UvA-DARE (Digital Academic Repository) Decision … · Marti-Keuning Eckhardt Stichting en het SGS-Achmea- fonds. ... CHAPTER 1 Introduction and thesis outline ... CHAPTER 4 How surgeons

UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Decision-making in orthopaedic surgery

Hageman, M.G.J.S.

Link to publication

Citation for published version (APA):Hageman, M. G. J. S. (2018). Decision-making in orthopaedic surgery

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: http://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.

Download date: 09 Jul 2018

http://dare.uva.nl/personal/pure/en/publications/decisionmaking-in-orthopaedic-surgery(5c292f53-9b31-40fc-a66a-f30346a91456).html

Curriculum Vitae

Michiel Hageman was born in Al Jubail, Saudi Arabia on April 7th, 1985. After a

short interlude in The Netherlands, Michiel lived with his family in Malaysia until

1992. Back in the Netherlands, after graduating from high school (VWO, Den Haag)

in 2004, he studied at the medical school of the University of Amsterdam. During

his study Michiel worked for the Bio-Implant Service (BIS) the Netherlands as

orthopaedic tissue-donation surgeon. In his final year of his bachelor he conducted

a research internship at the department of orthopaedic surgery of the Academic

Medical Center Amsterdam (prof. dr. C.N. van Dijk). The experiences at BIS, his

research internship and clinical internship at the AMC made him enthusiastic to

continue working in the medical field of orthopaedic surgery. After obtaining the

medical doctor’s degree in 2011, he worked as PhD student at the department of

Orthopaedic Hand and Upper Extremity of the Massachusetts General Hospital,

Boston – United States as well as the Slotervaart Ziekenhuis in Amsterdam, which

finally resulted in this thesis. During his time in Boston Michiel developed a special

interest in “Shared Decision Making” and “Decision Aids” to facilitate the decision-

making. Together with his friend and colleague Teun Teunis, Michiel launched

PATIENT+, dedicated to support shared decision-making with digital decision

aids. Subsequently Michiel and Teun wrote the book SAMEN Beslissen: waarom

moeilijk doen als het SAMEN kan? and were awarded the best value best health care

initatieve of 2017 (Doelmatigheidsprijs 2017).

In 2014, Michiel started his training for orthopaedic surgery at the department

of general surgery at the Onze Lieve Vrouwe Gasthuis (dr. M.Gerhards).

He continued his residency at the department of orthopaedic surgery at the AMC

(prof. dr. C.N. van Dijk) and Slotervaart Ziekenhuis (dr. H. van der Vis). During his

clinical work, Michiels’ interests to innovate and develop products to improve

health care further increased. At the end of 2017 he decided to focus solely on

PATIENT+. Michiel will lead and support the team of PATIENT+ to develop, integrate

and evaluate decision aids into innovative health care delivery systems.

ISBN 978 94 91549 88 5

omsl.proefschrift.Hageman.indd 1 26-02-18 14:06

Decision-making in Orthopaedic Surgery

M.G.J.S. Hageman

© 2018 M.G.J.S. Hageman, Amsterdam, the Netherlands

Design: Joen design, Wormer

This thesis was prepared at the Orthopaedic Hand and Upper Extremity Service, Massachusetts

General Hospital, Harvard Medical School, Boston, MA, United States of America and the

Department of Orthopaedic Surgery, Academic Medical Center, University of Amsterdam,

Amsterdam, the Netherlands.

All rights reserved. The copyright of the published and accepted articles has been transferred to

the respective publishers. No part of this publication may be reproduced, stored in a retrieval

system of any nature, or transmitted in any form or by any means, mechanically, by photocopying,

recording, or otherwise, without prior written permission from the author.

Hereby I want to gratefully acknowledge the research support I received from Anna Fonds|NOREF,

Marti-Keuning Eckhardt Stichting en het SGS-Achmea- fonds.

Decision-making in Orthopaedic Surgery

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad van doctor aan de Universiteit van Amsterdam

op gezag van de Rector Magnificus prof. dr. i.r. K.I.J. Maex

ten overstaan van een door het College voor Promoties ingestelde commissie,

in het openbaar te verdedigen in de Agnietenkapel

op dinsdag 17 april 2018, te 12.00 uur

door

Michiel Gerardus Johannes Staro Hageman

geboren te Al Jubail, Saoedi- Arabië

PROMOTIECOMMISSIE

Promotores: Prof. dr. C.N. van Dijk AMC - UVA

Prof. dr. D.C. Ring The University of Texas

at Austin

Overige leden: Dr. E.R.A. van Arkel MC Haaglanden

Dr. J.A.M. Bramer AMC - UVA

Prof. dr. S.E. Geerlings AMC - UVA

Prof. dr. I.C. Heyligers Universiteit Maastricht

Prof. dr. J.A.M. Kremer Radboud universiteit

Prof. dr. M. Maas AMC - UVA

Prof. dr. M.P. Schijven AMC - UVA

Faculteit der Geneeskunde

Voor Pilou, Nicoline, mijn zusjes en ouders

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

6

TABLE OF CONTENTS

7

PART 1 GENERAL INTRODUCTION

CHAPTER 1 Introduction and thesis outline

PART 2 DECISION-MAKING IN ORTHOPAEDIC SURGERY

CHAPTER 2 Variation in recommendation for surgical

treatment for compressive neuropathy

J Hand Surg Am. 2013 May;38(5):856-62.

CHAPTER 3 The factors influencing the decision-making of

operative treatment for proximal humeral

fractures

J Shoulder Elbow Surg. 2015 Jan;24(1):21-6.

CHAPTER 4 How surgeons make decisions when the evidence is

inconclusive

J Hand Surg Am. 2013 Jun;38(6):1202-8.

CHAPTER 5 Do previsit expectations correlate with satisfaction

of new patients presenting for evaluation with an

orthopaedic surgical practice?

Clin Orthop Relat Res. 2014 Apr;39(9):11999-014.

CHAPTER 6 Carpal tunnel syndrome: assessment of surgeon

and patient preferences and priorities for decision-

making

J Hand Surg Am. 2014 Sep;39(9):1799-1804.e1.

CHAPTER 7 RCT: The influence of decision aids on decisional

conflict and satisfaction of patient with hip or knee

osteoarthritis

Submitted to KSSTA. 2017 Nov.

10

11

14

15

29

43

57

69

83

8

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

CHAPTER 8 Do upper extremity trauma patients have different

preferences for shared decision-making than

patients with non-traumatic conditions?

Clin Orthop Relat Res. 2015 Nov;473(11):3542-8.

PART 3 GENERAL DISCUSSION

CHAPTER 9 Summary and discussion

CHAPTER 10 Dutch summary and discussion

PhD portfolio 130

List of publications 133

Acknowledgements 135

99

110

111

121

9

10

PART 1

GENERAL INTRODUCTION

11

CHAPTER 1

Introduction and thesis outline

There is substantial variation in the rates and type of operative and non-

operative treatment that cannot be explained by demographics, pathophysiology,

or comorbidities.1-3 There should be some variation in medical treatments. But it’s

difficult to justify variation from surgeon-to-surgeon. The observed variation should

derive entirely from variation in patient preferences based on their values. The

surgeon-to-surgeon variation demonstrates some important opportunities: we can

do better to assist patients in becoming aware of their values, ensuring that their

initial preferences are not based on misconceptions so that they can consider all the

available options, and then helping them choose the option that suits them.

Some of the variation is explained by differences in opinion among

surgeons. For example, some surgeons offer patients with symptoms of carpal

tunnel and normal electro-diagnostic testing surgery based on symptoms alone,

and others do not. It is unclear whether factors such as workers’ compensation,

litigation and less specific symptoms are associated with recommendation for

surgery.3-5 In the second chapter of this thesis we measured factors associated

with variation in recommendation for operative and non-operative treatment

for compressive neuropathy.

Another area of debate is the role of operative treatment for fractures of

the proximal humerus. Surgery is considered for approximately 1 in 5 patients, but

there is no consensus on which fractures benefit from surgery or which procedure

to perform.6 The data to date are limited and inconclusive.6,7 In chapter 3, we

measured the factors that influence agreement between surgeons on treatment

12

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

recommendations and the factors that lead a surgeon to recommend operative

treatment and type of surgery (ie, fixation vs arthroplasty).

Variation in recommendations for operative and non-operative

treatment seems greater for the least objectively verifiable issues. The “Evidence-

Based Guidelines” from the American Academy of Orthopaedic Surgeons are

largely inconclusive for lack of evidence.8 Well-designed, prospective, randomized

controlled trials frequently show no difference or a small and possibly not

clinically relevant difference between 2 treatments.9,10 The fourth chapter

measured how health care providers decide which option to recommend to their

patients when the evidence is inconclusive.

In 2010 among 2,500 common treatments 51% were classified as having

insufficient evidence, 23% likely to be beneficial, 7% requiring trade-offs between

benefits and harms, 5% unlikely to be beneficial, 3% likely to be ineffective or

harmful, and 11% as clearly beneficial.11 A decision aid could inform patients

about the best available evidence and ongoing areas of debate in order to limit

the effect of both patient and surgeon bias and improve the patient’s comfort and

participation in the decision.12 Chapter 5 measured the priorities and preferences

of patients and hand surgeons facing decisions about management of CTS.

Patient satisfaction measures are increasingly used to evaluate the

quality of medical service.13 Many factors play into satisfaction, including

patient’s understanding of their own health and patient’s rating of the quality

of their care and perhaps expectations.14,15 We therefore measured in chapter 6

how previsit expectations affected satisfaction in the orthopaedic practice.

It is thought that it is important to make a shared decision when the evidence

is inconclusive, when there is more than one reasonable option, when there is no clear

advantage in outcomes or when each benefit of harm may be valued differently.16

In shared decision-making the caregiver provides expertise and evidence, and the

patient and caregiver choose diagnostic and treatment options consistent with their

values and preferences.17,18 In chapter 7 we measured the effect of decision aids on the

magnitude of decisional conflict, anxiety, knowledge, satisfaction, physical function

and quality of life to patients with knee and hip osteoarthritis.

Patients with traumatic problems are thought to be less capable of and

less interested in participating in decisions because they feel vulnerable and

time-pressured. In addition, patients with greater symptoms of depression or

less self-efficacy might have less desire or confidence about participation in the

decision-making process and might prefer to fall back to a paternalistic style of

medical care and take a more passive role. In chapter 8, we measured patient

13

preferences for shared decision-making in relation to the acuity of the diagnosis

and to psychological factors.

The final chapter provides a summary and discussion followed by an overall

conclusion and future perspective based on the study results presented in this thesis.

REFERENCES

1. Frymoyer JW. Degenerative Spondylolisthesis: Diagnosis and Treatment. J Am Acad Orthop

Surg 1994;2:9-15.2. Duszak R, Jr., Behrman SW. National trends in percutaneous cholecystostomy between 1994

and 2009: perspectives from Medicare provider claims. J Am Coll Radiol 2012;9:474-9.3. Fanuele J, Koval KJ, Lurie J, Zhou W, Tosteson A, Ring D. Distal radial fracture treatment: what

you get may depend on your age and address. The Journal of bone and joint surgery American

volume 2009;91:1313-9.4. de Beer J, Petruccelli D, Gandhi R, Winemaker M. Primary total knee arthroplasty in patients

receiving workers’ compensation benefits. Can J Surg 2005;48:100-5.5. Harris I, Mulford J, Solomon M, van Gelder JM, Young J. Association between compensation

status and outcome after surgery: a meta-analysis. Jama 2005;293:1644-52.6. Handoll HH, Ollivere BJ, Rollins KE. Interventions for treating proximal humeral fractures in

adults. Cochrane Database Syst Rev 2012;12:CD000434.7. Misra A, Kapur R, Maffulli N. Complex proximal humeral fractures in adults--a systematic

review of management. Injury 2001;32:363-72.8. American Academy of Orthopaedic Surgeons Clinical Practice Guideline on the Diagnosis and

Treatment of Osteochondritis Dissecans Rosemont (IL). American Academy of Orthopaedic

Surgeons (AAOS); 2010.9. Gibbs L, Gambrill E. Evidence-based practice: Counterarguments to objections. . Resarch on

Social Work Practice 2002;12:452-76.10. Pawson R. Evidence Based Policy: In search of a method. Evaluation 2002;8:157-81.11. How much do we know. British Medical Journal 2010;Clinical Evidence 2010.12. Legare F, O’Connor AM, Graham ID, Wells GA, Tremblay S. Impact of the Ottawa Decision

Support Framework on the agreement and the difference between patients’ and physicians’

decisional conflict. Med Decis Making 2006;26:373-90.13. Hudak PL, Wright JG. The characteristics of patient satisfaction measures. Spine (Phila Pa

1976) 2000;25:3167-77.14. Hickson GB, Clayton EW, Entman SS, et al. Obstetricians’ prior malpractice experience and

patients’ satisfaction with care. Jama 1994;272:1583-7.15. Soroceanu A, Ching A, Abdu W, McGuire K. Relationship between preoperative expectations,

satisfaction, and functional outcomes in patients undergoing lumbar and cervical spine

surgery: a multicenter study. Spine (Phila Pa 1976) 2012;37:E103-8.16. Stiggelbout AM, Van der Weijden T, De Wit MP, et al. Shared decision making: really putting

patients at the centre of healthcare. BMJ 2012;344:e256.17. Slover J, Shue J, Koenig K. Shared decision-making in orthopaedic surgery. Clin Orthop Relat

Res 2012;470:1046-53.18. Stacey D, Bennett CL, Barry MJ, et al. Decision aids for people facing health treatment or

screening decisions. Cochrane Database Syst Rev 2011:CD001431.

14

PART 2

DECISION-MAKING IN ORTHOPAEDIC

SURGERY

15

Michiel G.J.S Hageman, MD, Stephanie J.E. Becker, MD, Arjan G.J. Bot, MD, Thierry Guitton, MD, PhD,

David Ring, MD, PhD, the Science of Variation Group.*

Orthopaedic Hand and Upper Extremity Service, Harvard Medical School, Massachusetts General

Hospital, Boston, MA, USA.

J Hand Surg Am. 2013 May;38(5):856-62.

CHAPTER 2

Variation in recommendation for

surgical treatment for compressive

neuropathy

16

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

ABSTRACT

Background It is our impression that there is substantial, unexplained variation in hand surgeon recommendations for treatment of peripheral mono-neuropathy. We tested the null hypothesis that specific patient and provider factors do not influence recommendations for surgery.Methods Using a web-based survey, hand surgeons recommended surgical or nonsurgical treatment for patients in two different scenarios. Six elements of the first scenario (symptoms, circumstances, mindset, diagnosis, objective testing, and expectations) had two possibilities that were each independently and randomly assigned to each rater. For the second scenario, two different scenarios were randomly assigned to each rater. Multivariable logistic regression sought factors associated with a recommendation for surgery.Results A total of 186 surgeons of the Science of Variation Group completed a survey regarding recommendation of surgery for two different patients based on clinical scenarios. Recommendations for surgery did not vary significantly according to provider characteristics.For the various elements in scenario 1, recommendation for surgery was more likely for patients who were self-employed and continued to work and who had objective electro-diagnostic abnormalities. For the two vignettes used in scenario 2, a recommendation for surgery was associated with abnormal electrophysiology.Conclusions The findings of this study suggest that – at least in a survey setting – surgeons prefer to offer peripheral nerve decompression to patients with abnormal electrophysiology, particularly those with effective coping strategies.

INTRODUCTION

Pathophysiology and demographics cannot explain the substantial geographic

variation in rates of surgery. Cholecystectomy for silent gallstones and lumbar

spine surgery are known examples of small area variation in surgical rates.1,2

The rates and types of surgical treatment of distal radius fractures in the

United States Medicare population also demonstrate small area variation based

primarily on sex and age.3

17

Another area of variation is differences in opinion. Some of the most debatable

issues in hand and upper extremity surgery are the least scientific, meaning

the least objectively verifiable. For instance, diagnosis and treatment of radial

tunnel and pronator syndromes (diagnoses defined in part by normal electro-

diagnostic testing) varies substantially: some surgeons make these diagnoses

and offer surgical treatment routinely, whereas others consider these diagnoses

illness constructs (an illness that exists only because we agree to behave as if

it exists) and do not find them useful for patients. There is debate regarding

whether idiopathic median neuropathy at the carpal tunnel should be considered

a syndrome (a constellation of symptoms and signs) or an objectively verifiable

median neuropathy at the carpal tunnel (pathophysiology/ disease). For example,

some surgeons offer patients with normal electro-diagnostic testing surgery

based on symptoms alone, and others do not. Workers’ compensation, litigation,

and less specific symptoms are associated with worse outcomes from surgery,4-10

but it is unclear whether these factors affect recommendations for surgery.10,11

In this study, we surveyed a large group of hand surgeons regarding

recommendations for surgery for peripheral nerve disorders. We tested the

null hypothesis that specific patient and provider factors do not influence

recommendations for surgery.

MATERIAL AND METHODS

A total of 235 surgeons of the Science of Variation Group were asked to complete

a survey regarding recommendation of surgery for two different patients based

on clinical scenarios.

The Science of Variation Group is an international collaboration

of practicing surgeon observers that studies variation in the definition,

interpretation, classification, and treatment of human illness. Collaborative

authorship and scientific curiosity and camaraderie are the only incentives for

participation.

The study protocol was approved by our institutional review board.

Incentives, other than acknowledgment as part of the Science of Variation

Group, were not provided. Of the total 235 surgeons, 186 completed the survey

(73%). (Table 1).

Evaluation

After logging in to the website, each observer entered his demographic and

18

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

professional information: sex, country or region of practice, years in independent

practice, supervision of trainees, and surgical subspecialty.

The observers were then presented with two scenarios and asked

whether they would recommend surgical treatment (yes or no). In the

first scenario (scenario 1), six elements of the scenario were randomized

independently. The following was a constant part of scenario 1: “A 55-year-

old woman, a journalist, presents with symptoms unresponsive to splinting,

medication, modification of activities, and hand therapy.” Afterward, information

about (1) symptoms, (2) circumstances, (3) mindset, (4) diagnosis, (5) objective

Table 1 Demographics n=186

n %

Sex Men 167 90 Women 19 10

Practice Asia 3 2 Canada 1 1 Europe 9 4 United Kingdom 4 2 United States 161 87 Other 8 4

Years In practice 0-5 55 30 6-10 38 20 11-20 56 30 21-30 37 20

Supervise Yes 122 66 No 64 34

Specialization Hand and wrist 180 97 Other 6 3

19

testing, and (6) expectations was presented. Each of these elements had two

alternatives, A and B, which were randomly assigned. The alternatives for

symptoms were (1A) symptoms consist of numbness of the thumb, index,

middle, and ring fingers that occasionally wake her from sleep, are present

most mornings, and also occur with hair drying, driving, and other bent-wrist

activities; and (1B) symptoms consist of forearm and wrist pain with typing and

occasional numbness of the entire hand. The alternatives for circumstances

were (2A) she is not currently working; she has an open workers’ compensation

claim that is in dispute; and she has hired a lawyer to represent her; and (2B) she

is self-employed and continues to work. The alternatives for mindset were (3A)

she can type for only 10 to 15 minutes at a time, and the pain is excruciating;

and (3B) nothing. The alternatives for diagnosis were (4A) a diagnosis of

carpal tunnel syndrome is made, and (4B) a diagnosis of pronator syndrome

is made. The alternatives for objective testing were (5A) electro-diagnostic

testing demonstrates motor and sensory nerve dysfunction consistent with

the diagnosis, and (5B) electro-diagnostic testing is normal. The alternatives for

expectations were (6A) her primary care doctor sent her to you for surgery; and

(6B) none.

The observers were then presented with a second scenario and asked

whether they would recommend surgical treatment (yes or no). In this case, one

of two complete scenarios was randomly assigned:

Scenario A: A 55-year-old woman, a journalist, presents with symptoms

unresponsive to splinting, medication, modification of activities, and hand

therapy. Symptoms consist of forearm and wrist pain with typing. She is

tender over the lateral side of the proximal forearm. She can type for only 10

to 15 minutes at a time. The pain is excruciating. She is not currently working.

She has an open workers’ compensation claim that is in dispute. She has hired

a lawyer to represent her. Electro-diagnostic testing demonstrates dysfunction

of the radial nerve in the proximal forearm. Her primary care doctor diagnosed

radial tunnel syndrome and sent her to you for surgery.

Scenario B: A 55-year-old woman, a journalist, presents with symptoms

unresponsive to splinting, medication, modification of activities, and hand

therapy. Symptoms consist of forearm and wrist pain with typing. She is

tender over the lateral side of the proximal forearm. She is self-employed and

continues to work. Electro-diagnostic testing is normal.

20

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

Table 2 Bivariate analysis for demographics combined with scenario 1 and 2 n=186

Independent parameters

Scenario 1 Sex

Location

Years in practise

Supervision

Specialization

Scenario 2 A Sex

Location

Years in practise

Supervision

Specialization

MenWomen

United StatesOther

0-56-1011-2021-30

YesNo

Hand and wristOther

MenWomen

United StatesOther

0-56-1011-2021-30

YesNo

Hand and wristOther

8714

9110

33202721

6734

974

38

5

637

129

139

2716

421

805

7015

22182916

5530

832

57

5

755

19101914

4220

611

0.073

0.12

0.64

0.054

0.54

0.54

0.68

0.94

0.60

0.79

Recommended TreatmentNon-operative Operative P-Value n n

continue >

21


Scenario 2 B Sex

Location

Years in practise

Supervision

Specialization

Scenario 2A + B Sex

Location

Years in practise

Supervision

Specialization

MenWomen

United StatesOther

0-56-1011-2021-30 YesNo

Hand and wristOther

MenWomen

United StatesOther

0-56-1011-2021-30

YesNo

Hand and wristOther

576

954

21112011

4122

459

9511

1591

33203320

6838

1015

153

315

3843

126

018

728

1070

22182317

5426

791

0.40

0.80

0.11

0.90

0.27

0.93

0.74

0.87

0.63

0.19


continued table 2

22

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

Statistical analysis

Our primary outcome measure for both scenarios was the decision to operate

or not. The association between the outcome measures, demographics, and the

elements of the scenarios were investigated using the chi-square test. All factors

that had p < 0.10 in bivariate analysis were inserted in a backward, stepwise

(likelihood) binary logistic regression to find the factors associated with the

decision to operate.

RESULTS

Observer demographics are recorded in Table 1. Region of practice, years in

practice, and type of specialization did not influence recommendation for

surgery for either scenario (Table 2).

Scenario 1 and treatment

In scenario 1, objective testing (electro-diagnostic tests demonstrate nerve

dysfunction) and circumstances (self-employed and continues to work) were

significantly associated with the decision to operate. Sex and supervision

satisfied the criteria for entry in the backward logistic regression. In the final

model with predictors for recommending surgery, only patients who were

self-employed and continued to work (OR 2.9, 95% CI 1.4 to 5.9) and who had

objective electro-diagnostic abnormalities (OR 12, 95% CI 5.8 to 25) were

retained in the model (Nagelkerke R2 0.38, p < 0.001) (Table 3).

Scenario 2 and treatment

The demographics of the surgeons were not significantly associated with

the decision to operate in either scenario A (patient is not working and has

abnormal electro-diagnostic testing), scenario B (self-employed and a normal

electro-diagnostic test), or both scenarios combined (Table 2). Surgeons who

were randomized to scenario A were significantly more likely to operate

(p < 0.001) (Table 4).

DISCUSSION

Our impression is that there is substantial variation in recommendations for

peripheral nerve surgery, particularly for more debatable diagnoses, such as

pronator syndrome. The findings of this study suggest that surgeons tend to

23

Table 3 Bivariate analysis - scenario 1 and 2 n=186


49

52

57

44

52

49

4952

25

76

55

46

49

36

32

53

45

40

4144

66

19

36

49

0.21

0.011

0.84

0.97

<0.001

0.10


Scenario 1 Symptoms

Symptoms consist of numbness of the thumb, index, long and ring fingers which occasionally waker her from sleeping.Symptoms consist of forearm and wrist pain with typing and occasional numbness of the entire hand.

CircumstancesShe is not working open WC claim in dispute, has hired a lawyer.Self-employed and continues to work.

MindsetShe can only type for 10-15 minutes at a time. The pain is excruciating.None.

DiagnosisA diagnosis of CTS is made.A diagnosis of pronator syndrome is made.

Objective testingEMG demonstrates motor and sensory nerve dysfunction consistent with the diagnosis.EMG is normal.

ExpectationsHer primary care doctor sent her to you for surgery.None.

continue >

24

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

continued table 3


43

63

62

18

<0.001


Scenario 2 Type A

A 55-year-old woman journalist presents with symptoms unresponsive to splinting, medication, modification of activities and hand therapy. Symptoms consist of forearm and wrist pain with typing. She is tender over the lateral side of the proximal forearm. She can only type for 10-15 minutes at a time. The pain is excruciating. She is not currently working. She has an open worker’s compensation claim that is in dispute. She has hired a lawyer to represent her. Electrodiagnotic testing is normal. Her primary care doctor diagnosed her with radial tunnel syndrome and sent her to you for surgery.

Type BA 55-year-old woman journalist presents with symptoms unresponsive to splinting, medication, modification of activities and hand therapy. Symptoms consist of forearm and wrist pain with typing. She is tender over the lateral side of the proximal forearm. She is self-employed and continues to work. Electrodiagnotic testing demonstrates dysfunction of the radial nerve in the proximal forearm.

25

Table 4 Logistic regression predicting likelihood of suggesting operative treatment n=186

Category / Variable*

Scenario 01 - Would you suggest operative treatment?

2.9

12

1.4

5.8

5.9

25

0.38

95.0% CI for OR Odds Ratio Lower Upper Nagelkerke R2

SociologicalSelf-employed and continues to workEMG demonstrates motor and sensory

Objective testing

nerve dysfunction consistent with the diagnosis

offer peripheral nerve decompression in patients who continue to work and

have abnormal electrophysiology, and that abnormal electrophysiology takes

priority. This was an unexpected finding, because if surgeons offer surgery

based primarily on reliable and valid objective testing, there should be limited

variation in treatment recommendations. On the other hand, our best statistical

models could account for only 38% of the variation in recommendations for

surgery, suggesting that other, unmeasured factors (such as reimbursement,

rapport, etcetera) account for substantial variation.

