea

n BM

I ins

ecur

e pa

tient

s=40

.6 (s

d=9.

1)M

ean

BMI s

ecur

e pa

tient

s=37

.2 (s

d=6.

4)

Mea

n w

eigh

t in

kg

inse

cure

pa

tient

s=11

6.0

(sd=

22.

3)M

ean

wei

ght

in k

g se

cure

pa

tient

s=10

4 (s

d= 2

3.7)

No

rela

tion

betw

een

atta

chm

ent

and

BMI a

t ba

selin

e.Se

cure

and

inse

cure

pat

ient

s di

d di

ffer

on

wei

ght

in k

g at

bas

elin

e p=

.03.

Low

Yes

for

wei

ght

in k

g,

no f

or B

MI

Bahr

ami e

t al

. 70

202

over

wei

ght

stud

ents

Cro

ss-

sect

iona

lIP

PA-R

-qu

estio

nnai

reM

ean

BMI 2

7.48

(sd

= 4

.81)

Att

achm

ent

qual

ity a

nd B

MI r

=-0

.27,

p=

.003

Low

Yes



2

Hin

tsan

en e

t al

. 32

1570

men

an

d w

omen

fr

om t

he

Car

diov

ascu

lair

Risk

Fin

ns s

tudy

Cro

ss-

sect

iona

lRQ

-qu

estio

nnai

reN

ot r

epor

ted

Fear

ful a

ttac

hmen

t an

d yo

uth

BMI

Beta

=.0

7, p

=.0

1Fe

arfu

l att

achm

ent

and

adul

thoo

d BM

I Be

ta=

.07,

p=

.007

Fear

ful a

ttac

hmen

t an

d W

HR

Beta

=.0

82, p

=.0

16Pr

eocc

upie

d at

tach

men

t an

d W

HR

Beta

=.1

32, p

=.0

02N

o as

soci

atio

ns b

etw

een

othe

r at

tach

men

t st

yles

and

BM

I wer

e fo

und.

Med

ium

Yes

for

preo

ccup

ied

atta

chm

ent

in m

en

and

for

fear

ful

atta

chm

ent,

and

no

for

othe

r at

tach

men

t st

yles

Goo

ssen

s et

al

. 33

601

prea

dole

scen

tsC

ross

-se

ctio

nal a

nd

long

itudi

nal

SC-

ques

tionn

aire

Base

d on

adj

uste

d BM

I sco

re

youn

gste

rs[(a

ctua

l BM

I/per

cent

ile 5

0 of

BM

I for

age

and

gen

der)

x

100]

71.

4% u

nder

wei

ght

(adj

uste

d BM

I≤85

) 83

% n

orm

al w

eigh

t (8

5<

adju

sted

BM

I<12

0)11

% o

verw

eigh

t (1

20≤

adju

sted

BM

I<14

0)2%

obe

se (a

djus

ted

BMI≥

140)

No

rela

tion

betw

een

atta

chm

ent

tow

ards

mot

her

and

base

line

BMI.

Att

achm

ent

inse

curit

y to

war

ds m

othe

r, -b

ut n

ot t

owar

ds f

athe

r- s

igni

fican

tly

pred

icte

d in

crea

se o

f BM

I one

yea

r la

ter

β=-.

07, p

<.0

1, a

djus

ted

for

base

line

BMI.

Hig

hYe

s, f

or a

ttac

hmen

t in

secu

rity

tow

ards

m

othe

r in

pre

dict

ing

BMI,

no f

or

atta

chm

ent

tow

ards

m

othe

r fo

r ba

selin

e BM

I or

tow

ards

fat

her

And

erso

n et

al

. 12

8750

chi

ldre

nLo

ngitu

dina

lO

bser

vatio

n of

the

m

othe

r-ch

ild

inte

ract

ion

with

the

TA

SS-

45 (m

odifi

ed

vers

ion

of A

QS

Prev

alen

ce o

besi

ty in

secu

re

23.1

%Pr

eval

ence

obe

sity

sec

ure

16.6

%O

besi

ty s

tatu

s at

4.5

yea

rs w

as

defin

ed r

elat

ive

to t

he 2

000

US

grow

th r

efer

ence

72.

The

odds

for

obe

sity

at

4.5

year

s w

as

1.30

(95%

CI,

1.05

-1.6

2) t

imes

hig

her

for

inse

cure

vs.

sec

ure

atta

chm

ent

(with

the

obs

erva

tion

of a

ttac

hmen

t at

24

mon

ths

of a

ge),

adju

sted

for

so

ciod

emog

raph

ic c

hara

cter

istic

s su

ch

as b

irth

wei

ght

Hig

hYe

s

R1R2R3R4R5R6R7R8R9


30 | Chapter 2

Stu

dy

Pop

ula

tio

nD

esig

nA

ttac

hm

ent

mea

sure

Ob

esit

y ch

arac

teri

zati

on

Ou

tco

me

Qu

alit

ySi

gn

ifica

nt

And

erso

n et

al

. 42

977

child

ren

Long

itudi

nal

Obs

erva

tion

of t

he

mot

her-

child

in

tera

ctio

n w

ith t

he A

QS

32.9

% “

secu

re”

of w

hich

13

.4 %

obe

se42

.7 %

“in

bet

wee

n gr

oup”

(n

ot s

ecur

e no

t in

secu

re) o

f w

hich

17.

5% o

bese

24.5

% “

inse

cure

” of

whi

ch

18.8

% o

bese

Ado

lesc

ent

obes

ity w

as

defin

ed a

s ge

nder

-spe

cific

BM

I ≥9

5th p

erce

ntile

of

the

Cen

ters

fo

r D

isea

se C

ontr

ol a

nd

Prev

entio

n gr

owth

ref

eren

ce 72

The

odds

for

ado

lesc

ent

obes

ity (w

ith

the

obse

rvat

ion

of a

ttac

hmen

t at

15

and

36 m

onth

s of

age

) was

1.2

9 (9

5%

CI,

0.85

-1.9

4) t

imes

hig

her

for

the

“in

betw

een

grou

p” v

s. “

secu

re”

and

1.23

(95%

CI,

0.75

-1.9

3) t

imes

hig

her

for

“ins

ecur

e” v

s. “

secu

re”,

aft

er

adju

stm

ent

for

gend

er a

nd b

irth

wei

ght

Hig

hYe

s

Mid

ei e

t al

. 3121

3 ad

oles

cent

sLo

ngitu

dina

lM

AQ

-qu

estio

nnai

reM

ean

WH

R=.7

9 (s

d=.0

5)M

ean

BMI=

22.7

(sd=

4.0)

Att

achm

ent

anxi

ety

vs. W

HR

(3 y

ears

af

ter

the

obse

rvat

ion

of a

ttac

hmen

t),

adju

sted

for

BM

I at

base

line

β=.1

15,

p=.0

6

Hig

hN

o

BMI=

body

mas

s in

dex

(kg/

m2 )

, W

HR=

Wai

st-t

o-hi

p ra

tio,

ECR-

R=ex

perie

nces

in c

lose

rel

atio

nshi

ps-r

evis

ed,

AH

Q=

atta

chm

ent

hist

ory

ques

tionn

aire

, RQ

=re

latio

nshi

p qu

estio

nnai

re,

IPPA

-R=

inve

ntor

y of

par

ent

and

peer

att

achm

ent-

revi

sed

vers

ion

for

child

ren,

MA

Q=

Mea

sure

men

t of

att

achm

ent

qual

ity,

SC=

Secu

rity

Scal

e,

AA

PR=

Adu

lt A

ttac

hmen

t Pr

otot

ype

Ratin

g, A

QS=

Att

achm

ent

Q-s

ort,

TA

SS-4

5=To

ddle

r A

ttac

hmen

t So

rt-4

5



2

The relationship questionnaire (RQ)14 is a single item measure made up for four short paragraphs,

each describing a prototypical attachment pattern as it applies in close adult peer relationships.

The RQ is relatively brief, has been implemented in multiple studies and demonstrated

independence from self-deceptive biases.34 A weakness, however, is that each attachment style is

measured with only one item, and therefore no internal consistency reliability can be determined.

The attachment history questionnaire (AHQ)35 provides self-report information about early

attachment-related events and peer relationships. It has 51 items, with responses rated on seven

point scales, which assess the frequency and intensity of behaviors by attachment figures. The

AHQ shows respectable reliability and validity, but the nature of the AHQ is relatively untested.

The adult attachment prototype rating (AAPR) is a measurement used during a semi-standardized

one- to two hour attachment interview and determines a patients’ attachment style. The AAPR

has demonstrated its reliability and validity in a variety of studies. The attachment Q-sort (AQS)

is used by a data collector during approximately 2 hours of observation of the mother-child

interaction. The data collector sorted 45 “cards” based on how well the behavior described

on the card applied to the mother-child interaction. From the AQS, a continuous measure of

attachment security was derived, which could range from -1 to 1, with higher values indicating

a secure child. IJzendoorn et al. showed that the AQS is a reliable an valid measure.36 The

Experiences in Close Relationships- Revised Scale (ECR-R) is a 36-item self-report measure of

adult attachment, which requires participants to reflect on their typical ways of relating in close/

romantic relationships. Reviews of self-report measures of adult attachment suggest that the

ECR-R has the best psychometric properties of the available measures.37 The measurement of

attachment quality (MAQ)38 is a 14-item, measure of attachment orientation. It has separate

scales to assess secure attachment tendencies and avoidant tendencies, and two scales reflecting

aspects of the anxious-ambivalent pattern. The inventory of parent and peer attachment-revised

version for children (IPPA-R),39 is a 28-item child report questionnaire measuring the quality of

attachment to parents (and peers) and how well they serve as a source of psychological secu rity.

The IPPA-R had good internal consistency and good convergent validity. However, it is possible

that children answer these ques tions in a more socially desirable manner as the questions are

more personalised. The security scale (SC)40 is a self-report questionnaire to measure attachment

toward mother and father. The SC has been found to be internally consistent and stable. Support

was found for the convergent and discriminant validity.41

Findings

The prevalence of obesity in insecurely attached children (23.1%, 95% CI, 19.9%-26.3%) and

securely attached children (16.6%, 95% CI, 15.3%-17.8%) was reported once.12

R1R2R3R4R5R6R7R8R9


32 | Chapter 2

Overall, the reviewed studies suggested a concurrent relationship between insecure attachment,

particularly attachment anxiety, and a higher BMI.12, 32, 33, 42-46 All longitudinal studies controlled

for baseline BMI or birth weight. The two longitudinal studies of Anderson et al.12, 42 that included

the largest study populations found an association between attachment insecurity at 2 years of

age and obesity two-an-a-half years later12 and in their adolescence.42 This was also confirmed in

a longitudinal study with 601 preadolescent children, in which attachment insecurity was found

to predict an increase of BMI one year later.33 No longitudinal studies were performed in adults.

Three studies reported mediational pathways between attachment and BMI.43, 44, 46 One study

suggested that the relationship between attachment and BMI was partially mediated by mood

(i.e., symptoms of anxiety and depression).44 Other studies indicated that eating self-efficacy43

and disinhibited eating46 were mediators of the relationship between attachment representations

and BMI.

However, not all studies observed a relation between attachment and BMI. One study did not

found a relation between attachment and waist-hip ratio (WHR).31

Quality of included studies

A detailed description of individual study quality is provided in Appendix 1. Four studies of high

quality and two studies of medium quality were identified. The other remaining studies were

considered of low quality.

Discussion

This paper reviewed empirical studies highlighting the association between attachment

representations and obesity. Both cross-sectional and longitudinal studies rather uniformly

suggest that individuals with more insecure attachment representations, in particular individuals

high on attachment anxiety, are at greater risk for obesity.

Although the three of the four reviewed longitudinal studies observed a temporal relationship

between early observations of attachment and later observations of obesity,12, 33, 42 solid

conclusions about causality cannot be inferred from observational longitudinal data. Moreover,

in two studies birth weight –but not body weight at the time of assessment of attachment– was

used as a control variable.12, 42 However, another study that controlled for earlier body weight



2

did also observe an association between attachment insecurity and an increase of body weight

one year later.33 Thus, although it cannot be excluded that the presence of obesity impacts on

attachment relationships or that both insecure attachment and obesity are influenced by a third

variable, the longitudinal studies do suggest that early attachment insecurity shows who is at risk

for later obesity.

Several complementary explanations for the association between attachment and obesity may

hold. One explanation is based on physiological responses to stress.24 Studies suggest that

attachment insecurity, and in particular attachment anxiety,47, 48 lead to hyperactivity of the

hypothalamic pituitary adrenal (HPA) axis and the release of glucocorticoids of which cortisol is the

most well-known.49 Hyperactivity of the HPA-axis can cause accumulation of depot fat in visceral

adipose tissues.50 This can alter glucose metabolism and promote insulin resistance which changes

the number of appetite-related hormones (e.g. leptin, ghrelin) and feeding neuropeptides. As a

result the secretion of Neuropeptide Y and ghrelin (hunger-stimulating hormone) may increase,

while the release of leptin (satiety-stimulating hormone) may decrease.51 By modifying glucose

metabolism and insulin sensitivity, eating and especially eating of high caloric food may reduce

the symptoms of stress.45, 46, 52, 53 Thus, in people with high attachment anxiety, the heightened

physical responses to stressors may have stimulated eating leading to obesity.

As stress responses not only depend on stressors but also on the appraisal of stressors and one’s

capability of dealing with stressors,54 a supplementary explanation for the association between

attachment representations and obesity can be found in the underdevelopment of emotion-

regulation processes. Confronted with a stressor, securely attached individuals seek proximity

to significant others. Previous studies suggest that satisfying interpersonal relationships may

reduce the impact of stressors on individuals’ health,55-57 in interaction with other mediators by

decreasing cortisol levels.58 Similarity high caloric foods may act to calm the stress-perceiving

areas of the brain as was shown in animal studies.59-61 That is, food intake, just as during satisfying

social interactions with significant others, leads to a release of oxytocin from the hypothalamus

which has an anxiolytic effect.62 Therefore, to compensate for poor emotion regulation skills, food

consumption may serve as a way of “self-medication” for more anxiously attached individuals by

releasing oxytocin and down-regulating negative affect.

A number of limitations to this review should be recognised. First, we reviewed 10 individual

studies, each of which with its own strengths, but also limitations such as small sample size or

representatives of the study population. Using the Newcastle-Ottawa Scale as a guide,30 only six

R1R2R3R4R5R6R7R8R9


34 | Chapter 2

of the ten studies were observed medium to high quality. Second a variety of measures were used

for attachment and obesity. Obesity was measured by BMI and waist-hip-ratio and attachment

was measured with nine different measures, which may have affected the results. However, the

relationship between attachment and obesity was rather uniformly observed, independent of

the attachment measure, which supports the strength of the association between attachment

representations and obesity. The assessment of attachment representations involved observation

of the child, the Adult Attachment Prototype Rating (AAPR) interview, and questionnaires

measuring attachment in a categorical or dimensional way. While categorical measures provide

clear textbook cases of the four prototypical attachment styles, dimensional measures describe

a two dimensional space which may be depicted linearly on a spectrum with attachment anxiety

at the one en and attachment avoidance at the other.63, 64 Both types of measures have their

advantages and disadvantages. Categorical measures often use responses to single items to

make classifications (e.g.65), which can lead to serious problems in conceptual analyses, statistical

power, and measurement precision,66 whereas dimensional measures do not guarantee that

measurement precision will be equally distributed across the domain of interest.67

One important step in future research will be to perform high quality longitudinal studies and

research on the predictive role of attachment on obesity. The prediction of adult attachment

behavior and obesity from attachment patterns in early childhood is needed to verify the

hypothesized etiological role of early attachment behavior. In these studies baseline assessments

of body weight should be included. Also the prediction of adult obesity from attachment in

adolescence is particularly useful, because adolescence is a significant period for the onset and

increase of obesity, especially among girls.68 Future studies employing a prospective design could

investigate the usefulness of interventions aimed at the guidance of more insecurely attached

patients both during weight loss treatment programs and during treatment of comorbidities.

The early stage of theory and research in the field of obesity indicates the potential importance

of considering attachment representations in obesity care. Implications can be twofold. On an

individual level attachment theory may guide us in indicating individuals at risk of obesity in order

to better customize prevention efforts to individual characteristics. At a wider level, attachment

theory can usefully contribute to finding an overall framework for future research and the

development of obesity care and services.



2

Reference List

1. James WP. WHO recognition of the global obesity epidemic. Int J Obes (Lond) 2008;32 Suppl 7:S120-S126.

2. Ogden J, Clementi C, Aylwin S, Patel A. Exploring the impact of obesity surgery on patients’ health status: a quantitative and qualitative study. Obes Surg 2005;15(2):266-272.

3. McWilliams LA, Bailey SJ. Associations between adult attachment ratings and health conditions: evidence from the National Comorbidity Survey Replication. Health Psychol 2010;29(4):446-453.


5. Francis LA, Susman EJ. Self-regulation and rapid weight gain in children from age 3 to 12 years. Arch Pediatr Adolesc Med 2009;163(4):297-302.

6. Ozier AD, Kendrick OW, Leeper JD, Knol LL, Perko M, Burnham J. Overweight and obesity are associated with emotion- and stress-related eating as measured by the eating and appraisal due to emotions and stress questionnaire. J Am Diet Assoc 2008;108(1):49-56.

7. Bowlby J. Attachment and Loss, retrospect and prospect. Am J Orthopsych 1969; 52, 664-678.

8. Platts H, Tyson M, Mason O. Adult attachment style and core beliefs: Are they linked. Clin Psychol Psychot 2002; 9: 332-348200.


10. Griffin D BK. Models of the self and other: Fundamental dimensions underlying measures of adult attachment. J Pers Soc Psychol 1994; 67: 430-445.



13. Crowell J, Fraley R, Shaver P. Measurement of individual differences in adolescent and adult attachment. In: Cassidy J, Shaver P (eds). Handbook of attachment theory and research. Guilford Press: New York, 1999, pp. 434-465.



16. Brennan KA, Wu S, Loev J. Adult Romantic attachment and Individual Differences in Attidudes Toward Physical contact in the Context of Adult Romantic Relationships. In: Simpson JA, Rholes WS, editors. Attachment theory and close relationships. New York: Guilford Press; 1998:394-428.







R1R2R3R4R5R6R7R8R9


36 | Chapter 2


24. Maunder RG, Lancee WJ, Nolan RP, Hunter JJ, Tannenbaum DW. The relationship of attachment insecurity to subjective stress and autonomic function during standardized acute stress in healthy adults. J Psychosom Res 2006;60(3):283-290.

25. Mikulincer M, Florian V. The relationship between adult attachment style and emotional and cognitive reactions to stressful events. In: Simpson JA, Rholes WS (eds). Attachment theory and close relationships. NY: Guilford: New York, 1998, pp. 143-147.



28. Raynes E, Auerbach C, Botyanski NC. Level of object representation and psychic structure deficit in obese persons. Psychol Rep 1989;64(1):291-294.

29. Coyne JC, Hagedoorn M, Thombs B. Most published and unpublished dissertations should be excluded from meta-analyses: comment on Moyer et al. Psychooncology 2011;20(2):224-225.

30. Wells GA, Shea B, O’Connell D et al. Newcastle-Ottawa Scale. 2006.

31. Midei AJ, Matthews KA. Social relationships and negative emotional traits are associated with central adiposity and arterial stiffness in healthy adolescents. Health Psychol 2009;28(3):347-353.

32. Hintsanen M, Jokela M, Pulkki-Raback L, Viikari JS, Keltiangas-Jarvinen L. Associations of youth and adulthood body-mass index and waist-hip ratio with attachment styles and dimensions. Curr Psychol 2010; 29:257-271.

33. Goossens L, Braet C, Van Durme K, Decaluwe V, Bosmans G. The parent-child relationship as predictor of eating pathology and weight gain in preadolescents. J Clin Child Adolesc Psychol 2012;41(4):445-457.

34. Leak GK, Parsons CJ. The suspectibility of three attachment style measures to socially desirable responding. Social Behavior and Personality 2001;29:21-30.

35. Pottharst K. Research explorations in adult attachment. New York: Peter lang Publishing; 1990.

36. Van Ijzendoorn MH, Vereijken CM, Bakermans-Kranenburg MJ, Riksen-Walraven JM. Assessing attachment security with the Attachment Q Sort: meta-analytic evidence for the validity of the observer AQS. Child Dev 2004;75(4):1188-1213.

37. Fraley RC, Waller NG, Brennan KA. An item response theory analysis of self-report measures of adult attachment. J Pers Soc Psychol 2000;78(2):350-365.

38. Carver C. Adult attachment and personality: Converging evidence and a new measure. Personality and Social Psychology 1997;23:865-883.

39. Gullone E, Robinson K. The Inventory of Parent and Peer Attachment-Revised (IPPA-R) for children: a psychometric evaluation investigation. Clinical Psychology and Psychotherapy 2005;12:67-79.

40. Kerns K, Klepac L, Cole A. Peer relationships and pre-adolescents’ perceptions of security in the child-mother relationship. Developmental Psychology 1996;32:457-466.

41. Kerns K, Schlegelmilch A, Morgan T, Abraham M. Assessing attachment in middle childhood. In: Kerns K, Rchardson R, editors. Attachment in middle hood. New York: NY: Guilford; 2005:46-70.

42. Anderson SE, Gooze RA, Lemeshow S, Whitaker RC. Quality of early maternal-child relationship and risk of adolescent obesity. Pediatrics 2012;129(1):132-140.

43. Bahrami F, Kelishadi R, Jafari N, Kaveh Z, Isanejad O. Association of children’s obesity with the quality of parental-child attachment and psychological variables. Acta Paediatr 2013;102(7):e321-e324.

44. Cooper MJ, Warren L. The relationship between body weight (body mass index) and attachment history in young women. Eat Behav 2011;12(1):94-96.



2

45. D’Argenio A, Mazzi C, Pecchioli L, Di LG, Siracusano A, Troisi A. Early trauma and adult obesity: is psychological dysfunction the mediating mechanism? Physiol Behav 2009;98(5):543-546.


47. Jaremka LM, Glaser R, Loving TJ, Malarkey WB, Stowell JR, Kiecolt-Glaser JK. Attachment anxiety is linked to alterations in cortisol production and cellular immunity. Psychol Sci 2013;24(3):272-279.

48. Powers SI, Pietromonaco PR, Gunlicks M, Sayer A. Dating couples’ attachment styles and patterns of cortisol reactivity and recovery in response to a relationship conflict. J Pers Soc Psychol 2006;90(4):613-628.

49. Kidd T, Hamer M, Steptoe A. Examining the association between adult attachment style and cortisol responses to acute stress. Psychoneuroendocrinology 2011;36(6):771-779.

50. Bjorntorp P. Do stress reactions cause abdominal obesity and comorbidities? Obes Rev 2001;2(2):73-86.

51. Torres SJ, Nowson CA. Relationship between stress, eating behavior, and obesity. Nutrition 2007;23(11-12):887-894.

52. Nemeroff CB. Early-Life Adversity, CRF Dysregulation, and Vulnerability to Mood and Anxiety Disorders. Psychopharmacol Bull 2004;38 Suppl 1:14-20.

53. Teicher MH, Andersen SL, Polcari A, Anderson CM, Navalta CP, Kim DM. The neurobiological consequences of early stress and childhood maltreatment. Neurosci Biobehav Rev 2003;27(1-2):33-44.

54. Lazarus RS. Emotion and adaptation. New York: NY: Oxford University Press; 1991.

55. George LK, Blazer DG, Hughes DC, Fowler N. Social support and the outcome of major depression. Br J Psychiatry 1989;154:478-485.

56. Hibbard JH, Pope CR. The quality of social roles as predictors of morbidity and mortality. Soc Sci Med 1993;36(3):217-225.

57. Stadler G, Snyder KA, Horn AB, Shrout PE, Bolger NP. Close relationships and health in daily life: a review and empirical data on intimacy and somatic symptoms. Psychosom Med 2012;74(4):398-409.

58. Heffner KL, Kiecolt-Glaser JK, Loving TJ, Glaser R, Malarkey WB. Spousal support satisfaction as a modifier of physiological responses to marital conflict in younger and older couples. J Behav Med 2004;27(3):233-254.

59. Dallman MF, Pecoraro N, Akana SF et al. Chronic stress and obesity: a new view of “comfort food”. Proc Natl Acad Sci U S A 2003;100(20):11696-11701.

60. Pecoraro N, Reyes F, Gomez F, Bhargava A, Dallman MF. Chronic stress promotes palatable feeding, which reduces signs of stress: feedforward and feedback effects of chronic stress. Endocrinology 2004;145(8):3754-3762.

61. Peters A, Pellerin L, Dallman MF et al. Causes of obesity: looking beyond the hypothalamus. Prog Neurobiol 2007;81(2):61-88.

62. Onaka T, Takayanagi Y, Yoshida M. Roles of oxytocin neurones in the control of stress, energy metabolism, and social behaviour. J Neuroendocrinol 2012;24(4):587-598.


64. Maunder RG, Hunter JJ. A prototype-based model of adult attachment for clinicians. Psychodyn Psychiatry 2012;40(4):549-573.

65. Hazan C, Shaver P. Romantic love conceptualized as an attachment process. J Pers Soc Psychol 1987;52(3):511-524.

66. Fraley RC, Waller NG. Attachment theory and close relationships. In: Simpson JA, Rholes WS, editors. Adult attachment patterns: A test of the typological model. New York: Guilford Press; 1998:77-114.


R1R2R3R4R5R6R7R8R9


38 | Chapter 2

68. Harding S, Maynard MJ, Cruickshank K, Teyhan A. Overweight, obesity and high blood pressure in an ethnically diverse sample of adolescents in Britain: the Medical Research Council DASH study. Int J Obes (Lond) 2008;32(1):82-90.

69. Kiesewetter S, Kopsel A, Mai K et al. Attachment style contributes to the outcome of a multimodal lifestyle intervention. Biopsychosoc Med 2012;6(1):3.


71. Van Winckel M, Van Mil E. Wanneer is dik té dik? [When is fat too fat?]. In: Braet C, Van Winckel M, editors. Behandelingsstrategieën bij kinderen met overgewicht [Treatmentstrategies in overweight children]. Houten/Diegem, The Netherlands: Bohn Stafleu Van Loghum; 2001:11-26.

72. Kuczmarski RJ, Ogden CL, Guo SS et al. 2000 CDC Growth Charts for the United States: methods and development. Vital Health Stat 11 2002;(246):1-190.



2

Appendix 1

Main items of quality assessment and their scoring using the Newcastle-Ottawa Scale (NOS).30

Selection (Maximum 5 stars)

1) Representatives of sample (1 star for quality is given if the cohort consisted of patients

truly or somewhat comparable to the general population).

2) Ascertainment of exposure (2 stars are given if data was derived from medical records,

trough observation or by structured interview, 1 star is given by the use of validated

self-reported measurement tool and presenting cronbach’s alpha)

3) Sample size (1 star is given if justified and satisfactory)

4) Non-respondents (1 star is given if comparability between respondents and non-

respondents characteristics is established, and the response rate is satisfactory).

Comparability (Maximum 2 stars)

5) Confounding (2 stars are given as body weight at the start of the study was presented

and controlled for and if the study controls for the most important factors and 1 star is

given if the study controls for the most important factor or when a study controls for

any additional factor)

Outcome (Maximum 3 stars)

6) Statistical test (1 star is given if the statistical test used to analyze the data is clearly

described and appropriate, and the measurement of the association is presented,

including confidence intervals and the probability level).