This study should be interpreted in light of the fact that the observers are

predominantly in academic practice and are largely from the United States. This

study was also limited to the designed scenarios, which may have influenced

the responses. One shortcoming is that physical examination findings were not

one of the 6 components of the scenario presented. This was intentional, based

on the need to limit the number of variables for statistical reasons and the fact

that the subjective aspect of physical examination findings is more difficult to

represent in a Web-based scenario; however, we plan to develop methods for

looking into this. Short patient scenario’s fail to capture all the elements of the

patient-physician interaction, but even studies that simplify this

complex interaction (which all such studies must inevitably do, to some extent)

can provide useful feedback. In our opinion, the elements that we studied likely

26

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

represent the critical aspects of the clinical encounter and form a basis from

which all surgeons and all patients can evaluate their particular circumstance.

Another shortcoming is that what surgeons say in surveys may or may not

reflect what they say to actual patients. Finally, although a large number of

surgeons participated, the group may be different in many ways from the

average hand surgeon and may not represent the full variation in opinions

among hand surgeons worldwide. We also want to mention that our reference

to compressive neuropathy is speculative, particularly for the more debatable

diagnoses such as pronator syndrome and radial tunnel syndrome.

In the typical patient-provider interaction, symptom intensity and

magnitude of disability (which can be grouped together as illness behavior)

seem to have a more substantial influence on treatment choices than

pathophysiology (e.g. electro-diagnostic abnormalities). For instance, if a patient

with carpal tunnel syndrome and electro-diagnostic abnormalities experiences

a decrease in symptoms with nonsurgical treatment (less illness), then neither

the patient nor the surgeon find surgical treatment appealing, even though

there is some evidence that carpal tunnel syndrome is a progressive disorder

that can cause permanent nerve damage (in other words, there has been no

change in the disease). Conversely, many surgeons will offer surgical treatment

on the basis of symptoms alone, even if electrophysiological testing is normal

(more illness than expected based on objective pathophysiology). In contrast

to a disease like type 1 diabetes mellitus – in which patients must take their

insulin even if they are feeling well or risk diabetic ketoacidosis – the idea of

surgery to stop pathophysiology and prevent nerve damage in carpal tunnel

syndrome seems counterintuitive to most patients and surgeons, and illness

behavior seems at times to have a relatively greater influence on treatment than

pathophysiology does. The results of our survey suggest that – at least among

a group of surgeons that is largely academic and primarily based in the United

States – these impressions may be incorrect because, on average, surgeons

indicated that they rely more on electro-diagnostic testing than we expected.

Another unexpected finding was the absence of an influence based

on the referring provider’s expectations. Anecdotally, depending on the

reimbursement setting, some of us have the impression that some surgeons

are concerned about satisfying the primary care doctor, without whose referrals

their practice would be less busy and, therefore, less profitable.

Finally, we expected to find a difference between debatable (e.g.

pronator and radial tunnel syndromes) and widely accepted (e.g. carpal tunnel

27

syndrome) diagnoses, but this did not have a measurable influence on surgeon

recommendations. Perhaps these diagnoses are less debated than we thought,

or perhaps these diagnoses are accepted if there are measurable electro-

diagnostic abnormalities.

According to this survey, patient circumstances (e.g. workers’

compensation in dispute) and objective testing have more influence than

symptoms, mindset, diagnosis, and expectations on surgeon recommendations

for peripheral nerve decompression. Additional studies are needed to identify

factors that influence actual day-today decision-making, the sources of

variation, and how informed, shared decision-making, using techniques such

as decision aids, might reduce this variation, increase patient satisfaction, and

provide optimal health care as resourcefully as possible.

*From the Orthopaedic Hand and Upper Extremity Service, Massachusetts General Hospital,

Boston, MA. The Science of Variation Group: Abhijeet L. Wahegaonkar, Aida E. Garcia G, Alan

Schefer, Alberto Pérez Castillo, Andrew L. Terrono, Andrew W. Gurman, MD, T. Apard, Barry Watkins,

Asif Ilyas, Bernard F. Hearon, MD, Brian P.D. Wills, MD, Bruce I. Wintman, Carrie Swigart, Catherine

Spath, Cesar Dario Oliveira Miranda, Charles A. Goldfarb, Charles Cassidy, Charles Metzger, Charles

Eaton, Chris Wilson, Christopher J. Walsh, Christopher J. Wilson, Christopher M. Jones, Colby Young,

Craig A. Bottke, MD, Daniel A Osei, D. Kay Kirkpatrick, Daniel Polatsch, David E. Tate, Jr, David L.

Nelson, MD, David M. Kalainov, David M. Lamey, MD, Doug Hanel, David M. Ostrowski, MD, David

R. Miller, Desirae M. McKee, David Ruchelsman, Ekkehard Bonatz, Eon K. Shin, Eric P. Hofmeister,

Evan S. Fischer, MD, F. ThomasD. Kaplan, C.H. Fernandes, Jamie E. Forigua, Fidel Ernesto Cayón

Cayón, Frank J. Raia, Frank L. Walter, Gary K. Frykman, MD, Gary M. Pess, MD, Gary R. Kuzma, Georg

M. Huemer, Gregory Dee Byrd, George W. Balfour, Gladys Cecilia Zambrano Caro, German Ricardo

Hernandez, Gregory DeSilva, H. Brent Bamberger, DO, H.W. Grunwald, Hal MccUtchan, Harrison

Solomon, MD, Hervey L. Kimball, J.E.B. Stuart, InesC.Lin, Jack Choueka, James G. Reid, James M.

Boler, Jay Pomerance, Jeff W. Johnson, Jeffrey Yao, Jim Calandruccio, Jennifer B.Green, Jennifer

Moriatis Wolf, Jessica A. Frankenhoff, Jerome W. Oakey, Jochen Fischer, John Howlett, John Jiuliano,

John M. Erickson, John McAuliffe, John P. Evans, John Taras, Jorge G. Boretto, Jonathan Isaacs, Jose

A. Ortiz, Jr, José Fernando Di Giovanni, Jose Nolla, Joshua M. Abzug, Julie Adams, L.C. Bainbridge,

Karel Chivers, Karl-Josef Prommersberger, Kevin J. Malone, Kendrick Lee, Lawrence S. Halperin, MD,

Lawrence Weiss, Leon Benson, Lewis B. Lane, Lior Paz, Lisa Lattanza, M. Jason Palmer, Louis Catalano

III, Marc J. Richard, Marco Rizzo, Martin Boyer, Maurizio Calcagni, Megan M. Wood, Michael

Baskies, Michael W. Grafe, Michael Behrman, Michael Jones, Michael Quinn, Michael Nancollas,

Michael W. Kessler, Miguel A. Pirela-Cruz, Milan M. Patel, NaquiraEscobarLuisFelipe, NeilG.Harness,

MD,NgoziM.Akabudike, NicholasJ.Horangic, Oleg M. Semenkin, Nicky L. Leung, Patrick T. McCulloch,

Patrick W. Owens, Paul A. Martineau, Paul Bettinger, Paul Guidera, Peter E. Hoepfner, Prasad

Sitaram, Peter H. DeNoble, Peter Jebson, Philip Coogan, Phani Dantuluri, R. Glenn Gaston, MD,

Ralf Nyszkiewicz, Ralph M. Costanzo, Ramon de Bedout, Randy Hauck, Renato M. Fricker, Richard

28

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

S. Gilbert, Richard L. Hutchison, Richard W. Barth, Rick Papandrea, Robert M. Szabo, Robert R. L.

Gray, Ross Nathan, Rozental, Sander Spruijt, Russell Shatford, Ryan Klinefelter, Samir Sodha, Ryan

P. Calfee, Sanjeev Kakar, Saul Kaplan, Scott F. Duncan, Scott Mitchell, Seth Dodds, Sidney M. Jacoby,

Stephen A. Kennedy, Stanley Casimir Marczyk, Stephen W. Dailey, MD, Steve Kronlage, Steven Alter,

Steven Beldner, Steven J. McCabe, Stuart M. Hilliard, Thomas J. Fischer, Taizoon Baxamusa, C. Taleb,

Thomas F. Varecka, Theresa Wyrick, Timothy G. Havenhill, Todd Siff, Victoria D. Knoll, Vipul P. Patel,

W. Arnnold Batson, Warren C. Hammert, and William J. Van Wyk, MD.

REFERENCES



and 2009: perspectives from Medicare provider claims. J Am Coll Radiol 2012;9:474-9.3. Fanuele J, Koval KJ, Lurie J, Zhou W, Tosteson A, Ring D. Distal radial fracture treatment:

what you get may depend on your age and address. The Journal of bone and joint surgery

American volume 2009;91:1313-9.4. Cassidy JD, Carroll LJ, Cote P, Lemstra M, Berglund A, Nygren A. Effect of eliminating

compensation for pain and suffering on the outcome of insurance claims for whiplash injury.

N Engl J Med 2000;342:1179-86.5. de Beer J, Petruccelli D, Gandhi R, Winemaker M. Primary total knee arthroplasty in patients

receiving workers’ compensation benefits. Can J Surg 2005;48:100-5.6. Harris I, Mulford J, Solomon M, van Gelder JM, Young J. Association between compensation

status and outcome after surgery: a meta-analysis. Jama 2005;293:1644-52.7. MacKenzie EJ, Bosse MJ, Kellam JF, et al. Early predictors of long-term work disability after

major limb trauma. J Trauma 2006;61:688-94.8. Scuderi C, Khedroo F. Herniation of the intervertebral disc; diagnosis, treatment and resume

of follow-up study. J Int Coll Surg 1955;23:194-204.9. Wong JY. Time off work in hand injury patients. J Hand Surg Am 2008;33:718-25.10. Day CS, Alexander M, Lal S, et al. Effects of workers’ compensation on the diagnosis and

surgical treatment of patients with hand and wrist disorders. J Bone Joint Surg Am

2010;92:2294-9.11. McGlaston TJ, Kim DW, Schrodel P, Deangelis JP, Ramappa AJ. Few insurance-based differences

in upper extremity elective surgery rates after healthcare reform. Clin Orthop Relat Res

2012;470:1917-24.

29

CHAPTER 3

The factors influencing the

decision-making of operative

treatment for proximal humeral

fractures

Hageman MG, Jayakumar P, King JD, Guitton TG, Doornberg JN, Ring D; Science of Variation Group.



J Shoulder Elbow Surg. 2015 Jan;24(1):e21-6.

30

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

ABSTRACT

Background The factors influencing the decision-making of operative treatment for fractures of the proximal humerus are debated. We hypothesized that there is no difference in treatment recommendations between surgeons shown radiographs alone and those shown radiographs and patient information. Secondarily, we addressed (1) factors associated with a recommendation for operative treatment, (2) factors associated with recommendation for arthroplasty, (3) concordance with the recommendations of the treating surgeons, and (4) factors affecting the inter-rater reliability of treatment recommendations.

Methods A total of 238 surgeons of the Science of Variation Group rated 40 radiographs of patients with proximal humerus fractures. Participants were randomized to receive information about the patient and mechanism of injury. The response variables included the choice of treatment (operative vs non-operative) and the percentage of matches with the actual treatment.

Results Participants who received patient information recommended operative treatment less than those who received no information. The patient information that had the greatest influence on treatment recommendations included age (55%) and fracture mechanism (32%). The only other factor associated with a recommendation for operative treatment was region of practice. There was no significant difference between participants who were and were not provided with information regarding agreement with the actual treatment (operative vs non-operative) provided by the treating surgeon.

Conclusions Patient information – older age in particular – is associated with a higher likelihood of recommending non-operative treatment than radiographs alone. Clinical information did not improve agreement of the Science of Variation Group with the actual treatment or the generally poor inter-observer agreement on treatment recommendations.

31

INTRODUCTION

The role of operative treatment for fractures of the proximal humerus is

debated. Surgery is considered for approximately one in five patients, but there

is no consensus on which fractures benefit from surgery or which procedure

to perform.1 The data to date are limited and inconclusive.1,2 A recent Cochrane

review found no statistically significant difference between operative and

non-operative treatment regarding patient-reported functional scores and

EuroQoL results at 1 year from 3 randomized control trials with a total of

153 participants.1 However, compared to non-operative treatment, operative

treatment had superior EuroQoL scores at two years of follow-up in two

randomized control trials with a total of 101 participants.1

Among a small group of surgeons at two level-1-trauma centers, Okike et

al3 identified younger age, operative treatment of other musculoskeletal injuries,

Arbeitsgemeinschaft f ür Ostesynthesefragen (AO) classification, translation-type

displacement, associated glenohumeral dislocation, and surgeon subspecialty

(upper extremity specialists were more likely to operate than traumatologists) as

factors associated with operative intervention. The use of arthroplasty rather than

internal fixation was associated with a higher Charlson score and more severe

Neer and AO classifications.3-5 Many of these factors relate to the radiographic

appearance of the fracture, whereas some relate to patient or surgeon factors.

We were curious about the factors that influence agreement between

surgeons on treatment recommendations and the factors that lead a surgeon to

recommend operative treatment and type of surgery (i.e. fixation vs arthroplasty). We

used the Science of Variation Group (SVOG), an international Web-based collaborative

of practicing surgeons, to test the primary null hypothesis that there is no difference

in treatment recommendations regarding operative vs non-operative treatment

between surgeons shown radiographs alone and those shown radiographs and

patient information such as age, sex, hand dominance, and fracture mechanism.

Secondarily, we addressed (1) factors associated with a recommendation for

operative treatment, (2) factors associated with recommendation for arthroplasty,

(3) concordance with the recommendations of the treating surgeons, and (4) factors

affecting inter-rater reliability of treatment recommendations.

MATERIALS AND METHODS

We asked the surgeons of the SOVG to complete a survey regarding the

recommendation of operative or non-operative treatment for a series of

32

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

proximal humeral fractures. The SOVG is an international collaboration of fully

trained surgeon observers that studies variation in the definition, interpretation,

classification, and treatment of human illness. Collaborative authorship,

scientific curiosity, and camaraderie are the only incentives for participation.

Participating members viewed the radiographs of 20 fractures of the

proximal humerus treated operatively and 20 treated non-operatively. These were the

radiographs used by the surgeon caring for the patient and were not standardized.

Participants were randomized to receive information about the patient, including sex,

age, American Society of Anesthesiologists (ASA) classification, and hand dominance,

and mechanism of injury or not, in a 1-to-1 allocation.

Evaluation

The 40 proximal humeral fractures were selected from a separate case-control

study in which 66 patients, 33 treated operatively and 33 treated non-

operatively, were matched for fracture type, age, sex, and ASA classification. The

treating surgeons classified those fractures as 2-part surgical neck fractures in

7 pairs of patients, 3-part fractures in 9 pairs, and 4-part fractures in 4 pairs.

Participants viewed anterior-posterior and lateral radiographs.

Each participant provided demographic and professional information:

sex, world region of practice, years in independent practice, supervision of

trainees, and surgical subspecialty. Each observer was asked two questions for

each set of images: (1) Would you recommend surgery? And, if so, (2) What is

your preference: open reduction and internal fixation, percutaneous pinning,

or arthroplasty? In addition, the observers who received patient information

were asked a third question: What information was most influential? They could

choose from the following answers: (1) patient characteristics (sex, age, ASA,

hand dominance), (2) fracture mechanism, (3) other (Table 1).

Of the 238 surgeons who completed the survey, 130 were randomized to

receive information and 108 were randomized to receive radiographs alone. The

cohorts were comparable (Table 2).


The response variables included the choice of treatment (operative vs

non-operative) and the percentage of matches with the actual treatment.

Associations between response variables and categorical explanatory variables

were assessed using X2 tests. Factors with p < 0.10 in bivariate analysis were

entered into a multiple logistic regression analysis.

Table I Patient characteristics

Parameters

Mean

60

65

56 n 9

24

1216

5

2310

1716

1018

41 7

26 9

24

Range

13 - 86

19 - 90

15 - 83

Range

33 - 92

40 - 94

0 - 29

7 - 89

P value

0.74

0.42

<0.01

0.59

0.22

0.40

0.62

0.40

0.52

0.59

SD

15

15

17 %

1436

1824

7.6

3515

2624

1628

52

1139

1436

SD

13

12

8.1

25 %

3317

2917

4.5

3020

2921

2220

7.8

7.642

3317

Nonoperative groupn=33

Operative fixation group n=33

Time

Age at injury (Y) Age at follow-up (Y)

Duration injury until operative treatment

Duration of follow-up (Months)

Sex

MenWomen

Neer classification

2 Part3 Part4 Part

Fracture mechanism

Type 1 (Slipped)Type 2 (High energy trauma)

Dominant arm effected

YesNo

Comorbidity scale

ASA 1ASA 2ASA 3ASA 4

Diabetes

YesNo

Smocking

YesNo

Sex

Neer Classification

Fracture mechanism

Dominant arm effected

Comorbidity scale

Diabetes

Smocking

Mean

59

62

14

35 n

1122

1911

3

2013

1914

1413

5

528

2211

34

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

The percentage of agreement (1) with other observers and (2) with the

original treating surgeon was calculated for surgeons who received information

or not, and the κ-multi-rater measure was also measured. The κ values were

interpreted with use of the guidelines proposed by Landis and Koch.6,7

An a priori power analysis indicated that a cohort of 200 surgeons

randomized equally to review radiographs with or without patient information

would provide 80% power to detect a mean difference of 0.10 in inter-observer

reliability based on κ-multi-rater, assuming a pooled standard deviation of 0.25

(moderate effect size: 0.10/0.25 = 0.40) using a parametric Z-test with a two-

tailed α-level of 0.05 and assuming an underlying normal distribution in the

patient population.

RESULTS

Participants who received patient information recommended operative

treatment less than those who received no information (61% vs 66%; p < 0.01;

Table 3). The only other factor associated with a recommendation for operative

treatment was region of practice: participants from Asia were more likely than

participants from Canada to recommend surgery (72% vs 51%; p < 0.01; Table 2).

In multivariable analysis, the only factor associated with a recommendation

of non-operative treatment was receiving clinical information in addition to

radiographs, which explained 10% of the variation in recommendation for non-

operative treatment (odds ratio, 4.3;r2 = 0.10; p < 0.01).

Participants provided with patient information were more likely to

recommend arthroplasty (24% vs 17%; p < 0.01) and less likely to recommend

open reduction and internal fixation (69% vs 76%; p < 0.01; Table 4). The patient

information that had the greatest influence on treatment recommendations

included age (55%) and fracture mechanism (32%; Table 5).

There was no significant difference between participants provided with

information and participants not provided information regarding agreement

with the actual treatment (operative vs non-operative) provided by the treating

surgeon (65% vs 67% on average; p = 0.11). Agreement with the actual treatment

provided by the treating surgeon varied significantly by location of practice

(p < .01; Table 6).

Inter-observer agreement regarding recommendations for operative

treatment was poor (average, 0.008; range, 0.007 to –0.008; Supplementary (Table 7).

κ

35

Table 2 Demographics of the participants n=228

Parameters

SexMenWomen

Location of practiceAsiaAustraliaCanadaEuropeUnited KingdomUnited States of AmericaOther

Years In practice0-56-1011-2021-30

Supervise TraineesYesNo

Fractures per year0-56-1011-20>20

SpecializationGeneral orthopaedicsOrthopaedic traumatologyShoulder and elbowHand and wristOther

Radiographs plus additional informationn (%)

122 (52) 8 (3.2)

9 (3.6) 4 (1.6) 7 (2.8) 40 (17) 5 (2.0) 59 (24) 12 (4.8)

41 (18) 29 (12) 34 (15) 26 (11)

119 (50) 11 (5.6)

11 (4.8) 32 (13) 41 (17) 50 (21)

7 (2.8) 58 (24) 27 (11) 32 (15) 6 (2.4)

Radiographs without additional informationn (%)

101 (42) 7 (2.8)

2 (0.8) 2 (0.8) 4 (1.6) 27 (12) 0 62 (25) 11 (4.4)

40 (17) 18 (7.3) 33 (13) 17 (6.9)

95 (39) 13 (5.6)

14 (5.6) 23 (9.7) 34 (14) 37 (15)

6 (2.4) 35 (14) 27 (11) 39 (16) 1 (0.4)

Table 3 Bivariable analysis about the participants’ recommendation

Operative treatment

%

64

Operative treatment

6166

6365

72695162546173

62626664

6365

63616565

6263626660

Operative treatment

%

46

Operative treatment

4350

4647

58533045374560

44444949

4647

47414947

4246444950

Non-operative

%

38

Non-operative

3934

3735

28314938463927

38383436

3735

37393535

3837383440

Non-operative

%

54

Non-operative

5750

5453

42477055635540

56565151

5453

53595153

5854565150

Min - Max

0.23-1

P-Value

<0.01

0.8

<0.01

0.24

0.48

0.57

0.41

Min - Max

0-1

P-Value

<0.01

0.93

<0.01

0.25

0.54

0.18

0.62

Operative treatment Nonoperative treatment

OverallObservers

Information group

Radiographs + informationRadiographs - additional information

SexMenWomen






Table 4 Comparison of percentage of arthroplasty as prefered osteosynthesis for the information and non information group

Parameters Arthroplasty

ORIF

Pin

95% CI

Information groupRadiographs plus additional informationRadiographs without additional information

Parameter


Parameter


mean24%17%

mean69%76%

mean5.9%6.8%

P-Value

<0.01

P-Value

<0.01

P-Value

0.54

Upper

0.1

Upper

0.75

Upper

0.02

Lower

0.04

Lower

0.71

Lower

-0.04

Table 5 Additional information used by the information group for decision making regarding the preferred treatment

SexAgeASAHand dominanceFracture mechanismOther

n9

7220124268

%7.255158.93252

37

Table 6 Bivariable analysis comparing the recommendation of the participant and the actual treatment

Operative treatment

%

80

Operative treatment

7783

8082

85867179717986

79778179

8078

78788081

8279808069

Operative treatment

%

46

Operative treatment

4350

4647

58533045374560

44444849

4647

47414947

4246444950

Non-operative

%

20

Non-operative

2317

2020

15142921292114

21231921

2022

22222019

1821202031

Non-operative

%

54

Non-operative

5750

5453

42477055635540

66665151

5453

52595153

5854575150

Min - Max

0.2-1.0

P-Value

<0.01

0.45

<0.01

0.39

0.52

0.62

0.23

Min - Max

0.0-1.0

P-Value

<0.01

0.93

<0.01

0.26

0.93

0.14

0.62

Operative treatment Nonoperative treatment

OverallObservers

Information group

Radiographs + informationRadiographs - information

SexMenWomen






39

Table 7 Agreement measurement

123456789

10111213141516171819202122232425262728293031323334353637383940

ASE

0.2390.0220.9860.4720.0310.1370.986

*0.0670.2390.0590.5540.1310.3410.0760.9860.6880.4720.2140.2390.131

*0.1030.0930.2910.0450.0840.1440.2550.6880.0340.0380.0840.3140.4720.2020.2720.2720.6880.108

Z-Value

-0.038-0.423-0.009-0.019-0.297-0.067-0.009

*-0.136-0.038-0.154-0.017-0.070-0.027-0.121-0.009-0.013-0.019-0.043-0.038-0.070

*-0.089-0.098-0.032-0.206-0.109-0.064-0.036-0.013-0.269-0.244-0.109-0.029-0.019-0.045-0.034-0.034-0.013-0.085

Kappa

-0.009-0.009-0.009-0.009-0.009-0.009-0.009

*-0.009-0.009-0.009-0.009-0.009-0.009-0.009-0.009-0.009-0.009-0.009-0.009-0.009

*-0.009-0.009-0.009-0.009-0.009-0.009-0.009-0.009-0.009-0.009-0.009-0.009-0.009-0.009-0.009-0.009-0.009-0.009

P-Value

0.9690.6720.9930.9850.7660.9470.993

*0.8920.9690.8770.9870.9440.9790.9030.9930.9890.9850.9660.9690.944

*0.9290.9220.9750.8370.9130.9490.9710.9890.7880.8070.9130.9770.9850.9640.9730.9730.9890.932

Radiographs plus additional information Radiographs without additional information

Z-Value

-0.048-0.308-0.013-0.030-0.137-0.638-0.071-0.056-0.098-0.050-0.114-0.018-0.040-0.022-0.059-0.053

*-0.008-0.689-0.087-0.074

*-0.091-0.104-0.023-0.056-0.567-0.046-0.037-0.021-0.121-0.211-0.058-0.035

*-0.058-0.118-0.155-0.017-0.056

P-Value

0.9620.7580.9890.9760.8910.5230.9440.9550.9220.9600.9090.9860.9680.9820.9530.958

*0.9940.4910.9310.941

*0.9280.9170.9820.9550.5710.9630.9710.9830.9030.8330.9540.972

*0.9540.9060.8770.9860.956

Kappa

-0.007-0.007-0.008-0.008-0.008-0.008-0.008-0.008-0.008-0.008-0.008-0.008-0.008-0.008-0.008-0.008

*-0.008-0.008-0.008-0.008

*-0.008-0.008-0.008-0.008-0.008-0.008-0.008-0.008-0.008-0.008-0.008-0.008

*-0.008-0.008-0.008-0.008-0.008

ASE

0.1530.2420.5580.2490.5510.1180.1060.1340.7680.1510.6640.4220.1930.3470.1310.148

*0.9880.1140.9050.107

*0.8670.7580.3460.1410.0140.1740.2200.3780.6650.0380.1400.232

*0.1400.6910.5250.4760.146

*=Based on negative interobserver variability no information was given regarding agreement.

40

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

DISCUSSION

The recommendations for managing proximal humeral fractures vary

substantially. Recent studies demonstrate the poor levels of reliability in the

treatment of these injuries.8,9 We were interested in the relative influence of

patient information and surgeon characteristics on the decision-making process

in treating proximal humeral fractures.

This study should be considered in light of its shortcomings. Most

regions were represented by small numbers of observers, and the findings may

not be representative of the average surgeon in those regions. The low κ-values

may be due to the κ-paradox: when the prevalence of an outcome is low, it could

cause an imbalance that generates a lower κ than one might expect based

on the agreement. Also, our study did not include fracture classification as an

explanatory variable, although inter-observer reliability of fracture classification

is limited amongst orthopedic surgeons.8,10 In addition, a consecutive selected

case series would have limited spectrum bias. However, a relatively even

distribution between operatively and non-operatively treated patients is needed

to avoid the κ-paradox. Readers should also keep in mind that observers were

shown the unstandardized anterior-posterior and lateral radiographs used by

the treating surgeons to direct management, thereby reflecting daily practice.

Some surgeons use other radiographic views and computed tomography scans,

including 3-dimensional reconstructions, more routinely in the decision process.

We found that patient information – older age in particular – is

associated with a higher likelihood of recommending non-operative treatment

than radiographs alone. This is consistent with the observations of Okike et al.3

We also identified regional variations in treatment recommendations. This was

in concordance with earlier published reports that demonstrated wide regional

variations for surgical treatment adjusted for age, sex, and race in populations

of elderly patients with proximal humeral fractures ranging from 0% in many

regions to almost 70% in Duluth, Minnesota, USA..11,12 Of interest, the sex,

level of experience in years in practice, number of proximal humeral fractures

treated per year, and specialization did not have a significant influence on the

decision to operate. This is in contrast to other studies, which report shoulder

and upper extremity specialists are more likely to choose operative intervention

than general orthopedic trauma specialists.11 The effect of specialty may vary

in specific centers with small samples of each type of surgeon, but our larger,

broader cohort suggests that specialty has relatively little influence.

The provision of clinical information also influenced the recommended

41

type of operative treatment, with arthroplasty (as with non-operative treatment)

favored by surgeons who received information about the patients. This

suggests that – with current biases – older, more infirm, and inactive patients

are less likely to be treated operatively and are more likely to be treated with

arthroplasty if they do have operative treatment.

Clinical information did not improve agreement of the SOVG

participants with the actual treatment or the generally poor inter-observer

agreement on treatment recommendations. The poor agreement may be

unreliable due to the κ paradox. A study by Petit et al12 documented moderate

inter-observer agreement on the surgical management (non-operative,

closed manipulation and reduction, open reduction and internal fixation, and

hemiarthrosplasty) of 38 proximal humeral fractures among 8 fellowship-

trained orthopedic surgeons.

Treatment recommendations for proximal humeral fractures are

influenced by patient information – older age in particular – but most of the

variation in recommendations remains unaccounted for. The highly variable

and inconsistent influence of patient factors on surgeon recommendations

belies variations in surgeon preferences and values that are likely at the root of

the substantial treatment variations documented in this and other studies. We

speculate that greater involvement of the patient in decision-making is likely to

decrease variation across caregivers.

CONCLUSION

Patient information – older age in particular – is associated with a higher

likelihood of recommending nonoperative treatment than radiographs alone.

Clinical information did not improve agreement of the SOVG with the actual

treatment or the generally poor inter-observer agreement on treatment

recommendations.

42

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

REFERENCES

1. Handoll HH, Ollivere BJ, Rollins KE. Interventions for treating proximal humeral fractures in

adults. Cochrane Database Syst Rev 2012;12:CD000434.2. Misra A, Kapur R, Maffulli N. Complex proximal humeral fractures in adults--a systematic

review of management. Injury 2001;32:363-72.3. Okike K, Lee OC, Makanji H, Harris MB, Vrahas MS. Factors associated with the decision for

operative versus non-operative treatment of displaced proximal humerus fractures in the

elderly. Injury 2013;44:448-55.4. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic

comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373-

83.5. Neer CS, 2nd. Displaced proximal humeral fractures. I. Classification and evaluation. The

Journal of bone and joint surgery American volume 1970;52:1077-89.6. Cohen J. A Coefficient of agreement for nominal scales. Educ Psychol Meas 1960;20:37-46.7. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics

1977;33:159-74.8. Bruinsma WE, Guitton TG, Warner JJ, Ring D, Science of Variation G. Interobserver reliability of

classification and characterization of proximal humeral fractures: a comparison of two and

three-dimensional CT. J Bone Joint Surg Am 2013;95:1600-4.9. Foroohar A, Tosti R, Richmond JM, Gaughan JP, Ilyas AM. Classification and treatment of

proximal humerus fractures: inter-observer reliability and agreement across imaging

modalities and experience. Journal of orthopaedic surgery and research 2011;6:38.10. Brorson S, Rasmussen JV, Frich LH, Olsen BS, Hrobjartsson A. Benefits and harms of locking

plate osteosynthesis in intraarticular (OTA Type C) fractures of the proximal humerus: a

systematic review. Injury 2012;43:999-1005.11. Bell JE, Leung BC, Spratt KF, et al. Trends and variation in incidence, surgical treatment,

and repeat surgery of proximal humeral fractures in the elderly. J Bone Joint Surg Am

2011;93:121-31.12. Sporer SM, Weinstein JN, Koval KJ. The geographic incidence and treatment variation of

common fractures of elderly patients. J Am Acad Orthop Surg 2006;14:246-55.

43

CHAPTER 4

How surgeons make decisions when

the evidence is inconclusive

Hageman MG, Guitton TG, Ring D; Science of Variation Group.



J Hand Surg Am. 2013 Jun;38(6):1202-8.

44

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

ABSTRACT

Background To address the factors that surgeons use to decide between two options for treatment when the evidence is inconclusive.

Methods We tested the null hypothesis that the factors surgeons use do not vary by training, demographics, and practice. A total of 337 surgeons rated the importance of seven factors when deciding between treatment and following the natural history of the disease and twelve factors when deciding between two operative treatments using a 5-point Likert scale between “very important” and “very unimportant.”

Results According to the percentages of statements rated very important or somewhat important, the most popular factors influencing recommendations when evidence is inconclusive between treatment and following the natural course of the illness were “works in my hands,” “familiarity with the treatment,” and “what my mentor taught me.” The most important factors when evidence shows no difference between 2 surgeries were “fewer complications,” “quicker recovery,” “burns fewer bridges,” “works in my hands” and “familiarity with the procedure.” Europeans rated “works in my hands” and “cheapest/most resourceful” of significantly greater importance and “what others are doing,” “highest reimbursement,” and “shorter procedure” of significantly lower importance than surgeons in the United States. Observers with fewer than ten years in independent practice rated “what my mentor taught me,” “what others are doing” and “highest reimbursement” of significantly lower importance compared to observers with ten or more years in independent practice.

Conclusions Surgeons deciding between two treatment options, when the evidence is inconclusive, fall back to factors that relate to their perspective and reflect their culture and circumstances, more so than factors related to the patient’s perspective, although this may be different for younger surgeons.

45

INTRODUCTION

Evidence-based medicine has been defined as “the conscientious, explicit, and

judicious use of current best evidence in making decisions about the care

of individuals and populations.” In practice, this involves an integration of

individual clinical expertise with the best available external clinical evidence

from systematic research.1,2 Patients and health care providers look to scientific

evidence to help guide their medical decisions.

The “Evidence-Based Guidelines” from the American Academy of

Orthopaedic Surgeons have been largely inconclusive for lack of evidence.3

Well-designed, prospective, randomized controlled trials frequently show

no difference or a small and possibly unimportant difference between two

treatments.4,5 Clinical evidence in 2010 classified 2,500 common treatments

as 51% having insufficient evidence, 23% likely to be beneficial, 7% requiring

trade-offs between benefits and harms, 5% unlikely to be beneficial, 3% likely

to be ineffective or harmful, and 11% being clearly beneficial.6 How do health

care providers decide which option to recommend to their patients when the

evidence is inconclusive?

This study addresses the factors that surgeons deciding between two

options fall back to when the data are inconclusive. Specifically, we tested

the null hypothesis that the top fallback principles do not vary by training,

demographics, and practice.

MATERIALS AND METHODS

Using an institutional review board–approved protocol, we asked the

400 surgeons of the Science of Variation Group to complete a survey about

decision-making in the face of inconclusive evidence, and 337 participated.

The Science of Variation Group is an international collaboration of fully trained

surgeon observers that studies variation in the definition, interpretation,

classification, and treatment of human illness. Collaborative authorship,

scientific curiosity, and camaraderie are the only incentives for participation.

Evaluation

The observers were first asked to enter their demographic and professional

information: sex, country or world region of practice, years in independent

practice, supervision of trainees, and surgical subspecialty. Next, the observers

were given the following context:

46

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

“The American Academy of Orthopaedic Surgeons Evidence-Based Guidelines have been largely inconclusive for lack of evidence. It is difficult to show a difference in a well-designed prospective randomized, controlled trial – most will show little or no difference between treatments. Therefore it is important to decide – before starting the study – what our fallback will be. How do we decide between treatment options when the data are either insufficient or otherwise inconclusive?”

In this context, participants were asked to rate the importance of seven

factors when deciding between operative treatment and palliative treatment

(e.g. the natural history of the disease) and twelve factors when deciding

between two operative treatments (Table 1), with a comment section for listing

additional factors. The ratings were based on a 5-point Likert-scale between

very important and very unimportant. The statements were developed by

brainstorming. One author created a list, and the other authors edited until all

authors felt that the list covered all potential fallback options.


Categorical data were presented as frequencies and percentages. The

statements were ranked from highest to lowest by adding the percentages of

the very important and somewhat important (Figs. 1,2). The write-in answers

were grouped by subject. In addition, the Likert-scale was translated to an

ordinal scale from 2 (very important) to -2 (very unimportant), and the mean on

each scale across the entire sample was calculated. We analyzed the influence

of nationality, years in practice, fractures treated per year, and specialization

on preferred fallbacks. The subcategory “years in practice” was dichotomized

to less than or equal to ten years and more than ten years of experience to

facilitate analysis. For continuous variables, we used a Mann Whitney U test to

compare two groups and a Kruskal-Wallis test for multiple groups. We evaluated

differences between subgroups with the Mann Whitney U-test.

Observer demographics

The demographics for the 338 respondents are listed in (Table 2).

47

Table 1 Geographic difference and factors

Variable

The importance, that a given treatment is better than the course of the illness without treatment

The importance when comparing two surgeries for a given problem:

Mean

Works in my handsFamiliarity with treatmentWhat my mentor taught meDo something vs. Do nothingWhat others are doingPatients requiring the procedureHighest reimbursement

Fewer complicationsQuick recoveryBurns fewer bridgesWorks in my handsFamiliarity with procedureCheapest/ most resourcefulShorter procedureAesthics: Smaller or fewer scarsWhat my mentor taught meWhat others are doingPatients requiring the procedureHighest reimbursement

US

-1.3 -0.84 -0.58 -0.29 -0.11 0.28 1.2

-1.8 -1.4 -1.4 -1.3 -1.2 -0.69 -0.51 -0.46 -0.39 -0.11 0.26 1.08

EU

-1.04 -1.04 -0.68 -0.53 -0.40 0.03 0.78

-1.86 -1.52 -1.15 -1.15 -1.21 -0.47 -0.76 -0.50 -0.53 -0.23 0.12 0.63

P-Value

0.02 0.07 0.38 0.14 0.02 0.06 <0.01

0.21 0.15 0.02 0.13 1.00 0.05 0.02 0.74 0.23 0.32 0.24 <0.01

Mean. Dif.

-0.26 0.20 0.10 0.24 0.29 0.25 0.40

0.07 0.12 -0.21 -0.15 0.00 -0.22 0.25 0.04 0.15 0.12 0.14 0.45

RESULTS

Statement rating

According to the percentages of statements rated very important or somewhat

important, the most popular fallbacks when evidence cannot demonstrate that

a given treatment is better than following the natural course of the illness are

noted in Figure 1. The top fallbacks when evidence shows no difference between

two surgeries are noted in Figure 2.

48

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

Very important

Somewhat important

Neutral

Somewhat unimportant

Very unimportant

Works in my hands

Familiarity with treatment

What my mentor taught me

Do something vs. Do nothing

What other are doing

Patients are requesting the procedure

Hight reimbursement

0 10 20 30 40 50 60 70 80 90 100

Figure 1

0 10 20 30 40 50 60 70 80 90 100

Fewer complications

Quick recovery

Burns fewer bridges

Works in my hands

Familiarity with procedure

Cheapest/most resourceful

Shorter procedure

Aesthics: Smaller or fewer scars


What other are doing

Patients requesting

Highest reimbursement

Figure 2

49

Table 2 Demographics n=338

n %

Sex Men 306 91 Women 32 9 Location of practice Asia 19 6 Australia 6 2 Canada 18 5 Europe 92 27 United Kingdom 11 3 United States of America 174 52 Other 17 5 Years In practice 0-5 106 31 6-10 72 21 11-20 102 30 21-30 57 17 Supervise Yes 279 83 No 58 17 Fractures per year 0-5 61 18 6-10 71 21 11-20 109 32 >20 99 29 Specialization General orthopaedics 21 6 Orthopaedic traumatology 120 36 Shoulder and elbow 54 16 Hand and wrist 129 38 Other 13 4

50

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

United States versus Europe

Using the average values on the numeric conversion of the Likert-scale,

Europeans rated “works in my hands,” “burns fewer bridges,” and “cheapest/

most resourceful” of significantly greater importance and “what others are

doing,” “highest reimbursement,” and “shorter procedure” of significantly lower

importance than surgeons in the United States (Table 1).

Years in practice

Observers with ten or fewer years in independent practice rated “what my

mentor taught me,” “what others are doing,” and “highest reimbursement” of

significantly lower importance compared to observers with more than ten years

in independent practice (Table 3).

Orthopedic specialty

General orthopedists rated “what my mentor taught me” of greater importance

than orthopedic traumatologists and hand and wrist surgeons. In addition,

general orthopedists rated “what others are doing” of greater importance than

shoulder and elbow surgeons and hand and wrist surgeons (Table 4).

Write-in answers

The most common write-in answers were “best available outcome/evidence-

based” (14 surgeons), “common sense and risk for patients” (5 surgeons), and

“shared decision-making or patient’s opinion” (4 surgeons) (Table 5).

DISCUSSION

Because evidence-based medicine is an amalgamation of individual clinical

expertise and best available evidence, the question arises, what is the basis for

provider recommendations when the best evidence is inconclusive? We found

that the most popular factors that surgeons use to make recommendations

when evidence is inconclusive relate primarily to the surgeon’s perspective

(e.g. “works in my hands,” “familiarity with the treatment,” “what my mentor

taught me”) rather than the patient’s perspective (e.g. “doing something vs

doing nothing,” “patients are requesting the procedure”). Exceptions include

“fewer complications” and “quicker recovery,” which benefit both the surgeon

and the patient. Highest reimbursement was also rated relatively unimportant,

particularly in Europe but across all countries and regions.

51

Table 3 Difference in experience

Variable


The importance when comparing two surgeries for a given problem:

Mean

Works in my handsFamiliarity with treatmentWhat my mentor taught meDo something vs. Do nothingWhat others are doingPatients requiring the procedureHighest reimbursement

Fewer complicationsQuick recoveryBurns fewer bridgesWorks in my handsFamiliarity with procedureCheapest/ most resourcefulShorter procedureAesthics: Smaller or fewer scarsWhat my mentor taught meWhat others are doingPatients requiring the procedureHighest reimbursement

<10

-1.2 -0.99 -0.84 -0.52 -0.35 0.14 0.92

-1.8 -1.4 -1.3 -1.3 -1.3 -0.61 -0.69 -0.47 -0.65 -0.25 0.13 0.71

>10

-1.3 -0.92 -0.47 -0.39 -0.01 0.27 1.1

-1.9 -1.5 -1.3 -1.3 -1.2 -0.65 -0.54 -0.47 -0.31 -0.05 0.30 1.1

P-Value

0.56 0.50 <0.01 0.34 <0.01 0.27 0.08

0.072 0.47 0.83 0.58 0.25 0.71 0.13 0.95 <0.01 0.058 0.14 <0.01

Mean. Dif.

0.05 -0.06 -0.37 -0.13 -0.34 -0.13 -0.21

0.09 0.05 -0.02 0.04 -0.90 0.04 -0.14 -0.010 -0.33 -0.20 -0.16 -0.36

This study should be interpreted in light of the fact that the 337

participating surgeons may not be representative of the average surgeon,

because many surgeons in the surveyed group are in academic practice. Also,

important options such as “I share the decision with the patient” were not

offered because it was our intention to study the recommendation of the

surgeon before accounting for the patient’s preferences. Finally, there is evidence

that incentives such as reimbursement can have a subconscious influence that

may not be accounted for by this survey.7

52

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

Table 4 Difference in specialty - Post hoc tukey test

Variable


The importance of the following fractors when comparing two surgeries for a given problem


What others are doing


General Orthopaedics



Shoulder and elbow

vs

Orthopaedic Traumatology

Hand and Wrist

Shoulder and elbow

Hand and wrist

Other

Hand and wrist

Other

Other

P-Value

0.033

0.033

0.033

0.033

0.048 0.011

0.003

0.037

Mean. Dif.

-0.62

-0.65

-0.62

-0.65

-0.92 -0.71

-1.20

-0.82

That health care providers fall back to their personal preferences based

on experience is no surprise.8 On the other hand, it is notable that factors

related to quality, safety, and efficiency such as “cheapest/most resourceful,”

“shorter procedure,” and “what others are doing” (in the sense of diminished

unwarranted variation) were rated relatively unimportant. The fact that

Europeans rated “cheapest/more resourceful” significantly more important

than Americans may reflect the prevalence of national health care in Europe,

leading to a greater awareness of the management of limited resources. In

53

Table 5 Others

Category


The importance when comparing two surgeries for a given problem

Best available outcome/ evidence basedCommen senseCombination of common sense and evidenceBasic principlesCost of the treatmentCost of the treatment vs. Doing nothingDo whats best for patiëntDoing something means to me, having the knowledgeExplore Complementary and Alternative Medicine inGuidelines/ ProtocolI would choose the procedure for myselfI have always taughtInnovationLeast bias from industry fundingRisk for patiëntPatients comfortpatients’ perception of the condition-/ shared decision making

Best available outcome/ evidence basedAvailable equimentConsesus local colleaguesCommen senseEquimentLong lasting effectLess painfulPatiënt functional demandsPhilosophical or legal medical aspects p.e.(americShared decision makingTrack recordWhat is “popular”

Total

14 5 1 1 1 1 1 1 1 3 2 1 1 1 5 1 4

10 1 1 2 1 2 1 1 1 1 1 1

54

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

contrast, surgeons from the United States rated “what others are doing,”

“highest reimbursement,” and “shorter procedure” more important than

European surgeons. It is not clear whether these factors relate most to quality

and efficiency or marketing and profitability of health care in a for-profit system,

or both.

Less experienced surgeons placed significantly less importance on “what

my mentor taught me,” “what others are doing,” and “highest reimbursement.”

This might reflect a change in mindset as the emphasis is placed on evidence

and as we continue to address the rising costs of health care.

The write-in answers revealed that surgeons prefer to fall back to the

“best available outcome/evidence-based” even when the scenario is that the

evidence is inconclusive. Patient-centered care/shared decision-making was also

mentioned, which is entirely applicable. The involvement of patients in decision-

making is particularly important when the evidence is inconclusive. Decision

aids (independent structured guides, either written, video, or web-based)

have been shown to decrease decision conflict and, for some illnesses, use of

resources.9-12 These merit additional study.

In other words, rather than studying the surgeon’s recommendation

before accounting for the patient’s preferences, it might have been preferable

for our survey to include the option of following the patient’s preference when

evidence is inconclusive. On the other hand, we have an obligation to consider

resources, safety, simplicity, consistency, efficiency, practicality, optimism,

and patient self-management as important goals in and of themselves, and

this is part of the expertise that we share with our patients. Patients look

to their surgeons for expertise regarding the optimal fallback options when

evidence is inconclusive. Perhaps – on the basis of the results of this survey

study – surgeons will be motivated to develop consensus regarding the fallback

principles that best support optimal health.

*The Science of Variation Group: A. Lee Osterman, A.B. Spoor, A.L. van der Zwan, Abhay Shrivastava,

Abhijeet L. Wahegaonkar, Aida E. Garcia G., M.A. Aita, Alberto Pérez Castillo, Alexander Marcus,

Amy Ladd, Andrew L. Terrono, Andrew P. Gutow, Andrew Schmidt, AngelaA. Wang, Anica Eschler,

Anna N. Miller, Annette K.B. Wikerøy, Antonio Barquet, April D. Armstrong, Arie B. van Vugt,

Asheesh Bedi, Ashok K. Shyam, Augustus D. Mazzocca, Axel Jubel, Babst Reto H., Betsy M. Nolan,

Bob Arciero, Van den Bremer, Brent Bamberger, Bret C. Peterson, Brett D. Crist, Brian J. Cross, Brian

L. Badman, C. Noel Henley, Carl Ekholm, Carrie Swigart, Chad Manke, Charalampos Zalavras,

55

Charles A. Goldfarb, Charles Cassidy, Charles Cornell, Charles L. Getz, Charles Metzger, Chris

Wilson, Christian Heiss, Christian J. Perrotto, Christopher J. Wall, Christopher J. Walsh, Christos

Garnavos, Chunyan Jiang, Craig Lomita, Craig M. Torosian, Daniel A. Rikli, Daniel B. Whelan, Daniel

C. Wascher, Daniel Hernandez, Daniel Polatsch, Daphne Beingessner, Darren Drosdowech, David

E. Tate, Jr, David Hak, David J. Rowland, David M. Kalainov, David Nelson, David Weiss, Desirae

M. McKee, D. F. P. van Deurzen, Donald Endrizzi, Konul Erol, Joachim P. Overbeck, Wolfgang Baer,

Eckart Schwab, Edgardo Ramos Maza, Edward Harvey, Edward K. Rodriguez, Elisabeth Prelog-

Igler, Emil H. Schemitsch, Eon K. Shin, Eric P. Hofmeister, F. Thomas D. Kaplan, F.J.P. Beeres, Fabio

Suarez, C.H. Fernandes, Fidel Ernesto Cayón Cayón, Filip Celestyn Dolatowski, Fischmeister Martin,

Francisco Javier Aguilar Sierra, Francisco Lopez-Gonzalez, Frank Walter, Franz Josef Seibert, Fred

Baumgaertel, Frede Frihagen, P.C. Fuchs, Georg M. Huemer, George Kontakis, George S. Athwal,

George S.M. Dyer, George Thomas, Georges Kohut, Gerald Williams, German Ricardo Hernandez,

Gladys Cecilia Zambrano Caro, Grant Garrigues, Greg Merrell, Gregory DeSilva, Gregory J. Della

Rocca, Gustavo Regazzi, Gustavo Borges Laurindo de Azevedo, Gustavo Mantovani Ruggiero, H. J.

Helling, Hal Mc Utchan, Hans Goost, Hans J. Kreder, PaulaM. Hasenboehler, Howard D. Routman,

Huub van der Heide, I. Kleinlugtenbelt, Iain McGraw, Ian Harris, Ibrahim Mohammad Ibrahim, Ines

C. Lin, A. Iossifidis, J. Andrew I. Trenholm, J. Carel Goslings, J. Michael Wiater, Jack Choueka, Jaimo

Ahn, James Kellam, Jan Biert, Jay Pomerance, Jeff W. Johnson, Jeffrey A. Greenberg, JeffreyYao,

JeffryT.Watson, JenniferL. Giuffre, JeremyHall, Jin-YoungPark, Jochen Fischer, Joel Murachovsky,

JohnHowlett, JohnMcAuliffe, John P. Evans, JohnTaras, Jonathan Braman, JonathanL. Hobby,

Jonathan Rosenfeld, Jorge Boretto, Jorge Orbay, Jorge Rubio, JoseA. Ortiz, Jr, Joseph Abboud,

Joseph M. Conflitti, Joseph P.A.M. Vroemen, Julie Adams, J.V.Clarke, K.Kabir, Karel Chivers, Karl-Josef

Prommersberger, Keith Segalman, Kendrick Lee, Kevin Eng, Kimberlly S. Chhor, K.J. Ponsen, Kyle

Jeray, l. Marsh, L.M.S.J. Poelhekke, Ladislav Mica, Lars C. Borris, Lawrence Halperin, Lawrence Weiss,

Leon Benson, Leon Elmans, Leonardo Alves de Mendonca, Jr, Leonardo Rocha, Leonid Katolik, Lisa

Lattanza, Lisa Taitsman, Lob Guenter, Louis Catalano III, Luis Antonio Buendia, Luke S. Austin, M.

Jason Palmer, M.R. de Vries, M.R. Krijnen, Maarten W.G.A. Bronkhorst, Mahmoud I. Abdel-Ghany,

M.A.J. Van de Sande, Marc Swiontkowski, Marco Rizzo, Marcus Lehnhardt, Marinis Pirpiris, Mark

Baratz, MarkD. Lazarus, MartinBoyer, Martin Richardson, Matej Kastelec, Matt Mormino, Matthew

D. Budge, Matthias Turina, Megan M. Wood, Michael Baskies, Michael Baumgaertner, Michael

Behrman, Michael Hausman, Michael Jones, Michael LeCroy, Michael Moskal, Michael Nancollas,

Michael Prayson, Michael W. Grafe, Michael W. Kessler, Michel P.J. Van den Bekerom, Mike

Mckee, Milind Merchant, Minos Tyllianakis, Naquira Escobar Luis Felipe, NealC. Chen, NeilSaran,

NeilWilson, Nicholas L. Shortt, Niels Schep, Nigel Rossiter, N.G. Lasanianos, Nikolaos Kanakaris,

Noah D. Weiss, Norah M. Harvey, P.V. van Eerten, Parag Melvanki, Patrick T. McCulloch, Paul A.

Martineau, Paul Appleton, Paul Guidera, Paul Levin, Peter Giannoudis, Peter J. Evans, Peter Jebson,

Peter Kloen, Peter Krause, Peter R.G. Brink, J.H. Peters, Philip Blazar, Philipp N. Streubel, Porcellini,

Prashanth Inna, S. Prashanth, PunitaV. Solanki, QiugenWang, M. Quell, R. Bryan Benafield, Jr, R.