7) Follow-up (1 star is given if study design is longitudinal)

8) Adequacy of follow up (1 star is given in case of complete follow-up, or subjects lost to

follow up unlikely to introduce bias - small number lost - > 30 % or description provided

of those lost)

A total of 8-10 stars was considered a high quality study; 5-7 stars a medium quality study; 4 stars

or less, a low quality study

R1R2R3R4R5R6R7R8R9


40 | Chapter 2

Table 1. Individual quality assessment of observational studies

First author

Repr

esen

tativ

enes

s

Asc

erta

inm

ent

of e

xpos

ure

Sam

ple

size

Non

-res

pond

ents

Con

foun

ding

Stat

istic

al t

est

Follo

w-u

p

Ade

quac

y of

fol

low

-up

Scor

e

Stud

y qu

ality

Cooper et al. 44 0 0 1 0 1 1 0 0 3 LowWilkinson et al. 46 0 1 1 0 1 1 0 0 4 LowD’Argenio et al. 45 1 1 1 0 1 1 0 0 5 MediumKiesewetter at al. 63, 69 0 2 0 0 1 1 0 0 4 LowBahrami et al. 70 0 1 1 0 0 1 0 0 3 LowHintsanen et al. 32 1 1 1 1 1 1 0 0 6 MediumGoossens et al. 33 1 1 1 1 2 1 1 1 9 HighAnderson et al. 12 1 2 1 0 2 1 1 1 9 HighAnderson et al. 42 1 2 1 1 2 1 1 1 10 HighMidei et al. 31 0 1 1 1 2 1 1 1 8 High

3Psychologists’ evaluation of bariatric surgery

candidates influenced by patients’ attachment

representations and symptoms of depression

and anxiety

Floor Aarts, Chris Hinnen, Victor EA Gerdes, Yair Acherman, Dees PM Brandjes

Published in Journal of Clinical Psychology in Medical Settings, 2014; 21 (1).

R1R2R3R4R5R6R7R8R9


42 | Chapter 3

Abstract

Background: This study examines whether patients self-reported attachment representations

and levels of depression and anxiety influenced psychologists’ evaluations of morbidly obese

patients applying for bariatric surgery.

Methods: A sample of 250 patients (mean age 44, 84% female) who were referred for bariatric

surgery completed questionnaires to measure adult attachment and levels of depression and

anxiety. Psychologists rated patients’ suitability for bariatric surgery using the Cleveland Clinic

Behavioural Rating System (CCBRS), unaware of the results of the completed questionnaires.

Results: Attachment anxiety (OR = 2.50, p = .01) and attachment avoidance (OR = 3.13, p = .001)

were found to be associated with less positive evaluations on the CCBRS by the psychologists,

and symptoms of depression and anxiety mediated this association.

Conclusion: This study strongly supports the notion that patients’ attachment representations

influence a psychologist’s evaluation in an indirect way by influencing the symptoms of depression

and anxiety patients report during an assessment interview. The clinical implications of these

findings are discussed.


Psychologists' evaluation of bariatric surgery candidates influenced by adult attachment | 43

3

Introduction

The prevalence of morbid obesity is increasing worldwide.1,2 Currently, morbid obesity is seen

as a chronic disease and a risk factor for several medical conditions and increased mortality.3-5

Since patients with morbid obesity respond poorly to traditional dietary and exercise weight-loss

regimens, a gastric bypass operation is currently the most effective method of ensuring significant

and sustained weight loss, improved health and quality of life.6-9

In spite of these promising results, approximately 15% of the patients are unable to benefit

significantly from a gastric bypass operation in terms of weight loss and quality of life.10 These

suboptimal results may be more associated with psychological factors such as psychiatric

problems (e.g. anxiety and depression), eating habits, past success or failure with weight loss

attempts, behavioral compliance, availability of support, and motivation than with surgical factors

such as pouch size.11 Research shows that extensive psychiatric problems, motivation, and lack

of compliance preoperatively, are important factors that may negatively impact the results of a

gastric bypass operation.12-14 Since psychological factors may influence the long-term outcome of

a gastric bypass operation, a pre-surgical psychological assessment to identify possible indicators

of suboptimal outcomes and postoperative psychological problems is a standard component of

the evaluation of candidates applying for bariatric surgery.15-18 However, to date, no studies have

investigated whether psychological risk factors for suboptimal outcomes such as anxiety and

depressive symptomathology actually do influence psychologists’ evaluation of gastric bypass

candidates.

In addition to levels of anxiety and depression, patients’ attachment representations may also

influence the psychologists’ evaluation before bariatric surgery. In accordance with attachment

theory19, insecure attachment representations (indicated by high attachment anxiety and/or high

attachment avoidance) have been found to be associated with poorer mentalization (i.e., the

ability to understand behaviors of oneself and others in terms of mental representations such as

thoughts and feelings),20 less perceived support21, less adherence to lifestyle changes,22, 23 poorer

coping abilities24 as well as more extensive psychological problems throughout life.25-29 In other

words, attachment may influence many of the factors psychologists are likely to incorporate into

their evaluation, including the symptoms of depression and anxiety patients’ experience. If this is

the case, patients’ attachment representations may in part influence psychologists’ evaluations

through the influence attachment may have on levels of anxiety and depression.

R1R2R3R4R5R6R7R8R9


44 | Chapter 3

Based on the aforementioned literature, we hypothesized that insecure attachment representations

would be associated with less positive evaluations by psychologists. Moreover, we hypothesized

that the association between patients’ attachment representations and psychologists’ evaluation

of the patients would be mediated by patients’ symptoms of depression and anxiety.

Methods

Study sample

A group of 250 consecutively referred candidates for bariatric surgery assessed by the Slotervaart

bariatric surgery clinic in Amsterdam, the Netherlands between February 2012 and July 2012

participated in the study. To be eligible for bariatric surgery, patients must have a BMI ≥ 40 or

≥ 35 kg/m² with a co-morbid obesity-related condition such as hypertension, sleep apnoea or

diabetes mellitus.30 Since we were interested in the data of the patients who were referred to the

clinic, none of the candidates were excluded from this study based on BMI, a particular diagnosis

or co-morbidity.

Procedures

All patients referred to the Slotervaart bariatric surgery clinic received pre-surgical multidisciplinary

assessments. The assessment process involved examinations by a dietician, internist, surgeon

and a psychologist. Before meeting with the psychologist, patients received a ‘take-home

package’ designed to assess the preoperative diet and exercise habits, co-morbidities and

sociodemographics. Although in normal pre-surgical psychological assessments (interviews and

questionnaires) inquiries into aspects such as psychological symptomathology, social support and

coping abilities are made, no standardized instruments are used to assess adult attachment or

levels of anxiety and depression. For this study two questionnaires were added that assessed

adult attachment and symptoms of depression and anxiety. After signing informed consent, the

patients completed their questionnaires at home and returned them upon their first visit with one

of the psychologists. Since our psychologists were asked not to view the additional questionnaires

and all confirmed they had not done so, we are confident that these questionnaires did not

influence their clinical judgement. Patients that failed to return the questionnaires upon their visit

with the psychologist were asked to fill out the questionnaire in the waiting room immediately

before their visit and to leave them with the counter assistant when completed. For the screening

of patients, five trained psychologists were available and evaluated a comparable amount of

patients. All psychologists have obtained a bachelor’s and master’s degree in Psychology and

completed internal training for bariatric surgery.



3

All questionnaires returned were treated strictly confidentially and received an identification

number through random allocation. The study was approved by the Medical Ethics Committee.

Measures

Adult attachment was assessed using the Experiences in Close Relationship Scale Revised (ECR-R).

Patients completed the 36-item ECR-R31 that assesses how individuals experience emotionally

intimate relationships. Assessing two broad dimensions, the ECR-R contains 18 items on

attachment anxiety (e.g., “I am afraid that I will lose the love of others” and “My desire to

be very close sometimes scares people away.”) and 18 items on attachment avoidance (e.g.,

“I prefer not to show others how I feel deep down” and “I am nervous when others get too

close to me.”) that theoretically underlie adult attachment.32, 33 Individuals rate how well each

statement describes their feelings in romantic relationships on a 5-point scale, ranging from

“strongly disagree” to “strongly agree.” Attachment anxiety and attachment avoidance are each

calculated as the mean score of 18 items. Individuals who score high on attachment anxiety

exhibit fear of rejection and abandonment and have feelings of unworthiness. Individuals

high in attachment avoidance are uncomfortable with intimacy and interdependence, have an

excessive need for self-reliance, and are reluctant to self-disclose. A combination of relatively

low attachment anxiety and low attachment avoidance is recognized as secure attachment. The

present data showed that Cronbach’s alpha for the attachment anxiety subscale was 0.88 and

Cronbach’s alpha for the attachment avoidance subscale was 0.90. Furthermore, both subscales

of the attachment measure are correlated, which means that there are obvious conceptual and

empirical commonalities between the two. However, a large part of the variance (66%) remains

as yet unexplained. In addition, the two scales were not designed to capture the same constructs

and are therefore treated as conceptually independent in analyses.

Psychologists’ evaluations were assessed using the Cleveland Clinic Behavioural Rating System

(CCBRS). Research shows that the multidimensional CCBRS is a reliable instrument in the

preoperative psychological evaluation of candidates for bariatric surgery.34, 35 The CCBRS

was developed to assess patients across the domains found to be important in preoperative

psychological assessments (consent, expectations, social support, mental health, substance use/

abuse/dependence, eating behaviors, adherence, coping/stressors and overall impression). The

CCBRS was completed by the evaluating psychologist immediately after the interview resulting

in an overall score ranging from 1 (poor) to 5 (excellent). The 5 ratings were operationalized as

follows: 5, excellent- no concerns and no psychological follow-up recommended unless future

problems develop; 4, good – if a problem is present, it is well managed and relative weaknesses

R1R2R3R4R5R6R7R8R9


46 | Chapter 3

or concerns may be addressed without significant intervention; 3, fair- concerns or risk factors are

present but reasonably well-controlled or managed, with a balance between the patient’s relative

strengths and weaknesses; 2, guarded- strongly recommend intervention before proceeding and

likely to require discussion in multidisciplinary rounds; and 1, poor – inappropriate with risks very

likely outweighing benefits (e.g., threatening or aggressive to staff; acutely psychotic). Results

suggest that the CCBRS is an instrument with very high internal consistency and internal-rater

reliability.34

The Hospital Anxiety and Depression Scale (HADS) consists of 14 items, divided into two subscales.

Seven items relate to anxiety (e.g., “I feel tense or wound up.” “I get sudden feelings of panic.”)

and seven items relate to depression (e.g., I have lost interest in my appearance.” “I feel as if

I am slowed down.”). Each item has four descriptive response options to be scored on a scale

of 0 to 3, with a value of 0 corresponding to “not exhibiting the symptom at all,” and a value

of 3 corresponding to “exhibiting the symptom to a high degree.” Scores for each of the two

sub-scales are derived by summation of its seven items. If one or more of its items were missing,

the subscale was disregarded. The lowest possible score for each subscale is 0 and the highest

possible score for each subscale is 21. The developers have suggested that aggregate sub-scale

scores of 0–7 represent non-cases, while scores of > 8 on the subscale indicate a current disorder

that would warrant clinical attention may be present.36 High internal consistency was found

for both subscales of the HADS in this study; HADS-anxiety Cronbach’s alpha = .85 and HADS-

depression Cronbach’s alpha = .79.

Statistical analysis

Statistical analyses were performed using the SPSS 19.0-software package. Descriptive statistics

were calculated for demographic variables and medical variables (see Table 1). Means (M)

and standard deviations (SD) were calculated for continuous variables with frequencies and

percentages used to describe categorical data.

Chi-square tests and the one-way-ANOVA were used to test the associations between demo-

graphic variables (e.g., age, gender, marital status), medical variables (e.g., hypertension,

diabetes) and evaluation by different psychologists on the one hand and the outcome variable

(CCBRS overall rating) on the other, to see which variables should be included as covariates in

further analyses.



3

To test our hypothesis the four criteria by Baron and Kenny (1986) should be met.37 The first

criterion states that attachment representations should be significantly associated with the

evaluation score by psychologists (i.e., CCBRS score). This association was tested using multinomial

logistic regression. Multinomial logistic regression is an extension of binary logistic regression. The

technique breaks up the regression analyses into a series of binary regressions comparing each

group of the CCBRS (poor/guarded and fair) to a baseline group (good).

The second criterion state that attachment anxiety and attachment avoidance should be

associated with symptoms of depression and anxiety. This association was tested with linear

regression analysis. The third criterion states that symptoms of depression and anxiety should

be associated with the evaluation score by psychologists. This association was also tested with

multinomial logistic regression. The fourth criterion states that after controlling for symptoms of

depression and anxiety the association between attachment anxiety and attachment avoidance,

on the one hand, and the evaluation score by psychologists on the other, should be reduced

or be no longer significant. This association was tested using multinomial logistic regression.

Sobel test38 was used to determine whether the reduction in the association between attachment

and the evaluation score by psychologists was significant. Findings support partial mediation

if the association between the independent variable (attachment) and the dependent variable

(CCBRS) is reduced but still significant, and full mediation if the mediated effect is no longer

significant. The criteria proposed by Baron and Kenny (1986) and the use of the Sobel tests are

well established.37, 38

Results

Our study sample included 250 patients, was predominantly female (84%), and the mean age

was 44 ± 10.9 years. More than half of the patients lived together with a partner (68%), and

only 19% had completed a Bachelor’s degree or higher. The majority of patients was currently

employed (65%). Median weight was 121.9 kg /299.83 lb (interquartile range 109.0 kg – 136.0

kg/ 240.3 lb – 299.83 lb) and median BMI was 42.4 (interquartile range 39.6 - 46.9).

R1R2R3R4R5R6R7R8R9


48 | Chapter 3

Table 1. Characteristics of patients (N =250)

Variable All patientsN (%)

CCBRSGuarded/Poor

N (%)

CCBRSFair

N (%)

CCBRSGoodN (%)

Mean age (SD) 43.9 (10.9) 37.9 (13.9) 43.3 (11.0) 46.2 (9.2)Mean BMI (SD) 43.6 (6.3) 46.2 (6.4) 44.1 (5.7) 42.8 (6.7)Mean weight kg (SD) 124.6 (24.6) 125.8 (21.4) 126.8 (23.1) 123.0 (22.1)Gender

Female 209 (83.6%) 27 (93.1%) 78 (83.0%) 85 (81.0%) Male 41 (16.4%) 2 (6.9%) 16 (17.0%) 20 (19.0%)

Marital status Married 166 (68.6%) 13 (44.8%) 60 (65.2%) 80 (79.2%) Single 50 (20.7%) 11 (37.9%) 21 (22.8%) 14 (13.9%) Divorced/Widow 26 (10.7%) 5 (17.2%) 11 (12.0%) 7 (6.9%)

Education Higher level of Education

(bachelor’s degree or higher)47 (19.0%) 4 (14.3%) 12 (12.9%) 28 (26.9%)

Medical comorbidities Diabetes 89 (35.6%) 12 (41.4%) 30 (31.9%) 41 (39.0%) Hypertension 105 (42.0%) 15 (51.7%) 34 (36.2%) 49 (46.7%) Arthralgia 160 (64.0%) 20 (69.0%) 63 (67.0%) 59 (56.2%) Sleep apnoea 51 (20.4%) 7 (24.1%) 23 (24.5%) 16 (15.2%)

Smoking 48 (19.2%) 13 (44.8%) 16 (17.0%) 15 (14.3%)Alcohol 20 (8.0%) 4 (13.8%) 9 (9.7%) 7 (6.7%)Drugs 4 (1.6%) 1 (3.4%) 2 (2.1%) 1 (1.0%)Job

Employed 163 (65.2%) 16 (55.2%) 62 (66.0%) 71 (67.6%)

Note: due to missing data the sum of N varies between 250 and 222

Pre-operative psychological assessment: CCBRS

Most candidates (87.3%, n = 199) were deemed psychologically acceptable for bariatric surgery,

with 41.2% (n = 94) rated as “fair”, and 46.1% (n = 105) rated as “good.” None of the patients

were deemed excellent. A significant subset (11.4%, n = 26) was considered guarded, and

additional treatment and/or requirements were deemed necessary before psychological clearance.

Only 1.3% (n = 3) of candidates were deemed “poor” and were evaluated as unable to achieve

the goals that would lead to clearance for surgery. Since only a small group of the patients were

rated poor by the specialized psychologists, we decided to combine the patients rated as poor

or guarded. Further analyses were performed with the CCBRS divided into three groups (poor/

guarded, fair and good).



3

Table 1 presents demographic characteristics stratified by CCBRS group. Patients who were rated

by the psychologist as poor/guarded on the CCBRS were more likely to be smokers (x² = 14.14,

p = .001), to live without a partner (x² = 13.51, p = .009), to be younger, (F(2) = 7.2, p = .001),

to have a lower level of education (x² = 6.69, p = .035), and to have a higher BMI, (F(2) = 3.6,

p = .03) than patients who were rated fair or good on the CCBRS.

Table 2. Correlation matrix of the four predictors that are central in this study

1 2 3 Mean SD1. Attachment anxiety 1.99 .772. Attachment avoidance .58* 2.14 .763. HADS-anxiety .52* .41* 5.56 4.094. HADS-depression .41* .35* .69* 5.80 3.98

*Correlation is significant at the 0.01 level (2-tailed).

Therefore, we adjusted for smoking, marital status, age, education and BMI in further analyses, to

ensure that the effect of the predictors were independent of these variables. The CCBRS ratings

did not differ significantly based on gender, evaluation by different psychologists, comorbidities,

or alcohol and drug use and were excluded from further analyses. Coefficients are considered

significant if the respective p-values are less than α = 0.05. Table 2 shows the correlations between

the four predictors that are central in this study.

Association of adult attachment and CCBRS mediated by symptoms of depression and anxiety

We tested whether symptoms of depression and anxiety were potential mediators for the

association between adult attachment and independent CBBRS ratings by psychologists (Figure

1). Table 3 shows all beta coefficients and p-values of the analyses. The first criterion that should

be met for symptoms of depression and anxiety to influence the association between attachment

representations and the CCBRS is that attachment anxiety and attachment avoidance should

be significantly associated with the CCBRS. Multinomial logistic regression analyses between

attachment representations and the CCBRS showed that higher scores on attachment anxiety

(OR = 2.50, 95% [CI 1.25, 4.99]) and attachment avoidance (OR = 3.13, 95% CI [1.56, 6.29])

predicted lower scores on the CCBRS for the group evaluated as poor/guarded than for the group

evaluated as good. Moreover, higher scores on attachment anxiety (OR = 2.71, 95% CI [1.69,

4.33]) and attachment avoidance (OR = 1.64, 95% CI [1.03, 2.58]) were also associated with

lower scores in group fair vs. group good (Unmediated effect column of Table 3). Thus patients

with a higher score on attachment anxiety and/or attachment avoidance received a less positive

evaluation from the specialized psychologist.

R1R2R3R4R5R6R7R8R9


50 | Chapter 3

Tab

le 3

. Med

iatio

n of

att

achm

ent

on t

he C

CBR

S by

sym

ptom

s of

dep

ress

ion/

anxi

ety

Un

med

iate

d e

ffec

t

Β

p

Path

A

Β

p

Path

B

Β

p

Med

iate

d e

ffec

t

Β

p

Sob

el’s

Z (p

val

ue)

Β

p

Att

achm

ent

anxi

ety

-> H

AD

S-an

xiet

y ->

Po

or/ G

uard

ed v

s. G

ood

.916

(p =

.01)

2.63

8 (p

< .0

01)

.130

(p <

.05)

.776

(p =

.07)

a2.

13 (p

< .0

5)

Att

achm

ent

anxi

ety

-> H

AD

S-an

xiet

y ->

Fa

ir vs

. Goo

d.9

98 (p

< .0

01)

2.63

8 (p

< .0

01)

.125

(p <

.01)

.878

(p =

.001

)b2.

93 (p

< .0

1)

Att

achm

ent

avoi

danc

e ->

HA

DS-

anxi

ety

->Po

or/ G

uard

ed v

s. G

ood

1.14

2 (p

= .0

01)

2.15

1 (p

< .0

01)

.130

(p <

.05)

1.02

6 (p

< .0

1)b

2.08

(p <

.05)

Att

achm

ent

avoi

danc

e ->

HA

DS-

anxi

ety

->Fa

ir vs

. Goo

d.4

91 (p

< .0

5)2.

151

(p <

.001

).1

25(p

< .0

1).2

49 (p

= .3

2)a

2.81

(p <

.01)

Att

achm

ent

anxi

ety

-> H

AD

S-de

pres

sion

->

Poor

/ Gua

rded

vs.

Goo

d.9

16 (p

= .0

1)1.

842

(p <

.001

).1

70 (p

< .0

1).6

54 (p

= .0

8)a

2.44

(p =

.01)

Att

achm

ent

anxi

ety

-> H

AD

S-de

pres

sion

->

Fair

vs. G

ood

.998

(p <

.001

)1.

842

(p <

.001

).1

53 (p

= .0

01)

.843

(p =

.001

)b2.

97 (p

< .0

1)

Att

achm

ent

avoi

danc

e ->

HA

DS-

depr

essi

on -

>

Poor

/ Gua

rded

vs.

Goo

d1.

142

(p =

.001

)1.

671

(p <

.001

).1

70 (p

< .0

1).8

78 (p

< .0

5)b

2.37

(p <

.05)

Att

achm

ent

avoi

danc

e ->

HA

DS-

depr

essi

on -

>Fa

ir vs

. Goo

d.4

91 (p

< .0

5)1.

671

(p <

.001

).1

53 (p

= .0

01)

.277

(p =

.26)

a2.

86 (p

< .0

1)

a fu

ll m

edia

tion

b pa

rtia

l med

iatio

n



3

Psychologists evaluationAdult attachment

Depressive and anxiety symptoms

path A path B

Unmediated effectFigure 1. Depression and anxiety symptoms as mediator of the effect of adult attachment on psychologists evaluation. Pathway values are reported in the results section.

The second criterion states that attachment representations should be significantly associated with

symptoms of depression and anxiety. Linear regression analyses showed that attachment anxiety

(β = 1.842, p < .001) and attachment avoidance (β = 1.671, p < .001) were associated with

symptoms of depression. Furthermore, attachment anxiety (β = 2.638, p < .001) and attachment

avoidance (β = 2.151, p < .001) were also associated with symptoms of anxiety. Thus, the second

criterion was also met (Path A column of Table 3).

Next, in accordance with the third criterion by Barron and Kenny, we investigated whether

symptoms of depression and anxiety were significantly associated with CCBRS. Multinomial

logistic regression analyses showed that higher scores on the depression scale (OR = 1.19, 95%

CI [1.05, 1.34]) and on the anxiety scale (OR = 1.14, 95% CI [1.02, 1.28]) predicted lower scores

on the CCBRS for the group evaluated as poor/guarded than for the group evaluated as good.

The analyses also showed that higher scores on the depression scale (OR = 1.17, 95% CI [1.07,

1.27]) and the anxiety scale (OR = 1.13, 95% CI [1.05, 1.23]) were significantly associated with

lower scores in the fair vs. the good group. In other words, having more symptoms of depression

and anxiety was associated with lower ratings on the CCBRS by the psychologist (Path B column

of Table 3).

Finally, the fourth criterion states that for symptoms of depression and anxiety to be mediators,

the association between attachment and CCBRS should be reduced or no longer be significant

when controlling for symptoms of depression or anxiety. As seen in the Mediated effect column

of Table 3, multinomial logistic regression analyses showed that the strength of the association

between attachment representations and the CCBRS decreased when symptoms of depression

or anxiety were taken into account as mediators. Sobel tests showed that the decreases in all

beta coefficients were significant (Sobel’s column of Table 3). Thus, the relationship between

attachment and the CCBRS is mediated by symptoms of depression and anxiety. The first row of

Table 3 is an example for full mediation, as the unmediated effect (p = .01) is significant and the

R1R2R3R4R5R6R7R8R9


52 | Chapter 3

mediated effect (when symptoms of depression or anxiety were taken into account) is no longer

significant (p = .07). The second row is an example for partial mediation, as the beta coefficients

of the mediated effect decreases (β = .878) with reference to the unmediated effect (β = .998),

but still is significant (p = .001).

Discussion

In accordance with our hypothesis, we found that for a group of 250 patients referred for bariatric

surgery, patients’ symptoms of depression and anxiety mediated the association between patients’

attachment representations and psychologists’ evaluation of the patients with the Cleveland

Clinic Behavioural Rating System (CCBRS).

Attachment anxiety as well as attachment avoidance was found to differentiate the psychologists’

evaluation of gastric bypass candidates. That is, higher scores on attachment anxiety and attachment

avoidance were associated with less positive evaluations by the specialized psychologists. To the

best of our knowledge, this has not been shown before. The results of our study confirmed that

both higher levels of attachment anxiety and attachment avoidance were significantly linked with

more symptoms of depression and anxiety. These results are in line with previous studies, which

show that an insecure attachment style may be viewed as a vulnerability factor in developing

psychological problems when confronted with stressors and illness.39-41 Significant differences in

the evaluation of gastric bypass candidates were also found to depend on patients’ symptoms

of depression and anxiety. That is, patients with more symptoms of depression and/or anxiety

were considered by the psychologist to be less suited for surgery. Finally, Sobel tests showed

that attachment anxiety and attachment avoidance were associated with more symptoms of

depression and anxiety, which in turn were associated with a less positive evaluation by the

psychologist. Thus, the effects of higher levels of attachment anxiety and attachment avoidance

on psychologists’ evaluation can be accounted for, in part and in some cases fully, by more

symptoms of depression and anxiety.

Very similar to other reports,13,42-45 most candidates were considered mentally suited for surgery,

with a very small percentage of patients requiring additional treatment and/or stabilization, and

an even smaller percentage of the patients were unable to receive clearance for surgery. None

of the patients were evaluated as excellent. This might indicate psychologists’ preference for less

extreme classifications or might be indicative of the frequency of at least minimal psychological

concerns in a bariatric population.



3

Findings of this study are interesting for the psychological screening of patients applying for

bariatric surgery. That is, patients’ attachment representations seem to influence psychologists’

evaluations in an indirect way by influencing the symptoms of depression and anxiety patients

report during an assessment interview. This may, however, not be the only path through which

attachment informs psychologists’ evaluations. Other ways may be through patients’ patterns

in interpersonal behavior, their mentalization capabilities and the coherence of their stories.20

Moreover, more insecurely attached patients may also be perceived as more difficult, which may

in turn influence psychologists’ evaluation.46

These results should be interpreted in the context of a few study limitations. First, the cross-

sectional design precludes conclusions regarding cause and effect. Second, the attachment and

anxiety and depression data were obtained exclusively from self-report. Third, although five

different psychologists evaluated patients suitability for surgery, there was a possibility that the

different psychologists may have evaluated patients differently. Therefore, we tested for possible

confounding, and no association between evaluation by different psychologists and the CCBRS

scores were found. A further limitation is that we included patients from one bariatric surgery

clinic only, so results may not be generalizable to bariatric surgery patients as a whole. Future

research should further investigate through which factors patients’ attachment representations

may influence psychologists’ evaluation before bariatric surgery.

In summary, this is the first study to empirically document that patients’ attachment

representations influences psychologists’ evaluations before bariatric surgery while being blind

for the attachment measure. Whether a less positive evaluation of more insecurely attached

patients is pertinent should be investigated in future longitudinal studies showing the relationship

between attachment and post-operative outcomes.

R1R2R3R4R5R6R7R8R9


54 | Chapter 3

Reference List

1. Fabricatore AN, Wadden TA. Obesity. Annu Rev Clin Psychol 2006;2:357-377.

2. Haslam DW, James WP. Obesity. Lancet 2005;366(9492):1197-1209.

3. Kopelman PG. Medical management of obesity. Br J Hosp Med (Lond) 2007;68(2):89-93.




7. Colquitt J, Clegg A, Loveman E, Royle P, Sidhu MK. Surgery for morbid obesity. Cochrane Database Syst Rev 2005;(4):CD003641.


9. Carlsson LM, Peltonen M, Ahlin S et al. Bariatric surgery and prevention of type 2 diabetes in Swedish obese subjects. N Engl J Med 2012;367(8):695-704.

10. Sarwer DB, Wadden TA, Fabricatore AN. Psychosocial and behavioral aspects of bariatric surgery. Obes Res 2005;13(4):639-648.

11. Hsu LK, Benotti PN, Dwyer J et al. Nonsurgical factors that influence the outcome of bariatric surgery: a review. Psychosom Med 1998;60(3):338-346.

12. Bauchowitz AU, Gonder-Frederick LA, Olbrisch ME et al. Psychosocial evaluation of bariatric surgery candidates: a survey of present practices. Psychosom Med 2005;67(5):825-832.

13. Fabricatore AN, Crerand CE, Wadden TA, Sarwer DB, Krasucki JL. How do mental health professionals evaluate candidates for bariatric surgery? Survey results. Obes Surg 2006;16(5):567-573.

14. Fried M, Hainer V, Basdevant A et al. Interdisciplinary European guidelines for surgery for severe (morbid) obesity. Obes Surg 2007;17(2):260-270.

15. Ritz SJ. The bariatric psychological evaluation: a heuristic for determining the suitability of the morbidly obese patient for weight loss surgery. Bariat Nurs Surg Pat 2006;1:97-105.






21. Hinnen C, Schreuder I, Jong E, Duijn MV, Dahmen R, van Gorp EC. The contribution of adult attachment and perceived social support to depressive symptoms in patients with HIV. AIDS Care 2012.

22. Kiesewetter S, Kopsel A, Mai K et al. Attachment style contributes to the outcome of a multimodal lifestyle intervention. Biopsychosoc Med 2012;6(1):3.

23. Ciechanowski P, Russo J, Katon W et al. Influence of patient attachment style on self-care and outcomes in diabetes. Psychosom Med 2004;66(5):720-728.

24. Mikulincer M, Florian V, Weller A. Attachment styles, coping strategies, and posttraumatic psychological distress: the impact of the Gulf War in Israel. J Pers Soc Psychol 1993;64(5):817-826.

25. Doron G, Moulding R, Nedeljkovic M, Kyrios M, Mikulincer M, Sar-El D. Adult attachment insecurities are associated with obsessive compulsive disorder. Psychol Psychother 2012;85(2):163-178.



3



28. Mikulincer M. Adult attachment style and individual differences in functional versus dysfunctional experiences of anger. J Pers Soc Psychol 1998;74(2):513-524.

29. Ward MJ, Lee SS, Polan HJ. Attachment and psychopathology in a community sample. Attach Hum Dev 2006;8(4):327-340.

30. Melissas J. IFSO guidelines for safety, quality, and excellence in bariatric surgery. Obes Surg 2008;18(5):497-500.

31. Mikulincer M, Shaver PR. An attachment perspective on psychopathology. World Psychiatry 2012;11(1):11-15.

32. Brennan KA, Clark CL, Shaver PR. Self-report measurement of adult attachment: An integrative overview. Attachment theory and close relationships. In J. A. Simpson & W. S. Rholes (1998) (Eds.) Attachment theory and close relationships:46-76. New York: Guilford Press.

33. Kurdek LA. On being insecure about the assessment of attachment styles. J Soc Pers Relat 2002;19:811-834.

34. Heinberg LJ, Ashton K, Windover A. Moving beyond dichotomous psychological evaluation: the Cleveland Clinic Behavioral Rating System for weight loss surgery. Surg Obes Relat Dis 2010;6(2):185-190.

35. Heinberg LJ, Ashton K, Windover A, Merrell J. Older bariatric surgery candidates: is there greater psychological risk than for young and midlife candidates? Surg Obes Relat Dis 2011.

36. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983;67(6):361-370.

37. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol 1986;51(6):1173-1182.

38. Sobel ME. Effect analysis and causation in lineair structural equation models. Psychometrika 1990;55:495-515.


40. Riggs SA, Vosvick M, Stallings S. Attachment style, stigma and psychological distress among HIV+ adults. J Health Psychol 2007;12(6):922-936.

41. Rodin G, Walsh A, Zimmermann C et al. The contribution of attachment security and social support to depressive symptoms in patients with metastatic cancer. Psychooncology 2007;16(12):1080-1091.

42. Pawlow LA, O’Neil PM, White MA, Byrne TK. Findings and outcomes of psychological evaluations of gastric bypass applicants. Surg Obes Relat Dis 2005;1(6):523-527.

43. Sadhasivam S, Larson CJ, Lambert PJ, Mathiason MA, Kothari SN. Refusals, denials, and patient choice: reasons prospective patients do not undergo bariatric surgery. Surg Obes Relat Dis 2007;3(5):531-535.

44. Tsuda S, Barrios L, Schneider B, Jones DB. Factors affecting rejection of bariatric patients from an academic weight loss program. Surg Obes Relat Dis 2009;5(2):199-202.

45. Walfish S, Vance D, Fabricatore AN. Psychological evaluation of bariatric surgery applicants: procedures and reasons for delay or denial of surgery. Obes Surg 2007;17(12):1578-1583.

46. Maunder RG, Panzer A, Viljoen M, Owen J, Human S, Hunter JJ. Physicians’ difficulty with emergency department patients is related to patients’ attachment style. Soc Sci Med 2006;63(2):552-562.

4Coping style as a mediator between attachment

and mental and physical health in patients

suffering from morbid obesity

Floor Aarts, Chris Hinnen, Victor EA Gerdes, Yair Acherman, Dees PM Brandjes

Published in International Journal of Psychiatry in Medicine 2014;47(1)

R1R2R3R4R5R6R7R8R9


58 | Chapter 4

Abstract

Background: The presence of mental health problems and limitations in physical functioning is

high in patients suffering from morbid obesity. The purpose of the current study was to examine

the mediating role of coping style in the relationship between attachment representations and

mental health and physical functioning in a morbidly obese population.

Methods: A total of 299 morbidly obese patients who were referred to the Slotervaart bariatric

surgery unit in Amsterdam, the Netherlands, completed self-report questionnaires assessing

adult attachment style (Experiences in Close Relationship – Revised Questionnaire), coping style

(Utrecht Coping List) and patients physical functioning and mental health (Short Form-36).

Results: Attachment anxiety (β = -.490, p < .001) and attachment avoidance (β = -.387, p < .001)

were both found to be related to mental health. In addition, attachment anxiety was also found

to be related to physical functioning (β = -.188, p < .001). Coping style partly mediated these

associations.

Conclusion: Findings suggest that coping mediates the association between attachment anxiety

and attachment avoidance on the one hand and mental health and physical functioning in

patients with morbid obesity on the other hand.


Attachment and mental and physical health | 59

4

Introduction

Morbid obesity has a negative impact on patients’ mental health and physical functioning,1

and has been shown to be related to physical and psychiatric co-morbidities such as diabetes,

hypertension, depression and eating disorders.2-4 Research shows that 35.6% of the morbidly

obese patients applying for bariatric surgery have diabetes, the lifetime prevalence of Axis I (DSM-

IV) clinical psychiatric diagnoses was 47.5%.5, 6 Patients suffering from morbid obesity often report

feelings of sadness and shame, lack of energy and limitations in self-care and mobility, which are

important reasons for people to consider bariatric surgery.7, 8 Although patients overweight may

account for these higher levels of mental health problems and limitations in physical functioning,

within the group of morbidly obese patients large differences exist which may in part be explained

by patients’ attachment style.

According to the attachment theory (Bowlby, 1969), internal working models of attachment are

the mechanisms by which the continuity of childhood experiences is thought to be maintained

over time and into adulthood.9 Bartholomew & Horowitz (1991) and Brennan et al. (1998)

demonstrated that these internal working models of attachment style can be categorized

as either attachment anxiety or attachment avoidance.10, 11 Individuals who score high on

attachment anxiety exhibit fear of rejection and abandonment, have feelings of unworthiness

and tend to turn to others in an anxious, clingy manner. Individuals high in attachment avoidance

are uncomfortable with intimacy and interdependence while maintaining a high sense of self-

worth 10, 11. Those low in attachment anxiety and attachment avoidance (i.e. secure attachment)

have been found to show self-confidence coupled with confidence about the availability and

responsiveness of others.12

Compared to more insecure attachment representations, more secure attachment representations

have been consistently found to be associated with better mental and physical functioning in

healthy people,13, 14 chronically ill patients15 and in morbidly obese patients.16 More securely

attached patients may have a more realistic view of the stressors and threats they are facing

and of their own resilience.17 Consequently, physical symptoms and complaints tend to be ‘real’

and connected to illness and injury.18 In contrast, patients high in attachment anxiety tend to

worry about health more than seems justified, are hypervigilant for threats and stressors and

report physical symptoms and complaints also when there is no indication for illness or injury.19, 20

Moreover, patients high in attachment avoidance try their best to suppress symptoms of distress

and may worry too little about their health.18

R1R2R3R4R5R6R7R8R9


60 | Chapter 4

In part, the association between attachment and mental and physical functioning may be

explained by differences in coping skills between more securely and more insecurely attached

patients. 9, 21 Kotler et al. (1994) and Wei et al. (2003) have found evidence for the mediating role

of coping between attachment representations and physical and mental functioning in student

samples.13, 21 However, no published studies have examined the association between attachment

style, various coping styles and mental and physical functioning in morbidly obese patients

seeking bariatric surgery. A complex patient population characterised by both physical as well as

psychiatric co-morbidities.

Coping, defined as the thoughts and actions we use to deal with stress, is found to be strongly

associated with mental health and physical functioning. Lazarus and Folkman (1984) classified

two types of coping responses: emotion focused coping and problem focused coping.22 The

seven coping strategies used in this study are derived from the Utrecht Coping List (UCL).23 The

coping strategies seeking social support, palliative reacting, avoiding, passive reacting, reassuring

thoughts and expression of emotions can be seen as emotion focused coping, whereas active

tackling can be seen as problem focused coping. In general, problem focused coping has been

found to be associated with better outcomes when dealing with situations where a high level

of control is perceived, while emotion focused coping seems more appropriate when stressors

are unalterable.24, 25 Moreover, passive ways of coping and expressing emotions is found to be

associated with lower mental and physical functioning, whereas a more active way of coping and

seeking support is shown to positively influence patients’ mental and physical functioning.26-28

Furthermore, individuals with more secure attachment representations (i.e., low on attachment

anxiety and attachment avoidance) who perceive others as available and responsive may not

only be comfortable seeking support but may also have learned that their own actions (active

tackling) are often effective in reducing distress and in solving problems.29, 30 On the other hand,

less securely attached patients (i.e., those who believe others will not be readily available) may

be more likely to feel overwhelmed and paralyzed (passive reacting) by problems and may

possibly cope by adopting external regulatory mechanisms (palliative reacting), such as smoking

and drinking.31, 32 Specifically, those people with more anxious attachment representations may

tend to express their fears and worries in a profound and clingy way (emotional expression) in

order to ensure support and care from others.33 This tendency is consistent with developmental

experiences where the primary caregiver was only helpful if the “signal” of neediness was strong

enough.14 As a consequence, these people learned to focus on and express negative emotions

while waiting for reassurance as the ability to sooth and distract oneself is underdeveloped.30



4

Instead, they have become hyper-vigilant to distress which manifest itself through worrying

and thinking about negative experiences and emotions in a repetitive and passive way (passive

reacting). Furthermore, patients with more avoidant attachment representations have a habitual

way of coping with negative experiences and emotions by distancing, avoiding and repressing.34-37

They have most likely received consistently unresponsive care giving,38 maintaining a high sense

of self-worth by defensively denying the value of close relationships and stressing the importance

of independence and self-reliance39 and therefore may be reluctant to seek support.40, 41

The purpose of the current study was to examine the relationships between attachment

representations and coping styles on the one hand and mental health and physical functioning

on the other, in a morbidly obese population. Based on the aforementioned literature three

main hypotheses were formulated: (1) more attachment anxiety and attachment avoidance are

associated with worse mental health and physical functioning; (2) lower attachment anxiety

and lower attachment avoidance (i.e., attachment security) are associated with more support

seeking and active tackling, whereas attachment anxiety and attachment avoidant are associated

with more passive reacting and palliative reacting; and (3) the association between patients’

attachment representations on the one hand and physical and mental functioning on the other

are mediated by patients’ coping style.

Method

Study sample

This study took place in the Slotervaart bariatric surgery unit, Amsterdam, the Netherlands

between February and August 2012. The total sample included 299 morbidly obese patients

referred for bariatric surgery. Patients between the ages of 18 and 60 years are eligible for gastric

bypass surgery if IFSO criteria are met: BMI above 40 or a BMI above 35 combined co morbidity

such as hypertension, diabetes, obstructive sleep apnea syndrome or arthrosis, and if they have

made serious attempts at losing weight. 42

Procedures

All patients referred to the Slotervaart bariatric surgery unit received a pre-surgical multidisciplinary

assessment. During this assessment patients received questionnaires to complete at home. These

questionnaires assessed patients’ attachment style, coping styles, physical functioning and mental

health. Patients were asked to bring the completed questionnaire to their next visit at the bariatric

surgery unit.

R1R2R3R4R5R6R7R8R9


62 | Chapter 4

After random allocation all completed and returned questionnaires received an identification

number and information gathered was treated as strictly confidential. The study was approved by

the Medical Ethical Committee. Research participants provided informed consent.

Measures

Attachment styles were assessed using the Experiences in Close Relationships- Revised scale

(ECR-R). The ECR-R is a 36-item self-report measure of adult attachment style, which requires

participants to reflect on their typical ways of relating in close/romantic relationships. Reviews

of self-report measures of adult attachment suggest that the ECR-R has the best psychometric

properties of the available measures.43 The ECR consists of two subscales, attachment anxiety

(e.g., I’m afraid that I will lose my partner’s love) and attachment avoidance (e.g., I prefer not to

show a partner how I feel deep down) and both dimensions are assessed with 18 items. Answers

are on a 5-point scale ranging from ‘strongly disagree’(1) to ‘strongly agree’(5). The present data

showed that Cronbach’s alpha for subscale attachment anxiety was 0.88 and the Cronbach’s

alpha for subscale attachment avoidance was 0.90.

Coping styles were measured using the Utrecht Coping List (UCL), a 47-item, self-report

questionnaire that measures seven empirically derived subscales that assess ‘active tackling’

(7 items, e.g. ‘putting things in a row’, ‘seeking a way to solve a problem’), ‘seeking social

support’ (6 items, e.g. ‘discussing the problem with friends or family’ and ‘asking somebody for

help’), ‘palliative reacting’ (8 items, e.g. ‘looking for distraction’ and looking for good company’),

‘avoiding’ (8 items, e.g. ‘avoiding difficult situations’ and ‘letting things go’), ‘passive reacting’ (7

items, e.g. ‘being overwhelmed by problems’), ‘reassuring thoughts’ (5 items, e.g. ‘imagining that

things could be worse’) and ‘expression of emotions’ (3 items e.g. ‘showing anger to the person

who is responsible for the problem’). Answers are on a 4-point scale ranging from ‘seldom or

never’ to ‘very frequently’. Prior research has shown that the UCL is a valid and reliable instrument

for measuring coping strategies and that it has fairly good internal consistency.23 In the present

data the different coping scales showed good internal consistency, Cronbach’s alpha for active

tackling was .85, for seeking social support .89, for palliative reacting .63, for avoiding .74,

for passive reacting .74, for reassuring thoughts .67, with the exception of the expression of

emotions scale, Cronbach’s alpha is .57. This might be due to the small number of items in this

scale, whereas most other scales consisted of at least five or six items.23



4

Physical functioning and mental health were evaluated using the SF-36, a widely used Health

Related Quality Of Life (HRQOL) measure. Its use in bariatric surgery patient populations is well-

established. For the domain physical functioning and mental health scores were coded, summed

up and transformed to a scale of 0 (worst health) to 100 (best health). The instrument has been

translated into Dutch and validated for the Dutch population.44 The physical functioning domain

and the mental health domain were used as outcome variables in the present study. The physical

functioning domain consists of 10 items (e.g. to what extent do you have limitations in lifting and

carrying groceries?) and answers are on a 3-point scale ranging from ‘extremely limited’ to ‘not

limited at all’. The mental health domain consists of 5 items ( e.g. in the last 4 weeks, how often

did you feel nervous?) and answers are on 6-point scale ranging from ‘constantly’ to ‘never’. The

present data showed that Cronbach’s alpha for physical functioning 0.88, and mental health

0.84, showed good internal consistency.

Statistical analyses

Statistical analyses were performed using SPSS 19.0. Independent T-tests and Pearson’s

correlations were used to explore possible confounding in the relationship between, on the

one hand, demographics (i.e., age, gender), BMI and education level, and, on the other hand,

coping styles, physical functioning and mental health. The mean scores on the SF-36 of physical

functioning and mental health were compared to age-matched Dutch general population norms

using t-tests.44

In order to determine whether coping styles are a mediator of attachment style, and physical

functioning and mental health, three regression equations were carried out: we first regressed

the mediator (coping style) on the independent variable (attachment style); second, we regressed

the dependent variable (physical functioning and mental health) on the independent variable

(attachment style); and third we regressed the dependent variable (physical functioning and

mental health) on both the independent variable (attachment style) and on the mediator (coping

styles). To establish mediation, we tested the three regression equations following the criteria

of Baron and Kenny (1986).45 All relationships that were found insignificant were excluded

from further analyses. Finally, we used unstandardized regression coefficients (B) and standard

errors for the approximate significance test of Sobel (1982) to test for the indirect effect of the

independent variable on the dependent variable via the mediator 45, 46. Standardized regression

coefficients (β) are presented in the text. The level of significance was set at p< .05. All tests were

two-tailed.

R1R2R3R4R5R6R7R8R9


64 | Chapter 4

Results

Descriptive statistics

A total of 299 patients seeking bariatric surgery were included in the study. Mean age of the

study population was 44 (SD = 11.0), and 85% of test subjects were women. Mean BMI was

44.1 (SD = 6.2) and only a small part of the patients had followed a higher education (20.5%).

Pearson’s correlations showed that age was not associated with most of the coping styles except

for palliative reacting (r = -.162, p = .006). Women (M = 2.47, SD = .62) reported to seek more

social support than men (M = 2.15, SD = .66), t(291) = 3.06, (p = .002), 95% CI: .11 - .51.

Women (M = 2.54, SD = .49) were also found to use more reassuring thoughts than men (M

= 2.30, SD = .45), t(292) = 3.11, (p = .002), 95% CI:.09;.40. Patients with a higher education

used more active tackling, t(285) = -3.51, (p = .001), 95% CI: -.41;- .11, and sought more social

support, t(289) = -2.16, (p = .032), 95% CI:-.38;-.02, than patients with a lower education.

Furthermore, patients’ BMI was not associated with one of the different coping styles.

Independent t-test showed gender differences for physical functioning t(297) = -1.981, (p =

.048). Similarly, significant differences in physical functioning t(295) = -2.820, (p = .005), 95%

CI:-15.69; -2.79, were found between patients with a higher and a lower education level.

Pearson correlation showed that both age (r = -.125, p = .030) and BMI (r = -.132, p = .023) were

associated with physical functioning. In other words women, patients with a lower education,

older patients and patients with a higher BMI scored lower on physical functioning than men,

patients with a higher education, younger patients and patients with a lower BMI. Table 1 shows

the correlations between the main study variables. No correlations between demographics and

mental health were found.

Mean scores for physical functioning (present study M = 55.0, norm M = 83.0, P < .001) and

for mental health (present study M = 71.3, norm M = 76.8, P = .001) were significantly lower

compared to those of persons of comparable age in the general Dutch population.44



4

Tab

le 1

. Cor

rela

tion

mat

rix o

f th

e m

ain

stud

y va

riabl

es.

2

34

56

78

910

111.

Att

ach

men

t an

xiet

y.5

86**

-.40

6**

.121

*.2

21**

-.27

5**

.532

**.1

20*

-.00

5-.

495*

*-.

210*

*2.

Att

ach

men

t av

oid

ance

1

-.29

3**

.115

.262

**-.

570*

*.4

57**

.027

-.08

4-.

379*

*-.

065

3. A

ctiv

e ta

cklin

g1

.110

-.38

7**

.327

**-.

480*

*-.

100

.310

**. 3

97**

.181

**4.

Pal

liati

ve r

eact

ing

1.1

89**

.216

**.2

19**

.110

.411

**-.

103

-.02

15.

Avo

idin

g

1-.

220*

*.4

63**

.023

.096

-.22

4**

-.07

16.

See

kin

g s

oci

al s

up

po

rt1

-.21

1**

.183

**.2

57**

.210

.093

7. P

assi

ve r

eact

ing

1

.256

**.0

04-.

657*

*-.

216*

*8.

Exp

ress

ion

of

emo

tio

ns

1.0

31-.

123*

.042

9. R

eass

uri

ng

th

ou

gh

ts1

.023

-.04

010

. Men

tal h

ealt

h1

.340

**11

. Ph

ysic

al f

un

ctio

nin

g1

Cor

rela

tion

is s

igni

fican

t at

the

0.0

5 le

vel (

2-ta

iled)

. **

Cor

rela

tion

is s

igni

fican

t at

the

0.0

1 le

vel (

2-ta

iled)

.

R1R2R3R4R5R6R7R8R9


66 | Chapter 4

Attachment, Coping and Health

We tested whether coping was a potential mediator for the effect of adult attachment and

mental and physical health (Figure 1). The first criterion that should be met for a coping style

to mediate the association between attachment anxiety and attachment avoidance on the one

hand and mental health and physical functioning on the other stipulates that attachment should

be significantly associated with mental health and physical functioning. The unmediated effect in

Table 2 shows that, after controlling for age, gender, education and BMI, a significant negative

association was found between attachment anxiety and physical functioning (β = -.188, p < .001)

and between attachment anxiety and mental health (β = -.490, p < .001). Furthermore, we also

found a strong relation between attachment avoidance and mental health (β = -.387, p < .001).

AttachmentAttachment anxiety

Attachment avoidance

HealthPhysical functioning

Mental health

CopingActive tackling

Seeking social support

Palliative reacting

Avoiding

Passive reacting

Reassuring thoughts

Expression of emotions

Unmediated effect

Path A Path B

Figure 1. Coping as mediator of the effect of adult attachment on health outcomes. Pathway values are reported in the Table 2.



4

Tab

le 2

. C

opin

g st

yle

as m

edia

tor

betw

een

atta

chm

ent

styl

e an

d th

e de

pend

ent

varia

bles

men

tal h

ealth

and

phy

sica

l fun

ctio

ning

, ad

just

ed f

or a

ge,

gend

er,

educ

atio

n an

d BM

I

Pred

icto

r ->

Med

iato

r ->

Ou

tco

me

Un

med

iate

d e

ffec

tB

p

Path

AB

p

Path

BB

p

Med

iate

d e

ffec

tB

p

Sob

el’s

Z

(p v

alu

e)B

p

Att

achm

ent

anxi

ety

-> A

ctiv

e ta

cklin

g->

• Ph

ysic

al f

unct

ioni

ng-5

.422

(p<

.001

)-.

266

(p<

.001

)5.

642

(p <

.05)

-.4.

857

(p <

.05)

-2.0

96 (p

< .0

5)•

Men

tal h

ealth

-10.

696

(p <

.001

)-.

266

(p<

.001

)12

.663

(p<

.001

)-8

.523

(p <

.001

)-4

.954

(p <

.001

)A

ttac

hmen

t av

oida

nce

-> A

ctiv

e ta

cklin

g->

• M

enta

l hea

lth-8

.632

(p <

.001

)-.

197

(p <

.001

)12

.663

(p <

.001

)-6

.694

(p <

.001

)-4

.063

(p <

.001

)A

ttac

hmen

t an

xiet

y ->

Avo

idin

g->

• M

enta

l hea

lth-1

0.69

6 (p

< .0

01)

.122

(p <

.001

)-7

.994

(p =

.001

)-1

0.34

8 (p

< .0

01)

-2.5

11 (p

< .0

5)A

ttac

hmen

t av

oida

nce

->A

void

ing-

>•

Men

tal h

ealth

-8.6

32 (p

< .0

01)

.148

(p <

.001

)-7

.994

(p =

.001

)-7

.951

(p <

.001

)-3

.089

(p <

.05)

Att

achm

ent

anxi

ety

-> S

eeki

ng s

ocia

l sup

port

->

• M

enta

l hea

lth-1

0.69

6 (p

< .0

01)

-.21

8 (p

< .0

01)

6.46

5 (p

< .0

01)

-10.

122

(p <

.001

)-2

.902

(p <

.05)

Att

achm

ent

avoi

danc

e ->

See

king

soc

ial s

uppo

rt -

>•

Men

tal h

ealth

-8.6

32 (p

< .0

01)

-.46

0 (p

< .0

01)

6.46

5 (p

< .0

01)

-8.7

64 (p

< .0

01)

-3.8

40 (p

< .0

01)

Att

achm

ent

anxi

ety

-> P

assi

ve r

eact

ing-

>•

Phys

ical

fun

ctio

ning

-5.4

22 (p

< .0

01)

.315

(p <

.001

)-9

.637

(p =

.001

)-2

.945

(p =

.130

)-3

.311

(p <

.001

)•

Men

tal h

ealth

-10.

696

(p <

.001

).3

15 (p

< .0

01)

-23.

901

(p <

.001

)-4

.265

(p <

.001

)-8

.343

(p <

.001

)A

ttac

hmen

t av

oida

nce

-> P

assi

ve r

eact

ing-

>•

Men

tal h

ealth

-8.6

32 (p

< .0

01)

.274

(p <

.001

)-2

3.90

1 (p

< .0

01)

-2.5

62 (p

< .0

5)-7

.219

(p <

.001

)A

ttac

hmen

t an

xiet

y ->

Exp

ress

ion

of e

mot

ions

• M

enta

l hea

lth-1

0.69

6 (p

< .0

01)

.076

(p <

.05)

-4.6

45 (p

< .0

5)-1

0.43

9 (p

< .0

01)

-1.5

12 (p

= .1

29)

R1R2R3R4R5R6R7R8R9


68 | Chapter 4

These results show that the first criterion for mediation was met for all the associations, except

for attachment avoidance and physical functioning.

The second criterion stipulates that attachment representations should be significantly associated

with coping style. Path A in table 2 shows that a lower score on attachment anxiety and

attachment avoidance are found to be associated with more active tackling and social support

seeking. A higher score on attachment anxiety and/or attachment avoidance are found to be

associated with more avoidance, passive reacting and palliative coping. Moreover, a higher score

on attachment anxiety is found to be associated with more emotional expression. The coping

style reassuring thoughts, was not associated with any attachment style and therefore excluded

from further analyses.