Haverlag, R. W. Peters, Rajat Varma, Ralf Nyszkiewicz, Ralph M. Costanzo, Ramon de Bedout, Ashish

S. Ranade, Raymond Malcolm Smith, Reid Abrams, Renato M. Fricker, Reza Omid, Richard Barth,

Richard Buckley, Richard Jenkinson, Richard S. GIlbert, Richard S. Page, Richard Wallensten, Robert

D. Zura, Robert J. Feibel, Robert R.L. Gray, Robert Tashijan, Robert Wagenmakers, Rodrigo Pesantez,

Roger van Riet, Rolf Norlin, Roman Pfeifer, Ronald Liem, Roy G. Kulick, Rozental, Rudolf W. Poolman,

Russell Shatford, Ryan Klinefelter, Ryan P. Calfee, Sam Moghtaderi, Samir Sodha, Sander Sprujt,

Sanjeev Kakar, Saul Kaplan, Schandelmaier, Scott Duncan, Sebastian Kluge, Sebastian Rodriguez-

Elizalde, SergioL. Checchia, Sergio Rowinski, Seth Dodds, Shep Hurwit, K. Sprengel, W.A.H. van der

Stappen, Steve Kronlage, Steven Beldner, StevenJ. McCabe, StevenJ. Morgan, StevenJ. Rhemrev,

56

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

StuartHilliard, Taco Gosens,Takashi Sasaki, C. Taleb, Tamir Pritsch, Theodoros Tosounidis, Theresa

Wyrick, Thomas DeCoster, Thomas Dienstknecht, Thomas G. Stackhouse, Thomas Hughes,

Thomas Wright, Thuan V. Ly, Timothy G. Havenhill, Timothy Omara, Todd Siff, Toni M. McLaurin,

Tony Wanich, Johannes M. Rueger, Frederico C.M. Vallim, Vani J. Sabesan, Vasileios S. Nikolaou,

Verhofstad, Victoria D. Knoll, Vidyadhar Telang, Vishwanath M. Iyer, Vispi Jokhi, W. Arnnold Batson,

W. Jaap Willems, Wade R. Smith, William Dias Belangero, J. Wolkenfelt, Yoram Weil.

REFERENCES

1. Guyatt G. Evidence-based medicine. ACP J Club 1991;114.2. Sackett DL, Straus S, Richard SR, Rosenberg W, Haynes RB. Evidence-based medicine: How to

practice and Teach EBM. London, Churchill Livingstone 2000.3. American Academy of Orthopaedic Surgeons Clinical Practice Guideline on the Diagnosis and

Treatment of Osteochondritis Dissecans Rosemont (IL). American Academy of Orthopaedic

Surgeons (AAOS); 2010.4. Gibbs L, Gambrill E. Evidence-based practice: Counterarguments to objections. . Resarch on

Social Work Practice 2002;12:452-76.5. Pawson R. Evidence Based Policy: In search of a method. Evaluation 2002;8:157-81.6. How much do we know. British Medical Journal 2010;Clinical Evidence 2010.7. Esposito TJ, Maier RV, Rivara FP, Carrico J. Why Surgeons Prefer Not to Care for Trauma

Patients. Arch Surg 1991;126:292-7.8. Thomas G, Pring R. Evidence-Based Practise in Education. Youblishercom 2004.9. Kennedy AD, Sculpher MJ, Coulter A, et al. Effects of decision aids for menorrhagia on

treatment choices, health outcomes, and costs: a randomized controlled trial. Jama

2002;288:2701-8.10. Ozanne EM, Annis C, Adduci K, Showstack J, Esserman L. Pilot trial of a computerized decision

aid for breast cancer prevention. Breast J 2007;13:147-54.11. Slover J, Shue J, Koenig K. Shared decision-making in orthopaedic surgery. Clin Orthop Relat



57

CHAPTER 5

Do pre-visit expectations correlate

with satisfaction of new patients

presenting for evaluation with an

orthopaedic surgical practice?

Hageman MG, Briet JP, Bossen JK, Blok RD, Ring D, Vranceanu M.



Clin Orthop Relat Res. 2014 Apr;39(9):11999-014.

58

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

ABSTRACT

Background Patient satisfaction is associated with increased compliance, improved treatment outcomes, and decreased risk of litigation. Factors such as patient understanding and psychological well-being are recognized influences on satisfaction. Less is known about the relationship between pre-visit expectations and satisfaction. Questions/purposes (1) Are there correlations among pre-visit expectations, met expectations, and patient satisfaction? (2) What are the categories of expectations, and which one(s) correlate with satisfaction?

Methods 86 new patients presenting to a hand surgery practice of a tertiary referral hospital with 70% direct primary care referrals, mostly with elective concerns, indicated their pre-visit expectations (Patient Intention Questionnaire [PIQ]). Immediately after the visit, the same patients rated the degree to which their pre-visit expectations were met (Expectation Met Questionnaire [EMQ]) and their satisfaction level (Medical Interview Satisfaction Scale). These tools have been used in primary care office settings and claim good psychometric properties, and although they have not been strictly validated for responsiveness and other test parameters, they have good face validity. We then conducted a multivariable backward linear regression to determine whether (1) scores on the PIQ; and (2) scores on the EMQ are associated with satisfaction.

Results Satisfaction correlated with met expectations (r = 0.36; p =0.001) but not with pre-visit expectations (r = 0.01, p=0.94). We identified five primary categories of pre-visit expectations that accounted for 50% of the variance in PIQ: (1) ‘‘Information and Explanation’’; (2) ‘‘Emotional and Understanding’’; (3) ‘‘Emotional Problems’’; (4) ‘‘Diagnostics’’; and (5) ‘‘Comforting’’. The only category of met expectations that correlated with satisfaction was Information and Explanation (r = 0.43; p=0.001).

Conclusions Among patients seeing a hand surgeon, met expectations correlate with satisfaction. In particular, patients with met expectations regarding information and explanation were more satisfied with their visit. Efforts to determine the most effective methods for conveying unexpected information warrant investigation.

59

INTRODUCTION

Patient satisfaction measures are increasingly used to evaluate the quality of

medical service.1 Patient satisfaction is associated with increased compliance,

improved treatment outcomes across a variety of medical settings2, decreased

risk of litigation3, and patient ratings of the quality of their care. Patient

satisfaction is affected by patient understanding of their own health and

psychological well-being.4 Socio-demographic factors can also affect patient

satisfaction.5-10

It is likely, however, that other factors – as yet unexplored – may

influence patient satisfaction with a medical encounter. Met expectations are

associated with better patient satisfaction in population surveys and primary

care settings.10-12 However, the relationship between pre-visit or pre-operative

expectations and satisfaction is inconsistent.11,13 A study among primary care

patients found that pre-visit expectations (whether they were realistic or

unrealistic) were not associated with satisfaction.11 Research among orthopaedic

patients undergoing surgery for lower back pain found that higher pre-visit

expectations of pain relief were associated with lower satisfaction, whereas

higher pre-visit expectations of improved function were associated with

higher satisfaction.13 Because of the inconsistencies across studies in terms of

the association between pre-visit expectations and satisfaction11,13 as well as

the paucity of research on met expectations outside of primary care settings,

we sought to evaluate the relationships among pre-visit expectations, met

expectations, and satisfaction with a hand surgery outpatient visit. Specifically,

we aimed to identify (1) patient pre-visit expectations (level and type) for a hand

surgery office visit; and (2) the association of pre-visit expectations and met

expectations with satisfaction.

This study attempts to answer the following questions: (1) Are there

correlations among pre-visit expectations, met expectations, and patient

satisfaction? (2) What are the categories of expectations, and which one(s)

correlate with satisfaction?


This was an observational cross-sectional study. Between September 2012 and

December 9, 2012, adult, English speaking patients presenting to the practice

of one of three orthopaedic hand surgeons (JJ, CM, DCR) for an initial evaluation

were invited to enroll under a protocol approved by our Human Research

60

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

Committee. The study was described in detail and the treating physician/study

staff obtained informed consent.

Participants/Study Subjects

One hundred two patients were enrolled in the study. Of these, six were

excluded as a result of lack of English proficiency, three declined participation

after enrollment, and seven patients did not complete the second part of the

questionnaire after their medical appointment, most claiming lack of time.

Analyses were done on a final sample of 86 patients (Table 1).

Before the medical encounter with the hand specialist, patients

completed the Patient Intentions Questionnaire (PIQ)14 and a demographics

and medical profile questionnaire. After the encounter patients, completed the

Expectations Met Questionnaire (EMQ)12 and the Medical Interview Satisfaction

Scale (MISS).12,15,16

Measurement tools

The PIQ 14 consists of 34 equally weighted statements measuring a patient’s pre-

visit expectations and specific goals for a primary care medical visit.12 Examples

include: ‘‘I want my GP [general practitioner] to understand the problem’’;

‘‘I want the GP to explain my emotional problems.’’ All items in the PIQ were

scored on a 3-point Likert-scale (agree, uncertain, or disagree). We modified the

term ‘‘GP’’ to ‘‘doctor’’ in the PIQ questions to match the study setting (Appendix

1 [Supplemental materials are available with the online version]). The PIQ score

represents the percentage of expectations endorsed before the visit divided by

the total potential pre-visit expectations.

The EMQ12 consists of the same 34 statements on the PIQ aimed to

determine if a patient’s expectations were met after the visit. For example,

‘‘The doctor understood the problem’’; ‘‘The doctor explained my emotional

problems.’’ Comparable with the PIQ, all items in the EMQ were scored on a

3-point Likert-scale (agree, uncertain, or disagree). The EMQ was scored as

percentage of met expectation per item in the PIQ as initially endorsed by the

patient.

The results of the PIQ and EMQ were divided into three groups according

to low pre-visit and met expectation (0%–35%), moderately and uncertain pre-

visit and met expectation (36%–80%), and highly pre-visit and met expectation

(81%–100%), consistent with previously developed methodology.10

The MISS15-17 includes 21 items measuring satisfaction with a medical

61

Table 1 Patients demographics n=86

Mean

4416

7.9

n4343

72

44012

487154421

14

568095511

333

4361 3

776

sd

162.9

2

%5050

87

55012

568165521

16

6590

106611

393

5071 4

907

Range

19-7710-222--10

Age, yEducation Overall health (sd)

SexMenWomen

RaceWhiteBlack or African AmericanAsianAmerican Indian or Alaskan NativeMore than one raceOther or unknown

Diagnosis Acute injuriesNon-specific arm painCarpal Tunnel SyndromeGanglionDequervain Trigger fingerDupuytrenOsteoarthritisOther

Work statusWorking full timeWorking part timeHomemakerRetiredUnemployed, able to workUnemployed, unable to workWorkers compensationCurrently on sick leave

Marital StatusSingleLiving with partnerMarriedSeparated/ DivorcedWidowed

PhysicianSurgeon 1Surgeon 2Surgeon 3

62

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

encounter (e.g. ‘‘The doctor seemed to take my problems seriously’’). All items

were scored on a 7-point Likert-scale from very unsatisfied to completely

satisfied. In the digital version of the MISS, question 6 (‘‘The doctor seemed

to be interested in me as a person’’) was constantly skipped for all patients as

a result of a mistake in how the questionnaire was adapted from the paper-

based questionnaire. This question was part of the Rapport subscale, which

is comprised of seven other similar questions, which represent this aspect

of the visit well (e.g. ‘‘The doctor seemed warm and friendly to me’’ and ‘‘The

doctor seemed to take my problems seriously’’). A mean satisfaction index score

was calculated by dividing the total satisfaction score by the total number of

answered questions.

The primary measures used in this study have good psychometric

properties as evidenced by internal consistency reliability a between 0.84 and

0.9710 as well as validation in patients with back pain18 and in primary care.10,12,15

Statistical Analysis

An a priori power analyses indicated that a sample of 84 patients total would

provide 80% statistical power with a= 0.05 for a moderate effect size of 0.5

based on an analysis of variance (ANOVA). Continuous data were presented as

means when normally distributed. When data were not normally distributed,

we reported the median with interquartile range. Mean imputation was used

to account for missing values. Four patients skipped one question in the PIQ and

four patients skipped one question in the EMQ. One patient missed one question

in the MISS questionnaire. To determine the categories of desired expectations

on the PIQ, we performed a factor analysis with the help of the statistical

orthogonal principal component analysis through the Varimax rotation. A

question was related to a specific factor if there was a loading of minimal 0.40

or more. This method was used and validated in prior research.10 We used the

Spearman correlations to test for correlation between continuous variables. The

strength of the correlation was interpreted by the following guidelines: small

strength (r = 0.10–0.29), medium strength (r = 0.30– 0.49), and large strength

(r = 0.50–1.0).19 We used ANOVA to test for differences in satisfaction by

categories of expectations met and by type of expectation on the PIQ. We

conducted a multivariable backward linear regression to determine whether (1)

score on the PIQ; and (2) score on the EMQ were associated with satisfaction.

We included all variables with p- 0.10 in bivariate analysis.

63

RESULTS

Correlations Among Pre-visit Expectations, Met Expectations, and Patient

Satisfaction

Satisfaction correlated with met expectations (r = 0.36, p= 0.001) but not with

pre-visit expectations (r = 0.01, p= 0.94). The best linear regression model for

greater satisfaction included met expectations alone and explained 27% of the

variance. Four (5%) patients had low pre-visit expectations, 74 (86%) moderate

pre-visit expectations, and eight (9%) had high pre-visit expectations.

The degree of met expectations was low in 4 patients (5%), moderate

in 33 (38%), and high in 49 (57%). Preliminary bivariate analysis identified

differences in satisfaction in patients with low, moderate, and high met

expectations.

Categories of Expectations and Correlations With Satisfaction

Factor analysis identified five primary categories of previsit expectation that

accounted for 50% of the variance in PIQ: (1) ‘‘Information and Explanation’’;

(2) ‘‘Emotional and Understanding’’; (3) ‘‘Emotional Problems’’;(4) ‘‘Diagnostics’’;

and (5) ‘‘Comforting’’. Cronbach’s a ranged from 0.76 to 0.90 indicating overall

good to excellent reliability for all factors. Patients’ goals for the visit with the

hand surgeon focused more on ‘‘Information and Explanation’’, ‘‘Comforting’’,

and ‘‘Diagnostics’’ than on ‘‘Emotional Understanding’’ and ‘‘Emotional

Problems’’ (Table 2). The only category of met expectations that correlated with

satisfaction was ‘‘Information and Explanation’’ (r = 0.43; p= 0.001)

(Table 3). Interestingly, among the pre-visit expectation categories, the category

‘‘Information and Explanation’’ was highly met, whereas the other four factors

were met to a moderate or low extent.

DISCUSSION

Patient satisfaction is an important measure, because it is associated with

increased compliance, improved treatment outcomes, and decreased risk of

litigation. Many factors play into satisfaction, including patient’s understanding

of their own health and patient’s rating of the quality of their care and perhaps

expectations. We therefore investigated how pre-visit and met expectations

affect satisfaction and looked for categories of expectations that influence

satisfaction. We found that high pre-visit expectations did not correlate with

64

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

satisfaction with a hand surgery outpatient visit but met expectations did.

Patients had the highest expectations about information and explanation

followed by diagnostics and comforting, both of which were endorsed more that

emotional support.

This study should be considered in light of its shortcomings. One

limitation of this study is that the PIQ was developed for primary care practice.

Little is known about the repeatability, responsiveness, and the floor/ceiling

effects. The clinical situation in a primary care practice may be different when

a patient is rating their primary doctor with whom they are quite familiar as

opposed to a specialist they have never met. Nevertheless, the high Cronbach’s

a values give us confidence in the methodology described by Williams et al12

using the factor analysis, which is also a reliable method in other settings,

including orthopaedic practices. Additional validation of these questionnaires in

an orthopaedic practice is merited. Another limitation is the absence of question

6 from the MISS, but we think there is sufficient overlap with other questions

evaluating rapport that this probably has little or no effect on the results.

In our study, there was no association between level of pre-visit

expectations and patient satisfaction. The association between pre-visit

Table 2 Desired and met expectations divided by categories

Factor desired

%

96

38

5.5

61

74

Factor not

desired and met

% 3

26

9

20

14

Factor desired and met

%

83

26

2

33

51

Factor not

desired and not

met% 1

37

86

19

12

Factor desired and not

met %

13

11

3

27

23

Total

%

100

100

100

100

100

Factor 1 (Information and explanation)

Factor 2 (Emotional and understanding)

Factor 3 (Emotional problems)

Factor 4 (Diagnostics)

Factor 5 (Comforting)

65

expectations and satisfaction appears to depend on setting, patient population,

and type of pre-visit expectations.2,11,13 The fact that the majority of patients

in this sample had moderate pre-visit expectations (few had low or high

expectations) may have limited our ability to test the association of pre-visit

expectations and satisfaction. The finding that met expectations correlate with

satisfaction in patients with upper extremity illness is consistent with prior

studies in other populations.10-12 For instance, satisfaction and expectations

were strongly correlated in studies of patients undergoing THA.20,21 This may be

a foregone conclusion because measures of met expectations and measures

of patient satisfaction may be assessing the same construct. Future research

should replicate these findings with a larger sample of patients, perhaps with

one or more diagnoses associated with a greater rate of high expectations.

As one might expect, the pre-visit expectations reported by patients

undergoing hand surgery focused more on ‘‘Information and Explanation’’,

‘‘Comforting’’, and ‘‘Diagnostics’’ than on ‘‘Emotional Understanding’’ and

‘‘Emotional Problems’’. The only category of pre-visit expectation that correlated

with satisfaction was ‘‘Information and Explanation’’. As a result, attempts to

improve patient satisfaction might focus on establishing appropriate pre-

Table 3 Correlation of percentage of met expectations with satisfaction

P-Value

<0.001

0.15

0.72

0.71

0.07

Pearson rho

0.43

0.15

0.04

0.04

0.19

Patient Satisfaction (MISS)

Percentage of met expectations

Factor 1 (Information and explanation)

Factor 2 (Emotional and understanding)

Factor 3 (Emotional problems)

Factor 4 (Diagnostics)

Factor 5 (Comforting)

66

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

visit expectations perhaps by corresponding directly with the primary care

doctor (‘‘curbside consult’’), providing evidenced-based information in an

understandable and meaningful form (e.g. decision aids) before the visit, and

even pre-visit triage and education.

It has been more difficult to determine factors associated with patient

satisfaction than factors associated with other aspects of the illness experience

such as symptoms and disability. Collective research suggests that satisfaction

relates to factors like patient understanding, depression, pain intensity22-24

as well as effective communication,25 but there is not a strong relationship

with pre-visit expectations. Given the sense of many physicians that pre-visit

expectations do seem to lead to disappointment, future research regarding

pre-visit expectations might benefit from a focus on a specific paradigm where

unrealistically high expectations are common while accounting for psychological

factors, effective communication skills, time spent waiting for the doctor, and

time spent with the doctor.

Acknowledgments: We thank Drs Chaitanya Mudgal and Jesse Jupiter for

allowing us to enroll their patients.

67

REFERENCES

1. Hudak PL, Wright JG. The characteristics of patient satisfaction measures. Spine (Phila Pa

1976) 2000;25:3167-77.2. Soroceanu A, Ching A, Abdu W, McGuire K. Relationship between preoperative expectations,

satisfaction, and functional outcomes in patients undergoing lumbar and cervical spine

surgery: a multicenter study. Spine (Phila Pa 1976) 2012;37:E103-8.3. Hickson GB, Clayton EW, Entman SS, et al. Obstetricians’ prior malpractice experience and

patients’ satisfaction with care. Jama 1994;272:1583-7.4. GE H, MA W, F H. Components and predictors of patient satisfaction. Br J Health Psychol

1996;1:65-85.5. Fox JG, Storms DM. A different approach to sociodemographic predictors of satisfaction with

health care. Soc Sci Med A 1981;15:557-64.6. Greene JY, Weinberger M, Mamlin JJ. Patient attitudes toward health care: expectations of

primary care in a clinic setting. Soc Sci Med Med Psychol Med Sociol 1980;14A:133-8.7. Hall JA, Feldstein M, Fretwell MD, Rowe JW, Epstein AM. Older patients’ health status and

satisfaction with medical care in an HMO population. Med Care 1990;28:261-70.8. Like R, Zyzanski SJ. Patient satisfaction with the clinical encounter: social psychological

determinants. Soc Sci Med 1987;24:351-7.9. Williams SJ, Calnan M. Key determinants of consumer satisfaction with general practice. Fam

Pract 1991;8:237-42.10. Zebiene E, Razgauskas E, Basys V, et al. Meeting patient’s expectations in primary care

consultations in Lithuania. Int J Qual Health Care 2004;16:83-9.11. Bowling A, Rowe G, McKee M. Patients’ experiences of their healthcare in relation to their

expectations and satisfaction: a population survey. Journal of the Royal Society of Medicine

2013;106:143-9.12. Williams S, Weinman J, Dale J, Newman S. Patient expectations: what do primary care

patients want from the GP and how far does meeting expectations affect patient

satisfaction? Fam Pract 1995;12:193-201.13. Iversen MD, Daltroy LH, Fossel AH, Katz JN. The prognostic importance of patient pre-

operative expectations of surgery for lumbar spinal stenosis. Patient Educ Couns

1998;34:169-78.14. Salmon P, J. Q. Patient’s intentions in primary care: measurement and preliminary

investigation. Psychol Health 1989;3:103-10.15. Kinnersley P, Stott N, Peters T, Harvey I, Hackett P. A comparison of methods for measuring

patient satisfaction with consultations in primary care. Fam Pract 1996;13:41-51.16. Meakin R, Weinman J. The ‘Medical Interview Satisfaction Scale’ (MISS-21) adapted for British

general practice. Fam Pract 2002;19:257-63.17. Wolf MHPSMJSA, Stiles W.B., . The medical interview Satisfaction Scale: Development of a

scale to measure patients’ perceptions on physician behaviour. J Behav Med 1978;1:391.18. Georgy EE, Carr EC, Breen AC. Back pain management in primary care: development and

validity of the Patients’ and Doctors’ Expectations Questionnaire. Quality in primary care

2013;21:113-22.19. Cohen JW. In: Hillsdale, ed. Statistical power anlays for behavioral sciences (2nd edn). NJ::

Lawrence Erlbaum Associates; 1988:79-81.20. Mancuso CA, Jout J, Salvati EA, Sculco TP. Fulfillment of patients’ expectations for total hip

arthroplasty. J Bone Joint Surg Am 2009;91:2073-8.

68

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

21. Mancuso CA, Salvati EA, Johanson NA, Peterson MG, Charlson ME. Patients’ expectations and

satisfaction with total hip arthroplasty. J Arthroplasty 1997;12:387-96.22. Archer KR, Castillo RC, Wegener ST, Abraham CM, Obremskey WT. Pain and satisfaction in

hospitalized trauma patients: the importance of self-efficacy and psychological distress. The

journal of trauma and acute care surgery 2012;72:1068-77.23. Lozano Calderon SA, Paiva A, Ring D. Patient satisfaction after open carpal tunnel release

correlates with depression. J Hand Surg Am 2008;33:303-7.24. O’Toole RV, Castillo RC, Pollak AN, MacKenzie EJ, Bosse MJ, Group LS. Determinants of patient

satisfaction after severe lower-extremity injuries. J Bone Joint Surg Am 2008;90:1206-11.25. Bartlett EE, Grayson M, Barker R, Levine DM, Golden A, Libber S. The effects of physician

communications skills on patient satisfaction; recall, and adherence. J Chronic Dis

1984;37:755-64.

69

CHAPTER 6

Carpal tunnel syndrome:

assessment of surgeon and patient

preferences and priorities for

decision-making

Hageman MG, Kinaci A, Ju K, Guitton TG, Mudgal CS, Ring D; Science of Variation Group.



J Hand Surg Am. 2014 Sep;39(9):1799-1804.e1.

70

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

ABSTRACT

Background This study tested the null hypothesis that there are no differences between the preferences of hand surgeons and those patients with carpal tunnel syndrome (CTS) facing decisions about management of CTS (i.e. the preferred content of a decision aid).

Methods 103 hand surgeons of the Science of Variation Group and 79 patients with CTS completed a survey about their priorities and preferences in decision-making regarding the management of CTS. The questionnaire was structured according the Ottawa Decision Support Framework for the development of a decision aid.

Results Important areas on which patient and hand surgeon interests differed included a preference for non-painful, non-operative treatment and confirmation of the diagnosis with electro-diagnostic testing. For patients, the main disadvantage of non-operative treatment was that it was likely to be only palliative and temporary. Patients preferred, on average, to take the lead in decision-making, whereas physicians preferred shared decision-making. Patients and physicians agreed on the value of support from family and other physicians in the decision-making process.

Conclusions There were some differences between patient and surgeon priorities and preferences regarding decision-making for CTS, particularly the risks and benefits of diagnostic and therapeutic procedures. Clinical relevance Information that helps inform patients of their options based on current best evidence might help patients understand their own preferences and values, reduce decisional conflict, limit surgeon-to-surgeon variations, and improve health.

INTRODUCTION

Decision aids (videos, web sites, or handouts that contained balanced

information about diagnostic and treatment options) can help patients

understand their values and preferences and more fully participate in decision-

71

making.1 The Ottawa Decision Support Framework (ODSF) is an evidence-based,

practical theory used to guide the development of decision aids. It uses a 3-step

process: measure the needs of patients and their providers, provide decision

support tailored to patients’ needs, and evaluate the decision-making process

and outcomes.1 The ODSF asserts that unresolved needs will affect decision

quality,2 which in turn can affect illness behavior, health outcomes, emotions,

and resource utilization.1,3

There are many misconceptions about carpal tunnel syndrome (CTS) and

its treatment. There are also many areas of debate including the role of electro-

diagnostic testing, the best operative technique, and the indications for surgery

for mild (normal electro-diagnostic testing) or severe (atrophy, static numbness)

disease. A decision aid could inform patients of the best available evidence and

ongoing areas of debate in order to limit the effect of both patient and surgeon

bias and improve the patient’s comfort and participation in the decision.

This study assessed the priorities and preferences of patients and

hand surgeons facing decisions about management of CTS. We tested the null

hypothesis that there are no differences in priorities and preferences of patients

with CTS and hand surgeons.


Using an institutional review board-approved protocol, we surveyed hand

surgeon members of the Science of Variation Group (SOVG) and 79 new patients

diagnosed with CTS after the first consultation with one treating physician

regarding factors that influence decision-making and their preferences about

decision aids. The patients were English speaking, 18 years or older, able to fill

out the questionnaire, and not pregnant with CTS eventually verified by electro-

diagnostic testing presenting between May 2012 and April 2013.

The study was described in detail to the patients, and the research

assistant obtained informed consent. One hundred three hand surgeon-

members of the SOVG completed the survey (Appendix A, available on the

Journal’s Web site at www.jhandsurg.org). The SOVG is an international

collaboration of hand surgeons. Incentives, other than acknowledgment as part

of the SOVG, were not provided. None of the surgeons were involved in the care

of the patients surveyed. After logging into the web site, each surgeon entered

identifying demographic and professional information: sex, country or region of

practice, years in practice, supervision of trainees, and surgical subspecialty. The

72

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

surgeons were then presented with an online survey based on the ODSF.4,5

91 patients were enrolled, but 1 patient was excluded for not being

able to navigate the online questionnaire and 11 patients declined participation.

The mean age of the 79 patients who completed the study was 55 years (SD = 16;

range, 20-90 y), and 29 patients (35%) were men (Appendix B, available on the

Journal’s Web site at www.jhandsurg.org).

Measurement tools

The survey was based on the ODSF. There is a general framework that measures

the following aspects of various treatment options: desirability; advantages

and disadvantages; probability of choosing; preferred way to arrive at a final

decision; who, if anyone, is usually involved in the decision-making process;

what would help to arrive at a final decision; ways to facilitate the decision-

making process; the type of information desired; and who should prepare the

information.

When surveying patients and caregivers with respect to a specific

disease, one simply inserts common diagnosis and treatment options into

the framework. For instance, for CTS we provided the widely used treatment

options of orthosis fabrication, corticosteroid injection, and surgery (Appendix C,

available on the Journal’s Web site at www.jhandsurg.org).


A post hoc power analysis showed that 103 subjects of the SOVG and 79 patients

with CTS with the observed effect size of 0.54 provided 93% power to detect a

significant difference using a 2-tailed Student t-test, setting alpha level at 0.05.