Next, in accordance with the third criterion, we investigated whether the different coping styles

were significantly associated with mental health and physical functioning. Regression analyses

showed that active tackling was positively and passive reacting negatively associated with

both physical functioning and mental health (see Path B Table 2). Moreover, social support was

positively associated with mental health, while avoidance and expression of emotions were both

negatively associated. Thus, the third criterion was also met.

Finally, the fourth criterion for coping style to be a mediator, requires the association between

attachment style and physical functioning or mental health to be reduced or to no longer be

significant after controlling for a specific coping style. The mediated effect in Table 2 shows that

this final criterion was also met. The strength of the association between attachment anxiety or

attachment avoidance and mental health and physical functioning, decreased when the specific

coping style was taken into account as a mediator. The Sobel test confirmed these findings and

showed that the decrease in all beta coefficients was significant (Table 2). The beta coefficients

of attachment anxiety and attachment avoidance in all relations decreased, but remained in

almost all mediating analyses significant when coping style was controlled for. This indicates

that the association between attachment style and mental health or physical functioning was

partly mediated by the mentioned coping styles. In contrast, a full mediation model was found

to describe the association between attachment anxiety and physical functioning, and the beta

coefficient was no longer significant when controlling for passive reacting.



4

Discussion

Although previous studies identified relations between attachment and mental health and physical

functioning,13, 15 our results extends these findings by suggesting that attachment representations

influences mental health and physical functioning through various coping styles in morbidly obese

patients seeking bariatric surgery. A population who constantly has to cope with the different

aspects of morbid a obesity, a chronic life threatening and limiting disease often combined with

complex psychological and medical problems (e.g. diabetes).

Our results suggest that more securely attached morbidly obese patients (i.e., those with lower

levels of attachment anxiety and attachment avoidance) reported more active problem solving

strategies as well as a willingness to seek support when needed. These findings confirm our

expectations and may reflect an adequate balance between two sides of a continuum as described

by Maunder and colleagues.14, 47 This continuum stretches from autonomy and confidence in one’s

own ability to solve problems to dependency and the need to trust others for support and care.

While more securely attached patients are able to integrate both sides, more insecurely attached

patients (i.e., those with higher levels of attachment anxiety and attachment avoidance) discard

one side of the continuum (either autonomy in the case of attachment anxiety or dependency

in the case of attachment avoidance) in favour of the other side. This is reflected by their coping

style, which, in this and other studies,30 was found to be characterised by not knowing what to

do and by feeling overwhelmed in stressful situations (i.e., avoidance and passive coping).

Active problem solving coping as well as support seeking were in the present study, as well as

in previous studies28 found to be associated with better mental health. Hence, these coping

strategies seem to help more securely attached patients maintain a positive outlook despite their

overweight. In contrast, avoidance coping, passive coping and expression of emotions were

found to be associated with worse mental health. This may help explain why more insecurely

attached patients may experience more mental problems than more securely attached patients.

Although we did not find any association between attachment avoidance and palliative reacting,

we did find an association for attachment anxiety.

Moreover, in the present study we also found that more anxiously attached patients (i.e., those

scoring higher on attachment anxiety) reported worse physical functioning. This was not true for

more avoidant attached patients (i.e. those scoring higher on attachment avoidance). This finding

is in accordance with the idea that more anxiously attached patients may respond to a stressor in

R1R2R3R4R5R6R7R8R9


70 | Chapter 4

a hypervigilant way, that they focus on physical complaints and that they may express their fears

and worries more eagerly in an attempt to guarantee the availability and responsiveness of those

needed as much as possible.14 More anxiously attached patients have been found to report more

(un)explained physical symptoms and to use the health care system more regularly.48

In the present study active as well as passive coping were not only found to be associated with

mental functioning but also with physical functioning. That is, more active coping and less passive

coping were found to be associated with better physical functioning These findings are in line

with other studies.27 An explanation for these findings might be that due to their passive way

of coping patients develop more stress-related physical symptoms (e.g. back pain), which in

turn may lead to worse physical functioning.48 Moreover, patients with a less active coping style

may tend to focus more on bodily sensations, which in turn may also result in more physical

complaints. Alternatively patients with more physical problems may be forced to use more passive

coping strategies, due to their physical complaints.

Furthermore, patients with morbid obesity were found to report more impaired physical

functioning and mental health compared to the general Dutch population.49, 50 We found that

more securely attached patients exhibited better mental health and physical functioning than

more insecurely attached patients. While the physical functioning scores of the more securely

attached patients were still lower than those of the general population, the scores for mental

health of more securely attached patients were comparable to those of the general population.

These results indicate that more secure attachment representations may serve as a buffer for

mental health of morbidly obese patients. In contrast, more insecure attachment representations

may exacerbate the impact of obesity on mental health and physical functioning.

These results should be interpreted in the context of limitations. As we only included patients

from the Slotervaart bariatric surgery unit, the results may not be generalizable to all patients

seeking bariatric surgery and the general population as a whole. In addition all data is gathered

through self-reported measurements. Determining the mediating role of coping styles in the

relation between attachment representations and mental health and physical functioning

could therefore benefit from the inclusion of clinical interviews. Furthermore, this study has a

cross-sectional design which prevents us from drawing conclusions about causality. Therefore

longitudinal studies are needed to evaluate our findings.



4

Despite these limitations, the value of our study lies in that it is the first to investigate the

relationship between attachment representations and mental health and physical functioning

in patients seeking bariatric surgery and how this is mediated by patients’ coping styles. In

considering bariatric surgery for patients, specialists currently use patients’ quality of life as an

important criterion, as they expect that patients’ quality of life will improve after the operation.

Although post-operative improvements in quality of life (including physical functioning and

mental health) are expected, our findings suggest that BMI predicts only a small part of the

differences in physical functioning and mental health. Therefore, our study argues in favour of

a greater consideration of patients’ attachment representations and coping behaviors when

considering physical functioning and mental health. Findings suggest not only that it is important

to consider attachment anxiety or attachment avoidance in understanding mental health and

physical functioning in patients with morbid obesity but also that coping style plays an important

role in these relationships. Future studies are needed to investigate whether patients attachment

representations and coping behaviors can predict physical functioning and mental health after

the surgery.

R1R2R3R4R5R6R7R8R9


72 | Chapter 4

Reference List

1. van Hout GC, van Oudheusden I, van Heck GL. Psychological profile of the morbidly obese. Obes Surg 2004;14(5):579-588.

2. Guh DP, Zhang W, Bansback N, Amarsi Z, Birmingham CL, Anis AH. The incidence of co-morbidities related to obesity and overweight: a systematic review and meta-analysis. BMC Public Health 2009;9:88.

3. Alciati A, Gesuele F, Rizzi A, Sarzi-Puttini P, Foschi D. Childhood parental loss and bipolar spectrum in obese bariatric surgery candidates. Int J Psychiatry Med 2011;41(2):155-171.

4. Carpiniello B, Pinna F, Pillai G et al. Psychiatric comorbidity and quality of life in obese patients. Results from a case-control study. Int J Psychiatry Med 2009;39(1):63-78.

5. Campos GM, Ciovica R, Rogers SJ et al. Spectrum and risk factors of complications after gastric bypass. Arch Surg 2007;142(10):969-975.

6. Halmi KA, Long M, Stunkard AJ, Mason E. Psychiatric diagnosis of morbidly obese gastric bypass patients. Am J Psychiatry 1980;137(4):470-472.

7. Bond DS, Unick JL, Jakicic JM et al. Physical activity and quality of life in severely obese individuals seeking bariatric surgery or lifestyle intervention. Health Qual Life Outcomes 2012;10:86.

8. Stewart AL, Brook RH. Effects of being overweight. Am J Public Health 1983;73(2):171-178.

9. Bowlby J. Attachment and Loss. Attachment. New York: Basic Books;1969.


11. Brennan KA CCSP. Self-report measurement of adult attachment: An integrative overview. In: Simpson JA, Rholes WS (eds). Attachment theory and close relationships. New York: Guilford Press 1998:46-76.

12. Cooper ML, Shaver PR, Collins NL. Attachment styles, emotion regulation, and adjustment in adolescence. J Pers Soc Psychol 1998;74(5):1380-1397.





17. Ciechanowski P, Katon WJ. The interpersonal experience of health care through the eyes of patients with diabetes. Soc Sci Med 2006;63(12):3067-3079.



20. Stuart S, Noyes R, Jr. Attachment and interpersonal communication in somatization. Psychosomatics 1999;40(1):34-43.

21. Wei M, Heppner PP, Mallinckrodt B. Perceived Coping as a Mediator Between Attachment and Psychological Distress: A Structural Equation Modeling Approach. J Couns Psych 2003;50(4):434-447.

22. Lazarus RS, Folkman S. Stress, appraisal and coping. New York: Springer Publishing company; 1984.

23. Schreurs P, van de Willige G, Tellegen B, Brosschot J. Handleiding Utrechtse Coping Lijst: UCL. Swets & Zeitlinger, Lisse, 1988.



4

24. Folkman S. Lifespan developmental psychology: Perspectives on stress and coping. In: Cummings EM, Greene AL, Karraker KH, editors. Coping across the lifespan: Theoretical issues. Erlbaum, Hillsdale, NJ: 1991:3-19.

25. Folkman S. Personal coping: Theory, research, and application. In: Carpenter BN, editor. Making the case for coping. Praeger, Westport, CT: 1992:31-46.

26. Hopman-Rock M, Kraaimaat FW, Bijlsma JW. Quality of life in elderly subjects with pain in the hip or knee. Qual Life Res 1997;6(1):67-76.

27. Scharloo M, Kaptein AA, Weinman J, Bergman W, Vermeer BJ, Rooijmans HG. Patients’ illness perceptions and coping as predictors of functional status in psoriasis: a 1-year follow-up. Br J Dermatol 2000;142(5):899-907.

28. van der Have M, Minderhoud IM, Kaptein AA et al. Substantial impact of illness perceptions on quality of life in patients with Crohn’s disease. J Crohns Colitis 2012.

29. Simpson JA, Rholes WS, Nelligan JS. Support seeking and support giving within couples in an anxiety-provoking situation: The role of attachment styles. Journal of Personality and Social Psychology 1992;62(3):434-446.

30. Mikulincer M, Florian V, Weller A. Attachment styles, coping strategies, and posttraumatic psychological distress: the impact of the Gulf War in Israel. J Pers Soc Psychol 1993;64(5):817-826.

31. Ciechanowski P, Russo JE, Katon WJ et al. Influence of patient attachment style on self-care and outcome in diabetes. Psychosomatic Medicine 2004;66:720-728.

32. Maunder RG, Hunter JJ. Attachment and psychosomatic medicine: Developmental cotributions to stress and disease. Psychosomatic Medicine 2001;63:556-567.

33. Mikulincer M, Florian V, Tolmacz R. Attachment styles and fear of personal death: A case study of affect regulation. Journal of Personality and Social Psychology 1990;58(2):273-280.





38. Mikail SF, Henderson PR, Tasca GA. An interpersonally based model of chronic pain: An application of attachment theory. Clinical Psychology Review 1994;14(1):1-16.

39. Mikulincer M. Adult attachment style and affect regulation: Strategic variations in self-appraisals. Journal of Personality and Social Psychology 1998;75(2):420-435.





44. Aaronson NK, Muller M, Cohen PD et al. Translation, validation, and norming of the Dutch language version of the SF-36 Health Survey in community and chronic disease populations. J Clin Epidemiol 1998;51(11):1055-1068.

45. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol 1986;51(6):1173-1182.

R1R2R3R4R5R6R7R8R9


74 | Chapter 4

46. Sobel ME. Effect analysis and causation in lineair structural equation models. Psychometrika 1990;55:495-515.


48. Maunder RG, Hunter JJ, Lancee WJ. The impact of attachment insecurity and sleep disturbance on symptoms and sick days in hospital-based health-care workers. J Psychosom Res 2011;70(1):11-17.

49. Algul A, Ates MA, Semiz UB et al. Evaluation of general psychopathology, subjective sleep quality, and health-related quality of life in patients with obesity. Int J Psychiatry Med 2009;39(3):297-312.

50. Lier HO, Biringer E, Hove O, Stubhaug B, Tangen T. Quality of life among patients undergoing bariatric surgery: associations with mental health- A 1 year follow-up study of bariatric surgery patients. Health Qual Life Outcomes 2011;9:79.

5Mental Health Care Utilization in Patients

Seeking Bariatric Surgery:

the Role of Attachment Behavior

Floor Aarts, Chris Hinnen, Victor EA Gerdes, Dees PM Brandjes, Rinie Geenen

Published in Bariatric Surgical Practice and Patient Care 2013;8(4)

R1R2R3R4R5R6R7R8R9


76 | Chapter 5

Abstract

Obesity may be a factor contributing to mental health in patients seeking bariatric surgery.

Whether a person uses mental health care for one’s psychological problems may have its roots in

attachment behavior. The present study (N = 260) identified that attachment anxiety was associated

with more mental health care visits (OR = 1.86, 95% CI = 1.11-2.54, p = .02), present use of

medication (OR = 2.30, 95% CI = 1.43-3.68, p = .001) and previously prescribed medication (OR

= 2.01, 95% CI = 1.13-3.57, p = .02). Furthermore, the use of previously prescribed medication

was especially prevalent in patients with high attachment anxiety and low attachment avoidance

(OR = 2.96, 95% CI = 1.35-6.50, p = .007). The observation that attachment anxiety is associated

with mental health care utilization indicates that it should be recognized and considered by health

care providers working with patients with morbid obesity for therapeutic and economic reasons.


Attachment behavior and mental health care utilization | 77

5

Introduction

Obesity has been recognized as a growing public health problem and it is associated with physical

problems such as type II diabetes and hypertension as well as mental problems such as depressed

mood.1, 2 Mental problems are particularly high among patients with morbid obesity seeking

bariatric surgery,3-7 and mental health care utilization has also been found to be high.8 Some

patients with mental problems are given mental health counseling prior to bariatric surgery

to improve their mental health status.9 The present study focuses on the association between

attachment behavior and mental health care utilization.

Attachment behavior -the habitual way of relating to other persons- plays a role in the etiology

of mental problems, and may influence the risk of individuals becoming obese and the probability

of individuals using mental health services. According to attachment theory,10-13 early interactions

with attachment figures influence how people think, feel and behave in adulthood.14 Anxiously

attached people seek support from others through amplifying distress, while avoidantly attached

people evade dependency on others.15 Confronted with a stressor, people with anxious attachment

representations have been found to increase caloric intake and physiological responses relevant

to eating (e.g. cortisol).6, 16, 17 Moreover, insecure attachment has been found to be associated

with obesity in both child- and adulthood18, 19 and with poor self-efficacy of eating management 20. In addition, insecure attachment was shown to be a vulnerability factor for mental health

problems in the general population21 and in bariatric surgery candidates.22

Mental health care may be used by patients with morbid obesity as a one-off after crisis,23

throughout life in case of chronic psychiatric comorbidity,9 as a preoperative psychological

intervention for bariatric surgery patients with significant psychological problems,24 and as pre-

treatment for bariatric surgery.9 Based on observations in the general population for health care

utilization,25-28 attachment anxiety in bariatric surgery patients is hypothesized to predict mental

health care utilization of any kind. Individuals with anxious attachment representations are

expected to use more mental care, because they have a negative view of the self are hypervigilant

to stressors, have little faith in their own ability to manage and tend to rely on others.29 In

contrast, individuals with avoidant attachment representations have a positive view of the self

and a negative view of others, have fear for intimacy, and have been found to be self-reliant,29

and are therefore expected to use less mental health care. Finally, although individuals with secure

attachment representations believe that they are worthy of care and attention, are comfortable

in seeking support and are confident that health care providers are capable and willing to provide

R1R2R3R4R5R6R7R8R9


78 | Chapter 5

care,29 we expect their use of mental health care to be low because they have a low risk at mental

disorders.27

Thus, the aim of our study was to examine the association between attachment representations

and mental health care use in patients with morbid obesity applying for bariatric surgery.


Study sample

Patients with morbid obesity between the ages of 18 and 60 referred to the Department of

Bariatric Surgery of the Slotervaart Hospital, Amsterdam, the Netherlands between February and

August 2012 were included in this study. Patients are eligible for gastric bypass surgery if they

have a Body Mass Index (BMI) above 40 kg/m2 or a BMI above 35 kg/m2 and co-morbidity such as

hypertension, diabetes, obstructive sleep apnea syndrome (OSAS) or osteoarthritis. Furthermore,

patients should have made serious attempts at losing weight.30 A total of 299 patients from the

Slotervaart bariatric surgery unit completed the questionnaires. Of these 299 patients, the 260

patients with complete datasets on variables needed in this study were included in analyses.

Procedures

Data were obtained from questionnaires filled out by patients during their pre-surgical

multidisciplinary assessment. Questionnaires consisted of questions on demographics, adult

attachment and mental health care utilization. After random allocation all questionnaires received

an identification number and information gathered was treated confidentially. The study was

approved by the Medical Ethical Committee of the Slotervaart hospital. Research participants

provided informed consent.

Instruments

Adult attachment was measured with the Experiences in Close Relationship scale Revised (ECR-R),

a continuous measurement of attachment.21, 31 The ECR-R comprises 36 items to assess how

individuals experience intimate relationships emotionally by employing two broad dimensions,

attachment anxiety (18 items) and attachment avoidance (18 items). Items were rated on a

5-point Likert scale, ranging from ‘strongly disagree’ to ‘strongly agree’. The present data showed

good internal consistency for both subscales, Cronbach’s alphas for attachment anxiety and

attachment avoidance subscale were 0.88 and 0.90.



5

Mental health care utilization of patients was measured with the question: Have you ever been

in contact with a social worker, psychologist or psychiatrist for professional help? Previously

prescribed medication was measured with: Have you ever used medication for mental problems

in the past? The question to measure current medication use was: Do you use medication

for mental problems currently? Questions could be answered by yes or no. Medication use at

presentation was retrieved from the electronic patient files.


Descriptive statistics were used to summarize demographics, attachment, and mental health care

utilization. Means (M) and standard deviations (SD) were calculated for continuous variables.

Frequencies and percentages were used to describe categorical data. Differences between patients

with and without complete datasets regarding demographics were investigated using one-way

analysis of variance and Pearson χ2. Logistic regression analysis was used to predict mental health

care visits, previously prescribed medication and present use of medication for mental problems

from attachment anxiety, attachment avoidance and the interaction between attachment anxiety

and attachment avoidance. Also the possible prediction of age, gender, BMI and education level

of the patient (person characteristics) was examined. However, only those demographic variables

that significantly correlated (p < .10) with at least one of the three variables indicating mental

health care use were included in the regression model.

In step 1 of the logistic regression, demographic variables (i.e. gender, age) were entered. In

step 2, attachment anxiety and attachment avoidance were entered. In step 3, we examined the

interaction term between attachment anxiety and attachment avoidance. Attachment anxiety and

attachment avoidance were centred on their grand mean (i.e., the overall mean was subtracted

from the values of a variable). To probe a significant interaction effect, logistic regression analyses

were repeated including only patients with score below and above the median on attachment

anxiety and attachment avoidance, respectively. Statistical analyses were performed using SPSS

19.0 software package. The level of significance was set at p < .05. All tests were two-tailed.

Results

Description of the sample

The mean age of the study population was 44 years (SD = 10.8); 84% of the research participants

was female, mean BMI was 44 kg/m2 (SD = 6.2) and 20% of the patients had followed higher

R1R2R3R4R5R6R7R8R9


80 | Chapter 5

education (bachelor’s degree or higher). Mean attachment anxiety was 2.01 (SD = .79) and mean

attachment avoidance was 2.13 (SD = .79).

No statistically significant differences were found between the patients with missing data and

those with complete datasets regarding age, gender, BMI or education level (data not shown).

Personal characteristics, attachment style and mental health utilization

In our sample of patients seeking bariatric surgery, 53% of the patients had ever been in contact

with a mental health care provider, 60 patients (23%) had ever used prescribed medication

for mental problems, and 29 patients (11%) currently used prescribed medication for mental

problems. Most of the patients with current medication (n = 23) used antidepressants and two

patients used antipsychotics. Furthermore, six patients used antidepressants or antipsychotics

combined with benzodiazepines.

Table 1 shows the results of logistic regression analysis. In step 1, neither age nor gender

were found significantly associated with the outcome variables. Almost significant (p < 0.10)

observations were that previously prescribed medication use was higher in older than younger

patients (p = .07) and that mental health care visits (p = .06), previously prescribed medication use

(p = .06) and present use of medication (p = .07) were higher for women than men. Step 2 showed

that attachment anxiety was associated with more mental health care visits (OR = 1.86, 95%

CI = 1.11-2.54, p = .02), previously prescribed medication (OR = 2.30, 95% CI = 1.43-3.68,

p = .001), and present use of medication (OR = 2.01, 95% CI = 1.13-3.57, p = .02). No significant

associations were found between attachment avoidance and mental health care utilization. In

step 3, the interaction of attachment anxiety and attachment avoidance predicted a significant

proportion of individual differences in previously prescribed medication (OR = .56, 95% CI = .33-

.94, p = .03). In the prediction of previously prescribed medication, neither the odds ratios for

attachment avoidance in patients below the median on attachment anxiety (OR = 1.32, 95% CI =

0.59-2.96, p = .51) or above (OR = 0.73, 95% CI = 0.42-1.25, p = .25) the median on attachment

anxiety were significant, nor the odds ratio for attachment anxiety in patients scoring high on

attachment avoidance (OR = 1.47, 95% CI = 0.89-2.42, p = .13). However, previously prescribed

medication was significantly predicted by attachment anxiety in patients with attachment

avoidance below the median (OR = 2.96, 95% CI = 1.35-6.50, p = .007) indicating that previously

prescribed medication was especially prominent in patients scoring high on attachment anxiety

and low on attachment avoidance.



5

Tab

le 1

. Re

gres

sion

ana

lyse

s pr

edic

ting

men

tal h

ealth

car

e vi

sits

, pr

evio

usly

pre

scrib

ed m

edic

atio

n an

d pr

esen

t us

e of

med

icat

ion

for

men

tal p

robl

ems

from

pe

rson

cha

ract

eris

tics

(ste

p 1)

att

achm

ent

anxi

ety,

att

achm

ent

avoi

danc

e (s

tep

2) a

nd t

he in

tera

ctio

n te

rm (s

tep

3)

Men

tal h

ealt

h c

are

visi

tsPr

evio

usl

y p

resc

rib

ed m

edic

atio

nPr

esen

t u

se o

f m

edic

atio

nSt

ep 1

OR

95%

CI

OR

95%

CI

OR

95%

CI

Age

1.01

(.99-

1.04

)1.

03*

(.99-

1.06

)1.

03(.9

9-1.

07)

Gen

der

(0=

fem

ale,

1=

mal

e).5

3*(.2

7-1.

04)

.38*

(.14-

1.03

).1

5*(.0

2-1.

17)

Step

2 A

ge1.

01(.9

9-1.

04)

1.03

*(.9

9-1.

02)

1.03

(.99-

1.07

) G

ende

r (0

=fe

mal

e, 1

=m

ale)

.49*

*(.2

5-.9

9).3

7*(.1

4-1.

02)

.16*

(.02-

1.20

) A

ttac

hmen

t an

xiet

y1.

68**

(1.1

1-2.

54)

2.30

***

(1.4

3-3.

68)

2.01

**(1

.13-

3.57

) A

ttac

hmen

t av

oida

nce

1.09

(.73-

1.64

).7

9(.4

8-1.

31)

.71

(.37-

1.37

)St

ep 3

Age

1.01

(.99-

1.04

)1.

03*

(.99-

1.06

)1.

03(.9

9-1.

07)

Gen

der

(0=

fem

ale,

1=

mal

e).4

9**

(.25-

.98)

.36*

(.13-

1.00

).1

6*(.0

2-1.

19)

Att

achm

ent

anxi

ety

1.77

***

(1.1

6-2.

73)

2.66

***

(1.6

4-4.

29)

2.22

***

(1.2

4-3.

96)

Att

achm

ent

avoi

danc

e1.

09(.7

3-1.

63)

.90

(.55-

1.47

).8

(.41-

1.56

)

Att

achm

ent

anxi

ety

* A

ttac

hmen

t

avo

idan

ce.8

0(.5

2-1.

21)

.56*

*(.3

3-.9

4).6

3(.3

2-1.

25)

*p<

.10,

**p

<.0

5, *

**p<

.01

OR=

odds

rat

io; C

I=co

nfide

nce

inte

rval

; BM

I=Bo

dy M

ass

Inde

x

R1R2R3R4R5R6R7R8R9


82 | Chapter 5

Discussion

Our study shows that more than half of the 260 patients (53%) referred for bariatric surgery has

ever been in contact with a mental health care provider. In addition, 1 out of every 4 to 5 patients

(23%) has ever used prescribed medication for mental problems, and 1 out of 9 patients (11%)

currently uses such medication.

The results of this study demonstrate that the use of mental health care is greatest in more

anxiously attached patients and that the use of previously prescribed medication is especially

prevalent in patients scoring high on attachment anxiety and low on attachment avoidance.

These findings are in agreement with attachment theory and may reflect that patients with more

anxious attachment representations seek mental care more often because they rely more for

support and care on others in combination with being more vulnerable for developing mental

problems and experiencing higher levels of negative affect.29 On the other hand, attachment

avoidance was not found to be associated with mental health care, which may reflect preference

to be self-reliant and reluctance to become interdependent. Although patients with avoidant

attachment representations may show considerable biological distress (e.g. increased blood

pressure), they appear calm and subjectively feel and report to be not distressed.21

Furthermore, previous research showed that more secure attachment representations are

associated with resilience and good psychological health.32 Although patients with morbid obesity

who are more securely attached may not be free of mental problems, they might possess more

effective psychosocial skills (e.g. social and communicative competences) and coping strategies

(e.g. social support, active problem solving).32 These skills and strategies may prevent them from

needing mental health care. In our study, more secure attachment representations might be

reflected in the combination of low scores on attachment anxiety and low scores on attachment

avoidance. This interaction was not associated with low or high mental health care use, perhaps

because psychiatric disorders were low in this group and in case of psychiatric disorders these

patients are comfortable in seeking support and are confident that health care providers are

capable and willing to provide support.29 The most use of mental health care was made by

patients scoring high on attachment anxiety and low on attachment avoidance. These patients

may have relatively high mental problems or even psychopathology and are dependent without

being reluctant to accept help form others.



5

Although the association between higher health care utilization and attachment anxiety has been

described in previous studies,26, 27 the present study adds to this literature by focusing specifically

on mental health care utilization in a population seeking bariatric surgery. Some aspects of this

study require comment. The main limitation of this study is its retrospective, cross-sectional

design preventing conclusions about the direction or prospective relation between variables.

Furthermore, our findings do not generalize beyond the population of patients with morbid

obesity seeking bariatric surgery or to other variables not rooted in attachment that may affect

obesity and the use of health care. We cannot exclude that a proportion of the patients may have

had a visit with a psychologist or psychiatrist as part of an earlier weight loss program instead

of treatment for mental problems. A final limitation is that we used self-reports of health care

utilization. Future prospective studies should include questions about the number and reasons of

visits at the different mental health care providers and should verify these visits with mental health

care providers. While this study provide descriptive information on which patients seems to utilize

the most mental health care, future studies are required to examine who needs and benefits from

mental health care on pre- and post-surgical level.

Conclusions

Overall, the results suggest that attachment behavior plays a role in mental health care utilization

of patients with morbid obesity who apply for bariatric surgery. Therefore, it is important for

health care providers working with patients with morbid obesity to have knowledge of the

attachment theory, to recognize anxious attachment representations and to be aware of these

patients’ desire of close relationships and hypervigilance for rejection as well as of the mental

vulnerability of this group. Anticipation on attachment representations may help prevent

unnecessary delay and may increase throughput of patients needing psychological treatment to

improve their mental health before they are allowed to receive bariatric surgery. The implication

is twofold. First of all, more anxiously attached patients may actually need more mental health

care than securely attached patients, and, secondly, their emotionally dependency on caregivers

and fear of rejection and abandonment may lead to unnecessary mental health care visits and

high costs. To deal with both problems regularly scheduled frequent brief visits or telephone

calls with health care providers may be required for these patients.29, 33 If a health care provider

–responsive to concerns34– is available at these scheduled moments before the patient asks for

it and independently of symptoms, anxiously attached patients may become less compulsive in

care-seeking outside these moments. Patients may experience that support occurs regardless

of whether or not they communicate to have symptoms.29 Furthermore, it is important for the

patients that they experience enough support and empathy from the health care provider as

R1R2R3R4R5R6R7R8R9


84 | Chapter 5

well as from more accessible resources such as family, friends or religion.35 Conclusively, the

observation that attachment anxiety is associated with mental health care utilization in morbidly

obese patients seeking bariatric surgery indicates that it should be recognized and considered by

health care providers for therapeutic and economic reasons.



5

Reference List



3. Martinez EP, Gonzalez ST, Vicente MM, van-der Hofstadt Roman CJ, Rodriguez-Marin J. Psychopathology in a sample of candidate patients for bariatric surgery. Int J Psychiatry Clin Pract 2012.

4. Zijlstra H, Larsen JK, Wouters EJM, van RB, Geenen R. The Long-Term Course of Quality of Life and the Prediction of Weight Outcome After Laparoscopic Adjustable Gastric Banding: A Prospective Study. Bariatr Surg Pract Patient Care 2013;8(1):18-22.

5. Fitzgibbon ML, Stolley MR, Kirschenbaum DS. Obese people who seek treatment have different characteristics than those who do not seek treatment. Health Psychol 1993;12(5):342-345.

6. Jaremka LM, Glaser R, Loving TJ, Malarkey WB, Stowell JR, Kiecolt-Glaser JK. Attachment anxiety is linked to alterations in cortisol production and cellular immunity. Psychol Sci 2013;24(3):272-279.

7. Wuehlhans B, Horbach T, de ZM. Psychiatric disorders in bariatric surgery candidates: a review of the literature and results of a German prebariatric surgery sample. Gen Hosp Psychiatry 2009;31(5):414-421.

8. Keating CL, Moodie ML, Bulfone L, Swinburn BA, Stevenson CE, Peeters A. Healthcare utilization and costs in severely obese subjects before bariatric surgery. Obesity (Silver Spring) 2012;20(12):2412-2419.

9. Sarwer DB, Cohn NI, Gibbons LM et al. Psychiatric diagnoses and psychiatric treatment among bariatric surgery candidates. Obes Surg 2004;14(9):1148-1156.

10. Bowlby J. Attachment and Loss: Vol 1. Attachment. New York: Basic Books;1969.

11. Bowlby J. Attachment and Loss, Vol 2. Separation, Anxiety and Anger. New York: Basic Books;1973.

12. Bowlby J. Attachment and Loss, Vol 3. Loss, Sadness and Depression. New York: Basic Books;1980.

13. Levy KN, Ellison WD, Scott LN, Bernecker SL. Attachment style. J Clin Psychol 2011;67(2):193-203.

14. Bowlby J. A secure base: clinical applications of attachment theory. Londen: Routledge; 1988.

15. Mikulincer M, Shaver PR. Adult attachment and affect regulation. In: Cassidy J, Shaver PR, editors. Handbook of attachment: Theory, research and clinical implications. New York: Guilford Press; 2008:503-531.


17. Torres SJ, Nowson CA. Relationship between stress, eating behavior, and obesity. Nutrition 2007;23(11-12):887-894.




21. Mikulincer M, Shaver PR. An attachment perspective on psychopathology. World Psychiatry 2012;11(1):11-15.


23. Wiltink J, Weber MM, Beutel ME. Mental co-morbidity, health care utilization and illness behaviour in overweight and obese subjects--results from a representative German community survey. Psychother Psychosom Med Psychol 2007;57(11):428-434.

R1R2R3R4R5R6R7R8R9


86 | Chapter 5

24. Friedman KE, Applegate KL, Grant J. Who is adherent with preoperative psychological treatment recommendations among weight loss surgery candidates? Surg Obes Relat Dis 2007;3(3):376-382.

25. Caspers KM, Yucuis R, Troutman B, Spinks R. Attachment as an organizer of behavior: implications for substance abuse problems and willingness to seek treatment. Subst Abuse Treat Prev Policy 2006;1:32.

26. Ciechanowski P, Sullivan M, Jensen M, Romano J, Summers H. The relationship of attachment style to depression, catastrophizing and health care utilization in patients with chronic pain. Pain 2003;104(3):627-637.


28. Feeney JA, Ryan SM. Attachment style and affect regulation: relationships with health behavior and family experiences of illness in a student sample. Health Psychol 1994;13(4):334-345.



31. Brennan KA, Clark CL, Shaver PR. Self-report measurement of adult attachment: An integrative overview. In: Simpson JA, Rholes WS (eds). Attachment theory and close relationships. New York: Guilford Press 1998:46-76.

32. Hooper LM, Tomek S, Newman CR. Using attachment theory in medical settings: implications for primary care physicians. J Ment Health 2012;21(1):23-37.


34. Thompson D, Ciechanowski PS. Attaching a new understanding to the patient-physician relationship in family practice. J Am Board Fam Pract 2003;16(3):219-226.

35. Adler HM. The sociophysiology of caring in the doctor-patient relationship. J Gen Intern Med 2002;17(11):874-881.

PART IIPostoperative: attachment representations

and effect on family members

6Attachment anxiety predicts poor adherence

to dietary recommendations:

an indirect effect on weight change one year

after gastric bypass surgery

Floor Aarts, Rinie Geenen, Victor E.A. Gerdes, Arnold van de Laar,

Dees P.M. Brandjes, Chris Hinnen


R1R2R3R4R5R6R7R8R9


90 | Chapter 6

Abstract

Introduction: Weight-loss after gastric bypass surgery depends on the adoption of healthy dietary

recommendations, which may be influenced by psychological problems and patients’ attachment

representations (habitual states of mind with respect to interpersonal relations). The present study

examines 1) the association of psychological problems and attachment representations with

dietary adherence, 2) the association between dietary adherence and weight-loss, and 3) dietary

adherence as mediator of the relation of psychological problems and attachment representations

with weight reduction after gastric bypass surgery.

Materials and Methods: This longitudinal study included 105 patients who had a laparoscopic

Roux-en-Y gastric bypass operation. Current and past psychological problems and attachment

representations were assessed before surgery. Dietary adherence was assessed 6 and 12 months

after surgery. Patients’ weight and height were collected from medical records. Multiple linear and

logistic regression analyses and mediation analyses using bootstrapping resampling procedures

were conducted.

Results: Of all examined predictor variables, attachment anxiety, i.e. fear of social rejection and

abandonment, was most strongly associated with low dietary adherence at both 6 months (p

= .009) and 12 months (p = .006) post-surgery. Dietary adherence 6 months post-surgery was

associated with weight-loss 1 year after the operation (p = .003). Dietary adherence at 6 months

(B = .51; 95% CI = .19 to 1.04) mediated the association between preoperative attachment

anxiety and postoperative weight-loss.

Conclusion: The results suggest that more anxiously attached patients are less adherent to

dietary recommendations 6 months after gastric bypass surgery, influencing weight-loss in a

negative way during the first year after surgery.


Attachment anxiety, adherence and weight loss | 91

6

Introduction

Bariatric surgery is the weight loss treatment of choice for patients with morbid obesity.1 The

majority of the patients lose 25-35% of their initial body weight within one year after gastric

bypass surgery.2 However, there is a small but considerable proportion of patients who are unable

to benefit optimally from a gastric bypass operation in terms of weight loss.3, 4

The amount of weight loss after gastric bypass surgery will depend to a large extent on the

degree to which the patient succeeds in adopting healthy dietary recommendations.4 Patients

who underwent gastric bypass surgery typically receive a number of diet recommendations such

as to limit snacking and drinking soda.3 Patients who fail to adhere to these recommendations

will lose less weight after a gastric bypass surgery and may regain more weight on the long-term.

For example, preoperative and postoperative binge eating or grazing have been shown to be

associated with poorer weight loss one year following gastric bypass surgery.5

Therefore, identifying factors that could influence adherence to dietary recommendations after

a gastric bypass operation is important. It may help to optimize the results of the operation. In

accordance, a standard component of the clinical evaluation of candidates applying for bariatric

surgery is a pre-surgical psychological assessment to identify possible indicators of suboptimal

adherence and outcomes.6-9 A history of psychological problems and current psychological

problems (e.g., symptoms of anxiety and depression) are among the preoperatively assessed

variables.6, 10 Psychological problems have been found to be associated with a less positive

evaluation of the eligibility of bariatric surgery candidates by psychologists11 and with less weight

loss after the initial year of the gastric bypass operation.12-14 These relationships may be explained

by the level of adherence to dietary recommendations, as those who were less adherent were

found to have more psychological problems than those who did follow dietary recommendations

more stringently.15

In addition to past and current psychological problems, different individual characteristics

such as one’s attachment representations may help explain differences in adherence to dietary

recommendations. According to attachment theory, early childhood experiences that centre around

the interaction with primary caregivers result in enduring expectations about the availability and

responsiveness of others.16 These attachment representations are conceptualized in adulthood as

mental states concerning anxiety about rejection and abandonment, and avoidance of intimacy

and interdependence.12-14 Attachments representations not only impact intimate relationships but

R1R2R3R4R5R6R7R8R9


92 | Chapter 6

also the relationship with more distant social sources such as one’s physician.17 Both attachment

anxiety and attachment avoidance have been found to be related to poorer adherence to medical

regiments in chronically ill patients.18, 19

More anxiously attached patients have been consistently shown to be more prone to distress

when confronted with stressors.20 In stressful situations people with high levels of attachment

anxiety may view themselves unable to deal with the stressors and may rely on smoking, alcohol

and high caloric food to regulate their emotions.21-23 In accordance, attachment anxiety has been

found to be associated with obesity in both children and adults.23, 24 Due to their high levels of

distress and their tendency to rely on external and behavioral modulators of affect such as high

caloric food, more anxiously attached patients can be expected to find it more difficult to adhere

to dietary recommendations after bariatric surgery.

More avoidantly attached patients, on the other hand, tend to dismiss symptoms of distress and

vulnerability.25 They stress the importance of independence and self-reliance, are reluctant to seek

support and feel uncomfortable trusting others, including health care providers.26, 27 Due to their

high level of self-reliance and low collaboration with health care providers, it can be expected that

they will be less adherent to dietary recommendations after bariatric surgery as well.

The present study has three objectives. First, we aim to get insight into determinants of individual

differences in adherence to dietary recommendations after a gastric bypass operation. We expect

current and past psychological problems as well as attachment anxiety and attachment avoidance

to be associated with poor adherence to dietary recommendations. Second, we aim to examine

the association between dietary adherence and weight loss in the first year after surgery. We

expect that poorer dietary adherence is associated with less weight reduction. Finally, we aim to

examine the mediating role of adherence to dietary recommendations between on the one hand,

current and past psychological problems, attachment anxiety and attachment avoidance, and on

the other hand, weight reduction one year after gastric bypass surgery.


Study sample

Included in analyses were 105 patients with morbid obesity between 18-60 years of age who

applied for a laparoscopic Roux-en-Y gastric bypass operation in Slotervaart hospital between

February and August 2012. The inclusion flow chart is presented in Figure 1. Of the 310 patients



6

who applied for bariatric screening, 190 were operated between April and December 2012 and

131 of these patients agreed to participate in this study. Eventually, 105 patients completed the

study and 26 were lost to follow-up.

Patients applying for bariatric screening

(n=310)

Not operated during study inclusion period

(n=68)

Not eligible for operation(n=44)

Operated(n=190)

Decided not to have surgery (n=8)

Agreed to participate in the study

(n=131)

Declined participation in the study(n=59)

Completed the study(n=105)

Dropped-out(n=26)

Figure 1. Flowchart

Procedures


assessments by a dietician, internist, surgeon and a psychologist including self-report

questionnaires, semi-structured interviews, and assessments of weight and height (BMI, Body

Mass Index), preoperative diet and exercise habits, co-morbidity and sociodemographics.

For this study, questionnaires to assess patients’ attachment representations and previous and

current psychological problems were added to the standard set of preoperative measures. At 6

and 12 months postoperatively, adherence to dietary recommendations and BMI were assessed.

The study was conducted according to the guidelines of the Declaration of Helsinki and approved

by the Medical Ethical Committee of the Slotervaart Hospital. Informed consent was obtained

from all participants.

R1R2R3R4R5R6R7R8R9


94 | Chapter 6

Instruments

Patients’ height and weight (without shoes) were measured in the hospital. Weight was measured

at the approximately same time of the day, with the same pair of scales and rounded off to the

nearest 0.1 kg. BMI was calculated by weight in kilograms divided by the square of the height in

meters (kg/m2).

Adherence with dietary recommendations was assessed using a single item with three possible

responses: a) “I generally followed the dietary recommendations”, b) “I almost followed the

dietary recommendations” or c) “I did not follow dietary recommendations”.

Past psychological problems was assessed with a single item: “Have you ever been in contact with

a social worker, psychologist or psychiatrist for professional help?” Response alternatives were

“yes” and “no”.

Current psychological problems were measured with the Hospital Anxiety and Depression Scale

(HADS), which consists of 14 items divided into two subscales. Seven items relate to anxiety (e.g.,

“I feel tense or wound up.”) and seven items relate to depression (e.g., “I have lost interest in

my appearance.”). Each item has four descriptive response options to be scored on a scale of 0

to 3, with a value of 0 corresponding to “not exhibiting the symptom at all,” and a value of 3

corresponding to “exhibiting the symptom to a high degree.” Scores for each of the two sub-

scales are derived by summation of its seven items. If one or more of its items were missing,

the subscale was disregarded. The lowest possible score for each subscale is 0 and the highest

possible score for each subscale is 21. The developers have suggested that aggregate sub-scale

scores of 0–7 represent non-cases, while scores of > 8 on the subscale indicate that a current

disorder warranting clinical attention may be present.28 High internal consistency was found for

both subscales of the HADS in this study; HADS-anxiety Cronbach’s alpha = .85 and HADS-

depression Cronbach’s alpha = .79.

Adult attachment representations were assessed using the Experiences in Close Relationships-

Revised Scale (ECR-R). The ECR-R is a 36-item self-report measure of adult attachment, which

requires participants to reflect on their typical ways of relating in close relationships. Reviews

of self-report measures of adult attachment suggest that the ECR-R has the best psychometric

properties of the available measures.29 The ECR consists of two continues subscales, attachment

anxiety (e.g., “I’m afraid that I will lose my partner’s love”) and attachment avoidance (e.g., “I

prefer not to show a partner how I feel deep down”). Both dimensions are assessed with 18

items. Answers are on a 5-point scale ranging from ‘strongly disagree’ (1) to ‘strongly agree’ (5).



6

In the present study, Cronbach’s alpha for subscale attachment anxiety was 0.88 and for subscale

attachment avoidance 0.90.

Medical variables (e.g., hypertension, diabetes) were collected from patients’ medical records.

Demographic variables (e.g., age, gender, marital status) at the first assessment were self-reported

by the patients.


Continuous variables are presented as means (M) and standard deviations (SD) and categorical

data as frequencies and percentages. Assumptions of normality were checked for the dependent

variables. Missing items in psychometric rating scales were substituted by the individual

respondent’s mean score on the respective scales, when missing items did not constitute more

than half of the answered items.30

Independent samples t-tests, Pearson correlations and Chi-square tests were used to determine

which covariates had to be controlled for. The variables age, gender, marital status (married: yes/

no), education level (bachelors’ degree or higher: yes/no), comorbidities, type of operation (i.e.

gastric bypass or redo), were correlated with BMI at baseline and dietary adherence at 6 and 12

months after surgery. A variable was considered a potential covariate in case of a correlation

significant at p < .10.

Preliminary analysis showed that no patients reported not to have followed the dietary advices

at all. Therefore, logistic regression analyses were performed with the outcomes: “I generally

followed the dietary recommendations” and “I almost followed the dietary recommendations”.

In order to test the hypotheses concerning the association between current and past

psychological problems and attachment anxiety and attachment avoidance, on the one hand,

and adherence to dietary recommendations at 6 and at 12 months after bariatric surgery, on

the other hand, univariate logistic regression analyses were used. In order to determine which

variable predicted dietary adherence the most forward logistic repression analyses were used.

To test the hypothesis concerning the association between, on the one hand, dietary

recommendations at 6 and 12 months and, on the other hand, BMI at 12 months multiple

linear regression analyses were used, adjusted for baseline BMI. The percentage total weight loss

(%TWL) was calculated as (100x (baseline BMI - BMI at 12 months) / baseline BMI).

R1R2R3R4R5R6R7R8R9


96 | Chapter 6

Finally, to examine dietary adherence as a potential mediator of the association between the

strongest predictor(s) of dietary adherence and weight loss (BMI at 12 months following surgery

adjusted for baseline BMI) (Figure 2), a bootstrapping resampling method with bias-corrected

confidence estimates was done.31, 32 In the present study, the 95% bootstrap confidence interval

of the indirect effects was calculated based on 5000 bootstrap resamples 33, 34. If the confidence

interval does not overlap zero, the effect is said to be statistically significant. In all analyses,

significance levels were set at p < .05. Data were analyzed using SPSS 19.

Unmediated effectAttachmentanxiety

Adherence to dietary recommendations

Weight loss one year following gastric bypass

surgery

Figure 2. Dietary adherence at 6 months as mediator of the effect of attachment anxiety on weight loss one year following gastric bypass surgery. Pathway values are reported in the results section.

Results

Characteristics

Independent samples t-test showed that patients who dropped-out (n = 26) –as compared

to those who completed the study– scored higher on attachment anxiety [t(124) = -11.979,

(p < .001), 95% CI: -2.34; - 1.66] and attachment avoidance [t(126) = -3.139, (p = .002), 95% CI:

-.83; - 1.19]. Drop-outs did not differ significantly on any other variables in this study.

Our study sample included 105 patients, predominantly female (81%), with a mean age of 45 ±

9.1 years. Most patients lived with a partner (84%) and were employed (76%); about a quarter

of the sample (27%) had received higher education (bachelors’ degree or higher). Before surgery,

mean weight was 123.7 ± 19.7 kg and mean BMI was 42.7 (6.1) kg/m2. Most of the patients

underwent primary gastric bypass surgery (86%) and 14% of the patients had a revision gastric

bypass surgery with removal of gastric banding during the same operation. All patients were

operated laparoscopically.



6

Pearson correlations showed that the p-value of the correlation between BMI at baseline and a

younger age was < .10: r = -.18, p = .06. BMI was not significantly related to gender, diabetes,

marital status, level of education and type of operation; these variables were excluded from

further analyses.

For dietary adherence at 6 and 12 months after surgery, Chi-square tests showed that patients

with diabetes (χ2=4.1, p = .04) were less adherent to dietary recommendations. No other variables

were found to be related to dietary adherence. At 6 months after surgery 70% and at 12 months

58% of the patients reported to be adherent to their dietary recommendations. Table 1 shows

the correlations between the five predictors in this study. Low to moderate correlations were

observed between attachment representations and psychological problems.

Table 1. Pearson correlations between of the five predictors in the study

1 2 3 41. Past psychological problems2. Current anxiety symptoms .133. Current depressive symptoms .10 .64***4. Attachment anxiety .44*** .24* .34***5. Attachment avoidance .27** .37*** .37** .49***

*p<.05, **p<.01, ***p<.001

Psychological problems, attachment and dietary adherence

Univariate logistic regression analyses, controlled for diabetes, showed that past psychological

problems (OR = 5.04, p = .007), current anxiety symptoms (OR = 1.16, p = .03), current depressive

symptoms (OR = 1.18, p = .02), attachment anxiety (OR = 4.76, p < .001) –but not attachment

avoidance (OR = 1.63, p = .13)– were associated with dietary adherence at 6 months (path A,

Figure 2). Similarly, past psychological problems (OR = 3.29, p = .01) current anxiety symptoms

(OR = 1.17, p = .008), current depressive symptoms (OR = 1.14, p = .04), attachment anxiety

(OR = 2.38, p = .009) –but not attachment avoidance (OR = 1.18, p = .56)– were associated

with dietary adherence at 12 months. Forward logistic regression analyses showed that of these

variables, attachment anxiety was the strongest predictor of dietary adherence at both 6 months

(Nagelkerke R2 = .30, OR = 4.92, p = .009) and 12 months (Nagelkerke R2 = .13, OR = 2.61,

p=.006).

R1R2R3R4R5R6R7R8R9


98 | Chapter 6

Diet adherence and weight loss

Multiple linear regression analyses, controlled for age and for baseline BMI, showed that poorer

adherence to dietary recommendations at 6 months (β = .23, p = .003) (path B, Figure 2), was

associated with less weight loss 12 month after surgery explaining 58% of the variance. Patients

who were adherent at 6 months showed a %TWL of 30.5 ± 6.5, while the patients who were less

adherent to dietary recommendations showed a %TWL of 24.7 ± 6.5 in the first year following

surgery. In contrast no association was found between adherence to dietary recommendations at

12 months (β = .09, p = .243) and weight loss at 12 months.

Mediation analysis attachment anxiety, dietary adherence and weight loss

Attachment anxiety, the strongest predictor of dietary adherence, was examined in mediation

analysis using the bootstrapping resampling method. Analyses adjusted for age and baseline

BMI showed that dietary adherence at 6 months mediated the association between attachment

anxiety and BMI (B = .51; 95% CI = .19 to 1.02). Since there was no relationship between dietary

adherence at 12 months and weight loss at 12 months, no mediating effect was found for dietary

adherence at 12 months.

Conclusion

Attachment anxiety was observed to be a main predictor of poor dietary adherence. In addition,

poor dietary adherence in the first 6 months after surgery showed an association with the amount

of weight loss 1 year after the operation. Overall we found support for the indirect effect of

attachment anxiety on weight loss in the year following a gastric bypass operation, mediated by

difficulty with adherence to dietary recommendations at 6 months.

Of the predictors that were examined in this study (past psychological problems, current

psychological problems, attachment anxiety, and attachment avoidance), attachment anxiety

was found to be the strongest predictor of poor adherence to dietary recommendations. A

possible explanation for this association can be found in the tendency of more anxiously attached

individuals to experience high levels of distress and to have problems with down-regulating

their negative affect.35 Eating may function as an emotion regulatory mechanism which may be

especially difficult to give up for more anxiously attached patients.36 This notion is in accordance

with previous studies showing an association between inadequate emotion regulation strategies

and eating disorders.37-40 A physiological mechanism may explain eating in response to stress and

distress. Animal studies have shown that the consumption of high caloric foods acts to calm the

stress-perceiving areas of the brain41-43 possibly by the release of oxytocin from the hypothalamus



6

which has an anxiolytic effect.44 Thus, high caloric foods may help especially more anxiously

attached individuals to down-regulate their high levels of distress, thereby obstructing adherence

to dietary recommendations after gastric bypass surgery.

In contrast to expectations, no relationship was found between attachment avoidance and

adherence to dietary recommendations. This may be explained by a differential way of coping

between more anxious and more avoidantly attached individuals. While more anxiously attached

patients respond to stressors in a hypervigilant manner, more avoidantly attached patients tend to

respond by distancing, avoiding and repressing negative emotions.45-48 Perhaps for some of them

negative emotions occur less often, or their self-reliant way of coping sometimes even helps them

to follow dietary recommendations. Previous studies showing low adherence to medical regiment

by more avoidantly attached patients took the quality of the patient-physician relationship

into account.19 That is, more avoidantly attached patients may show to be less adherent only

when the relationship with the healthcare professional is less satisfying. Thus, although in some

circumstances attachment avoidance may obstruct healthy eating behavior, our results suggest

that attachment avoidance per se is not a vulnerability factor to low dietary adherence after

gastric bypass surgery.

Although there is no doubt that gastric bypass surgery is an effective treatment for the majority of

patients with morbid obesity, our results indicate that the amount of weight loss after surgery will

to some extent depend on the degree to which the patient succeeds in adopting healthy dietary

recommendations in the first 6 months after surgery, which is in agreement with other reports.4,

5 However, no relationship was found between adherence to dietary recommendations at 12

months and postoperative weight reduction. This finding suggests that adherence to dietary

recommendations is more predictive of future weight loss than of current weight loss. A future

study with a design using more repeated measurements during a longer time interval is needed

to examine this hypothesis.

A methodological asset of this study is the prospective design, but some weaker points should also

be indicated. First, our data of attachment and dietary adherence were obtained by self-report.

Second, dietary adherence was measured with single items which may have lead to problems in

measurement precision.49 Therefore, future studies should use more objective measures for dietary

adherence. Third, dietary recommendations were not controlled in this study, that is, patients

who showed more and less weight reduction postoperatively may have received more and less

dietary advice. Fourth, results do not necessarily generalizable to bariatric surgery patients as a

whole or to patients subjected to other operations or with a still higher weight before surgery.

R1R2R3R4R5R6R7R8R9


100 | Chapter 6

Fifth, selection bias is suggested as drop-out rates in this study were higher for more anxiously

and more avoidantly attached patients. Sixth, patients with a redo surgery were included as they

failed in terms of successful weight loss after laparoscopic adjustable gastric banding. However,

we did not find an association between type of surgery and weight loss or adherence to dietary

recommendations in this study.

Finally, while a dimensional and self-report measure of attachment has theoretical and statistical

advantages, the use of a categorical measure would have had clinical advantages 50. In future

studies a categorical measure based on a more thorough investigation such as the adult attachment

interview51 may be considered.52 A categorical measure makes it possible to further determine the

relevance of considering attachment when designing interventions aimed at optimizing the result

of a gastric bypass surgery.

In summary, in the year following a gastric bypass operation more anxiously attached patients are

indicated to have greater difficulty to adhere to dietary recommendations and consequently are

at greater risk of not being able to profit fully from a gastric bypass operation.



6

Reference List



3. Elkins G, Whitfield P, Marcus J, Symmonds R, Rodriguez J, Cook T. Noncompliance with behavioral recommendations following bariatric surgery. Obes Surg 2005;15(4):546-551.

4. Karmali S, Brar B, Shi X, Sharma AM, de Gara C, Birch DW. Weight recidivism post-bariatric surgery: a systematic review. Obes Surg 2013;23(11):1922-1933.

5. Colles SL, Dixon JB, O’Brien PE. Grazing and loss of control related to eating: two high-risk factors following bariatric surgery. Obesity (Silver Spring) 2008;16(3):615-622.