Continuous data were presented as the mean when normally distributed. The

Student t-test and chi-square test were used to assess the association between

continuous or categorical preferences and independent variables, such as

patient and surgeons. The Fisher exact test was used instead of the chi-square if

the sample sizes were smaller than 5.

RESULTS

Patients found all treatment options – corticosteroid injection in particular –

less desirable than surgeons did (Table 1). When citing advantages of treatment,

patients ranked “Does not involve surgery” highly, whereas surgeons considered

“No major risk or side effects” as most important. Surgeons also added many

73

Table 1 Comparison of the desirability and likability of the different treatment opportunities between patients and physicians

Comparison of the desirability of the different treatment options

How likely to choose the following treatment options

Table 2 Advantage of the treatment options

Patients

Patients

Physician

Physician

P-Value

P-Value

SplintingCorticosteroid injectionSurgery

SplintingCorticosteroid injectionSurgery

SplintingDoes not involve surgeryNo major risk or side effectsAbility to stop the treatment at any timeOther

Corticosteroid injectionDoes not involve surgeryNo major risk or side effectsOther

EMGIt can help confirm the diagnosisDocuments baseline nerve functionOther

Carpal tunnel release surgeryHighest succes rateNo major risk or side effectsOther

mean

2.8 1.8 2.5

3.4 2.4 4.2

n

4026

5

6

461214

7600

6760

sd

0.13 0.093 1.3

1.6 1.3 1.1

%

5234

6.5

7.8

641719

1000.00.0

928

0.0

sd

0.078 0.066 0.88

0.97 1.4 0.094

%

2765

5.9

2.0

464015

3659

5.0

573211

<0.01 <0.01 <0.01

<0.01 <0.01 0.28

<0.01

<0.01

<0.01

<0.01

mean

3.3 2.7 3.1

4.4 3.2 4.3

n

2766

6

2

464015

3660

5

583211

74

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

write-in explanations for their answers including “diagnostic and therapeutic

benefits” for corticosteroid injections and “most effective and reliable treatment”

for operative treatment. From the patients’ perspective, the disadvantage of

non-operative treatment was that it was likely palliative and temporary and the

disadvantage of operative treatment was pain. From the surgeons’ perspective,

the disadvantage of an orthosis was that it was palliative and some patients do

not like to wear them. For surgeons, the disadvantage of a corticosteroid shot

was pain and potential risk of nerve injury, and the disadvantage of surgery was

the risk and recovery time (Table 2). Surgeons were more likely to choose orthosis

Table 3 Disadvantage of the treatment options

Patients Physician P-Value

SplintingWearing the splint can be uncomfortableDoesn’t solve the problem, only manages it Will not prevent progression of the diseaseOther

Corticosteroid injectionIt’s only temporaryThe shot hurtsSmall risk of skin discolorationOther

EMGPainfulTime consumingExpensiveOther

Carpal tunnel release surgeryPainSmall risk of nerve damageSmall risk of infectionThe scar

n

13

45

15

0

48870

3123

40

2735

81

18

62

21

0.0

761311

0.0

5340

6.9 0.0

384911

1.4

36

47

13

5.0

599

28 4.0

48122912

3241

9.418

<0.01

<0.01

<0.01

<0.01

n

36

47

13

5

609

284

48122912

2735

815

75

Table 4 What is the best way to arrive at a final decision regarding the treatment plan

Table 5 Who, if anyone, is usually involved in the decision-making process

Patients

Patients

Physician

Physician

P-Value

P-Value

The health provider decides for the patient

The health provider advises and the patient and provider make a shared decision

The health provider advises and the patient decides

SpouseFamilyFriendPrimary care physician

n 5

27

47

n

2723

214

% 6

34

59

%

4135

3.021

% 0

74

26

%

3531

8.426

<0.01

0.46

n 0

76

27

n

3329

825

fabrication and corticosteroid injection than patients, but they were comparably

likely to choose surgery (Table 1).

Patients disliked the pain of electro-diagnostic testing but valued

confirming the diagnosis (Table 2). Some surgeons disliked the time involved,

saw it as a waste of resources, felt it added confusion and was not helpful for

determining treatment, and lacked confidence in the test results owing to the

technical variations and subjective aspects of the tests. Providers used it mostly

as a baseline but also to help confirm or rule out CTS (Table 3).

Patients preferred to be advised by the surgeon and decide for

themselves, whereas surgeons preferred a shared decision (Table 4). Patients and

surgeons had similar thoughts about who should be involved in the decision

(Table 5). Patients placed more value on their preferences and support from

76

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

others in decision-making, and surgeons placed more emphasis on specific risks

and benefits (Table 6). Patients valued second opinions more than surgeons

(Table 6). Surgeons valued a video format for information more than patients,

who largely favored web-based information (Table 6). Patients and surgeons

considered guidance in deliberation (e.g. decision aids) as the most important

content for patients. In addition, patients found information about treatment

options more important than surgeons (Table 7).

Surgeons and patients agreed that experts should prepare the decision

aids (Table 7).

Table 6 Help with final decision

Which of the following would help arriving at a final decision on one treatment option?

Ways to facilitate the decision-making process

Format to facilitate the decision-making process


The health provider’s recommendation

Information on the various treatment options

Information on the incidence of specific benefits and risks

Personal preferencesInformation on how others go about

decidingSupport from others

Second opinionDiscussion groupsInformation materials

BookletWebVideo

n

23

11

17

142

11

246

43

1322

8

%

29

14

22

18 2.6

14

33 8.2

59

305119

%

24

19

41

10 4.0

1.0

8.91477

232553

<0.01

<0.01

<0.01

n

24

19

41

104

1

91478

222451

77

DISCUSSION

In this study of patient- and surgeon-preferred content of a decision aid for CTS,

areas in which patients with CTS and surgeons agreed included the frequency of

choosing operative carpal tunnel release and the support from family, spouse,

and primary care physician in the decision-making process. We did, however,

find differences and rejected our null hypothesis. Important areas in which

patients and hand surgeons had different perspectives included the likelihood

of choosing an orthosis or corticosteroid injection as the preferred therapeutic

option; the opportunity to confirm the diagnosis with electro-diagnostic

testing; and the choice of non-operative, non-painful treatments. Patients

also considered the disadvantage of non-operative treatment that it was likely

palliative and temporary, whereas for surgeons, the disadvantage of an orthosis

was that some patients found it cumbersome. In addition, for patients, the

disadvantage of operative treatment was pain, whereas for surgeons, it was the

Table 7 What should the information material contain

What should the information material contain?

Who should prepare the material?


BasicTreatment opportunitiesBenefits of treatmentRisks of treatmentProbabilitiesPersonal implicationsGuidens in delibaration

ExpertSocietyGovermentInsurance companyConsumer associationFor profit organisationNon-profit organization

n

282687

10

25510182

%

4.7

19 4.7

14191623

6012

2.4 0.0 2.4

19 4.8

%

1.0 1.0 1.0

10222046

65

0.0 2.2 2.2 6.6

24 0.0

<0.01

<0.01

n

111

10222046

590226

220

78

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

actual scar and other surgical risks. Finally, patients preferred a more individual

role in decision-making, whereas physicians preferred a shared role in the

decision-making process.

This study had several shortcomings. Although we had relatively large

samples, these results may not be generalizable to the average surgeon or

patient and may differ across cultures. Patients at different stages of disease or

decision-making might have different preferences for decision-making – a factor

that we did not study. It is also possible that patients treated in more intimate

health delivery settings such as those in community-based private practice

offices (as opposed to the large tertiary care hospital-based practices in this

study) may exhibit different opinions than those seen in our study population.

Surgeon expertise and patient impressions are often in conflict.

Examples include the patient’s sense that CTS is a consequence of her or his

actions when best evidence suggests it is largely genetic6; that CTS with static

numbness and weakness has only been present for months when such advanced

disease develops over years to decades; and that orthoses and injections can

cure. Prior studies showed that decision aids can normalize and depersonalize

these conflicts, provide patients information they can reflect on at their own

pace and in their own way, and ensure that patients are empowered to make

decisions based on an better understanding of their disease.7 In particular, our

study identified several areas in which patients would like more information

about CTS – the specific risks and benefits of various diagnostic and therapeutic

options, health provider recommendations, and help clarifying their own

preferences.

There are many areas of debate among hand surgeons regarding the

management of CTS. Where current best evidence allows room for debate,

patients can benefit from an understanding of the range of options and the

source of the debate. When the best option is a matter of debate among

hand surgeons, patient preferences and values should take priority in the

decision-making process. A decision aid can help patients understand their

own preferences and values so that they can make a decision that they will

not second-guess. There is evidence that this lowers decisional conflict, which

might lead to increased satisfaction with care and decreased symptoms and

disability.8,9

Our study revealed the apparent contradiction that patients preferred

non-operative treatment but not injections. Perhaps patients want the problem

fixed with as little discomfort and inconvenience as possible. One hypothesis

79

worth testing is that patients may be more willing to invest in discomfort and

hope when a treatment is disease modifying and when the potential benefits

outweigh the risks.

Our study may help with the development of decision aids that will help

patients understand their choices, feel supported, and become more involved

in their recovery. We found that patients and surgeons have different priorities

and preferences regarding decision-making, particularly the risks and benefits of

diagnostic and therapeutic procedures. A decision aid that helps inform patients

of their options based on current best evidence and helps patients understand

their own preferences and values could improve health by reducing decisional

conflict and encouraging patients to take a more active role in their recovery.

Future studies will address the effectiveness of the decision aid developed

based on this study on decisional conflict, anxiety, symptoms, disability, and

satisfaction for patients with CTS.

*The Science of Variation Group: Joshua M. Abzug, Julie Adams, Gallo Fabio Arbelaez, T. Aspard,

George W. Balfour, H. Brent Bamberger, Romero Jose Camilo Barreto, Michael Baskies, W. Arnold

Batson, Taizoon Baxamusa, Ramon de Bedout, Steven Beldner, Prosper Benhaim, Leon Benson, G.

Jorge Boretto, Martin Boyer, Gregory Dee Byrd, Ryan P. Calfee, Gladys Cecilia Zambrano, Charles

Cassidy, Louis Catalano III, Karel Chivers, Ralph M. Costanzo, Phani Dantuluri, Gregory DeSilva, Seth

Dodds, John P. Evans, Naquira Escobar Luis Felipe, C.H. Fernandes, Thomas J. Fischer, Jochen Fischer,

M. Renato Fricker, Gary K. Frykman, Aida E. Garcia, R. Glenn Gaston, José Fernando Di Giovanni,

Charles A. Goldfarb, Michael W. Grafe, H.W. Grunwald, Warren C. Hammert, Randy Hauck, Ricardo

German Hernandez, Eric Hofmeister, Richard L. Hutchison, Asif Ilyas, Jonathan Isaacs, Sidney

M. Jacoby, Peter Jebson, Christopher M. Jones, Michael Jones, Sanjeev Kakar, David M. Kalainov,

Thomas D. Kaplan, Saul Kaplan, Leonid Katolik, Stephen A. Kennedy, Michael W. Kessler, Hervey L.

Kimball, G. A. Kraan, Paul A. Martineau, John McAuliffe, Steven J. McCabe, Desirae M. McKee, Greg

Merrell, Charles Metzger, Michael Nancollas, David L. Nelson, Ralf Nyszkiewicz, Jose A. Ortiz, Patrick

W. Owens, Jason M. Palmer, Lior Paz, Gary Pess, Daniel Polatsch, Frank J. Raia, Marc J. Richard,

Marco Rizzo, Rozental, David Ruchelsman, Oleg M. Semenkin, Aguilar Javier Francisco Sierra, Todd

Siff, Samir Sodha, Catherine Spath, Sander Spruijt, Thomas F. Stackhouse, Carrie Swigart, Robert

Szabo, John Taras, Jason Tavakolian, Andrew Terrono, Thomas F. Varecka, Abhijeet L. Wahegaonkar,

Christopher J. Walsh, Frank L. Walter, Lawrence Weiss, Brian P. D. Wills, Chris Wilson, Christopher

Wilson, Jennifer Moriatis Wolf, Megan Wood, and Colby Young.

80

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

Appendix A Demographic informationt of the observers

Parameters n %Sex Men 92 89 Women 11 11 Location of practice Asia 2 1.9 Canada 1 1.0 Europe 6 5.8 United Kingdom 2 1.9 United States of America 84 82 Other 8 7.8 Years In practice 0-5 34 33 6-10 23 22 11-20 26 25 21-30 20 19 Supervise Yes 77 75 No 26 25 Specialization Hand surgeons 102 99 Other 1 1

81

Appendix B Demographic information of the patients n= 84

Parameter Mean SD Range Age (y) 55 16 20-90Education (y of School, n=84) 15 2.9 1-22 Number % Sex

Man 29 35 Woman 55 65

Marital status

Single 14 17 Living with partner 3 3.6 Married 50 60 Separated/Divorced 11 13 Widowed 6 7.2

Work status (n=81) Working full time 40 49 Working part time 8 9.9 Homemaker 3 3.7 Retired 16 20 Unemployed, able to work 4 4.9 Unemployed, unable to work 10 12

Physician

Surgeon 01 10 12 Surgeon 02 21 25 Surgeon 03 53 63

82

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

REFERENCES

1. Legare F, O’Connor AM, Graham ID, Wells GA, Tremblay S. Impact of the Ottawa Decision

Support Framework on the agreement and the difference between patients’ and physicians’

decisional conflict. Med Decis Making 2006;26:373-90.

2. Fischloff B, Slovic P, S. L. Knowing what you want: measuring labile values. . Hillsdale (NJ):

Lawrence Erlbaum Associates Inc.; 1980.

3. O’Connor AM, Tugwell P, Wells GA, et al. A decision aid for women considering hormone

therapy after menopause: decision support framework and evaluation. Patient Educ Couns

1998;33:267-79.

4. Jacobsen MJ, O’Connor AM. Population Needs Assessment. Available from (last entered on

June 3rd 2013) wwwohrica/decisionaid 2007.

5. O’Connor AM S, D, & Jacobsen MJ. Ottawa Decision Support Tutorial (ODST): Improving

Practitioners’ Decision Support Skills Ottawa Hospital Research Institute: Patient Decision

Aids, 2011. Web. 2011 Nov 30.

6. Lozano-Calderon S, Anthony S, Ring D. The quality and strength of evidence for etiology:

example of carpal tunnel syndrome. J Hand Surg [Am] 2008;33:525-38.

7. Stacey D, Bennett CL, Barry MJ, et al. Decision aids for people facing health treatment or


8. Barry M, Cherkin DC, Chang Y, Fowler FJ, Skates SA. A randomzed trial of a multimedia

shared decision-making program for men facing a treatment decision for benign prostatic

hyperplasia. Disease Management and Clinical Outcomes 1997;1:5-14.

9. Kennedy AD, Sculpher MJ, Coulter A, et al. Effects of decision aids for menorrhagia on


2002;288:2701-8.

83

CHAPTER 7

Randomized controlled trial:

the influence of decision aids on

decisional conflict and satisfaction

of patient with hip or knee

osteoarthritis

Hageman MG, MD (1), Poolman RW, MD, PhD (2), Du Long J, MD (1), Vuijk D, MD (1), Vervest T,

MD, PhD (3), Kerkhoffs GMMJ, MD, PhD, Prof. (1), Haverkamp D, MD, PhD (4).

(1) Department of Orthopaedic Surgery at Academic Medical Center, Amsterdam, NL.

(2) Department of Orthopaedic Surgery at OLVG-Oost, JointResearch, Amsterdam, NL.

(3) Department of Orthopaedic Surgery at Tergooi ziekenhuis, Hilversum, NL.

(4) Department of Orthopaedic Surgery at Slotervaart ziekenhuis, Slotervaart Center of

Orthopedic Research and Education (SCORE), Amsterdam, NL.

Submitted to KSSTA Nov 2017.

84

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

ABSTRACT

Background There is an increasing interest in shared decision making. Decision aids are thought to be able to support patients and healthcare providers to make a shared decision, especially when there is more than one reasonable option, when there is no clear advantage in outcomes or when each benefit of harm may be valued differently. We hypothesized that there is no difference in decisional conflict comparing patients managed with a decision aid compared to patients without. Our secondary hypothesis was that there is no difference between patients managed with a decision aid compared to patients without, with respect to anxiety, knowledge, satisfaction, preferred treatment at enrollment and physical function and quality of life at 26 weeks follow-up. Methods This multi center randomized controlled trial included patients with knee or hip osteoarthritis who had not consulted an orthopedic surgeon for the same complain in the past. At the first encounter, patients enrolled in the control group were treated according standard care, while patients enrolled in the intervention group were managed with a decision aid. After the encounter patients were asked to complete a survey about decisional conflict (DCS), preferred treatment, gained knowledge, physical function (KOOS/HOOS), pain (NRS), anxiety (PASS-20) quality of life (EQ-5D) and satisfaction. The long-term follow-up was carried out after 26 weeks and evaluated the HOOS/KOOS, PASS, EQ-5D, satisfaction and the preferred treatment. Results At the first encounter, there was a significant difference in the total-DCS, knowledge and satisfaction score comparing the intervention group managed with the decision aid to the control group managed without. At follow-up there was no significant difference between both groups on the physical function, pain and quality of life scores.

Conclusions Our RCT showed that immediately after the consult about their treatment patients managed with a decision aid had less decisional conflict, more knowledge and were more satisfied with the delivered care compared to patients who were managed according to standard care. After 26 weeks there was no difference in decisional conflict between the patients managed with a decision aid compared to patients managed without.

85

INTRODUCTION

Patients with osteoarthritis of the hip and knee have various treatment options

ranging from watchful waiting to surgery. At certain stages of disease patients

and doctors have a choice. It is thought that for those diagnoses patients and

physicians could make a decision together, known as shared decision-making

(SDM).1,2

To support patients to make a shared-decision, healthcare provides are

thought to provide accurate, comprehensive and neutral information about

the treatment options.3 In addition patients have to share their values when

reflecting on the possible outcomes. However, clear communication between

the patient and physician leading to a shared decision is challenging. Especially,

since time is limited during the clinical encounter and the complex information

need to be understood before patients are able to clarify their values.4 This may

compromise the outcome of the surgery and may result in disappointment and

even regret.

Decision aids are made to support patients and physicians in the

decision-making about the optimal treatment for their diagnoses.5,6 A decision

aid provides the patient evidence based information and helps them to

consider what benefits and harms are important to them. The decision aids

are additionally to the information explained by the physician, rather than to

replace the consultation process.7

Previous studies about the effect decision aids in general medicine and

orthopaedic surgery showed promising results regarding decisional conflict,

anxiety, knowledge and satisfaction.3,7,8 For example, Achaval et al. examined the

effect of an education booklet, video booklet and decision tool on the decisional

conflict among patient with knee osteoarthritis in a tertiary referral center

in the United States of America.8 It showed a significant overall reduction in

decisional conflict. Furthermore, Arterburn et al showed that the introduction

of a decision aid changed patients preference for joint replacement, resulting in

26 percent fewer hip replacement surgeries, 38% fewer knee replacements.9

The objective of this study was determining the effect of decision

aids on the magnitude of decisional conflict, anxiety, knowledge, satisfaction,

physical function and quality of life to patients with knee and hip osteoarthritis.

We hypothesized that there is no difference in decisional conflict comparing

patients managed with a decision aid compared to patients without. Our

secondary hypothesis was that there is no difference between patients managed

with a decision aid compared to patients without, with respect to anxiety,

86

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

knowledge, satisfaction, preferred treatment at enrollment and physical

function and quality of life, 26 weeks after the treatment decision was made.


Study design and setting

This multicenter hypothesis blinded randomized controlled trial was carried out

at secondary and tertiary referral hospitals after approval from the Institutional

Research Board. Adult patients (18 years or older) with osteoarthritis of knee or

hip, Dutch fluency and literacy, who had not consulted an orthopedic surgeon

for the same complaints in the past, were invited to participate. The waiting

time for the operation was similar in the participating hospitals therefore the

first effects of the operative treatment and the non-operative treatment could

be expected after 26 weeks. For that reason the long term follow-up was carried

out after 26 weeks.

Participants/study subjects

New patients diagnosed with osteoarthritis of the knee or hip were asked to

consider participation in this randomized controlled trial.

Description of experiment, treatment or surgery

The online decision aids were developed by patients and physicians according

to the International Patient Decision Aids Standards and based on a previous

study, carried out by this research group, assessing patients and physicians

needs when deciding about the optimal treatment.4,11 The decision aids were

comprised of 5 steps comparing operative (total joint prosthesis) versus non-

operative treatment (lifestyle advice, painkillers and corticosteroid injections).

Step 1: Informed the patient about the osteoarthritis and the treatment options.

Step 2: Informed the patient about the benefits, harms, scientific uncertainties

and probabilities of outcome. The treatment options were elaborated in a table,

which showed the clinical outcomes after operation or non-operative treatment

regarding satisfaction, physical function, pain and complications in text and

were clarified by icon-based risk graphs Step 3: The most important points

were tested through a short quiz. Step 4: Values clarification, asking patients to

consider which benefits and risks matter most to them and Step 5: Asked the

participant to reflect on responses and make a decision regarding the treatment.

Prior to the implementation of this study an implementation workshop was

87

given to support the treating physicians in the implementation process of the

decision aid. The control group was managed without decision aids and received

standard care, reflecting routine practice by attending physicians.

Description of follow-up routine

After 26 weeks we called the patients for follow-up by telephone. If the patients

did not answer their phones at three different time points, we sent them the

questionnaires by mail or email.

Variables, outcome measures, data sources, and bias

After the consult about their treatment patients were asked to complete a

survey about decisional conflict (DCS), preferred treatment, gained knowledge,

physical function (KOOS/HOOS), pain (NRS), anxiety (PASS-20) quality of life

(EQ-5D) and satisfaction.1,10,12-14 The questionnaires took by email or phone

call approximately twenty minutes to complete. Follow-up evaluated the

final decision, DCS, HOOS/KOOS, PASS, EQ-5D and satisfaction. The Decisional

Conflict Scale (DCS) is a reliable and valid measure of personal perceptions

of: a) uncertainty in the face of options, b) modifiable factors contributing to

uncertainty such as feeling uninformed, unclear about personal values, or

unsupported in decision-making; and c) effective decision-making such as

feeling the choice is informed, values-based, likely to be implemented, and

expressing satisfaction with the choice.15 It consists of sixteen questions, with a

total score ranging from 0 (no decision conflict) to 100 (highest level of decision

conflict). The knowledge questionnaire was comprised of four questions and

was used to measure the patients’ knowledge of treatment options and risks.

The following questions were survey: Question 1: “Could painkillers sufficiently

reduce complaints due to osteoarthritis?”; Question 2: “Is the primary goal

of operative treatment to reduce pain?”; Question 3: “Do the majority of the

prosthesis last longer than 10 years? Question” 4: “Should joint replacement

be the first choice, when conservative treatments did not work? The score

ranged from 0 (no correct answers) to 4 (all correct answers)”.15 The decision

questionnaires contained two separate questionnaires. One questionnaire

inquired what phase of decision-making patients were in and which treatment

they preferred. The second questionnaire was to inquire whether patients have

made their definitive decision. The Hip disability and Osteoarthritis Outcome

Score (HOOS) measured patients’ symptoms, pain, activity limitations in daily

living, function in sport and recreation and quality of life. This score consisted

88

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

of forty questions, divided in the subscales mentioned above. For every subscale

there is a score ranging from 0 (extreme symptoms) to 100 (no symptoms).12

The Knee Injury and Osteoarthritis Outcome Score (KOOS) measured patients’

symptoms, pain, activity limitations daily living, function in sport and recreation

and quality of life. This score consisted of 42 questions, divided in the subscales

mentioned above. For every subscale there was a score ranging from 0 (extreme

symptoms) to 100 (no symptoms).1 To measure pain intensity we used question

8 of the KOOS or HOOS, which asked whether the patient has had knee or hip

pain during the last week.1,12 The short Pain Anxiety Symptoms Scale (PASS-

20) was used to measure patients’ pain-related anxiety and fear. It consisted

of twenty questions with a score ranging from 0 (no anxiety and fear) to 100

(extreme anxiety and fear).16 The EuroQol 5 Dimensions (EQ-5D) questionnaire

was used to measure health-related quality of life. It consisted of 5 questions

concerning mobility, self-care, usual activities, pain/discomfort and anxiety/

depression.13 It was also comprised of a Visual Analogue Scale (VAS) on which the

patients could score their health condition ranging from 0 (worst imaginable

health condition) to 100 (best imaginable health condition). The satisfaction

questionnaire consisted of three questions to measure patients’ satisfaction

with given information, the clinic and the physician. Patients could score each

question from 0 (no satisfaction) to 10 (complete satisfaction).

Demographics, description of study population

The baseline characteristics were similar in the two study groups. The

intervention group comprised 33 men and 33 women, who were on average

68-years-old (SD: 11). The control group comprised 30 men and 35 women, who

were on average 66-years-old (SD: 10). No significant differences were noted

with respect to physical function, pain and quality of life (Table 1).

Accounting for all patients / study subjects

In approximately eighteen months 145 participants were identified as being

eligible to participate in this study. All participants consented to participate

and were randomized. Among the 145 participants, 10 patients in the

intervention group and 4 patients in the control group did not complete the

first questionnaire due to time constraints leaving a total of 131 participants,

in which 66 patients were assigned to the intervention group and 65 patients

to the control group. 54 patients did not respond at follow up or could not be

contacted after three written requests and three requests by phone. (Table 1)

89

Table 1 Demographics

Intervention Control

SexMen

AgeLevel of educationDuration of pain (in weeks?)

Marital statusSingleUnmarriedMarriedDivorcedWidowed

Working statusWorking, full timeWorking, part timeSick leaveRetiredUnemployed, able to workUnemployed, unable to work

LocationLeft hipright hipboth hipsleft kneeright kneeboth knees

Tried non-operative treatment beforeYes

HospitalHospital 1Hospital 2Hospital 3Hospital 4

No.33

Mean681555

No.16

830

19

No.16

71

3413

No.9

153

1318

6

No.25

No.9

212610

%50

Sd11

2.276

%251347

214

%2611

255

25

%1423

52028

9

%39

%14323915

%46

Sd10

2.075

%201358

36

%1319

356

55

%1537

92014

5

%28

%3

354220

No.30

Mean661547

No.13

837

24

No.8

122

3633

No.1024

613

93

No.18

No.2

232713

90

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

STATISTICAL ANALYSIS

Based on previous studies, scores of 25 or lower on the Decisional Conflict Scale

(DCS) (0-100) are associated with low decisional conflict and following-through

with decisions. On the other hand, scores of 39 or higher are associated with

heightened mental conflict resulting in a delay in decision-making.15 Thus, our

study examined whether the use of decision aids results in an increased rate

of patients following-through with decisions. A sample size of 128 patients

was chosen for patients with osteoarthritis of the knee or hip to detect an

effect size of 50% on the decisional conflict scale with a type I error of 0.05 and

a type II error of 0.20 based on a two-tailed prediction. Continuous data was

presented as the mean and standard deviation when normally distributed.