6. Ritz SJ. The bariatric psychological evaluation: a heuristic for determining the suitability of the morbidly obese patient for weight loss surgery. Bariatr Surg Pract Patient Care 2006; 1: 97-105.




10. Fabricatore AN, Crerand CE, Wadden TA, Sarwer DB, Krasucki JL. How do mental health professionals evaluate candidates for bariatric surgery? Survey results. Obes Surg 2006;16(5):567-573.

11. Aarts F, Hinnen C, Gerdes VEA, Acherman Y, Brandjes DPM. Psychologists’ evaluation of bariatric surgery candidates influenced by patients’ attachment representations and symptoms of depression and anxiety. J Clin Psychol Med Set 2014; 21(1):116-123.


13. Brennan KA, Clark CL, Shaver PR. Self-report measurement of adult attachment: An integrative overview. Attachment theory and close relationships. In: J. A. Simpson & W. S. Rholes, eds. Attachment theory and close relationships: 46-76. New York: Guilford Press 1998.

14. Crowell JA, Fraley RC, Shaver PR. Measures of individual differences in adolescent and adult attachment. In: J. Cassidy & P. R. Shaver, eds. Handbook of attachment: Theory, research, and clinical applications: 434-465. New York: Guilford Press 1999.

15. Gorin AA, Raftopoulos I. Effect of mood and eating disorders on the short-term outcome of laparoscopic Roux-en-Y gastric bypass. Obes Surg 2009;19(12):1685-1690.

16. Bowlby J. Attachment and Loss: retrospect and prospect. Am J Orthopsych 1969; 52(4), 664-678.

17. Consedine NS, Magai C. Attachment and emotion experience in later life: the view from emotions theory. Attach Hum Dev 2003;5(2):165-187.

18. Bennett JK, Fuertes JN, Keitel M, Phillips R. The role of patient attachment and working alliance on patient adherence, satisfaction, and health-related quality of life in lupus treatment. Patient Educ Couns 2011;85(1):53-59.

19. Ciechanowski PS, Katon WJ, Russo JE, Walker EA. The patient-provider relationship: attachment theory and adherence to treatment in diabetes. Am J Psychiatry 2001;158(1):29-35.


21. Maunder RG, Hunter JJ. Attachment relationships as determinants of physical health. J Am Acad Psychoanal Dyn Psychiatry 2008;36(1):11-32.







28. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983;67(6):361-370.


30. Schafer JL, Graham JW. Missing data: our view of the state of the art. Psychol Methods 2002;7(2):147-177.

31. Preacher KJ, Hayes AF. SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behav Res Methods Instrum Comput 2004;36(4):717-731.

32. MacKinnon DP, Lockwood CM, Williams J. Confidence Limits for the Indirect Effect: Distribution of the Product and Resampling Methods. Multivariate Behav Res 2004;39(1):99.

33. Preacher KJ, Hayes AF. Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav Res Methods 2008;40(3):879-891.

34. Hayes AF. Beyond Baron and Kenny: Statistical Mediation Analysis in the New Millennium. Communication Monographs 2009;76(4):408-420.

35. Maunder RG, Lancee WJ, Nolan RP, Hunter JJ, Tannenbaum DW. The relationship of attachment insecurity to subjective stress and autonomic function during standardized acute stress in healthy adults. J Psychosom Res 2006;60(3):283-290.

36. Raynes E, Auerbach C, Botyanski NC. Level of object representation and psychic structure deficit in obese persons. Psychol Rep 1989;64(1):291-294.

37. Evers C, Marijn Stok F, de Ridder DT. Feeding your feelings: emotion regulation strategies and emotional eating. Pers Soc Psychol Bull 2010;36(6):792-804.

38. Spoor ST, Bekker MH, Van Strien T, van Heck GL. Relations between negative affect, coping, and emotional eating. Appetite 2007;48(3):368-376.

39. Vandewalle J, Moens E, Braet C. Comprehending emotional eating in obese youngsters: the role of parental rejection and emotion regulation. Int J Obes (Lond) 2013.

40. Zijlstra H, van MH, Devaere L, Larsen JK, van RB, Geenen R. Emotion processing and regulation in women with morbid obesity who apply for bariatric surgery. Psychol Health 2012;27(12):1375-1387.

41. Dallman MF, Pecoraro N, Akana SF et al. Chronic stress and obesity: a new view of “comfort food”. Proc Natl Acad Sci U S A 2003;100(20):11696-11701.

42. Pecoraro N, Reyes F, Gomez F, Bhargava A, Dallman MF. Chronic stress promotes palatable feeding, which reduces signs of stress: feedforward and feedback effects of chronic stress. Endocrinology 2004;145(8):3754-3762.

43. Peters A, Pellerin L, Dallman MF et al. Causes of obesity: looking beyond the hypothalamus. Prog Neurobiol 2007;81(2):61-88.

44. Onaka T, Takayanagi Y, Yoshida M. Roles of oxytocin neurones in the control of stress, energy metabolism, and social behaviour. J Neuroendocrinol 2012;24(4):587-598.





49. Fraley RC, Waller NG. Attachment theory and close relationships. In: Simpson JA, Rholes WS, editors. Adult attachment patterns: A test of the typological model. New York: Guilford Press; 1998:77-114.


51. Heinicke CM, Levine MS. Clinical applications of the adult attachment interview. New York: 2008.

52. van IJzendoorn MH. Adult attachment representations, parental responsiveness, and infant attachment: a meta-analysis on the predictive validity of the Adult Attachment Interview. Psychol Bull 1995;117(3):387-403.

7The significance of attachment representations

for quality of life one year following gastric

bypass surgery: a longitudinal analysis

Floor Aarts, Rinie Geenen, Victor E.A. Gerdes, Dees P.M. Brandjes, Chris Hinnen


R1R2R3R4R5R6R7R8R9


106 | Chapter 7

Abstract

Objective: Quality of life after bariatric surgery may among other things depend on patients’

attachment representations such as anxiety about rejection and abandonment (attachment

anxiety) and avoidance of intimacy and interdependence (attachment avoidance). The aim of

this study was to examine whether attachment representations, independent of body mass index

(BMI), are associated with the level and course of physical functioning and mental well-being

after gastric bypass surgery.

Methods: This longitudinal study included 105 patients applying for a Roux-en-Y gastric bypass

operation. Patients’ attachment representations (ECR-R) were measured before surgery and

quality of life dimensions (physical functioning and mental well-being, SF-36) were measured

before surgery and 1, 3, 6 and 12 months following surgery. Linear mixed effect models were

used in analyses.

Results: Physical functioning (p < .001) improved and mental well-being worsened (p = .002)

in the postoperative interval. Both attachment anxiety (p = .005) and attachment avoidance

(p < .001) were associated with a lower level of mental well-being, but not with the postoperative

course of quality of life.

Conclusion: Our study suggests that bariatric surgery leads to improvement in physical

functioning but not mental well-being. Results highlight that patients with insecure attachment

representations should be protected against unrealistic expectations regarding improvement of

mental well-being after surgery.


Significance of attachment for quality of life following gastric bypass surgery | 107

7

Introduction

Gastric bypass surgery is the weight loss treatment of choice for obese individuals with a body

mass index (BMI) of more than 35 kg/m2 in the presence of weight-related comorbidities.1 Several

studies indicated improvement in quality of life following gastric bypass surgery.2-4 In the present

study we propose that improvement in quality of life after bariatric surgery may in addition to the

amount of weight loss, depend on individual difference characteristics5 such as one’s attachment

representations.

Attachment theory describes the significance of attachment representations, i.e. the enduring

beliefs and expectations about the availability and responsiveness about the self (e.g., as

worthy of love and care) and about others (e.g., as trustworthy and caring).6 In adulthood these

beliefs and expectations are conceptualized as a set of mental states concerning anxiety about

rejection and abandonment, and avoidance of intimacy and interdependence.6-8 Patients high on

attachment anxiety have a sense of vulnerability and hypervigilance for threats, resulting in high

levels of perceived stress and distress.9 Despite their strong desire for closeness and reassurance,

research shows that support is hardly effective in reducing distress in these people.10 In contrast,

patients high on attachment avoidance feel uncomfortable in close relationships, perceive

others as unavailable and unable to provide adequate support when needed, and therefore

value independency and self-reliance.11, 12 The importance of attachment theory is supported

by the finding that both attachment anxiety and attachment avoidance have been found to

be associated with impaired quality of life in healthy people13, 14 and chronically ill patients.15 In

a cross-sectional study in morbidly obese bariatric surgery candidates an association between

attachment avoidance and poor mental health quality of life was observed.16 It is as yet unknown

whether attachment representations might impact the postoperative level and course of quality

of life. Quality of life takes into account patients’ physical functioning and mental well-being,

which are the main elements of health.

The aim of the present longitudinal study was to investigate whether attachment anxiety and

attachment avoidance, independent of the course of body mass index (BMI), are associated with

the level and course of physical functioning and mental well-being in the first year after a gastric

bypass operation.

R1R2R3R4R5R6R7R8R9


108 | Chapter 7

Methods

Study sample

Included in analyses were 105 patients with morbid obesity between 18-60 years of age who

had undergone Roux-en-Y gastric bypass operation in Slotervaart hospital, Amsterdam between

February and August 2012. The flow chart is presented in Figure 1. Of the 310 patients who

applied for bariatric screening, 190 were operated between April and December 2012 and 131

of these patients agreed to participate in this study. Eventually, 105 patients completed the study

and 26 were lost to follow-up.

Patients applying for bariatric screening

(n=310)

Not operated during study inclusion period

(n=68)

Not eligible for operation(n=44)

Operated(n=190)

Decided not to have surgery (n=8)


(n=131)

Declined participation in the study(n=59)

Completed the study(n=105)

Dropped-out(n=26)

Figure 1. Flowchart

Procedures


assessment by a dietician, internist, surgeon and a psychologist including self-report questionnaires,

semi-structured interviews, and assessments of body weight and height (BMI, body mass index),

preoperative diet and exercise habits, co-morbidity and sociodemographics.



7

For this study questionnaires were added to the standard set of preoperative measures assessing

patients’ attachment representations preoperatively and quality of life both pre- as postoperatively.


by the Medical Ethical Committee of the Slotervaart Hospital. Informed consent was obtained

from all participants.

Instruments

Quality of life was measured using the SF-36.17 The physical component summary (as a reflection

of physical functioning) and mental component summary (as a reflection of mental well-being)

of the SF-36 were used as outcome variables in the present study.18 The validity of the Dutch

version of SF-36 has been tested and has good construct validity, high internal consistency, and

high test-retest stability; norm data were obtained from the Dutch Health Survey in community

and chronic disease populations.19

Adult attachment was assessed using the Experiences in Close Relationships-Revised Scale (ECR-R).

The ECR-R is a 36-item self-report measure of adult attachment, which requires participants to

reflect on their typical ways of relating in close/romantic relationships. Reviews of self-report

measures of adult attachment suggest that the ECR-R has the best psychometric properties of

the available measures.20 The ECR consists of two continuous subscales, attachment anxiety

(e.g., “I’m afraid that I will lose my partner’s love”) and attachment avoidance (e.g., “I prefer

not to show a partner how I feel deep down”). Both dimensions are assessed with 18 items.

Answers are on a 5-point scale ranging from ‘strongly disagree’ (1) to ‘strongly agree’ (5). In the

present study, Cronbach’s alpha for subscale attachment anxiety was 0.88 and for the subscale

attachment avoidance 0.90.

Body weight and height as well as medical variables (e.g., hypertension, diabetes) were acquired

from patients’ medical records. Demographic variables (e.g., age, gender, marital status) at the

first assessment were self-reported by the patients.


Continuous variables are presented as means and standard deviations and categorical data

as frequencies and percentages. Assumptions of normality were checked for the dependent

variables. The score distributions of the quality of life dimensions were sufficiently normal to allow

parametric tests. Missing items in psychometric rating scales were substituted by the individual

respondent’s mean score on the respective scales, when missing items did not constitute more

than half of the answered items.21

R1R2R3R4R5R6R7R8R9


110 | Chapter 7

To examine which variables should be included as covariates in further analyses, independent

samples t-tests and Pearson correlations were used to test the associations between demographic

variables (i.e., age, gender, marital status, level of education), medical variables (i.e., hypertension,

diabetes, osteoarthritis, sleep apnoea, type of operation) on the one hand and the outcome

variables (i.e., physical functioning and mental well-being) at baseline on the other hand.

Linear mixed model analyses were used to examine whether the level and course of physical

functioning and mental well-being after the gastric bypass operation were predicted by

preoperative levels of attachment anxiety and attachment avoidance. Maximum likelihood

estimation was selected and a random intercept was added to all models to take account of

and maintain individual differences in baseline values. Three models were specified and tested

separately for both domains of quality of life.

In Model 1, linear mixed models examined the effect of time, that is, the change in outcome

measurements across baseline and the one, three, six and twelve month follow-ups after

surgery. Besides time, no other variables were added to this model.

Model 2 tested the main effect of attachment scores, that is, whether attachment anxiety and

attachment avoidance were associated with levels of physical functioning and mental well-being.

In this model, covariates that correlated with the dependent variable and the level of BMI over

time were included to adjust for these variables.

Since previous studies have shown that the course of BMI is associated with the course of quality

of life, we included the interaction term of BMI by time reflecting the course of BMI in model

3.22 In this way we adjusted for the course of BMI. Moreover, covariates that correlated with

the dependent variable were included in the model to adjust for these variables. Thus, model 3

tested whether attachment anxiety and attachment avoidance were associated with the course

of quality of life across the repeated measurements. This was done by predicting the quality

of life dimension from the interaction term of the attachment dimension by time. Significant

interactions were probed by plotting regression lines for individuals with low (-1 SD) and high (+1

SD) levels on the two interaction terms while filling out mean values for all the other variables.23

All analyses were performed with SPSS, version 19.0 (for Windows). The level of significance was

set at p < .05. All tests were two-tailed.



7

Results

Independent samples t-tests showed that patients who dropped-out (n = 26) and those who

completed the study differed on attachment anxiety t(124) = -11.80, (p < .001), 95% CI: -2.34,

- 1.66 and attachment avoidance t(126) = -3.14, (p = .002), 95% CI: -.83, - 1.19. That is, more

anxiously and more avoidantly attached patients dropped-out of the study. Drop-outs did not

differ significantly on any other variables in this study.

Our study sample included 105 patients, who were predominantly female (81%) and had a mean

age of 45 years ± 9.1 years. Most patients lived with a partner (84%) and about a quarter of the

sample (27%) had received high education (bachelors’ degree or higher). The majority of patients

were employed (76%). Before the operation, the mean weight was 123.7 ± 19.7 kg and mean

BMI was 42.8 ± 6.1 kg/m2. Most of the patients had gastric bypass surgery for the first time

(86%) and 14% of the patients had a redo surgery (gastric bypass and removal of gastric band).

Table 1. Pearson correlations between of the five main variables in the study

1 2 3 41. Attachment anxiety2. Attachment avoidance .49**3. Physical functioning -.13 -.074. Mental well-being -.42** -.42** -.055. BMI .05 .00 -.21* .06

*p<.05, **p<.01

Table 1 shows the correlations between the main study variables. Independent samples t-test

showed that being a woman t(99) = -3.27, (p = .001), 95% CI: -12.33, -3.02, not having diabetes

t(99) = 3.15, (p = .002), 95% CI: 2.29, 10.11, having osteoarthritis t(99) = -5.486, (p < .001),

95% CI: -13.18, -6.18 and not having a partner t(99) = -2.02, (p = .046), 95% CI: -8.84, -.07

were associated with poorer physical functioning. Furthermore, having a lower education level

t(98) = -2.39, (p = .02), 95% CI: -7.36, -.69 and not having a partner t(99) = -2.12, (p = .04), 95%

CI: -7.50, -.25 were associated with poorer mental well-being. Thus, in model 2 and 3, tests of

physical functioning included the covariates gender, diabetes, osteoarthritis, and marital status,

and tests of mental well-being included the covariates education level and marital status.

Table 2 shows the means and standard deviations of the attachment variables, quality of life

dimensions, and BMI at the repeated assessments.

R1R2R3R4R5R6R7R8R9


112 | Chapter 7

Table 2. Means and standard deviations of the attachment variables, quality of life dimensions, and BMI at

the repeated assessments.

Baseline

M (SD)

1 month AfterM (SD)

3 months AfterM (SD)

6 months AfterM (SD)

1 year AfterM (SD)

Attachment anxiety 1.9 (.7) -- -- -- -- -- -- -- --Attachment avoidance 2.1 (.76) -- -- -- -- -- -- -- --Physical functioning 37.6 (9.5) 46.1 (8.3) 51.6 (8.4) 53.2 (7.6) 54.6 (7.1)Mental well-being 51.9 (7.8) 48.3 (8.7) 50.8 (8.9) 50.9 (8.7) 49.7 (9.3)BMI 42.8 (6.1) 38.0 (5.6) 34.8 (5.1) 33.5 (5.4) 30.1 (4.7)

Figure 2 shows the mean scores of the quality of life dimensions over time. Model 1 of the

linear mixed model analyses showed that physical functioning improved across the five repeated

measurements (F = 128.6, p < .001), while mental well-being worsened between baseline and

1 month (p < .001) and between baseline and one year after surgery (F = 4.28, p = .01). The

postoperative levels of physical functioning and mental well-being after surgery were close to the

norm of 50 as seen in the general population.19

Physical functioningMental well-beingbefore 37,58 51,881 month 46,14 48,283 months 51,56 50,756 months 53,15 50,91-yr after 54,62 49,73

30

40

50

60

before 1 month 3 months 6 months 1-yr after

Qua

lity

of L

ife

Operation

Physicalfunctioning

Mental well-being

Figure 2. Mean scores of the quality of life dimensions, physical functioning and mental well-being before gastric bypass surgery, and 1, 3, 6 and 12 months following surgery



7

Model 2 examined whether baseline assessments of attachment anxiety and attachment avoi-

dance, controlled for covariates and the main effect of the level BMI over time, were associated

with the level of quality of life across repeated measurements. The level of BMI over time was

associated with the level of physical functioning (F = 14.43, p < .001) but not with the level

of mental well-being (F = .22, p = .639). Both attachment anxiety (F = 8.34, p = .005) and

attachment avoidance (F = 13.74, p < .001) were associated with the level of mental well-being

but neither attachment anxiety (F = .38, p = .537) nor attachment avoidance (F = .46, p =

.498) were associated with the level of physical functioning. Regarding the relation between

attachment representations and mental well-being, for each unit a patient scored higher on

attachment anxiety, the average level of mental well-being across repeated measurements was

2.5 units lower (CI: -4.27, .81). Moreover, for each unit a patient scored higher on attachment

avoidance, the average level of mental well-being across repeated measurements was 3.3 units

lower (CI: -5.05, -1.53).

Model 3 examined whether attachment representations were associated with the course of the

quality of life dimensions by testing the interaction between the attachment representations and

the repeated measurements at the quality of life dimensions. Analyses showed that attachment

representations were not associated with the course of physical functioning or mental well-being.

However, we did find that the postoperative course of physical functioning was predicted by the

interaction between BMI and time (F = 16.34, p < .001). The interaction is shown in Figure 3. As

compared to patients with a lower mean BMI (-1 SD in this sample), patients with a high mean

BMI (+1 SD in this sample) improved towards a similar level of physical functioning one year after

the operation despite their clearly lower level of physical functioning before the operation.

R1R2R3R4R5R6R7R8R9


114 | Chapter 7

low BMI high BMIbefore 46,02 38,561-yr after 55,35 56,42

30

40

50

60

before 1-yr after

Phys

ical

func

tioni

ng

Operation

low BMI

high BMI

Figure 3. Physical functioning as predicted from having a low vs. high average body mass index (BMI) and time of measurement (before vs. 1-yr after the operation)

Discussion

This study showed significant improvement in physical functioning but not in mental well-being

within the first year after a gastric bypass operation. Both attachment anxiety and attachment

avoidance were associated with a lower level of postoperative mental well-being. We did not find

an association between attachment anxiety and attachment avoidance and the postoperative

level of physical functioning. Also no association between attachment representations and the

postoperative course of physical functioning and mental well-being was found.

The finding that patients benefit more in terms of physical functioning than mental well-being is in

accordance with previous studies.22, 24 Mental well-being seems to be a rather constant factor that

is not easily influenced by weight reduction. As shown in a meta-analysis, physical functioning

deviates far more from normal in bariatric surgery candidates than mental well-being, and the

level of mental well-being is hardly dependent on body weight.25 In line with the results of this

meta-analysis, our study showed that a loss of body weight was associated with an improvement



7

of physical functioning. However, the present study extends these results by indicating that to

improve mental well-being a focus on processes related to attachment representations may be

more important than a reduction of body weight. That is, insecurely attached patients reported

lower levels of mental well-being before bariatric surgery and remained more dissatisfied after

bariatric surgery than more securely attached patients.

The prospective design is an asset of this study, but a weaker point is that the study might be

biased towards patients with relatively good attachment representations, because patients with

poorer attachment representations dropped out. Future studies should examine the significance

of attachment representations for quality of life on the longer-term, as adult attachment may be

helpful in the selection of patients for surgery and the guidance of patients after surgery.

In summary, we found significant improvements in physical functioning −but not mental well-

being− one year following gastric bypass surgery. While weight and weight loss are predictors of

physical functioning, attachment representations predict levels of mental well-being.

In order to improve mental well-being in insecurely attached patients with morbid obesity other

strategies than weight reduction should be considered. A recent review provides preliminary

evidence that psychotherapy may be effective in promoting well-being in insecurely attached

patients by reframing and reappraising past interpersonal experiences.26 Also, rather than

changing attachment representations which may be an intensive and time consuming task,

attachment tailored care by healthcare workers may also promote well-being and prevent

an unnecessary increase in distress and turmoil.27 All patients, especially those with insecure

attachment representations, should be protected against unrealistic expectations of improvement

in mental well-being after surgery.

R1R2R3R4R5R6R7R8R9


116 | Chapter 7

Reference List


2. Karlsson J, Taft C, Ryden A, Sjostrom L, Sullivan M. Ten-year trends in health-related quality of life after surgical and conventional treatment for severe obesity: the SOS intervention study. Int J Obes (Lond) 2007;31(8):1248-1261.

3. Suter M, Donadini A, Romy S, Demartines N, Giusti V. Laparoscopic Roux-en-Y gastric bypass: significant long-term weight loss, improvement of obesity-related comorbidities and quality of life. Ann Surg 2011;254(2):267-273.

4. Aftab H, Risstad H, Sovik TT et al. Five-year outcome after gastric bypass for morbid obesity in a Norwegian cohort. Surg Obes Relat Dis 2013.

5. Loving TJ, Smets EM. Romantic relationships and health. In: Simpson JA, Campbell L, editors. The Oxford handbook of close relationships. New York: Oxford Univeristy Press: 2013:617-637.

6. Bowlby J. Attachment and Loss: retrospect and prospect. Am J Orthopsych 1969; 52(4), 664-678.

7. Crowell JA, Fraley RC, Shaver PR. Measures of individual differences in adolescent and adult attachment. In: J. Cassidy & P. R. Shaver, eds. Handbook of attachment: Theory, research, and clinical applications: 434-465. New York: Guilford Press 1999.

8. Griffin D BK. Models of the self and other: Fundamental dimensions underlying measures of adult attachment. Journal of personality and social psychology. J Pers Soc Psychol 1994; 67(3): 430–445.



11. Florian V, Mikulincer M, Bucholtz I. Effects of adult attachment style on the perception and search for social support. J Psychol 1995;129(6):665-676.






17. VanderZee KI, Sanderman R, Heyink JW, de HH. Psychometric qualities of the RAND 36-Item Health Survey 1.0: a multidimensional measure of general health status. Int J Behav Med 1996;3(2):104-122.

18. Ware JE, Kosinski M, KSD. SF-36 physical and mental health summary scales: A user’s manual. Boston, MA: The Health Institute 1994.

19. Aaronson NK, Muller M, Cohen PD et al. Translation, validation, and norming of the Dutch language version of the SF-36 Health Survey in community and chronic disease populations. J Clin Epidemiol 1998;51(11):1055-1068.


21. Schafer JL, Graham JW. Missing data: our view of the state of the art. Psychol Methods 2002;7(2):147-177.



7

22. Karlsson J, Sjostrom L, Sullivan M. Swedish obese subjects (SOS)--an intervention study of obesity. Two-year follow-up of health-related quality of life (HRQL) and eating behavior after gastric surgery for severe obesity. Int J Obes Relat Metab Disord 1998;22(2):113-126.

23. Aiken LS, West SG. Multiple regression: Testing an interpreting interactions. Newbury Park: CA: Sage; 1991.

24. Julia C, Ciangura C, Capuron L et al. Quality of life after Roux-en-Y gastric bypass and changes in body mass index and obesity-related comorbidities. Diabetes Metab 2013;39(2):148-154.

25. van Nunen AM, Wouters EJ, Vingerhoets AJ, Hox JJ, Geenen R. The health-related quality of life of obese persons seeking or not seeking surgical or non-surgical treatment: a meta-analysis. Obes Surg 2007;17(10):1357-1366.

26. Taylor P, Rietzschel J, Danquah A, Berry K. Changes in attachment representations during psychological therapy. Psychother Res 2014.


8Gastric bypass may promote weight loss in

overweight partners the first year after surgery

Floor Aarts, Nalini N.E. Radhakishun, Mariska van Vliet, Rinie Geenen, Ines A. von Rosenstiel,

Chris Hinnen, Jos H. Beijnen, Dees P.M. Brandjes, Michaela Diamant, Victor E.A. Gerdes


R1R2R3R4R5R6R7R8R9


120 | Chapter 8

Abstract

Introduction: Both obesity and eating behavior clusters in families, with family members

imitating each others’ lifestyle. Following bariatric surgery, patients are expected to implement

diet and lifestyle changes. We hypothesize that cohabitating family members will lose weight and

improve their eating behavior within one year after surgery.

Materials and Methods: In this observational prospective study, repeated assessments were

taken in 92 morbidly obese patients who had gastric bypass surgery and their family members

(88 partners, 20 children >18 years and 25 children between 12-17 years) who lived in the same

household. Family members were asked to assess their weight and height before and three, six

and twelve months following bariatric surgery of the patient, and filled out the Dutch Eating

Behavior Questionnaire (DEBQ).

Results: Between baseline and one year following surgery, 49 (66.2%) partners of patients who

underwent gastric bypass surgery lost weight, 6 (8.1%) remained stable and 19 (25.7%) gained

weight. Using linear mixed model analysis, body mass index (BMI) of partners (p = .002), particular

of overweight (p < .001) partners –but not children– showed a small, significant decrease over

time. No significant changes in eating behavior for partners or children were found.

Discussion: The study indicates that gastric bypass surgery may have a ripple effect with body

weight in partners of patients decreasing over time. However, there is considerable variation in

the postoperative weight loss of partners.


Family effect gastric bypass | 121

8

Introduction

The development of obesity is multifactorial with an important role for sedentary lifestyle,

hypercaloric diet and family factors.1 Parental weight has proven to be one of the most important

independent predictors of childhood obesity (17.9% increased risk for obesity with overweight

parents vs. 7.9% with normal weight parents),2 and consequently of obesity in adulthood.3, 4 While

parents and children share both genetic and environmental factors, data from the Framingham

Heart Study showed that if one’s partner becomes obese, the likelihood that the other partner

will become obese is increased by 37%.5

For patients with morbid obesity gastric bypass surgery is one of the most reliable and effective

treatments when previous interventions for achieving and maintaining weight loss have been

unsuccessful.6 The benefits of a gastric bypass operation include weight loss and reduction

of comorbidities and overall mortality.6, 7 Following gastric bypass, patients are instructed to

implement diet and lifestyle changes which may lead to partners and children mimicking the

altered behaviors of the patients undergoing gastric bypass surgery.8

Although some studies have been performed in partners and children of patients who underwent

gastric bypass surgery,9, 10 only one paper examined changes in weight and health behavior in

patients and their family members in the interval before and one year after the operation.8 This

study did find a decrease in weight in obese adult family members (n=50), but did not detect

significant differences in obese children, perhaps due to a small sample size (n=15). Furthermore,

family members received lifestyle counselling (e.g., dietary advice, multivitamins, physical exercise

advice) which will have influenced the effect of the gastric bypass operation on family members.

The present study adds to findings in previous studies by only monitoring weight change of

family members of patients undergoing gastric bypass surgery without adding an intervention

aimed to reduce weight loss in partners, by including twice as many partners and children, by

also investigating the change in eating behavior of family members, and by taking measurements

preoperatively and at three time points postoperatively.

We hypothesized that cohabitating family members of patients undergoing gastric bypass surgery

will lose weight and improve their eating behavior during the first year after the operation.

Therefore we set out a prospective observational study to evaluate weight, height and eating

behavior of the family members of patients undergoing gastric bypass surgery both before, as

well as three, six and twelve months after the operation.

R1R2R3R4R5R6R7R8R9


122 | Chapter 8


Study sample

Patients who underwent gastric bypass surgery in the Slotervaart Hospital (Amsterdam, The

Netherlands), and their cohabitating family members (partner and/or children, (12-85 years)) were

asked to participate in the study. Partners and children were excluded in case of not holding a

pair of scales or height measurement equipment (n = 0), an endocrine disorder (n = 2), treatment

with medications that may cause significant weight gain or loss such as glucocorticoids (n = 4),

participation in an organized weight reduction program (n = 0), participation in a clinical trial

within the last three months prior to screening (n = 0), previous surgical treatment for obesity

(n = 8), history of depressive disorder or other psychiatric disorders (n = 0), pregnancy (n = 1),

language barrier (n = 1) or mental incapacity (n = 0).

Study design

The design was an observational prospective study with repeated measurements before the

operation and three, six and twelve months postoperatively. Figure 1 shows the flowchart of the

inclusion process. At baseline, 177 families were eligible for participation in the study, of which

92 families agreed to participate. One year after bariatric surgery, 80 families of patients who had

gastric bypass surgery (77 partners, 20 children > 18 years and 19 children between 12-17 years)

completed the study and 4 families were lost to follow-up.

The patients’ weight and height were retrieved from the electronic patient files before surgery

and 1 year after the surgery. Family members’ weight and height were self-recorded at home and

participants were asked to fill out eating behavior questionnaires before and three, six and twelve

months postoperatively. Families were contacted by telephone to remind them of and instruct

them about the measurements and questionnaires.


by the Ethics Review Board of the hospital. Written informed consent was obtained from all

participants, and, if between 12-17 years, also from their parents/guardians.



8

Assessed for eligibility

n=177


n=92

Not interested n=69

Excluded n=16

3 months

Missing: 21

Lost to follow up: 3

n=68

6 months

Missing: 23


n=65

12 months

Missing: 8


n=80

Figure 1. Study flowchart

Eating Behavior Questionnaire

Individual’s eating behavior was assessed using the validated Dutch Eating Behavior Questionnaire

(DEBQ).11 The questionnaire includes 13 items on emotional eating (e.g., ‘‘Does worrying make

you feel like eating?’’), 10 items on external eating (e.g.,‘‘Does walking past a candy store make

you feel like eating?’), and 10 items on restrained eating (e.g.,‘‘Do you intentionally eat food

that helps you lose weight?’’). The questionnaire has a five point Likert rating format, ranging

from 1 (never) to 5 (very often). The questionnaire has shown to possess fair internal consistency,

satisfactory test-retest stability and adequate construct and discriminant validity.11 In the current

study, Cronbach’s alpha for emotional eating was .95, for external eating .76 and for restrained

eating .94. The theoretical value of alpha varies from zero to 1. An alpha between .6 and .7 is

represents acceptable internal consistency, between .7 and .9 good internal consistency, and an

alpha ≥ .9 represents excellent internal consistency.

R1R2R3R4R5R6R7R8R9


124 | Chapter 8

Weight and height

Height was measured by participants at home without shoes, at approximately the same time

of the day, with their back against the wall looking straight ahead. Participants were instructed

to measure weight at home with an empty bladder, without shoes, only wearing light clothing,

at approximately the same time of the day, with the same pair of scales. BMI was calculated as

weight divided by squared height (kg/m²) for adults. Normal weight was defined as BMI < 25 kg/

m², overweight as BMI ≥ 25 and < 30 kg/m² obesity as BMI ≥30 kg/m².

For children BMI was standardized using z-scores according to Dutch reference values. A z-BMI

from 1.1 to 2.3 was classified as overweight and a z-BMI of 2.3 or higher as obese.12 The amount

of weight loss was calculated by subtracting baseline weight of the weight measured at one

year following surgery. Then three groups were defined: 1) a stable group showing no changes

in body weight, 2) a group losing weight (> -0.1 kg) and 3) a group gaining weight (> +0.1 kg).

Statistical analysis

Descriptive statistics were computed for demographics, weight and eating behavior. Means and

standard deviations were calculated for continuous variables; frequencies and percentages were

used to describe categorical data. Results were stratified by type of family member (partners,

children >18 years, children 12-17 years). The score distributions of the dependent variables, BMI

and eating behavior (emotional eating, external eating and restrained eating) were sufficiently

normal to allow parametric tests. Variables with a positively skewed distribution were log-

transformed or square root transformed before analysis. The univariate correlation between age,

sex and outcome variables (z-BMI and eating behavior) were examined. Demographic variables

that significantly correlated with the outcome variables were included in the model.

Linear mixed models analyses with random slopes were used to assess the change in outcome

measurements at baseline and three, six and twelve months after gastric bypass surgery for BMI,

emotional eating, external eating and restrained eating. A p-value of < 0.05 was considered

significant. All analyses were performed with SPSS, version 19.0 (for Windows).



8

Results

Baseline characteristics are presented in Table 1. Complete data was obtained of 83 partners and

47 children (26 children > 18 years, 21 children between 12-17 years) of patients who underwent

gastric bypass surgery. Partners were predominantly male (80.7%) with a mean age of 49.3 years.

The percentage of boys was nearly 50% (48.1% in children > 18 and 47.6% in children between

12 and 17 years of age) of all children.

Table 1. Baseline characteristics stratified by family member

Patients Partners Children >18 Children 12-17N 88 83 26 21Male (%) 16 (18.2%) 67 (80.7%) 12 (46.2) 10 (47.6%)Age (years) 47.3±8.4 49.3±9.4 22.6±4.2 15.1±1.4Height (m) 1.71±.08 1.78±.09 1.78±.13 1.69±.06Weight (kg) 127.9±20.2 90.6±18.8 82.9±20.9 61.3±11.9BMI (kg/m2) 43.7±5.9 28.4±5.4 25.9±4.8 -- --Z-BMI -- -- -- -- -- -- .84±.96Emotional eating -- -- 1.6±.59 2.0±.87 1.8±.62External eating -- -- 2.7±.47 2.7±.56 2.6±.56Restrained eating -- -- 2.5±.91 2.4±1.1 2.1±.79Overweight (%) 0 40 (48.2) 9 (34.6) 1 (4.8)Obese (%) 88 (100) 24 (28.9) 5 (19.2) 1 (4.8)

BMI - body mass index. overweight adults: BMI ≥ 25 kg/m² & < 30 kg/m²overweight children Z-BMI > 2.1 & < 2.30, obese adults: BMI ≥30 kg/m²; obese children: Z-BMI > 2.3 corresponding to ≥ 95th percentile for age and gender.

Pearson’s correlations showed that for partners higher age was associated with restrained eating

(r = .26, p < .05) and that a higher BMI was associated with emotional, external (both r = .26,

p < .05) and restrained eating (r = .30, p < .01). For children > 18 years, a higher age was

associated with BMI (r = .57, p < .001) but no correlations between BMI and eating behaviors

were found. Furthermore, for children between 12 and 17 years z-BMI was correlated with both

external (r = .56, p < .05) and restrained eating (r = .48, p = .05). No sex-related differences were

noted in adults or children.

At baseline, 72.5% of adult family members (≥ 18 years) were either overweight (45.0%) or

obese (27.5%) which are higher rates as seen in the general Dutch population (37% overweight,

11% obese).13 In children between 12-17 years, 9.6% was overweight (4.8%) or obese (4.8%)

R1R2R3R4R5R6R7R8R9


126 | Chapter 8

which is more comparable to the Dutch population (10% overweight, 3% obese).13 Furthermore,

emotional eating, external eating and restrained eating of partners and children were comparable

with the general Dutch population.11

In total, 49 (66.2%) partners of patients who underwent gastric bypass surgery lost weight one

year after the operation, 6 (8.1%) remained stable and 19 (25.7%) gained weight. Ten (50%)

children with an age > 18 years lost weight one year after the operation, none remained stable

and 10 (50%) gained weight. Furthermore, 6 (33.3%) children with an age of 12-17 years lost

weight, none remained stable and 12 (66.7%) gained weight.

Table 2 shows the means and standard deviations of BMI and eating behavior over time for the

family members. Mixed model analyses showed that body weight (p = .002) and BMI of partners

(p = .002) decreased significantly over time. Both body weight (all ps < .01) and BMI (all ps < .05)

improved significantly between all time points. Between the baseline assessment and the 1-year

postoperative follow-up, the median BMI decrease was .41 kg/m2, interquartile range (IQR) 1.39,

total range -5.25 kg/m2 to +5.26 kg/m2 and there was a median weight loss in kg of 1.3 kg, IQR

4.38, range -12.00 kg to +16.30 kg in partners one year after surgery.

When divided by weight status (normal weight, overweight and obese), mixed model analyses

showed that body weight (M = 87.32 SD = 11.74 before surgery, M = 85.62 SD = 11.61 one year

after surgery, p < .001) and BMI (M = 27.43 SD = 1.53 before surgery, M = 26.76 SD = 1.74 one

year after surgery, p < .001) of overweight partners decreased significantly over time between all

time points (all ps < .01). No significant improvements were found for normal weight and obese

partners (data not shown).

Changes in body weight and BMI in children > 18 years of age were not significant over time.

In children between 12-17 years of age, there was no significant change in z-BMI over time. No

significant changes were found for emotional, external or restrained eating, for partners, or for

children.



8

Table 2. Changes over time for family members of patients who underwent gastric bypass surgery.

BaselineM (SD)

3 months M (SD)

6 months M (SD)

1 year M (SD)

P (over time)

PartnersBody weight (kg) 90.58 (18.78) 88.72 (18.51)* 90.21 (16.3)* 88.87 (16.69)* <.01BMI (kg/m2) 28.49 (5.40) 27.68 (5.40)* 28.03 (4.86)* 28.00 (5.72)* <.01Eating behavior

emotional eating 1.64 (.59) 1.61 (.67) 1.60 (.56) 1.69 (.67) .623 external eating 2.68 (.47) 2.59 (.48) 2.56 (.54) 2.61 (.54) .410

restrained eating 2.50 (.91) 2.64 (.95) 2.55 (.82) 2.62 (.85) .244Children > 18 yearsBody weight (kg) 82.88 (20.87) 83.65 (19.23) 86.56 (19.62) 80.90 (16.42) .206BMI (kg/m2) 25.86 (4.79) 25.10 (4.33) 25.98 (5.39) 24.99 (3.67) .176Eating behavior


restrained eating 2.41 (1.05) 2.64 (1.06) 2.30 (1.11) 2.44 (1.11) .115Children 12-17 yearsBody weight (kg) 61.31 (11.95) 60.10 (11.36) 62.74 (13.86) 64.45 (13.77) <.05Z-BMI (kg/m2) .84 (.96) .57 (.87) .51 (1.32) .60 (1.23) .191Eating behavior


restrained eating 1.95 (.79) 1.80 (.75) 1.88 (.92) 1.84 (.69 .802

M= Means, SD= standard deviations, *p<.05 compared to baseline value. BMI- body mass index, Z-BMI: standard deviation score of BMI according to Dutch reference values. Eating behavior assessed by the Dutch Eating Behavior Questionnaire.

Conclusion

This study shows a positive change in the body weight in 2 out of every 3 partners living with

patients who underwent gastric bypass surgery. At baseline, almost half of the adult family

members in our cohort were overweight and one third obese, which, as expected, is significantly

higher than seen in the general Dutch population. A small but significant reduction in BMI of

partners, and in particular overweight partners, one year after bariatric surgery was observed.

This is in concordance with a previous study also observing no change in the weight in lean

partners.8 However, they did find significant weight loss in 21 obese partners (106 to 103 kg,

p < .01), but, in contrast to our results, not in overweight partners.8 However, another study

observed a mean increase of weight in 30 obese partners of patients who underwent gastric

bypass surgery.9 They hypothesized that a so called “garbage can effect” could cause this: due to

the limited oral intake of the partner who had gastric bypass surgery, the food that is left is eaten

by the partner causing an increase in weight.9 In our cohort, 66.2% of the partners lost weight,

R1R2R3R4R5R6R7R8R9


128 | Chapter 8

8.1% remained stable and 25.7% gained weight one year after gastric bypass of the partner. The

weight increase in a quarter of the partners in our cohort could be caused by this effect as well.

However, our data give more support to the hypothesis that the overweight partners mimicked

the more healthy eating behavior of the patient who underwent gastric bypass resulting in a

significant weight decrease over time.

Comparable to the only available study addressing this question,8 we found no difference in

z-BMI or eating behavior in children, aged 12-17 years. This could be due to the physiological

i.e. a higher energy need due to the puberty-related growth spurt or psychological development

of children going through puberty; i.e. they undergo progressive individualization, accompanied

by newly derived independence, which may leave the adolescents less sensitive to family-guided

behavioral changes; instead children’s eating behavior is affected by eating behavior of peers

proximate to the child.14

Indeed, we could not detect a change in eating behavior over time in children or partners of

patients who underwent gastric bypass surgery. In contrast, a previous study showed in 24 adult

family members of gastric bypass patients an improvement in uncontrolled and emotional eating

one year after surgery (p < .05).8 This discrepancy in outcome could not be explained in our study;

perhaps it is caused by differences in sample characteristics or the eating behavior questionnaire

used.

Although changes in body weight of partners9 and children8 of patients who underwent gastric

bypass surgery have been described in previous studies, the present study adds to this literature

by including twice as many family members and performing measurements at four different time

points; an interesting aspect of any weight loss intervention. Nevertheless, some limitations are

to be acknowledged. First, family members knew that they were participating in a study and had

to monitor their weight, which may have had impact on the family members’ weight loss. The

periodical contact by telephone to remind them and instruct them about the measurements and

questionnaires may have promoted their cooperation and weight loss. Second, comparable to a

previous study,9 weight, height and eating behavior of family members were obtained through

self-assessment. Although family members received extensive instructions regarding these

measurements, this may still have caused measurement bias, in particular for children in the age

between 12 and 17 who may still be growing. From the literature is known that in self-reports

individuals overestimate their height and underestimate their weight,15, 16 but since we were

interested in BMI change over time, measurement bias may not have influenced our results to a



8

large extent. Third, the follow up of the study was one year, even though the trend of weight loss

seams stable, it would be preferable to extend the follow-up period to examine if this effect is

permanent. Finally, we did not include a control group also being monitored in terms of weight,

which precludes strong interference of our observational results.

In conclusion, partners of patients after gastric bypass surgery showed a decrease of weight in

the one year postoperative interval. Considering their higher than normal weight, the group of

partners may be a target group for weight loss interventions in the future. Prospective studies

with a control group that also monitors weight but with partners who did not have bariatric

surgery are needed to verify whether the bariatric surgery was the cause of weight loss in partners

of gastric bypass patients.

R1R2R3R4R5R6R7R8R9


130 | Chapter 8

Reference List

1. Simopoulos AP. Characteristics of obesity: an overview. Ann N Y Acad Sci 1987;499:4-13.

2. Davis MM, McGonagle K, Schoeni RF, Stafford F. Grandparental and parental obesity influences on childhood overweight: implications for primary care practice. J Am Board Fam Med 2008;21(6):549-554.

3. Agras WS, Mascola AJ. Risk factors for childhood overweight. Curr Opin Pediatr 2005;17(5):648-652.

4. Keane E, Layte R, Harrington J, Kearney PM, Perry IJ. Measured parental weight status and familial socio-economic status correlates with childhood overweight and obesity at age 9. PLoS One 2012;7(8):e43503.

5. Christakis NA, Fowler JH. The spread of obesity in a large social network over 32 years. N Engl J Med 2007;357(4):370-379.

6. Christou NV, Sampalis JS, Liberman M et al. Surgery decreases long-term mortality, morbidity, and health care use in morbidly obese patients. Ann Surg 2004;240(3):416-423.

7. Adams TD, Gress RE, Smith SC et al. Long-term mortality after gastric bypass surgery. N Engl J Med 2007;357(8):753-761.

8. Woodard GA, Encarnacion B, Peraza J, Hernandez-Boussard T, Morton J. Halo effect for bariatric surgery: collateral weight loss in patients’ family members. Arch Surg 2011;146(10):1185-1190.

9. Madan AK, Turman KA, Tichansky DS. Weight changes in spouses of gastric bypass patients. Obes Surg 2005;15(2):191-194.

10. Watowicz RP, Taylor CA, Eneli IU. Lifestyle behaviors of obese children following parental weight loss surgery. Obes Surg 2013;23(2):173-178.

11. van Strien T, Frijters JER, Bergers GPA, Defares PB. The Dutch eating behaviour questionnaire (DEBQ) for assessment of restrained, emotional, and external eating behaviour. International Journal of Eating Disorders, 1986;5, 295-315.

12. Fredriks AM, van BS, Wit JM, Verloove-Vanhorick SP. Body index measurements in 1996-7 compared with 1980. Arch Dis Child 2000;82(2):107-112.

13. van Hilten O, Voorrips LE. Gezondheid en zorg in beeld. In: Centraal Bureau voor statistiek (CBS), editor. In Gezondheid en Zorg in Cijfers. Den Haag/Heerlen: 2012.

14. Wouters EJ, Larsen JK, Kremers SP, Dagnelie PC, Geenen R. Peer influence on snacking behavior in adolescence. Appetite 2010;55(1):11-17.

15. Pursey K, Burrows TL, Stanwell P, Collins CE. How Accurate is Web-Based Self-Reported Height, Weight, and Body Mass Index in Young Adults? J Med Internet Res 2014;16(1):e4.

16. Connor Gorber S, Tremblay M, Moher D, Gorber B. A comparison of direct vs. self-report measures for assessing height, weight and body mass index: a systematic review. Obes Rev 2007;8(4):307-326.

PART IIISummary and Appendices


Summary | 133

Summary

The aim of this thesis was to examine social and emotional aspects of bariatric surgery and

obesity. Particular attention was given to the role of patients’ attachment representations. It

describes in two parts, the role of patients’ attachment representations preoperatively and the

postoperative situation by examining attachment representations as a predictor of the treatment

outcome of gastric bypass surgery for morbid obesity and the effect of gastric bypass surgery on

weight and eating behavior of cohabitating family members.

The first study described in Chapter 2 is a systematic review focusing on the relationship between

attachment representations and obesity. Peer-reviewed literature published between 1990 and

2013 was derived from PubMed, PsycINFO and reference lists of included papers. Ten studies

met the selection criteria. Six studies investigated an adult population whereas four studies

investigated children. For the evaluation of attachment representations, nine different methods

were used comprising both categorical and dimensional measures of attachment. Overall the

reviewed studies suggested a relationship between obesity and attachment insecurity, particularly

attachment anxiety, the anxiety about rejection and abandonment by others. Currently, less claims

can be made with regard to attachment avoidance. Possible reasons for an impact of attachment

representations on obesity are heightened physiological responses to stressful situations and the

underdevelopment of emotion-regulation. However, more research is needed to come to solid

conclusions. Despite the early stage of research and understanding in the field of obesity, there

may be a potential in considering attachment theory in obesity care.

Part I: Attachment representations, obesity and preoperative assessment

Chapter 3 is a cross-sectional study, examining whether patients’ self-reported attachment

representations and levels of depression and anxiety influenced psychologists’ evaluations of

morbidly obese patients applying for bariatric surgery. A group of 250 consecutively referred

candidates for bariatric surgery were assessed by the Slotervaart bariatric surgery clinic in

Amsterdam, the Netherlands between February 2012 and July 2012. Attachment anxiety (OR

= 2.50, p = .01) and attachment avoidance (OR = 3.13, p = .001) were found to be associated

with less positive psychological evaluations by psychologists, and symptoms of depression and

anxiety mediated this association. Findings of this study are useful for the psychological screening

of patients applying for bariatric surgery. That is, patients’ attachment representations influence

a psychologist’s evaluation in an indirect way by influencing the symptoms of depression and

anxiety patients report during an assessment interview.

R1R2R3R4R5R6R7R8R9


134 | Summary

Chapter 4 examined the mediating role of coping styles in the relation between attachment

and mental health and physical functioning in patients applying for bariatric surgery. The study

consisted of 299 patients referred for bariatric surgery to the Slotervaart bariatric surgery unit,

Amsterdam, the Netherlands between February and August 2012. Attachment anxiety (β = -.490,

p < .001) and attachment avoidance (β = -.387, p < .001) were found to be related to mental

health. In addition, attachment anxiety was also found to be related to physical functioning

(β = -.188, p < .001). Coping style were indicated to partly mediate these associations. Findings

suggest not only that it is important to consider attachment anxiety or attachment avoidance in

understanding mental health and physical functioning in patients with morbid obesity but also

that coping style plays a role in these relationships.

In Chapter 5 we investigated the association between attachment representations and mental

health care use in patients with morbid obesity applying for bariatric surgery. This study (N =

260) identified that attachment anxiety was associated with more mental health care visits (OR

= 1.86, 95% CI = 1.11-2.54, p = .02), present use of medication (OR = 2.30, 95% CI = 1.43-

3.68, p = .001) and previously prescribed medication (OR = 2.01, 95% CI = 1.13-3.57, p = .02).

Furthermore, the use of previously prescribed medication was especially prevalent in patients

with high attachment anxiety and low attachment avoidance (OR = 2.96, 95% CI = 1.35-6.50,

p = .007). In conclusion, the results of this study suggest that attachment behavior plays a role

in mental health care utilization of patients with morbid obesity who apply for bariatric surgery.

Therefore, it is important for health care providers working with patients with morbid obesity to

have knowledge of the attachment theory, to recognize anxious attachment representations and

to be aware of these patients’ desire of close relationships and hypervigilance for rejection as well

as of the mental vulnerability of this group.

Part II: Postoperative: attachment representations and effect on family members

The primary aim of Chapter 6 was to examine whether the association between on the one hand

current and past psychological problems, attachment anxiety and attachment avoidance, and on

the other hand weight reduction one year after gastric bypass surgery is explained by patients’

adherence to dietary recommendations. This longitudinal study included 105 patients applying

for a Roux-en-Y gastric bypass operation. Although there is no doubt that gastric bypass surgery

is an effective treatment for the majority of patients with morbid obesity, our results indicate that

the amount of weight loss one year after surgery will to some extent depend on the degree to

which the patient succeeds in adopting healthy dietary recommendations at 6 months. Of all

examined predictor variables, attachment anxiety was most strongly associated with low dietary


Summary | 135

adherence at both 6 months (p = .009) and 12 months (p = .006) post-surgery. Mediation analyses

using resampling procedures indicated that in the year following a gastric bypass operation more

anxiously attached patients have greater difficulty with adherence to dietary recommendations at

6 months and consequently lose less weight.

Chapter 7 is a longitudinal cohort study, examining whether attachment anxiety and attachment

avoidance, independent of body mass index (BMI), predict the level and course of physical

functioning and mental well-being after gastric bypass surgery. This longitudinal study included

the same 105 patients applying for a Roux-en-Y gastric bypass operation as described in Chapter

6. Analyses showed significant improvement of quality of life for physical functioning (M = 37.6

SD = 9.5 before surgery, M = 54.6 SD = 7.1 one year after surgery, F = 128.6, p < .001), but

not for mental well-being (M = 51.9 SD = 7.8 before surgery, M = 49.7 SD = 9.3 after surgery,

F = 4.28, p = .01) within the first year after a gastric bypass operation. Both attachment anxiety

and attachment avoidance were associated with a lower level of mental well-being, but not with

the postoperative course of mental well-being and physical functioning. Our study suggests that

surgery and weight loss lead to improvement in physical functioning but not mental well-being

that is associated more clearly with attachment representations. Results highlight that patients

with weak attachment representations should be protected against unrealistic expectations

regarding improvement of mental well-being after surgery.

Chapter 8 describes a observational longitudinal study examining weight and eating behavior

changes in cohabitating family members of patients who underwent gastric bypass surgery during

the first year after the operation. In this study 92 morbidly obese patients undergoing gastric

bypass surgery and their cohabitating family members were followed (88 partners, 20 children

> 18 years and 25 children between 12-17 years). Results showed that between baseline and

one year following surgery, 49 (66.2%) partners of patients who had gastric bypass surgery lost

weight, 6 (8.1%) remained stable and 19 (25.7%) gained weight. Furthermore, body weight and

body mass index (BMI) of partners (p = .002), and in particular overweight partners (body weight

M = 87.32 SD = 11.74 before surgery, M = 85.62 SD = 11.61 one year after surgery, p < .001, and

BMI M = 27.43 SD = 1.53 before surgery, M = 26.76 SD = 1.74 one year after surgery, p < .001) –

but not children– showed a small, significant decrease over time. No significant changes in eating

behavior for partners or children were found. Gastric bypass surgery showed a ripple effect, as

weight and BMI of partners of patients decreased over time. However, there was a considerable

variation in the effect on partners. The results indicate that overweight family members may

be a target group for weight loss interventions in the future. Larger prospective studies with

R1R2R3R4R5R6R7R8R9


136 | Summary

standardized measurements in a case-control design are needed to assess the characteristics of

the group partners (of gastric bypass patients) that lose weight after the surgery.


Samenvatting (Dutch Summary) | 137

Nederlandse samenvatting

Inleiding

Obesitas is een groot gezondheidsprobleem in welvarende landen en wordt gekenmerkt door

extreme vetophoping in het lichaam. Obesitas gaat vaak gepaard met andere ziektes zoals

diabetes, hart- en vaatziekten en psychische problematiek, en een verhoogde kans op vroegtijdig

overlijden. De meest gebruikte classificatie van obesitas voor volwassenen is BMI (Body Mass

Index); het lichaamsgewicht (kg) gedeeld door het kwadraat van de lichaamslengte in meters (kg/

m2). Een BMI tussen de 19-25 kg/m2 duidt op een normaal gewicht, tussen de 25-30 kg/m2 op

overgewicht, tussen de 30-40 kg/m2 op obesitas en een BMI van 40 kg/m2 of meer op morbide

obesitas.