When comparing the intervention and the control group regarding continuous

dependent variables and dichotomous independent variables Student t-test was

used for normally distributed data and the Mann-Whitney U-test for skewed

data. The Kruskal Wallis test was used for ordinal data. In bivariate analysis,

the association between continuous dependent and continuous independent

variables was investigated using Spearman correlation. Associations with a

P-value less than 0.05 were considered statistically significant.

RESULTS

There was a significant difference (p < 0.001) in the total-DCS, comparing the

intervention group (mean=25) and the control group (mean=39) immediately

after the first encounter. The intervention group also scored significantly lower

on all DCS-subscales about (information, values clarity, support, uncertainty

and effective decision making) than the control group (Table 2). At the 26 weeks

follow-up there was no significant difference (p=0.17) in the DCS-total score

comparing the intervention group (mean=31) and the control group (mean=35).

Only on the subscale values clarity there was a significant difference (p=0.027)

between the intervention group and control group, favoring the intervention

group (mean=33 vs. mean=47, respectively) (Table 3).

Patients managed with a decision scored significantly higher on the

knowledge scale than the patients managed without (p < 0.01), after the first

encounter. At 26 weeks follow-up there was no significant difference in the

phase of decision-making, treatment preference and final choice (Table 2).

Patient allocated to the intervention group were significantly more

satisfied about the given information (mean=8.6 vs. mean=7.6; p < 0.001) and

91

Table 2 Outcomes after the first encounter

Intervention Control P-Value

Decisional conflict scaleInformed subscoreValues clarity subscoreSupport subscoreUncertainty subscoreEffective decision subscoreTotal score

SatisfactionInformationVisit outpatient clinicPhysician

Anxiety (Pass)

Knowledge

Stage of decision makingHave not begun to think about the treatment optionsHave not begun to think about the treatment options, but I am interested to do soI am considering the treatment options nowI am close to select an optionI have already made a decision, but am still willing to reconsiderI have already made a decision and I am unlikely to change my mind

What treatment option do you prefer?Watchful waitingLifestyle changesPhysiotherapyPainkillersCorticosteroid injectionProsthesisOther

Mean

322527232025

8.68.38.9

20

3.7

No.

1

0

103

15

37

51

2237

262

Sd

201613161512

1.11.50.9

17

0.6

%

1.5

0

154.523

56

7.61.533

4.51139

3

Sd

202216151311

1.81.71.7

19

0.9

%

3.1

9.2

141.515

57

4.64.620

4.61942

6.2

0.033 <0.001 <0.001 <0.001 0.001 0.000 0.000 0.30 0.01 0.29 0.0073

0.11

0.46

Mean

395045352839

7.68.08.3

23

3.3

No.

2

6

91

10

37

33

133

1227

4

continue >

92

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

their physician (mean=8.9 vs mean=8.3; p= 0.01) compared to the patients who

were allocated to the control group at enrolment, after the first encounter. On

the other hand, there was no significant difference in satisfaction about the visit

to the outpatient clinic (p= 0.30). At 26 weeks follow-up the results showed that

the patient managed with a decision aid were still significantly more satisfied

about the information given (p= 0.0038) but not about the visit of outpatient

clinic (p = 0.35) or their physician (p = 0.34) (Table 2). The results showed no

difference in preference among the different treatment options, after the first

encounter and 26 weeks follow-up (Table 2).

There was no difference in the magnitude of experienced anxiety

(p=0.29), after the first encounter and at follow-up. At 26 weeks follow-up there

was also no significant difference between both groups on the physical function,

pain and quality of life scores (Table 4).


Did you make a final choice

Yes

If yes, what did you chooseWatchful waitingLifestyle changesPhysiotherapyPainkillersCorticosteroid injectionProsthesisOther

Mean

50

41

1616

211

Sd

76

82

322

1242

2

Sd

79

3.95.916

3.92249

0

0.84

0.26

Mean

51

2382

1125

0

continued table 2

93

Table 3 Outcomes at follow up


Decisional conflict scaleInformed subscoreValues clarity subscoreSupport subscoreUncertainty subscoreEffective decision subscoreTotal score

SatisfactionInformationVisit outpatient clinicPhysician

Anxiety (Pass)

If yes, what did you chooseWatchful waitingLifestyle changesPhysiotherapyCorticosteroid injectionProsthesisOther

Mean

363337292331

8.47.98.2

17

4012

200

Sd

262625171515

0.91.71.7

14

150

3.77.474

0

Sd

232717151411

2.11.51.6

14

117.17.13.664

7.1

0.38 0.027 0.17 0.22 0.55 0.17 0.0038 0.35 0.34 0.92

0.59

Mean

314744342535

7.17.88.0

16

3221

182

Table 4 Clinical outcomes at follow up

Intervention Control

Hoos pain totalHoos symptoms totalKoos pain totalKoos symptoms total EQ5D MobileEQ5D CareEQ5D ActivityEQ5D PainEQ5D Anxiety

Mean

72767256

1.61.11.41.71.1

Sd

23171613

0.490.330.500.610.31

Sd

23272011

0.420.500.500.680.32

Mean

72697057

1.81.31.41.81.1

94

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

DISCUSSION

We found that patients, making a shared decision, with a decision aid had less

decisional conflict, more knowledge and increased clarity of values. The results

also showed that this did not influence the stage of decision-making, treatment

preference, physical outcome or quality of life on the long term.

The results of this study should be evaluated in the light of its

strengths and shortcomings. A strenght is that this randomized controlled trial

was carried out at tertiary and secondary referral centers, with patients from

multicultural background. The first shortcoming is that patients with knee or

hip osteoarthritis may experience different levels of decisional conflict with

respect to their diagnosis. On the other hand subsequent analysis showed no

significant difference between both groups at baseline. The second shortcoming

is the high percentage of loss to follow-up. As a result no strong conclusion

can be drawn from the follow-up results. The third shortcoming was that the

implementation workshop was given only once, while not every physician

participated. The physicians who participated in the workshop also managed the

patients in the control group. As a result the standard care may have improved

after the workshop and the patients in de control group could have been better

informed compared to the patients before the implementation workshop.

Another shortcoming was that there were 14 patients who were excluded due to

time constraints at the outpatient clinic. Patients in the intervention group may

have stopped not only due to the length of the questionnaires, but also due to

the time the decision aid took to work through (approximately 15-20 min). This

may have influenced the outcomes as well. Also 54 patients (27 patients in the

intervention group and 27 in the control group) did not respond to the follow

up or could not be contacted after three written requests and three requests by

phone.

The finding that patients managed with a decision aid experienced

less decisional conflict at enrollment compared to patients managed with usual

care is in concordance with previous studies.7,8,16,17 The systematic review of

Stacey et al included 115 RCT’s comparing decision aids to usual care, showing a

significant average decrease in the level of decisional conflict. The average score

of 25 in the decision aid group is associated with low decisional conflict and

following-through with decisions, while the score of 39 in the control groups

is associated with increased mental conflict resulting in delayed decision-

making.15 The reduced sub-scores also showed that patients felt more informed

about the options, clearer about their values and comfortable with their choices.

95

Besides for values clarity, this difference was not maintained at follow-up.

This may be explained by the fact that effect of decision aids decreases over

time when the final decision was made. This was also found in a randomized

controlled trial about total knee arthroplasty by Stacey et al with one year

follow-up.7

The decision aid appeared to have a positive effect on the level

of satisfaction about the given information and the treating physician at

enrollment, although the satisfaction about the physician decreased over time.

Among the 14 studies comparing decision aids to usual care, 10 measured

satisfaction with the decision-making process, 3 measured satisfaction

received and 1 measured satisfaction counseling. Of those 14 studies, 5 showed

statistically significant improvement in satisfaction with decision making

process and the information provided.17 Montori et al. found no difference in

satisfaction with the given information between patient, clinicians had higher

level of satisfaction. The finding that decision aids did not influence the level

of anxiety is also consistent with previous studies.17 None of the previous

studies demonstrated differences in effect on patients stated anxiety at one

month, three months or one year. One might expect that patients who are

better informed about potential drawbacks might be more anxious about the

outcome than patients, who are less informed, however this is not reflected

by the results. We found no significant change in the preference of surgery

between both groups. Previous studies reported mixed results. For example

Deyo et al. found no difference in preference of surgery for herniated disc in

the detailed versus simple decision aid.2 On the other hand Arterburn et al.

described a decrease in elective surgery after implementation of a decision aid

after one year.9 As in other studies, we found no difference in physical outcome

or quality of life. Stacey et al. reports 12 studies measuring various condition

specific health outcomes. 10 studies compared decision aids versus usual care

and one a detailed decision aid versus a simple decision aid. 9 of the 12 studies

found no significant effect on physical outcome. It is not surprising that no

effect on health outcome is found in our study, since there is a comparable

preference in treatment options and previous results showed that patients

physical outcomes improve in the operatively and non-operatively treated

groups. The positive effect of decision aids on reducing decisional conflict,

improving patient’s knowledge about risks and benefits, being informed and

feeling clear about their values provides supports for using them in clinical

practice. Both knowledge and understanding of probable risk are important

96

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

to make an informed decision making possible. Further research is needed to

develop decision aids based on individual characteristics, resulting in more

advanced complication and success rates based. Additional research is required

to determine the not only the effect of personalized decision aids, but also the

cost-effectiveness and adherence of decision aids.

In conclusion we found that our patients managed with a decision aid

had less decisional conflict, more knowledge and increased clarity of values at

the initial encounter with their attending surgeon.

97

REFERENCES

1. de Groot IB, Favejee MM, Reijman M, Verhaar JA, Terwee CB. The Dutch version of the Knee

Injury and Osteoarthritis Outcome Score: a validation study. Health and quality of life

outcomes 2008;6:16.2. Deyo RA, Cherkin DC, Weinstein J, Howe J, Ciol M, Mulley AG, Jr. Involving patients in clinical

decisions: impact of an interactive video program on use of back surgery. Medical care

2000;38:959-69.3. Skou ST, Roos EM, Laursen MB, et al. A Randomized, Controlled Trial of Total Knee

Replacement. N Engl J Med 2015;373:1597-606.4. du Long J, Hageman M, Vuijk D, Rakic A, Haverkamp D. Facing the decision about the

treatment of hip or knee osteoarthritis: What are patients’ needs? Knee Surg Sports

Traumatol Arthrosc 2016.5. Slover J, Shue J, Koenig K. Shared decision-making in orthopaedic surgery. Clin Orthop Relat


screening decisions. Cochrane Database Syst Rev 2011:CD001431.7. Stacey D, Briere N, Robitaille H, Fraser K, Desroches S, Legare F. A systematic process for

creating and appraising clinical vignettes to illustrate interprofessional shared decision

making. J Interprof Care 2014;28:453-9.8. de Achaval S, Fraenkel L, Volk RJ, Cox V, Suarez-Almazor ME. Impact of educational and patient

decision aids on decisional conflict associated with total knee arthroplasty. Arthritis Care Res

(Hoboken) 2012;64:229-37.9. Arterburn D, Wellman R, Westbrook E, et al. Introducing decision aids at Group Health

was linked to sharply lower hip and knee surgery rates and costs. Health Aff (Millwood)

2012;31:2094-104.10. O’Connor AM. Validation of a decisional conflict scale. Medical decision making : an

international journal of the Society for Medical Decision Making 1995;15:25-30.11. EuroQol G. EuroQol--a new facility for the measurement of health-related quality of life.

Health policy 1990;16:199-208.12. de Groot IB, Reijman M, Terwee CB, et al. Validation of the Dutch version of the Hip disability

and Osteoarthritis Outcome Score. Osteoarthritis Cartilage 2009;17:132.13. Elwyn G, O’Connor A, Stacey D, et al. Developing a quality criteria framework for patient

decision aids: online international Delphi consensus process. BMJ 2006;333:417.14. International Patient Decision Aid Standards (IPDAS) Collaboration. 2013. at ipdas.ohri.ca/.)15. O’Connor AM, Tugwell P, Wells GA, et al. A decision aid for women considering hormone

therapy after menopause: decision support framework and evaluation. Patient Educ Couns

1998;33:267-79.16. McCracken LM, Dhingra L. A short version of the Pain Anxiety Symptoms Scale (PASS-20):

preliminary development and validity. Pain research & management : the journal of the Canadian

Pain Society = journal de la societe canadienne pour le traitement de la douleur 2002;7:45-50.17. Stacey D, Hawker G, Dervin G, et al. Decision aid for patients considering total knee

arthroplasty with preference report for surgeons: a pilot randomized controlled trial. BMC

Musculoskelet Disord 2014;15:54.

98

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

99

CHAPTER 8

Do upper extremity trauma patients

have different preferences for shared

decision-making than patients with

non-traumatic conditions?

Hageman MG, Makarawung DJ, Briet JP, van Dijk CN, Ring D.



Clin Orthop Relat Res. 2015 Nov; 473(11):3542-8.

100

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

ABSTRACT

Background Shared decision-making is a combination of expertise, available scientific evidence, and the preferences of the patient and surgeon. Some surgeons contend that patients are less capable of participating in decisions about traumatic conditions than non-traumatic conditions.Questions/purposes (1) Do patients with non-traumatic conditions have different preferences for shared decision-making when compared with those who sustained acute trauma? (2) Do disability, symptoms of depression, and self-efficacy correlate with preference for shared decision-making?

Methods In this prospective, comparative trial, we evaluated a total of 133 patients presenting to the outpatient practices of two university-based hand surgeons with traumatic or non-traumatic hand and upper extremity illnesses or conditions. Each patient completed questionnaires measuring their preferred role in healthcare decision-making (Control Preferences Scale [CPS]), symptoms of depression (Patients’ Health Questionnaire), and pain self-efficacy (confidence that one can achieve one’s goals despite pain; measured using the Pain Self-efficacy Questionnaire). Patients also completed a short version of the Disabilities of the Arm, Shoulder, and Hand questionnaire and an ordinal rating of pain intensity.

Results There was no difference in decision-making preferences between patients with traumatic (CPS: 3 ± 2) and non-traumatic conditions (CPS: 3 ± 1 mean difference = 0.2 [95% confidence interval, 0.4 to 0.7], p = 0.78) with most patients (95 versus 38) preferring shared decision-making. More educated patients preferred a more active role in decision-making (beta = 0.1, r = 0.08, p = 0.001); however, differences in levels of disability, pain and function, depression, and pain-related self-efficacy were not associated with differences in patients’ preferences in terms of shared decision-making.

Conclusions Patients who sustained trauma have on average the same preference for shared decision-making compared with patients who sustained no trauma. Now that we know the findings of this study, clinicians

101

might be motivated to share their expertise about the treatment options, potential outcomes, benefits, and harms with the patient and to discuss their preference as well in a semi-acute setting, resulting in a shared decision.

INTRODUCTION

In shared decision-making the caregiver provides expertise and evidence, and

the patient and caregiver choose diagnostic and treatment options consistent

with their values and preferences.1 There is evidence that empowering patients

to participate in decision-making with the help of decision aids (videos, web

sites, or pamphlets that help patients understand their options and become

aware of their preferences) results in increased satisfaction and physical

function and reduced decisional conflict, anxiety, and resource utilization.2

Patient preferences for involvement in decision-making may vary by age, sex,

socioeconomic status, type of illness, and illness behavior, and perhaps the

gravity or acuity of the decision.3,4

Many surgeons hold the opinion that patients with traumatic problems

are less capable of and less interested in participating in decisions because they

feel vulnerable and time-pressured. Although to our knowledge this has not been

studied, many of our colleagues insist that patients with a painful acute fracture

cannot fully participate in the decision-making process and need the doctor to

recommend treatment. In addition, patients with greater symptoms of depression

or less self-efficacy might have less desire or confidence about participation in the

decision-making process and might prefer to fall back to a paternalistic style of

medical care and take a more passive role. Depressed mood and ineffective coping

strategies can make people feel more resigned, passive, and helpless. We therefore

wished to assess hand surgery patient preferences for shared decision-making in

relation to the acuity of the diagnosis and to psychological factors.

This study tested the following hypotheses: (1) Do patients with non-

traumatic conditions have different preferences for shared decision-making

when compared with those who sustained acute trauma? (2) Do disability,

symptoms of depression, and self-efficacy correlate with preference for shared

decision-making?

102

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery


After approval from our institutional research board, all new, non-pregnant,

English-speaking patients 18 years or older presenting to one of two hand

surgeons (DR, CM) were asked to participate in this prospective study. The

researcher informed the patient about the study details and informed consent

was obtained. Patients were enrolled between November 2012 and April 2013.

We asked 135 patients to participate in the study: 1 (0.7%) declined

and 134 were enrolled before seeing the treating physician. One patient

was excluded from the study as a result of invalid answers on one of the

questionnaires. The analyses were conducted on 133 patients (68 men

and 65 women) with a mean age of 47 ± 17 years (range, 18–86 years). The

demographics of trauma and non-trauma cohorts were comparable (Table1).

There was also no difference in levels of education comparing the trauma (mean,

16 years; range, 9–16 years) and non-trauma cohorts (mean, 15 years; range,

0–20 years; p = 0.10). Conditions categorized as traumatic included: fracture,

laceration, sprain, tendon injury, and amputation. All other diagnoses were

considered non-traumatic; examples included arthrosis, carpal tunnel, trigger

finger, and another discrete diagnosis.

Mearsurement Tools

At the time of enrollment, patients completed a demographic survey, including

level of education, and the following questionnaires: the Control Preferences

Scale (CPS), the short version of the Disabilities of the Arm, Shoulder and Hand

questionnaire (QuickDASH), the Pain Self-efficacy Questionnaire (PSEQ), the

short version of the Patients Health Questionnaire (PHQ-2), and an 11-point

ordinal pain intensity score.

After the encounter with the physician, the research assistant registered

whether the patient was a trauma or non-trauma patient. Education, as the

number of years of school, was measured on a continuous scale with graduation

from high school scored as 12.

The CPS is a validated measure of a patient’s preferred role in healthcare

decision-making.5 Patients rank-order five possible approaches to decision-

making, resulting in a score that is scaled from 1 (most active role) to 6 (most

passive role). A score of 3 or lower indicates a preference for shared decision-

making.5

The QuickDASH is a short version of the DASH and is used to determine

arm-specific disability.6,7 It consists of 11 questions, which are answered on a

Table 1 Demographics

Trauma cohort Non-trauma cohort

ParameterAgeEducation

SexWomenMen

Marital statusSingleLiving with PartnerMarriedSeparated/DivorcedWidowed

Work statusWorking, full timeWorking, part timeHome makerRetiredUnemployed and able to workUnemployed and unable to workOn worker compensationCurrently on sick leave

DiagnosisAcute injuryNon specific arm painTrigger fingerCarpal tunnel syndromeGanglion cystArthroseDequervain’sDupuytrenEpicondylitisBursitisGiant cell tumorCubital Tunnel syndromeOther

PhysicianPhysician IPhysician II

Health outcomesQuick DASHPainPSEQPHQ

Mean4516

3037

32

124

54

425073424

67

1750

434.542

1.4

Range 18-86

9-26

4555

48

1.536

9.06.0

63

7.50

104.55.63.06.0

50

2575

2.3-863.9-5.1

2-600-6

Range20-86

0-20

5347

30

4.648

9.17.6

61

9.13.017

6.13.01.50.0

6.7121315

6.718

5.05.06.71.71.73.015

3664

0-804.7-6.0

0-600-6

Mean 4915

3531

20

332

65

4062

114210 47894

113411129

2442

315.447

1103

104

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

5-point Likert-scale. The total score is scaled to range from 0 (no disability) to 100

(most severe disability).

The PSEQ is a questionnaire designed to assess a patient’s confidence

that they can achieve their goals despite pain.8,9 It involves ten items, which

can be scored by the patient on a 7-point Likert-scale, ranging from 0 (not at all

confident) to 6 (completely confident). The outcome score is calculated by adding

up the items on a scale ranging from 0 to 70. A higher score indicates greater

confidence. Mean imputation was used for two missing values.

The PHQ-2 was used to evaluate symptoms of depression. The PHQ-2

is a validated two-question measure of symptoms of depression.10,11 The two

questions are answered on a 4-point Likert-scale ranging from 0 (not at all) to

3 (nearly everyday) and the overall score ranges from 0 to 6.

The Numeric Rating Scale is an 11-point ordinal measure of pain

intensity.

Statistical Analysis

An a priori power analysis for our primary study question determined that

64 patients in the trauma cohort and 64 patients in the non-trauma cohort

would provide 80% power to detect a 0.50 SD (medium) difference in average

CPS score with α = 0.05 using a two-tailed Student’s t-test. We enrolled 135

patients to have at least 64 patients for each cohort accounting for dropouts and

incomplete questionnaires.

In bivariate analysis, Pearson’s correlation was used for continuous

variables. The strength of a correlation between 0.10 to 0.29, 0.30 to 0.49, and

0.50 to 1.0 is interpreted as small, medium, and large correlation, respectively.12

The Student’s t-test was used for the CPS (ordinal variable) when comparing

between two groups; and analysis of variance was used to compare differences

in CPS (again, ordinal variable) when more than two groups were present such

as based on marital status. Variables with p < 0.10 were inserted in a backward,

stepwise, multivariable linear regression analysis of factors associated with

CPS. When categorical variables were inserted in multivariable analysis, dummy

codes were generated when there were more than two categories.

RESULTS

There was no difference between trauma (mean CPS: 3; SD: 2) and non-trauma

patients’ (mean CPS: 3; SD: 1) preferred level of shared decision-making (mean

105

Table 2 Bivariable analyses

Control preference scale

ParameterNonelective versus Elective patients

Trauma cohortNon-trauma cohort

SexWomenMen

Marital statusSingleLiving with PartnerMarriedSeparated/DivorcedWidowed

Work statusWorking, full timeWorking, part timeHome makerRetiredUnemployed and able to workUnemployed and unable to workOn worker compensationCurrently on sick leave

DiagnosisAcute injuryNon specific arm painTrigger fingerCarpal tunnel syndromGanglion cystArthroseDequervain’sDupuytrenEpicondylitisBursitisGiant cell tumorCubital Tunnel syndromOther

PhysicianPhysician IPhysician II

Mean

3.12.9

3.03.0

3.22.02.92.93.3

2.92.54.53.42.73.25.02.8

3.13.13.62.43.52.62.32.32.2

36

2.52.5

33

SD

1.71.4

1.51.6

1.71.41.51.70.87

1.51.62.11.51.71.81.71.7

1.71.60.71.71.71.30.961.21.2

--

0.710.58

1.61.5

P-Value

0.78

0.96

0.39

0.27

0.62

0.78

continue >

106

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

difference = 0.2 [95% confidence interval, 0.4 to 0.7], p = 0.78; Table 2). Again,

scores of 3 or lower on the CPS suggest a desire on the part of the patient to

engage in shared decision-making.

More educated patients had a greater desire to participate in decision-

making (coefficient = 0.27, p < 0.01); but age, duration of complaint, disability

level, pain or pain self-efficacy, and symptoms of depression were not associated

with preferences for shared decision-making (Table 2).

DISCUSSION

Assuming that patients with acute injury are less interested or capable of

participating in decision-making risks devaluing their preferences. We found

that patients with acute hand and upper extremity trauma prefer to be as

engaged in decision-making as patients with non-traumatic conditions. As

education levels increased, patients’ desires to participate in shared decision-

making also increased, which is consistent with prior research.13-15 Coping

strategies and symptoms of depression did not affect decision-making

preferences.

This study should be considered in light of its shortcomings. First, the

setting was limited to hand and upper extremity conditions. These findings may

only generalize to other conditions or other practice settings, but that seems

unlikely. It is possible that for some specific conditions, however, such as very

severe trauma, the findings would be different. On the other hand, the lack of

Control preference scale

Health outcomesAge

EducationDuration of injuryQuick DASHPainPSEQPHQ

Coefficient

-0.03-0.27-0.0040.150.082

-0.10.00

P-Value

0.69<0.01

0.960.080.350.240.96

continued table 2

107

correlation between the duration since injury and the CPS suggests that time

pressure does not have a strong influence.

Patients have similar levels of desire for shared decision-making,

regardless of whether the condition was traumatic or non-traumatic. Decision-

making preferences were addressed in a study of Korean patients with carpal

tunnel syndrome.16 33% of patients felt less involved in the decision-making

regarding carpal tunnel release than they desired. 76% of patients who preferred

shared decision-making had lower scores on the DASH questionnaire compared

with those who preferred a fully active or fully passive role.16 There is some

evidence that decision aids can help patients achieve their preferred role in

decision-making.17,18 In general, patients who actively contribute to their health

care have better functional outcome, choose less invasive treatments, and are

more satisfied with their options.13,19-21 Patients’ outcomes and their satisfaction

seem to be enhanced by higher levels of patient engagement. Providing patients

with their desired level of involvement in decision-making is an important part

of improving patient engagement and clinical results.

It may be surprising that the magnitude of education is the only factor

associated with the desire to participate in shared decision-making and that

age, duration of complaint, magnitude of disability level, pain intensity, and

psychological factors did not have a measurable influence. There is a bias that

shared decision-making is more acceptable to younger patients22, but the

finding that age is not associated with preferences for participation in the

decision-making process agrees with prior studies.3 Furthermore, one might

guess that depressed mood and ineffective coping strategies might make

people feel more resigned, passive, and helpless; our findings suggest that these

factors do not influence preferences for participation in the decision-making

process. Many surgeons are of the opinion that injured patients must rely and

prefer to rely on the surgeon’s advice and feel less capable of participating

in decision-making (as a result of pain, limited time to decide, etcetera) than

patients with non-traumatic problems. One might also assume that older

patients prefer a more paternalistic style and that patients with greater stress,

distress, and less effective coping strategies will be more passive. This study in

combination with prior studies demonstrate that shared decision-making is

preferred by both trauma and non-trauma patients without obvious differences

between those two groups of patients.23-25 Patients, regardless of their level of

education, deserve to participate in shared decision-making, but to give less

well-educated patients the confidence to do so, appropriate tools need to be

108

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

developed. A decision aid appropriate for low levels of health literacy might

increase a less educated patient’s confidence that they can participate in

decision-making. In our opinion, it is safe to assume that all patients prefer to

participate in decision-making unless they suggest otherwise. Surgeons should

provide accurate, balanced, dispassionate information to patients so that they

can understand their preferences. We believe that most surgeons would agree

that, given the uncertainty about the best management of many problems,

the preferences of the patient should feature prominently in decision-making.