Omdat dieet- en beweegprogramma’s weinig effectief lijken te zijn voor mensen met morbide

obesitas, richt deze groep zich na een aantal serieuze afvalpogingen vaak tot bariatrische

chirurgie. Bariatrische chirurgie (operatieve behandeling met als doel gewicht te verliezen) is

op dit moment de meest effectieve en aangewezen behandeling voor mensen met morbide

obesitas. Eén van de meest toegepaste vormen van bariatrische chirurgie is een gastric bypass

operatie (maagverkleining), waarbij de maag operatief wordt verkleind en verbonden met de

dunne darm, die op deze wijze functioneel wordt ingekort.

Drie kwart van de patiënten die een gastric bypass operatie ondergaat lijkt te profiteren van een

operatie in de zin van gewichtsverlies en kwaliteit van leven. Echter, bij één op de vier patiënten

blijft het gewichtsverlies en de verbetering van kwaliteit van leven achterwege terwijl de operatie

technisch goed gelukt is. De uitkomst van de operatie hangt niet alleen af van het slagen van de

operatie, maar ook van de mate waarin de patiënt in staat is zich dieetadviezen eigen te maken

en kwaliteit van leven te veranderen.

Naast psychologische problematiek verwachten we dat de gehechtheid van de patiënt van invloed

kan zijn op het wel dan niet kunnen volgen van dieetadviezen. Ook verwachten we dat inzicht in

gehechtheid helpt om kwaliteit van leven van de patiënt te kunnen voorspellen.

In de hechtingstheorie, ontwikkeld door John Bowlby, gaat men er vanuit dat mensen een

aangeboren behoefte hebben om de nabijheid van hun naasten te zoeken (hechtingsfiguren) in

bedreigende of stressvolle situaties. De nabijheid van anderen zorgt ervoor dat er een vorm van

veiligheid en rust wordt gecreëerd waardoor er met de dreiging en stress kan worden omgegaan.

R1R2R3R4R5R6R7R8R9


138 | Samenvatting (Dutch Summary)

Eerdere ervaringen die mensen hebben met de aanwezigheid en responsiviteit van anderen leiden

tot bepaalde gedachten en verwachtingen (interne werkmodellen) van zichzelf (zoals het waard

zijn om liefde en zorg te ontvangen) en anderen (zoals de verwachting dat anderen te vertrouwen

zijn en voor me zullen zorgen). Wanneer iemand vertrouwen heeft in zichzelf en in anderen, en

goed kan omgaan met stressvolle situaties, dan is waarschijnlijk sprake van veilige gehechtheid. Is

het natuurlijke hechtingsproces verstoord en de emotionele behoefte tijdens eerdere ervaringen

niet vervuld, dan raak iemand onveilig gehecht. Vaak gebeurt dit al in de kindertijd, wanneer de

opvoeder niet in staat is geweest om op emotioneel gebied voor het kind te zorgen, waardoor

het kind bijvoorbeeld voortdurend vernederd, genegeerd of afgewezen werd. Mensen verliezen

hierdoor het vertrouwen in zichzelf en richten zich daardoor sterk op anderen en raken onveilig

angstig gehecht, of verliezen het vertrouwen in anderen, onderdrukken de behoefte aan de

nabijheid van anderen en hebben het idee alleen te kunnen vertouwen op zichzelf en raken

onveilig vermijdend gehecht. Onderzoek heeft uitgewezen dat onveilig gehechte mensen (angstig

of vermijdend) eerder stress ervaren en minder goed in staat zijn om met negatieve gedachten en

emoties om te gaan dan veilig gehechte mensen.

Het doel van dit proefschrift is ten eerste om beter inzicht te krijgen in de rol van gehechtheid

in patiënten die zich aanmelden voor bariatrische chirurgie, ten tweede om te onderzoeken of

gehechtheid kan voorspellen waarom er verminderde uitkomsten optreden bij patiënten na een

gastric bypass operatie en ten derde om in beeld te brengen of er een indirect effect is van een

gastric bypass operatie op het gewicht en eetgedrag van de familieleden van de patiënt.

Samenvatting

Dit proefschrift is opgebouwd uit twee delen: een deel met onderzoeken voor de operatie en

een deel met onderzoeken na de operatie. Het eerste deel beschrijft de rol van gehechtheid bij

patiënten die zich aanmelden voor bariatrische chirurgie, het tweede deel beschrijft gehechtheid

als mogelijke voorspeller voor de uitkomst van de gastric bypass operatie, en het mogelijke effect

van een gastric bypass operatie op het eetgedrag en gewicht van familieleden.

Hoofdstuk 1 geeft een inleiding van het proefschrift en bestaat uit een overzicht van het

onderzoek in dit proefschrift en achtergrondinformatie over obesitas en de hechtingstheorie.

Hoofdstuk 2 geeft een systematisch overzicht van onderzoek naar het verband tussen gehechtheid

en obesitas. De databases PubMed en PsycINFO werden doorzocht om alle gepubliceerde

artikelen tussen 1990 en 2013 die dit verband onderzochten te verzamelen. Uiteindelijk zijn



10 onderzoeken meegenomen voor dit overzicht. Over het algemeen vonden we dat onveilige

gehechtheid en met name angstige gehechtheid samenhangen met obesitas. Mogelijke redenen

waarom gehechtheid samenhangt met obesitas kunnen gevonden worden in de aanwezigheid

van verhoogde fysiologische reacties op stressvolle situaties en de onderontwikkeling van emotie-

regulatie. Alleen wanneer in de toekomst goed uitgevoerde longitudinale studies worden gedaan,

kan de relevantie van onze bevindingen, vooral in termen van preventie en behandeling van

obesitas, worden vastgesteld.

Deel I: Preoperatief

Hoofdstuk 3 betreft een onderzoek waarbij het verband tussen gehechtheid van de patiënt en

de beoordeelding door de psycholoog voor de mate van geschiktheid voor bariatrische chirurgie

in beeld werd gebracht. Vervolgens is onderzocht in hoeverre de beoordeling van de psycholoog

verklaard werd door de aanwezigheid van symptomen van angst en depressie bij de patiënt. De

onderzoekgroep bestond uit 250 patiënten die waren doorverwezen voor bariatrische chirurgie

in het Slotervaartziekenhuis. De resultaten lieten zien dat patiënten die werden gekenmerkt

door onveilige gehechtheid door de psycholoog werden beoordeeld als minder geschikt voor de

operatie. Ook vonden we aanwijzingen dat dit deels te verklaren was door de symptomen angst

en depressie die de patiënt die door de psycholoog worden opgepikt.

In Hoofdstuk 4 onderzochten we of de manier van omgaan met tegenslag (coping) een rol

speelt in het verband tussen gehechtheid en mentale gezondheid en fysiek functioneren bij

patiënten die waren doorverwezen voor bariatrische chirurgie in het Slotervaartziekenhuis. We

legden 299 patiënten vragenlijsten voor over de relatie met hun naasten, omgang met tegenslag,

en fysiek en mentaal functioneren. De resultaten lieten zien dat meer onveilig gehechte patiënten

een minder goede mentale gezondheid hadden dan meer veilig gehechte patiënten en dat meer

angstig gehechte patiënten ook veel meer fysieke beperkingen ervoeren, wat deels te verklaren is

door de manier van omgang met tegenslag. Onze bevindingen laten niet alleen het belang zien

van de rol van gehechtheid in de mentale en fysieke gezondheid van de patiënt, maar ook dat de

coping stijl een rol zou kunnen spelen.

In Hoofdstuk 5 onderzochten we het verband tussen hechtingsgedrag en het gebruik

van mentale gezondheidzorg. De onderzoeksgroep bestond uit 260 patiënten die waren

doorverwezen voor bariatrische chirurgie in het Slotervaartziekenhuis. We vonden dat meer

angstig gehechte patiënten in het verleden vaker psychologische hulp kregen en vaker gebruik

maakten van psychofarmaca en dat ook het huidige gebruik van psychofarmaca aanzienlijk

R1R2R3R4R5R6R7R8R9


140 | Samenvatting (Dutch Summary)

hoger was in meer angstig gehechte patiënten dan in veilig of vermijdend gehechte patiënten.

Omdat gehechtheid een rol lijkt te spelen in het gebruik van geestelijke gezondheidszorg, is het

nuttig dat behandelend professionals goede kennis hebben over de hechtingstheorie, zodat ze

angstige gehechtheid kunnen herkennen, en zich bewust worden van de behoeften van deze

patiëntengroep. Ook is het belangrijk dat ze zich bewust zijn van de overgevoeligheid bij deze

patiënten voor afwijzing en de kwetsbaarheid voor de ontwikkeling van psychische problemen.

Deel II: Postoperatief

Het hoofddoel van Hoofdstuk 6 was te onderzoeken in hoeverre het verband tussen enerzijds

psychologische problemen in verleden en heden en gehechtheid en anderzijds gewichtsverlies

verklaard kan worden door het in meer of mindere mate trouw zijn aan dieetadviezen.

In totaal deden 105 patiënten die waren doorverwezen voor bariatrische chirurgie in het

Slotervaartziekenhuis mee in de studie. Hoewel er geen twijfel is dat een gastric bypass operatie

effectief is voor het grootste gedeelte van de patiënten met morbide obesitas, bieden onze

resultaten een aanwijzing dat de mate van gewichtsverlies na een jaar samenhangt met het

in meer of mindere mate volgen van dieetadviezen op 6 maanden. Van alle voorspellende

variabelen in dit onderzoek was angstige gehechtheid de sterkste voorspeller voor het minder

goed volgen van dieetadviezen. Toen we dit vervolgens in een medierend model analyseerden

vonden we dat meer angstig gehechte patiënten meer moeite lijken te hebben met het volgen

van dieetadviezen, wat vervolgens leidt tot minder gewichtsverlies postoperatief.

Hoofdstuk 7 is een longitudinaal onderzoek, waarbij in beeld werd gebracht in hoeverre

gehechtheid, onafhankelijk van het lichaamsgewicht (body mass index,BMI), samenhangt met het

niveau en verloop van kwaliteit van leven een jaar na de gastric bypass operatie. In totaal deden

105 patiënten mee aan het onderzoek; zij waren doorverwezen voor bariatrische chirurgie in het

Slotervaartziekenhuis deden mee aan het onderzoek. We vonden een significante verbetering

in kwaliteit van leven voor lichamelijk functioneren maar niet voor emotioneel welbevinden een

jaar na de operatie. Zowel angstig als vermijdende gehechtheid hingen samen met een lager

postoperatief niveau van emotioneel welbevinden. Echter, we vonden geen verband tussen

gehechtheid en het postoperatieve niveau van lichamelijk functioneren en gehechtheid en het

verloop van lichamelijk functioneren en emotioneel welbevinden. Terwijl gewicht en gewichtsverlies

voorspellers zijn voor lichamelijk functioneren, lijkt gehechtheid vooral emotioneel welbevinden

te voorspellen. Dit onderzoek benadrukt dat onveiligere gehechte patiënten beschermd moeten

worden tegen onrealistische verwachtingen over verbeterd emotioneel welbevinden na de

operatie.



Hoofdstuk 8 beschrijft een observationele longitudinale studie waarbij we een jaar lang

veranderingen in gewicht en eetgedrag bij familieleden van patiënten die een gastric bypass

operatie hebben ondergaan in beeld brachten. In totaal deden 92 patiënten en hun familieleden

die deel uit maakten van hetzelfde huishouden mee aan de studie. Deze familieleden betreffen

88 partners, 20 kinderen van 18 of ouder en 25 kinderen tussen de 12-17 jaar. Resultaten

lieten zien dat een jaar na de operatie 49 (66.2%) partners van patiënten die een gastric bypass

operatie hadden ondergaan een gewichtsafname hadden, 6 (8.1%) stabiel bleven in gewicht

en 19 (25.7%) aankwamen. Verder zagen we dat het lichaamsgewicht (BMI) van de partners

(p=.002) en met name partners met overgewicht (p<.001) –maar niet kinderen– een kleine daling

gedurende het jaar na de operatie liet zien. Geen significante verschillen werden gevonden voor

eetgedrag. Concluderend, partners van patiënten die een gastric bypass operatie ondergaan

vormen wellicht een interessante doelgroep voor een interventie voor gewichtsverlies. Voordat

we deze conclusie kunnen trekken zijn grotere prospectieve studies nodig die gebruik maken van

gestandaardiseerde meetinstrumenten en een controlegroep om de kenmerken van partners die

gewichtsverlies laten zien in kaart te brengen.

R1R2R3R4R5R6R7R8R9


142 | Dankwoord


Dankwoord | 143

Dankwoord

Aangezien dit het eerst en meest gelezen hoofdstuk van mijn proefschrift zal zijn, wil ik graag

deze ruimte gebruiken om degene te bedanken die direct dan wel indirect –door er voor me te

zijn- betrokken waren bij de totstandkoming van mijn proefschrift.

Allereerst wil ik mijn twee promotoren prof. dr. Dees Brandjes en prof. dr. Rinie Geenen

bedanken. Beste Dees, zonder jou was mijn functie er wellicht niet (meer) geweest. Hoewel je

gedurende mijn promotieperiode je met name bezig hebt moeten houden met ‘de redding’ van

het Slotervaartziekenhuis, gaf je me toch vaak het gevoel dat je over mijn schouder meekeek

door het vertrouwen dat je in me had. Ik waardeer het dat we het samen toch nog voor elkaar

hebben weten te krijgen dat ik in juni kan promoveren. Beste Rinie, jou heb ik helaas pas in de

loop van mijn promotietraject als promotor aan boord gehaald. Direct nam je met jouw rustige

beschouwende maar kritische blik bij de interpretatie van de resultaten en het schrijven van de

artikelen een belangrijke rol in. Ik heb enorme bewondering voor jouw vermogen om in chaotische

stukken tekst orde te creëren. Je kritische blik zorgde keer op keer weer voor verfijning. Dank

voor deze prettige samenwerking!

Dr. Chris Hinnen en dr. Victor Gerdes beiden dank voor de begeleiding, de leerzame tijd en de

balans tussen de Psychologie en Geneeskunde. Beste Chris, wat heb ik het getroffen met jou als

co-promotor. Ik bewonder je bevlogenheid en kennis op het gebied van de psychologie. Jouw

vlotte respons op mijn e-mails werkte als een katalysator en hebben ervoor gezorgd dat het

tempo dat ik er graag in houd op hoog niveau bleef. Beste Victor, ik vond het mooi dat ik jouw

interesse in de psychologie als clinicus zag groeien gedurende mijn promotie. Jij durfde het aan

om met mij als promovenda dit promotietraject op te starten. Je gaf mij de ruimte om zelfstandig

te werken en hield me scherp. In het bijzonder was het prettig om zo nu en dan mijn hart bij je

te kunnen luchten, waarna mijn zorgen als wolken voor de zon verdwenen.

De leden van de promotiecommissie en de gastopponenten wil ik bedanken dat zij zitting

hebben willen nemen in mijn promotiecommissie en mijn proefschrift hebben beoordeeld op

wetenschappelijke waarde.

In het bijzonder wil ik alle patiënten die mee hebben gedaan aan de in dit proefschrift beschreven

studies ontzettend bedanken, zonder hun medewerking hadden deze onderzoeken niet plaats

kunnen vinden.

R1R2R3R4R5R6R7R8R9


144 | Dankwoord

Alle co-auteurs die hebben meegeschreven aan mijn stukken, Yair Acherman, Arnold van de Laar,

Mariska van Vliet, Ines von Rosenstiel, Jos Beijnen, Michaela Diamant en Nalini Radhakishun heel

erg bedankt. Lieve Nalini, wat heb ik het getroffen om samen met jou het ‘familieledenonderzoek’

te mogen uitvoeren. Jouw positieve benadering en gezelligheid hebben het onderzoek nog

leuker gemaakt.

Ook gaat mijn dank uit naar de psychologen van de medische psychologie van het

Slotervaartziekenhuis die hebben bijgedragen aan de inclusie van de patiënten voor het gros

van de studies. En in het bijzonder, Madelon Groenteman, Sanne Koemans, Willemijn Dekker

en Liedeke Duijverman. Kees, ook jij bedankt voor je hulp bij de inclusie en je interesse in het

onderzoek. Ook het ondersteunend personeel van het Slotervaartziekenhuis, zoals de postkamer,

secretaresses, en de centrale, dank voor jullie hulp. Jacqueline, jij bedankt voor het binnen halen

van mij bij de interne geneeskunde. Mede door jou ben ik van secretaresse uitgebloeid tot

onderzoeker. Beste Winnie, jou wil ik bedanken voor de artikelen die je telkens weer binnen no-

time voor me hebt opgezocht en de assistentie bij zoekopdrachten en het werken met Reference

Manager. Heel erg fijn!

Hoewel we altijd met een gezellig klein clubje aan promovendi op 9B in het Slotervaartziekenhuis

zaten, was het toch altijd wel een komen en gaan van onderzoekers. Mijn eerste onderzoeksmaatjes,

Danka en Patrick. Wat een top tijd had ik met jullie op ons kleine kamertje en wat een gemis toen

jullie beide gepromoveerd waren en het pand hadden verlaten. Lieve Marein/Marinus! SCALE-

maatje, maar ook gezellig ski- borrel- en dinermaatje. Lief en leed konden worden gedeeld. Erg

jammer dat je, met goed geluk, nog maar één keer per week in het SLZ was te vinden. Lieve

Laura, wat was ik blij toen jij na een aantal eenzame maanden onze kamer kwam bezetten. Dank

voor de gezelligheid en steuntjes in de rug die we elkaar konden geven. Thomas, een goede

aanwinst in ons onderzoeksteam die gedurende lange tijd alleen nog maar bestond uit vrouwen.

Mede dankzij jou, fungerend als mijn chauffeur naar het provinciale zuiden, heeft mijn promotie

toch nog in juni plaats kunnen vinden. ‘I’ll owe you big time’ –onbeperkt drank gedurende het

feest voor jou! Ook Funda, Maaike en Noëlle dank dat we er voor elkaar waren op de leuke

maar ook minder leuke momenten tijdens ons promoveren en niet te vergeten onze toch bijna

maandelijkse etentjes. Onderzoekers, het was gezellig!

Ook zijn er nog een aantal andere mensen in het Slotervaartziekenhuis die ik graag wil bedanken.

De internisten en arts-assistenten van 9B die in tijden van mijn promotie op de afdeling werkte.


Dankwoord | 145

Dank voor de leuke sfeer op 9B en de gezellige uitjes, zoals het zeilen, de borrels en niet te

vergeten de ski-trip naar Sauze d’Oulx. Dames van de researchpoli, lieve Olga en Monique.

Hoewel ik nooit met jullie heb hoeven samen te werken bevonden we ons toch regelmatig met

onze research werkzaamheden op dezelfde verdieping. Dank voor jullie gezellige praatjes, de

kopjes thee en jullie brede interesse. Derk-Jan jij ook bedankt voor de gezelligheid en grappen en

grollen wanneer je even spontaan bij ons langskwam.

Mijn lieve vrienden en vriendinnen, ontzettend veel dank dat jullie mij de afgelopen jaren hebben

bijgestaan. Lieve Indra, wij gaan echt ‘way back’ en hopelijk ook nog jaren ‘way forward’!! Wij

begrijpen elkaar met een blik of een half woord. Dank voor je trouwe vriendschap, je bent me

veel waard. Lieve Mirte, lieve Janneke wat heb ik samen met jullie genoten van het Amsterdamse

leven, maar ook van het leven daarbuiten, de geweldige verre reizen die we gemaakt hebben

en de gedeelde “bodemloze sushi behoefte”. Lieve Maj, jij ook bedankt voor je gezelligheid,

luisterend oor en creatieve uitjes. Lieve dierbare vriendinnetjes uit Arnhem, Inger & Astrid,

Pauline, Claudia, Famke en Britt, ook al zijn we uitgewaaierd over het land (en het buitenland)

het blijft altijd heerlijk als vanouds met jullie. Lieve HBO-v vriendinnetjes, Marjo en Maaike. Die

maandelijkse etentjes in het voor ons centraal gelegen Utrecht houden we erin! Heerlijk dineren

terwijl we onze belevenissen van de gehele maand aan elkaar vertellen, altijd veel te gezellig en

veel te weinig tijd! Lieve Annet en alle andere Amsterdamse bootcamp- en hardloopmatties,

dank voor het samen ‘tot het gaatje’ gaan en de oh zo gezellige derde helft!! Merlijn dank voor

het lezen van een aantal van mijn stukken als ‘native speaker’ en Coen jij enorm bedankt voor het

ontwerpen van de kaft mijn proefschrift! Lieve Remy, dat ik jou nou nog in mijn laatste maanden

van mijn promotie heb mogen ontmoeten! Ik kijk uit naar alle momenten die we samen nog

mogen delen.

Lieve pap en mam, dank jullie wel voor jullie eeuwige steun en liefde, het loslaten en de oneindige

interesse. Altijd staan jullie voor me klaar! Lieve Arnout, grote broer, dank voor je droge humor,

je interesse en gezelligheid. Het was een pittig jaar voor ons vieren, waar we ons met elkaar goed

doorheen hebben weten te slaan. Wat ben ik ontzettend dankbaar, lieve pap, voor je goede

herstel. Dat we nog vele gezonde jaren en mooie momenten met elkaar als familie mogen delen.

R1R2R3R4R5R6R7R8R9


146 | PhD Portfolio


PhD Portfolio | 147

PhD portfolioECTS

General courses

2011 Good Clinical Practice 0,9

2011 Electronic Data Capture training 0,9

2011 Training peripheral infusion 0,5

2012 Introductory biostatistics for researchers, Julius Centre Utrecht 2,9

Presentations

2011 The Influence of Psychological Factors on the Outcome of Gastric Bypass

Operation – Protocol presentation. Scientific meeting, Slotervaartziekenhuis.

0,5

2012 The evaluation of bariatric surgery candidates by psychologists. Scientific

meeting, Slotervaartziekenhuis, Amsterdam.

0,5

2013 Psychologists’ evaluation of bariatric surgery candidates influenced by patients’

attachment representations and symptoms of depression and anxiety.

Association for Researchers in Psychology and Health (ARPH) conference.

0,5

2013 Attachment representations: the gastric bypass patient. Internal Medicine,

Slotervaartziekenhuis Researchdag.

0,5

2014 Psychological attachment in obesity. The significance for bariatric surgery.

Internal Medicine, Slotervaartziekenhuis Researchdag.

0,5

Conferences

2012 the 30th Obesity Society Annual Scientific Meeting, Texas. 1,5

2012 23th symposium of the Netherlands Association for the Study of Obesity

(NASO), AMC, Amsterdam

0,25

2013 2nd annual conference of the Association for Researchers in Psychology and

Health (ARPH)

0,5

Other

1-2014 Trial nurse, SCALE-study, effect of liraglutide on body weight in non-diabetic

obese subjects or overweight subject with co-morbidities

10

1-2014 Trial nurse, CAPITA-study, Community-Acquired Pneumonia Immunization

Trial in Adults

2

Total ECTS 21,45

R1R2R3R4R5R6R7R8R9


148 | List of Publications


List of Publications | 149

List of Publications

Aarts F, Geenen R, Gerdes VEA, Brandjes DPM, Hinnen C. The significance of attachment

representations for obesity: a systematic review. Submitted.

Aarts F, Hinnen C, Gerdes VEA, Acherman Y, Brandjes DPM. Psychologists’ evaluation of bariatric

surgery candidates influenced by patients’ attachment representations and symptoms of

depression and anxiety. J Clin Psychol Med Set 2014; 21(1):116-123.

Aarts F, Hinnen C, Gerdes VEA, Acherman Y, Brandjes DPM. Coping style as a mediator between

attachment and mental and physical health in patients suffering from morbid obesity. Int J

Psychiatry Med 2014;47(1):75-91.

Aarts F, Hinnen C, Gerdes VEA, Brandjes DPM, Geenen R. Mental Health Care Utilization in

Patients Seeking Bariatric Surgery: the Role of Attachment Behavior. Bariatr Surg Pract Patient

Care 2013;8(4):134-138.

Aarts F, Geenen R, Gerdes VEA, Van de Laar A, Brandjes DPM, Hinnen C. Attachment anxiety

predicts poor adherence to dietary recommendations: an indirect effect on weight change one

year after gastric bypass surgery. Submitted.

Aarts F, Geenen R, Gerdes VEA, Brandjes DPM, Hinnen C. The significance of attachment

representations for quality of life one year following gastric bypass surgery: a longitudinal analysis.

Submitted.

Aarts F, Radhakishun NNE, Van Vliet M, Geenen R, Von Rosenstiel IA, Hinnen C, Beijnen JH,

Brandjes DPM, Diamant M, Gerdes VEA. Gastric bypass may promote weight loss in overweight

partners the first year after surgery. Submitted.

Hoek W, Aarts F, Schuurmans J, Cuijpers P. Who are we missing? Non-participation in an Internet

intervention trial for depression and anxiety in adolescents. Eur Child Adoles Psy 2012;21(10):593-

595.

Celik F, Squizzato A, Aarts F, Groote ME, Fugazzola C, Gerdes VE. Imaging for the diagnosis of

pulmonary embolism in very obese patients; a survey among internists and radiologists in Italy

and the Netherlands. Thromb Res. 2013;131(4):189-190.

R1R2R3R4R5R6R7R8R9


150 | Curriculum Vitae


Curriculum Vitae | 151

Curriculum Vitae

Floor Aarts werd op 4 juni 1986 geboren als dochter van Aad Aarts en Fenna Aarts - van Heek

en zusje van de vier jaar oudere Arnout. Van 1999 tot 2004 genoot zij het Hoger Algemeen

Voorgezet Onderwijs aan het Lorentz Lyceum in Arnhem. Hierna startte zij haar verpleegkunde

opleiding aan de Hogeschool van Arnhem en Nijmegen, die in 2008 met goed succes werd

afgerond. Alvorens te gaan werken besloot zij haar horizon verder te verbreden, en begon aan

haar premaster Gezondheidswetenschappen. In 2010 slaagde zij voor haar master Prevention

and Public Health aan de Vrije Universiteit van Amsterdam. Na een jaar van reizen en werken,

begon zij in juni 2011 aan haar promotie traject in het Slotervaartziekenhuis onder begeleiding

van prof dr. Dees Brandjes, prof dr. Rinie Geenen (Universiteit van Utrecht), dr. Chris Hinnen en dr.

Victor Gerdes, waarvan dit proefschrift het eindresultaat is.

Floor Aarts was born on June 4th 1986 as the daughter of Aad Aarts and Fenna Aarts - van Heek,

and sister of the four years older Arnout. From 1999 till 2004 she followed secondary school

at Lorentz Lyceum in Arnhem. In 2004 she started her Bachelors in Nursing at the University of

Applied Sciences of Arnhem and Nijmegen, and graduated in 2008. Before starting the working

life, Floor decided to broaden her horizon and started her Premaster in Health Sciences. In 2010

she graduated for her Master’s degree in Prevention and Public Health at the Free University of

Amsterdam. After a year of travelling and working, she started her PhD thesis in the Slotervaart

Hospital in Amsterdam in June 2011, under the supervision of prof dr. Dees Brandjes, prof dr.

Rinie Geenen (University of Utrecht), dr. Chris Hinnen and dr. Victor Gerdes, of which this thesis

is the final result.

Kaft Floor - Psychological Attachment in Obesity Final.pdf 1 24-04-14 14:45 Uitnodiging Floor - Psychological Attachment in Obesity (drukbestand).pdf 1 18-04-14 15:36

Documents

Proefschrift Aarts