Future research should help determine the best way to inform patients so that

they feel adequately involved in the decision-making process and surgeon-to-

surgeon variation in management is minimized. We think decision aids hold

promise for achieving these goals and plan to develop aids and test their impact

on decisional conflict, surgeon-to-surgeon variation, satisfaction with patient

care, symptoms, and disability. Future studies on greater scale are warranted to

assess if decision aids improve health outcome by encouraging patients to take a

more active role in their recovery and reduce variation.

REFERENCES

1. Stiggelbout AM, Van der Weijden T, De Wit MP, et al. Shared decision making: really putting

patients at the centre of healthcare. BMJ 2012;344:e256.2. Stacey D, Bennett CL, Barry MJ, et al. Decision aids for people facing health treatment or

screening decisions. Cochrane Database Syst Rev 2011:CD001431.3. Charles C, Gafni A, Whelan T. Decision-making in the physician-patient encounter: revisiting

the shared treatment decision-making model. Soc Sci Med 1999;49:651-61.4. Légaré F. Establishing patient decision aids in primary care: Update on the knowledge base.

Z Evid Fortbild Qual Gesundhwes 2008;102:427-30.5. Degner LF, Sloan JA, Venkatesh P. The Control Preferences Scale. The Canadian journal of

nursing research = Revue canadienne de recherche en sciences infirmieres 1997;29:21-43.6. Beaton DE, Katz JN, Fossel AH, Wright JG, Tarasuk V, Bombardier C. Measuring the whole or

the parts? Validity, reliability, and responsiveness of the Disabilities of the Arm, Shoulder

and Hand outcome measure in different regions of the upper extremity. J Hand Ther

2001;14:128-46.7. Gummesson C, Ward MM, Atroshi I. The shortened disabilities of the arm, shoulder and hand

questionnaire (QuickDASH): validity and reliability based on responses within the full-length

DASH. BMC musculoskeletal disorders 2006;7:44.8. Asghari A, Nicholas MK. Pain self-efficacy beliefs and pain behaviour. A prospective study.

Pain 2001;94:85-100.9. Nicholas MK. The pain self-efficacy questionnaire: Taking pain into account. Eur J Pain

2007;11:153-63.

109

10. Kroenke K, Spitzer RL, Williams JB. The Patient Health Questionnaire-2: validity of a two-item

depression screener. Medical care 2003;41:1284-92.11. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure.

J Gen Intern Med 2001;16:606-13.12. Cohen JW. In: Hillsdale, ed. Statistical power anlays for behavioral sciences (2nd edn). NJ::

Lawrence Erlbaum Associates; 1988:79-81.13. Hack TF, Degner LF, Dyck DG. Relationship between preferences for decisional control and

illness information among women with breast cancer: a quantitative and qualitative

analysis. Soc Sci Med 1994;39:279-89.14. Hibbard JH, Cunningham PJ. How engaged are consumers in their health and health care,

and why does it matter? Research brief 2008:1-9.15. Uldry E, Schafer M, Saadi A, Rousson V, Demartines N. Patients’ Preferences on Information

and Involvement in Decision Making for Gastrointestinal Surgery. World J Surg 2013.16. Gong HS, Huh JK, Lee JH, Kim MB, Chung MS, Baek GH. Patients’ preferred and retrospectively

perceived levels of involvement during decision-making regarding carpal tunnel release. J

Bone Joint Surg Am 2011;93:1527-33.17. Kennedy AD, Sculpher MJ, Coulter A, et al. Effects of decision aids for menorrhagia on


2002;288:2701-8.18. Murray E, Davis H, Tai SS, Coulter A, Gray A, Haines A. Randomised controlled trial of an

interactive multimedia decision aid on hormone replacement therapy in primary care. BMJ

2001;323:490-3.19. Degner LF, Kristjanson LJ, Bowman D, et al. Information needs and decisional preferences in

women with breast cancer. Jama 1997;277:1485-92.20. Golin C, DiMatteo MR, Duan N, Leake B, Gelberg L. Impoverished diabetic patients whose

doctors facilitate their participation in medical decision making are more satisfied with their

care. J Gen Intern Med 2002;17:857-66.21. Legare F, Stacey D, Briere N, et al. Healthcare providers’ intentions to engage in an

interprofessional approach to shared decision-making in home care programs: a mixed

methods study. J Interprof Care 2013;27:214-22.22. Frosch DL, Kaplan RM. Shared decision making in clinical medicine: past research and future

directions. Am J Prev Med 1999;17:285-94.23. Hutchinson RH, Barrie JL. The effects of shared decision making in the conservative

management of stable ankle fractures. Injury 2015.24. Slover J, Shue J, Koenig K. Shared decision-making in orthopaedic surgery. Clin Orthop Relat

Res 2012;470:1046-53.25. Youm J, Chenok KE, Belkora J, Chiu V, Bozic KJ. The emerging case for shared decision making

in orthopaedics. Instr Course Lect 2013;62:587-94.

110

PART 3

GENERAL DISCUSSION

111

CHAPTER 9

Summary and Discussion

112

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

SUMMARY AND DISCUSSION

There is substantial variation in rates and type of operative and non-operative

treatment that cannot be explained by demographics, pathophysiology, or

comorbidities alone.1-3 Although there should be some variation in medical

treatments, variation from surgeon-to-surgeon is more difficult to justify

than variation from patient-to-patient. The objectives of this thesis were to

address several aspects of orthopaedic decision-making from the surgeon and

patient perspectives: factors associated with variation in recommendation for

operative and non-operative treatment, how health care providers decide which

option to recommend to their patients when the evidence is inconclusive, the

priorities and preferences facing decisions, pre-visit expectation associated with

satisfaction of delivered care, the effect of decision aids, and patient preferences

for shared decision-making in relation to the acuity of the diagnosis and to

psychological factors.

To measure the factors leading to substantial, unexplained variation in

hand surgeon recommendations for treatment of peripheral mono-neuropathy,

chapter 2 tested the null hypothesis that specific patient and provider factors

do not influence recommendations for surgery. Using a web-based survey,

hand surgeons recommended surgical or nonsurgical treatment for patients

in two different scenarios. Six elements of the first scenario (symptoms,

circumstances, mindset, diagnosis, objective testing, and expectations) had

two possibilities that were each independently and randomly assigned to each

rater. For the second scenario, two different scenarios were randomly assigned

to each rater. Multivariable logistic regression sought factors associated with a

recommendation for surgery.

A total of 186 surgeons of the Science of Variation Group completed a

survey regarding recommendation of surgery for two different patients based

on clinical scenarios. Recommendations for surgery did not vary significantly

according to provider characteristics. For the various elements in scenario

1, recommendation for surgery was more likely for patients who were self-

employed and continued to work and who had objective electro-diagnostic

abnormalities. For the two vignettes used in scenario 2, a recommendation for

surgery was associated with abnormal electrophysiology.

The findings of this study suggest that – at least in a survey setting

– surgeons prefer to offer peripheral nerve decompression to patients with

113

abnormal electrophysiology, particularly those with effective coping strategies.

This was an unexpected finding, because if surgeons offer surgery based

primarily on reliable and valid objective testing, there should be limited

variation in treatment recommendations. It may be that the surgeon-observers

in the SOVG are not representative of the average surgeon.

Recommendations for managing proximal humeral fractures vary

substantially.4,5 Recent studies demonstrate substantial surgeon-to-surgeon

inconsistencies in the treatment of these injuries.4,5 We were interested in

the relative influence of patient information and surgeon characteristics

on the decision-making process in treating proximal humeral fractures.

We hypothesized in chapter 3 that there is no difference in treatment

recommendations between surgeons shown radiographs alone and those

shown radiographs and patient information.

We surveyed a total of 238 surgeons who rated forty radiographs of

patients with proximal humerus fractures. Participants were randomized to

receive information about the patient and mechanism of injury. The response

variables included the choice of treatment (operative vs non-operative) and the

percentage of matches with the actual treatment.

Participants who received patient information recommended operative

treatment less than those who received no information. The patient information

that had the greatest influence on treatment recommendations included age

(55%) and fracture mechanism (32%). The only other factor associated with a

recommendation for operative treatment was region of practice. There was no

significant difference between participants who were and were not provided

with information regarding agreement with the actual treatment (operative vs

non-operative) provided by the treating surgeon.

Treatment recommendations for proximal humeral fractures are

influenced by patient information – older age in particular – but most of the

variation in recommendations remains unaccounted for. The highly variable and

inconsistent influence of patient factors on surgeon recommendations belies

variations in surgeon preferences and values that are likely at the root of the

substantial treatment variations documented in this and other studies.

Because evidence-based medicine is an amalgam of individual clinical

expertise and best available evidence, the question arises what is the basis for

provider recommendations when the best evidence is inconclusive? Chapter 4

114

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

tested the null hypothesis that the factors surgeons use do not vary by training,

demographics, and practice. A total of 337 surgeons rated the importance of

seven factors when deciding between treatment and following the natural

history of the disease and twelve factors when deciding between two operative

treatments using a 5-point Likert-scale between “very important” and “very

unimportant.”

We found that the factors that surgeons use most to make

recommendations when evidence is inconclusive relate primarily to the

surgeon’s perspective (e.g. “works in my hands,” “familiarity with the treatment,”

“what my mentor taught me”) rather than the patient’s perspective (e.g.

“doing something vs doing nothing,” “patients are requesting the procedure”).

Exceptions include “fewer complications” and “quicker recovery”. Highest

reimbursement was also rated relatively unimportant, particularly in Europe but

across all countries and regions. The write-in answers revealed that surgeons

prefer to fall back to the “best available outcome/evidence-based” even when

the scenario is that the evidence is inconclusive. Patient-centered care/shared

decision-making was also mentioned.

That health care providers fall back to their personal preferences based

on experience is no surprise.6 Especially, since patients look to their surgeon’s

expertise regarding the optimal fallback options when evidence is inconclusive.

However, where current best evidence allows room for debate, patients can

benefit from an understanding of the range of options and the source of the

debate. The involvement of patients in decision-making is particularly important

when the evidence is inconclusive because patient may value the potential

outcomes of the treatment differently.

The study described in chapter 5 assessed the priorities and preferences

of patients and hand surgeons facing decisions about management of CTS.

We tested the null hypothesis that there are no differences in priorities and

preferences of patients with CTS and hand surgeons.

One hundred three hand surgeons of the Science of Variation Group and

79 patients with CTS completed a survey about their priorities and preferences

in decision-making regarding the management of CTS. The questionnaire

was structured according the Ottawa Decision Support Framework for the

development of a decision aid.7

Important areas on which patient and hand surgeon interests

differed included a preference for non-painful, non-operative treatment and

115

confirmation of the diagnosis with electro-diagnostic testing. For patients, the

main disadvantage of non-operative treatment was that it was likely to be only

palliative and temporary. Patients preferred, on average, to take the lead in

decision-making, whereas physicians preferred shared decision-making. Patients

and physicians agreed on the value of support from family and other physicians

in the decision-making process.

There were some differences between patient and surgeon priorities

and preferences regarding decision-making for CTS, particularly the risks and

benefits of diagnostic and therapeutic procedures. Clinical relevant information

that helps inform patients of their options based on current best evidence might

help patients understand their own preferences and values, reduce decisional

conflict, limit surgeon-to-surgeon variations, and improve health. In particular,

our study identified several areas in which patients would like more information

about CTS: the specific risks and benefits of various diagnostic and therapeutic

options, health provider recommendations, and help clarifying their own

preferences.

The study described in chapter 6 assessed the correlations among

pre-visit expectations, met expectations, and patient satisfaction and what

categories of expectations correlated with satisfaction.

86 new patients presenting to a hand surgery practice of a tertiary

referral hospital with 70% direct primary care referrals, mostly with elective

concerns, indicated their pre-visit expectations (Patient Intention Questionnaire

[PIQ]). Immediately after the visit, the same patients rated the degree to which

their pre-visit expectations were met (Expectation Met Questionnaire [EMQ])

and their satisfaction level (Medical Interview Satisfaction Scale). These tools

have been used in primary care office settings and claim good psychometric

properties, and although they have not been strictly validated for responsiveness

and other test parameters, they have good face validity. We then conducted a

multivariable backward linear regression to determine whether (1) scores on the

PIQ; and (2) scores on the EMQ are associated with satisfaction.

Satisfaction correlated with met expectations but not with pre-visit

expectations. We identified five primary categories of pre-visit expectations that

accounted for 50% of the variance in PIQ: (1) ‘‘Information and Explanation’’; (2)

‘‘Emotional and Understanding’’; (3) ‘‘Emotional Problems’’; (4) ‘‘Diagnostics’’;

and (5) ‘‘Comforting’’. The only category of met expectations that correlated with

satisfaction was Information and Explanation.

116

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

Among patients seeing a hand surgeon, met expectations correlate with

satisfaction. In particular, patients with met expectations regarding information

and explanation were more satisfied with their visit. Efforts to determine

the most effective methods for conveying unexpected information warrant

investigation.

Prior studies showed that decision aids can normalize and depersonalize

these conflicts, provide patients information they can reflect on at their own

pace and in their own way, and ensure that patients are empowered to make

decisions based on an better understanding of their disease.8 We hypothesized

in chapter 7 that there is no difference in decisional conflict comparing patients

with hip- or knee osteoarthroses managed with a decision aid compared to

patients without. Our secondary hypothesis was that there is no difference

between patients managed with a decision aid compared to patients without,

with respect to anxiety, knowledge, satisfaction, preferred treatment at

enrollment and physical function and quality of life, 26 weeks after the

treatment decision was made.

This multi center randomized controlled trial included patients with

knee or hip osteoarthritis who had not consulted an orthopedic surgeon for

the same complaint in the past. At the first encounter, patients enrolled in the

control group were treated according standard care, while patients enrolled in

the intervention group were managed with a decision aid. After the encounter

patients were asked to complete a survey about decisional conflict (DCS),

preferred treatment, gained knowledge, physical function (KOOS/HOOS), pain

(NRS), anxiety (PASS-20) quality of life (EQ-5D) and satisfaction.9-14 The long-term

follow-up was carried out after 26 weeks and evaluated the HOOS/KOOS, PASS,

EQ-5D, satisfaction and the preferred treatment.

The results showed that decision aids help to inform patients about their

options based on current best evidence and helps patients understand their

own preferences and values by reducing decisional conflict. Future studies on

greater scale are warranted to assess if decision aids improve health outcome

by encouraging patients to take a more active role in their recovery and reduce

variation.

Some surgeons contend that patients are less capable of participating

in decisions about traumatic conditions than non-traumatic conditions. We

tested in chapter 8 whether patients with non-traumatic conditions have

117

different preferences for shared decision-making when compared with those

who sustained acute trauma. We also tested whether disability, symptoms

of depression, and self-efficacy correlate with preference for shared decision-

making?

In this prospective, comparative trial, we evaluated a total of 133

patients presenting to the outpatient practices of two university-based hand

surgeons with traumatic or non-traumatic hand and upper extremity illnesses

or conditions. Each patient completed questionnaires measuring their preferred

role in healthcare decision-making (Control Preferences Scale [CPS])15, symptoms

of depression (Patients’ Health Questionnaire)12,16, and pain self-efficacy

(confidence that one can achieve one’s goals despite pain; measured using the

Pain Self-efficacy Questionnaire)17. Patients also completed a short version of the

Disabilities of the Arm, Shoulder, and Hand questionnaire and an ordinal rating

of pain intensity18.

The results showed that there was no difference in decision-making

preferences between patients with traumatic and non-traumatic conditions

with most patients preferring shared decision-making. More educated patients

preferred a more active role in decision-making; however, differences in levels of

disability, pain and function, depression, and pain-related self-efficacy were not

associated with differences in patients’ preferences in terms of shared decision-

making.

Patients who sustained trauma have on average the same preference for

shared decision-making compared with patients who sustained no trauma.

Conclusions and future directions

This thesis identified several opportunities for increasing patient involvement

in decision-making and satisfaction, reducing decision conflict, and limiting

surgeon-to-surgeon variations in care, all of which merit additional study.

First, objective testing should have more influence than symptoms, mindset,

diagnosis, circumstances, and expectations on surgeon recommendations.

Second, surgery is best used to address verifiable objective pathophysiology

where current best evidence confirms that the benefits of surgery outweigh

the harms. The measured influence of patient circumstances on surgeon

recommendations demonstrates how surgeon biases can have inordinate

influence. Third, the evidence that surgeons fall back to their comfort zone,

independent of factors related to the patient’s perspective suggests that if

surgeons were more comfortable with discomfort, they might be more likely

118

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

to acquiesce to patient preferences in areas with no clear best choice. Fourth,

focus on establishing appropriate pre-visit expectations by providing evidenced-

based information in an understandable and meaningful form (e.g. decision

aids) before the visit. Finally, where current best evidence allows room for

debate, patients decision aids might also help limit decision conflict, support

the conversation about the available options and improve health outcomes by

encouraging patients to take a more active role in their care. It might even limit

surgeon-to-surgeon variations, increase safety, efficiency, and resourcefulness.

While the findings of this thesis suggest that decision aids might be

effective in supporting orthopaedic decision-making from the surgeon and

patients perspectives, further implementation studies are needed to determine

how to incorporate decision aids in care pathways on greater scale and to test

the influence on practice variance and health outcomes.

REFERENCES:





American volume 2009;91:1313-9.4. Bruinsma WE, Guitton TG, Warner JJ, Ring D, Science of Variation G. Interobserver reliability of




modalities and experience. Journal of orthopaedic surgery and research 2011;6:38.6. Thomas G, Pring R. Evidence-Based Practise in Education. Youblishercom 2004.7. O’Connor AM S, D, & Jacobsen MJ. Ottawa Decision Support Tutorial (ODST): Improving


Aids, 2011. Web. 2011 Nov 30.8. Stacey D, Bennett CL, Barry MJ, et al. Decision aids for people facing health treatment or

screening decisions. Cochrane Database Syst Rev 2011:CD001431.9. 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;51:1055-68.

119

10. de Groot IB, Favejee MM, Reijman M, Verhaar JA, Terwee CB. The Dutch version of the Knee


outcomes 2008;6:16.11. EuroQol G. EuroQol--a new facility for the measurement of health-related quality of life.

Health policy 1990;16:199-208.12. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure.

J Gen Intern Med 2001;16:606-13.13. McCracken LM, Dhingra L. A short version of the Pain Anxiety Symptoms Scale (PASS-20):

preliminary development and validity. Pain research & management : the journal of the

Canadian Pain Society = journal de la societe canadienne pour le traitement de la douleur

2002;7:45-50.14. van Oldenrijk J, Sierevelt IN, Haverkamp D, Harmse IW, Poolman RW. Re: Validation of the

Dutch version of the Hip disability and Osteoarthritis Outcome Score (HOOS). Osteoarthritis

Cartilage 2009;17:133-4.15. Degner LF, Sloan JA, Venkatesh P. The Control Preferences Scale. The Canadian journal of

nursing research = Revue canadienne de recherche en sciences infirmieres 1997;29:21-43.16. Kroenke K, Spitzer RL, Williams JB. The Patient Health Questionnaire-2: validity of a two-item

depression screener. Medical care 2003;41:1284-92.17. Nicholas MK. The pain self-efficacy questionnaire: Taking pain into account. Eur J Pain



DASH. BMC musculoskeletal disorders 2006;7:44.

120

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

121

CHAPTER 10

Dutch summary and discussion

Nederlandse samenvatting en

discussie

122

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

SAMENVATTING EN DISCUSSIE

Er bestaat grote praktijkvariatie in de frequentie en het soort operatieve en niet-

operatieve behandelingen die niet kan worden verklaard door demografische

gegevens, pathofysiologie of co-morbiditeit.1-3 Hoewel er altijd enige variatie

in medische behandelingen bestaat, is die van chirurg tot chirurg moeilijker

te rechtvaardigen dan die van patiënt tot patiënt. De doelstelling van dit

proefschrift was om diverse aspecten van besluitvorming te adresseren vanuit

zowel het perspectief van de patiënt als dat van de chirurg. Het gaat om

de factoren die verband houden met de variatie in aanbevelingen voor een

operatieve of niet-operatieve behandeling, met de vraag hoe zorgverleners voor

een behandeloptie beslissen als de bewijslast vanuit de literatuur ontoereikend

is, met de prioriteiten en voorkeuren die patiënten en chirurgen hebben

wanneer zij een besluit moeten nemen, met de relatie tussen de verwachtingen

omtrent de zorg en de mate van tevredenheid over de zorg die is ontvangen, met

het effect van keuzehulpen en met de voorkeuren van patiënten ten aanzien van

samen beslissen, in relatie tot psychologische factoren en de diagnose.

Om zicht te krijgen op de factoren die leiden tot een grote, onverklaarde

variatie tussen handchirurgen in hun aanbevelingen over de behandeling van

perifere neuropathie, hebben we in hoofdstuk 2 de nulhypothese getest dat

bepaalde patiënt- en chirurgafhankelijke variabelen geen invloed hebben op die

aanbevelingen.

Door gebruik te maken van een online-vragenlijst gaven handchirurgen

een operatief of niet-operatief behandeladvies in twee verschillende scenario’s.

In het eerste werden zes elementen (demografische variabelen, symptomen,

coping, diagnose, aanvullende diagnostiek en verwachtingen) willekeurig

gecombineerd en toegewezen aan de ondervraagden. In het tweede scenario

werden twee vooraf gedefinieerde scenario’s willekeurig toegewezen aan

een beoordelaar. Door middel van een voorspellend rekenmodel (multivariate

regressie) werd getracht factoren te identificeren die een relatie hadden met

het advies om te opereren. In totaal vulden 186 handchirurgen, betrokken bij de

Science of Variation Group, de vragenlijsten over de verschillende scenario’s in.

De resultaten lieten zien dat adviezen voor een operatieve behandeling

niet significant verschilden tussen de ondervraagde chirurgen. Operatieve

behandeling werd in scenario 1 eerder geadviseerd bij patiënten die zelfstandig

werkten en doorwerkten en bij wie sprake was van objectieve elektro-

123

diagnostische afwijkingen. In de twee opties van scenario 2 hing een advies om

te opereren samen elektro-diagnostische afwijkingen.

De bevindingen van deze studies suggereren dat – althans in deze

testsetting – chirurgen liever een operatief behandeladvies geven aan patiënten

met afwijkende elektro-diagnostische resultaten, vooral die patiënten met

een sterke coping-strategie. Dit was een onverwachte bevinding. Wanneer

chirurgen bij voorkeur op basis van betrouwbare en gevalideerde testuitkomsten

aanbevelingen zouden doen, zou er een beperkter verschil in behandeladviezen

moeten zijn.

Ook de aanbevelingen voor de behandeling van proximale humerus

fracturen verschillen sterk.4,5 Recente studies tonen behoorlijke inconsistenties

tussen chirurgen daarin aan. We waren benieuwd naar de invloed van

informatie over de patiënt en de kenmerken van chirurgen op de besluitvorming

rond deze aandoening. In hoofdstuk 3 gingen we uit van de hypothese dat er

geen significant verschil zou worden gevonden tussen de behandeladviezen van

chirurgen die uitsluitend een röntgenfoto van de proximale humerus zouden

krijgen, versus chirurgen die naast de röntgenfoto ook informatie over de

patiënt zouden ontvangen.

Door middel van een online-vragenlijst beoordeelden 238 chirurgen

40 patiënten met een proximale humerus fractuur op basis van een röntgenfoto.

Door willekeurige selectie kregen zij hier wel of geen aanvullende informatie

bij over de patiënt en de toedracht van het ongeval. De uitkomstvariabelen

betroffen het behandeladvies (operatief versus niet-operatief) en het percentage

waarin dat behandeladvies overeenkwam met de uiteindelijke behandeling.

Chirurgen die aanvullende informatie over de patiënt ontvingen,

adviseerden minder vaak een operatie dan de chirurgen die alleen de

röntgenfoto kregen. De patiëntgegevens die de meeste invloed op het

behandeladvies hadden, waren leeftijd (55%) en fractuurmechanisme (32%).

De enige andere factor van invloed was de regio waar de chirurg werkzaam

was. Wat betreft de overeenkomst tussen behandeladvies en uiteindelijke

behandeling was er geen significant verschil tussen chirurgen die wel en

chirurgen die geen aanvullende informatie hadden ontvangen.

Het behandeladvies voor een proximale humerus fractuur wordt

beïnvloed door patiëntinformatie, leeftijd in het bijzonder, maar het grootste

deel van de praktijkvariatie blijft onverklaard. Patiëntinformatie leidt niet tot

een grotere mate van overeenstemming over het behandeladvies tussen de

124

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

ondervraagden onderling, noch tussen de ondervraagden en de uiteindelijke

behandeling.

Omdat evidence based medicine de optelsom is van de individuele

expertise van de arts en het best beschikbare bewijs rijst de vraag hoe

zorgverleners beslissen wanneer dat best beschikbare bewijs ontoereikend is.

Hoofdstuk 4 toetst de nulhypothese dat de factoren die van invloed zijn op de

besluitvorming, niet variëren naar gelang training, demografische gegevens of

specialisme.

In deze studie beoordeelden in totaal 337 chirurgen het belang van zeven

factoren wanneer ze moesten kiezen tussen een behandeling en de ziekte op

zijn beloop laten, en twaalf factoren wanneer ze moesten kiezen tussen twee

operatieve behandelingen. Ze gebruikten daarvoor een 5-punts Likert-scale,

variërend van “zeer onbelangrijk” tot “zeer belangrijk”.

De resultaten laten zien dat de factoren die het meest van invloed

waren op de beslissing van de chirurg, te maken hadden met zijn individuele

perspectief (bijvoorbeeld: “wat werkt in mijn handen”, “bekendheid met de

behandeling”, “wat mijn mentor mij heeft geleerd”) in plaats van met de

perspectieven van de patiënt (bijvoorbeeld: “iets doen versus niets doen”,

“de patiënt vraagt om de behandeling”). Uitzonderingen waren “minder

complicaties” en “sneller herstel”. Financiële compensatie werd ook als

relatief onbelangrijk beoordeeld, met name in Europa, maar ook daarbuiten.

De antwoorden op de openvragen toonden aan dat chirurgen bij voorkeur

terugvallen op de “best beschikbare bewijslast uit de literatuur”, zelfs als die

bewijslast in het scenario als ontoereikend was beschreven. “Patient centered

care” en “gezamenlijke besluitvorming” werden ook genoemd.

Dat zorgverleners terugvallen op hun persoonlijke voorkeur, gebaseerd

op hun ervaring, is geen verrassing.6 Vooral omdat ook patiënten naar die

ervaring kijken wanneer de bewijslast ontoereikend is. Toch kunnen patiënten,

juist als die bewijslast ruimte biedt voor discussie, baat hebben bij een beter

begrip van alle opties en van de bron voor de discussie. Patiënten bij de

besluitvorming betrekken is met name in die situaties belangrijk. Dan kan

het immers zijn dat zij de potentiële behandeluitkomsten verschillend zullen

waarderen.

De studie beschreven in hoofdstuk 5 beoordeelt de waarden

en voorkeuren van patiënten en handchirurgen met betrekking tot de

125

besluitvorming over de behandeling van carpaal tunnel syndroom (CTS). We

testten de nulhypothese dat er geen verschil is tussen de waarden en voorkeuren

van de patiënten en die van de handchirurgen.

103 handchirurgen, betrokken bij de Science of Variation Group, en 79

patiënten met CTS vulden een vragenlijst in over hun waarden en voorkeuren

bij de besluitvorming over de behandeling. De vragenlijst was gebaseerd op het

Ottawa Decision Support Framework voor de ontwikkeling van keuzehulpen.7

Belangrijke gebieden waarop patiënten en chirurgen van mening

verschilden, betroffen de voorkeur voor pijnloze, niet-operatieve behandeling en

de bevestiging van de diagnose door middel van elektro-diagnostische testen.

Patiënten vonden het grootste nadeel van de niet-operatieve behandeling

dat de uitkomst waarschijnlijk van tijdelijk aard was en mogelijk slechts

palliatief. Over het algemeen gaven patiënten aan de leiding te willen hebben

bij het nemen van de beslissing. De handchirurgen gaven de voorkeur aan

gezamenlijke besluitvorming. Patiënten en handchirurgen waren het eens over

de waarde van de ondersteuning van familie en andere zorgverleners bij het

besluitvormingsproces.

Er werden wel enkele verschillen waargenomen tussen de voorkeuren

van patiënten en handchirurgen met betrekking tot het besluitvormingsproces.

Die gingen vooral over de risico’s en voordelen van diagnostische en

therapeutische procedures.

Klinisch relevante informatie, gebaseerd op de meest recente bewijzen

uit de literatuur, zou patiënten kunnen helpen om hun voorkeuren en waarden

beter te begrijpen, keuzestress te verminderen, praktijkvariatie tussen chirurgen

te beperken en gezondheidsuitkomsten te vergroten. Onze studie identificeerde

enkele gebieden waarover patiënten meer informatie wensten: specifieke risico’s

en voor- en nadelen van de verschillende diagnostische en therapeutische opties,

het advies van de zorgverlener en zijn hulp bij de waarde-exploratie.

De studie beschreven in hoofdstuk 6 beoordeelde het verband

tussen de verwachtingen van de patiënt voorafgaand aan het consult, de

patiënttevredenheid en de categorieën binnen de verwachtingen die met de

tevredenheid samenhingen.

86 nieuwe patiënten, waarvan 70% direct door de huisarts naar de

handchirurg waren verwezen, en de meeste met een niet-spoedeisende

zorgvraag, deelden hun verwachting voorafgaand aan het consult (Patient

intention Questionnaire [PIQ]). Direct erna gaven dezelfde patiënten hun

126

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

mening over de vraag in hoeverre aan hun verwachtingen was voldaan

(Expectation Met Questionnaire [EMQ]) en de mate van hun tevredenheid

over het consult (Medical Interview Satisfaction Scale). Door middel van een

multivariate regressie-analyse hebben we vervolgens bepaald in hoeverre de

PIQ- en EMQ-scores verband houden met de mate van tevredenheid.

De resultaten toonden aan dat er wel een relatie is tussen de

tevredenheid en de mate waarin is voldaan aan de verwachtingen, maar niet

met de verwachtingen die patiënten voorafgaand aan het consult hadden.

We hebben vijf categorieën verwachtingen voorafgaand aan een consult

geïdentificeerd. Die verklaarden 50% van de variatie in de PIQ: (1) “Informatie en

uitleg”; (2) “Emotie en begrip”; (3) “Emotionele problemen”; (4) “Diagnostiek”,

en (5) “Comfort”. De enige categorie van ‘beantwoorde verwachtingen’ die

correleerde aan de mate van tevredenheid, was “Informatie en uitleg”.

De mate waarin verwachtingen zijn beantwoord is dus gerelateerd

aan de mate van tevredenheid. Vooral patiënten van wie de verwachtingen

over “Informatie en uitleg” werden beantwoord, waren meer tevreden met het

consult. Om te bepalen wat de meest effectieve methode is om informatie over

te brengen die de patiënt niet verwacht, is meer onderzoek nodig.

Eerdere studies hebben aangetoond dat keuzehulpen tegenstellingen

in de spreekkamer kunnen normaliseren en minder persoonlijk kunnen maken.

Ze geven patiënten informatie waar zij op hun eigen manier en in hun eigen

tempo over na kunnen denken. Bovendien rusten keuzehulpen patiënten toe

om een beslissing te nemen die gebaseerd is op een beter begrip van hun ziekte

of aandoening. In hoofdstuk 7 stelden we de hypothese dat er geen verschil in

keuzestress is tussen patiënten met heup- of knieslijtage die de ondersteuning

van een keuzehulp kregen en patiënten die dat niet kregen. Onze tweede

hypothese was dat er geen verschil is tussen deze twee groepen met betrekking

tot angst, kennis, tevredenheid, de behandeling die de voorkeur kreeg bij

inschrijving, de fysieke functie en kwaliteit van leven op de langere termijn

(26 weken nadat het besluit over de behandeling was genomen).

In deze gerandomiseerde onderzoeken met controlegroepen, die in

diverse instellingen werden gehouden, werden alleen patiënten met heup- of

knieslijtage betrokken die niet al eens met deze klacht bij een orthopedisch

chirurg waren geweest. Gedurende het eerste consult kregen de patiënten uit

de controlegroep de standaardzorg. De patiënten uit de interventiegroep kregen

de beschikking over een keuzehulp. Direct na het consult en nog eens na

127

26 weken werden de mate van keuzestress (DCS), angst (PASS-20), kennis, fysieke

functie (HOOS/KOOS), pijn (NRS), tevredenheid en kwaliteit van leven (EQ-5D)

gemeten.8-13

De resultaten toonde aan dat keuzehulpen patiënten helpen hun diagnose

en behandelopties te begrijpen, aan te geven wat zij belangrijk vinden en

keuzestress te verminderen. Toekomstig onderzoek op grote schaal is nodig om te

beoordelen wat de invloeden zijn van keuzehulpen op gezondheidsuitkomsten, de

betrokkenheid van patiënten gedurende hun herstel en praktijkvariatie.

Sommige zorgverleners zijn van mening dat patiënten met een

spoedeisende hulpvraag minder in staat zijn deel te nemen aan gezamenlijke

besluitvorming dan patiënten met een niet-spoedeisende hulpvraag. In

hoofdstuk 8 hebben we getoetst of deze patiënten een andere voorkeur hebben

voor gezamenlijke besluitvorming dan patiënten met een niet-spoedeisende

zorgvraag.

Aan deze prospectieve, cohortvergelijkende studie hebben 133 patiënten

met een spoedeisende- of niet-spoedeisende aandoening aan de bovenste

extremiteit deelgenomen. Elke patiënt vulde een vragenlijst in over zijn

voorkeursrol in het gezamenlijke besluitvormingsproces (controle preferences

scale [CPS])14, symptomen van depressiviteit (Patients’ Health Questionnaire)11,15

en over zijn coping-strategie (Pain self-efficacy)16. Patiënten vulden ook een

vragenlijst in over hun fysieke functie (Disability Arm, Shoulder and Hand

Questionnaire) en de mate van pijn die zij ervoeren17.

De resultaten toonden aan dat er geen verschil was in de mate van

voorkeur tussen patiënten die een spoedeisende- of niet-spoedeisende

zorgvraag hadden. Patiënten met een hogere opleiding gaven meer dan

lageropgeleide patiënten de voorkeur aan een actievere rol. Hun fysieke functie,

depressiviteit en coping-strategie vertoonden geen relatie met verschillende

voorkeursrollen in het besluitvormingsproces.

De resultaten laten zien dat patiënten met een spoedeisende

zorgvraag een vergelijkbare hoge mate van voorkeur voor een actieve rol in

het besluitvormingsproces hebben als patiënten met een niet-spoedeisende

zorgvraag. Hogeropgeleide patiënten gaven de voorkeur aan een actievere rol

in de besluitvorming, maar er was geen verband met verschillen in pijn en

functioneren, depressie en coping-strategie. Patiënten met een spoedeisende

zorgvraag hebben dus over het algemeen dezelfde voorkeur voor gezamenlijke

besluitvorming als patiënten met een niet-spoedeisende zorgvraag.

128

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

Conclusie mogelijk toekomstig onderzoek

Dit proefschrift beschrijft diverse mogelijkheden om de betrokkenheid van de

patiënt in het besluitvormingsproces over zijn behandeling te vergroten, de

tevredenheid over de geleverde zorg te verhogen, keuzestress te verminderen en

praktijkvariatie te beperken. Die verdienen allemaal aanvullende studie.

Ten eerste concluderen we dat objectieve klinische testen, van grotere

invloed zouden moeten zijn op de aanbevelingen van de zorgverlener dan

symptomen, mindset, diagnoses, omstandigheden en verwachtingen van de

patiënt.

Ten tweede, De gemeten invloed van patiëntgebonden factoren op het

advies van chirurgen toont aan dat het vooroordeel van de chirurg buitensporige

invloed kan hebben.

Ten derde, het gegeven dat chirurgen terugvallen op hun comfort zone,

ongeacht factoren gerelateerd aan het perspectief van de patiënt, suggereert

dat als chirurgen minder moeite zouden hebben om uit hun comfort zone te

treden, zij bij gebrek aan bewijs voor een optimale behandeling de voorkeur van

patiënten vaker zouden honoreren.

Ten vierde, reële verwachtingen over de uitkomst van een behandeling

kunnen worden bevorderd door patiënten evidence-based informatie in

een duidelijke en betekenisvolle vorm te geven (bijvoorbeeld in de vorm van

keuzehulpen), en door begeleiding.

Tot slot, als er sprake is van een gebrek aan bewijs en als uitkomsten

van behandelingen door patiënten verschillend kunnen worden gewaardeerd,

dan kunnen keuzehulpen keuzestress verminderen, het besluitvormingsproces

bevorderen, totdat één van de behandelopties de voorkeur heeft, de kwaliteit van

zorg bevorderen doordat patiënten een actievere rol vervullen in hun herstel en

de mate van tevredenheid bevordert.

Door patiënten actief te laten participeren in het besluitvormingsproces

kan ongewenste praktijkvariatie tussen chirurgen worden teruggebracht en

kunnen de veiligheid, efficiëntie en duurzaamheid van de zorg worden bevorderd.

Ondanks dat de bevindingen van dit proefschrift suggereren dat

keuzehulpen effectief kunnen zijn in de orthopedische praktijk, is toekomstig

onderzoek op grote schaal nodig om te beoordelen wat de invloeden zijn van

keuzehulpen op gezondheidsuitkomsten, de betrokkenheid van patiënten en

praktijkvariatie.

129

REFERENTIES:





American volume 2009;91:1313-9.4. Bruinsma WE, Guitton TG, Warner JJ, Ring D, Science of Variation G. Interobserver reliability of




modalities and experience. Journal of orthopaedic surgery and research 2011;6:38.6. Thomas G, Pring R. Evidence-Based Practise in Education. Youblishercom 2004.7. O’Connor AM S, D, & Jacobsen MJ. Ottawa Decision Support Tutorial (ODST): Improving


Aids, 2011. Web. 2011 Nov 30.8. 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;51:1055-68.9. de Groot IB, Favejee MM, Reijman M, Verhaar JA, Terwee CB. The Dutch version of the Knee


outcomes 2008;6:16.10. EuroQol G. EuroQol--a new facility for the measurement of health-related quality of life.

Health policy 1990;16:199-208.11. Kroenke K, Spitzer RL, Williams JB. The Patient Health Questionnaire-2: validity of a two-item

depression screener. Medical care 2003;41:1284-92.12. McCracken LM, Dhingra L. A short version of the Pain Anxiety Symptoms Scale (PASS-20):

preliminary development and validity. Pain research & management : the journal of the

Canadian Pain Society = journal de la societe canadienne pour le traitement de la douleur

2002;7:45-50.13. van Oldenrijk J, Sierevelt IN, Haverkamp D, Harmse IW, Poolman RW. Re: Validation of the

Dutch version of the Hip disability and Osteoarthritis Outcome Score (HOOS). Osteoarthritis

Cartilage 2009;17:133-4.14. Degner LF, Sloan JA, Venkatesh P. The Control Preferences Scale. The Canadian journal of

nursing research = Revue canadienne de recherche en sciences infirmieres 1997;29:21-43.15. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure.

J Gen Intern Med 2001;16:606-13.16. Nicholas MK. The pain self-efficacy questionnaire: Taking pain into account. Eur J Pain



DASH. BMC musculoskeletal disorders 2006;7:44.

130

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

PhD-portfolio

Name PhD-student: Michiel G.J.S. Hageman

PhD-period: November 2011 - July 2014

Name PhD-supervisor: Prof. dr. D. Ring, Prof. dr. C.N. van Dijk

1 PhD-training

General coursesThe Principles and Practice of Clinical Research Data

management (MGH). Ethics and Clinical Research Protocols (MGH).IRB and QI Roundtable Series: Consent Form Writing (MGH).How to Give a Presentation (MGH).What does the IRB Really Want? How to Write Human

Studies Protocol (MGH).Specific coursesBasic Biostatistics to Clinical Research (MGH/HMS).Introduction to Clinical Investigation Training Course (HMS).Design and Implementation of Clinical Trials (HMS).Design and Conduct of Clinical Trials (MGH/HMS).Certificate in Applied Biostatistics: (HMS).Applied Biostatistics for Clinical Trials (MGH/HMS).An Introduction to the Enhanced RPDR Query Tool (MGH).Seminars, workshops and master classesClinical Research 101.Orientation Program: Clinical research at MGH.Dr. Baratz visiting MGH.Dr. Szabo visiting MGH.Dr. Morrey visiting MGH.PROMIS/Assessment Center Course.Redcap Course.Podium presentations/ international conferencesNVT: Proximal Humeral Fractures: Operative versus

Conservative treatment. M. Hageman, D. Meijer, S. Stufkens, J. Ultee, J. Doornberg, E. Steller.

Smith Day: Variation in recommendations for operative treatment for compressive neuropathy. M. Hageman, S. Becker, A. Bot, T. Guitton, D. Ring.

Year

2011

2011201220122012

2011201120122012201220132013

2011201120122012201220122012

2012

2013

2013

Workload(Hours/ECTS)

0.1

0.20.10.10.1

112225

0.5

0.20.20.20.20.10.10.1

1

1

1

continue >

131

Smith Day: Spectrum and Trends in Complaints to the Patient Advocate. P. van Dijk, M. Hageman, J. King, C. Overbeek, D. Ring.

NEHS: How surgeons make decisions when the evidence is inconclusive.

Hageman MG, Guitton TG, Ring D; Science of Variation Group.Harvard Orthopaedic Trauma day: Predictors of Readmission

within 30 days of Orthopaedic Surgery. M. Hageman, T. Voskuyl, J. Bossen, J. Blauth, M. Smith, D. Ring.

AAHS: How surgeons make decisions when the evidence is inconclusive. Naples Florida, USA. Hageman MG, Guitton TG, Ring D; Science of Variation Group.

NOV: The effect of decion aids on patient with hip or knee osteoarthritis. M. Hageman, R. Poolman, J. Du Long, T. Vervest, D. Haverkamp.

Poster presentationASSH: Internet Self-diagnosis in Hand Surgery. M. Hageman,

J. Anderson, R. Blok, J. Bossen, D. Ring.

Year

2013

2013

2013

20132013

2016

2013

Workload(Hours/ECTS)

1

1

1

11

1

1

continued

132

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

2 Teaching

3 Parameters of Esteem

Tutoring, mentoring and supervisingJillian Gruber John King Pim van Dijk Jeroen Bossen Charlotte HoogstinsChristiaan SwellengrebelJan Paul BriëtSjoerd NotaMariano MenendezJade AndersonAhmet KinaciMark van SuchtelenStijn BekkersRobin BlokDennis MakarawungTimothy VoskuijlThijs OosterhoffSilke SpitEmily Thornton Joost StrookerAnne-Carolin DöringCeleste OverbeekPrakash JayakumarNick WickramasingheJos MellemaStein JanssenTeun TeunisDirk ter MeulenSaroj GolayRajesh ReddyJsamijn du LongAlexander RakicDick Vuijk

GrantsStichting Anna fonds| NOREFStichting Marti Keuning Eckhardt fondsStichting Achmea Gezondheidszorg

Year

201220122012201220122012201220122013201320132012201220122012201220122012201320132013201320132013201320132013201320132013201420142014

Year

201120112013

Workload

0.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.50.5

133

List of Publications

1. Barber LA, Hageman MG, King JD, Bekkers S, Bot AG, Ring D. The influence of patients’

participation in research on their satisfaction. J Hand Surg Am. 2014;39:1591-1594 e1593.

2. Becker SJ, Briet JP, Hageman MG, Ring D. Death, Taxes, and Trapeziometacarpal Arthrosis. Clin

Orthop Relat Res. 2013.

3. Beulen L, van den Berg M, Faas BH, Feenstra I, Hageman M, van Vugt JM, Bekker MN. The effect

of a decision aid on informed decision-making in the era of non-invasive prenatal testing: a

randomised controlled trial. Eur J Hum Genet. 2016.

4. Bossen JK, Hageman MG, King JD, Ring DC. Does Rewording MRI Reports Improve Patient

Understanding and Emotional Response to a Clinical Report? Clin Orthop Relat Res. 2013.

5. Briet JP, Bot AG, Hageman MG, Menendez ME, Mudgal CS, Ring DC. The Pain Self-Efficacy

Questionnaire, Validation of an Abbreviated Two-Item Questionnaire. Psychosomatics. 2014.

6. Briet JP, Hageman MG, Blok R, Ring D. When do patients with hand illness seek online health

consultations and what do they ask? Clin Orthop Relat Res. 2014;472:1246-1250.

7. Briet JP, Hageman MG, Overbeek CL, Mudgal C, Ring DC, Vranceanu AM. Factors Associated

With Met Expectations in Patients With Hand and Upper Extremity Disorders: A Pilot Study.

Psychosomatics. 2016;57:401-408.

8. Bruinsma W, Kodde I, de Muinck Keizer RJ, Kloen P, Lindenhovius AL, Vroemen JP, Haverlag

R, van den Bekerom MP, Bolhuis HW, Bullens PH, Meylaerts SA, van der Zwaal P, Steller

PE, Hageman M, Ring DC, den Hartog D, Hammacher ER, King GJ, Athwal GS, Faber KJ,

Drosdowech D, Grewal R, Goslings JC, Schep NW, Eygendaal D. A randomized controlled trial

of nonoperative treatment versus open reduction and internal fixation for stable, displaced,

partial articular fractures of the radial head: the RAMBO trial. BMC Musculoskelet Disord.

2014;15:147.

9. Doring AC, Nota SP, Hageman MG, Ring DC. Measurement of upper extremity disability

using the Patient-Reported Outcomes Measurement Information System. J Hand Surg Am.

2014;39:1160-1165.

10. du Long J, Hageman M, Vuijk D, Rakic A, Haverkamp D. Facing the decision about the

treatment of hip or knee osteoarthritis: What are patients’ needs? Knee Surg Sports

Traumatol Arthrosc. 2016;24:1710-1716.

11. Finger A, Teunis T, Hageman MG, Thornton ER, Neuhaus V, Ring D. Do patients prefer optional

follow-up for simple upper extremity fractures: A pilot study. Injury. 2016.

12. Gruber JS, Hageman M, Neuhaus V, Mudgal CS, Jupiter JB, Ring D. Patient activation and

disability in upper extremity illness. J Hand Surg Am. 2014;39:1378-1383 e1373.

13. Hageman MG, Anderson J, Blok R, Bossen JK, Ring D. Internet self-diagnosis in hand surgery.

Hand (N Y). 2015;10:565-569.

14. Hageman MG, Bossen JK, King JD, Ring D. Surgeon confidence in an outpatient setting. Hand

(N Y). 2013;8:430-433.

15. Hageman MG, Bossen JK, Neuhaus V, Mudgal CS, Ring D, Science of Variation G. Assessment of

Decisional Conflict about the Treatment of carpal tunnel syndrome, Comparing Patients and

Physicians. Arch Bone Jt Surg. 2016;4:150-155.

16. Hageman MG, Bossen JK, Smith RM, Ring D. Predictors of readmission in orthopaedic trauma

surgery. J Orthop Trauma. 2014;28:e247-249.

17. Hageman MG, Briet JP, Bossen JK, Blok RD, Ring DC, Vranceanu AM. Do Previsit Expectations

Correlate With Satisfaction of New Patients Presenting for Evaluation With an Orthopaedic

Surgical Practice? Clin Orthop Relat Res. 2014.

134

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

18. Hageman MG, Briet JP, Oosterhoff TC, Bot AG, Ring D, Vranceanu AM. The Correlation of

Cognitive Flexibility with Pain Intensity and Magnitude of Disability in Upper Extremity

Illness. Journal of hand and microsurgery. 2014;6:59-64.

19. Hageman MG, Jayakumar P, King JD, Guitton TG, Doornberg JN, Ring D, Science of Variation

G. The factors influencing the decision making of operative treatment for proximal humeral

fractures. J Shoulder Elbow Surg. 2014.

20. Hageman MG, Reddy R, Makarawung DJ, Briet JP, van Dijk CN, Ring D. Do Upper Extremity

Trauma Patients Have Different Preferences for Shared Decision-making Than Patients With

Nontraumatic Conditions? Clin Orthop Relat Res. 2015.

21. Hageman MG, Ring DC, Gregory PJ, Rubash HE, Harmon L. Do 360-degree Feedback Survey

Results Relate to Patient Satisfaction Measures? Clin Orthop Relat Res. 2014.

22. Janssen SJ, Ter Meulen DP, Hageman MG, Earp BE, Ring D. Quantitative 3-dimensional CT

analyses of fractures of the middle phalanx base. Hand (N Y). 2015;10:210-214.

23. Janssen SJ, Ter Meulen DP, Nota SP, Hageman MG, Ring D. Does verbal and nonverbal

communication of pain correlate with disability? Psychosomatics. 2015;56:338-344.

24. Janssen SJ, Teunis T, ter Meulen DP, Hageman MG, Ring D. Estimation of base of middle

phalanx size using anatomical landmarks. J Hand Surg Am. 2014;39:1544-1548.

25. Kortlever JT, Janssen SJ, Molleman J, Hageman MG, Ring D. Discrete Pathophysiology is

Uncommon in Patients with Nonspecific Arm Pain. Arch Bone Jt Surg. 2016;4:213-219.

26. Mellema JJ, O’Connor CM, Overbeek CL, Hageman MG, Ring D. The effect of feedback

regarding coping strategies and illness behavior on hand surgery patient satisfaction and

communication: a randomized controlled trial. Hand (N Y). 2015;10:503-511.

27. Menendez ME, Bot AG, Hageman MG, Neuhaus V, Mudgal CS, Ring D. Computerized adaptive

testing of psychological factors: relation to upper-extremity disability. J Bone Joint Surg Am.

2013;95:e149.

28. Neuhaus V, King J, Hageman MG, Ring DC. Charlson comorbidity indices and in-hospital

deaths in patients with hip fractures. Clin Orthop Relat Res. 2013;471:1712-1719.

29. Nota SP, Spit SA, Oosterhoff TC, Hageman MG, Ring DC, Vranceanu AM. Is Social Support

Associated With Upper Extremity Disability? Clin Orthop Relat Res. 2016;474:1830-1836.

30. Nota SP, Spit SA, Voskuyl T, Bot AG, Hageman MG, Ring D. Opioid Use, Satisfaction, and Pain

Intensity After Orthopaedic Surgery. Psychosomatics. 2015;56:479-485.

31. Overbeek CL, Nota SP, Jayakumar P, Hageman MG, Ring D. The PROMIS Physical Function

Correlates With the QuickDASH in Patients With Upper Extremity Illness. Clin Orthop Relat

Res. 2014.

32. Strooker JA, Nota SP, Hageman MG, Ring DC. Patients With Greater Symptom Intensity and

More Disability are More Likely to be Surprised by a Hand Surgeon’s Advice. Clin Orthop Relat

Res. 2014.

33. Ten Have IA, van den Bekerom MP, van Deurzen DF, Hageman MG. Role of decision aids in

orthopaedic surgery. World J Orthop. 2015;6:864-866.

34. ter Meulen DP, Janssen SJ, Hageman MG, Ring DC. Quantitative three-dimensional computed

tomography analysis of glenoid fracture patterns according to the AO/OTA classification.

J Shoulder Elbow Surg. 2016;25:269-275.

35. Ter Meulen DP, Nota SP, Hageman MG, Ring DC. Progression of Heterotopic Ossification

around the Elbow after Trauma. Arch Bone Jt Surg. 2016;4:228-230.

36. Ubbink DT, Hageman MG, Legemate DA. Shared Decision-Making in Surgery. Surgical

technology international. 2015;26:31-36.

37. Voskuijl T, Hageman M, Ring D. Higher Charlson Comorbidity Index Scores are associated with

135

readmission after orthopaedic surgery. Clin Orthop Relat Res. 2014;472:1638-1644.

38. Vranceanu AM, Hageman M, Strooker J, ter Meulen D, Vrahas M, Ring D. A preliminary RCT of a

mind body skills based intervention addressing mood and coping strategies in patients with

acute orthopaedic trauma. Injury. 2015;46:552-557.

39. Wickramasinghe NR, Duckworth AD, Clement ND, Hageman MG, McQueen MM, Ring D. Acute

Median Neuropathy and Carpal Tunnel Release in Perilunate Injuries Can We Predict Who

Gets a Median Neuropathy? Journal of hand and microsurgery. 2015;7:237-240.

Acknowledgements

I have been undeservedly lucky to work with people who are incredibly talented,

who were willing to share their wisdom and gracefulness and pass it off as my

own. Many people helped me carrying out scientific endeavors and supported

me while living abroad. Besides all the patients, who participated I would like to

say special thanks to:

Prof. David Ring, the mentor, editor and principal investigator of this

thesis. David Ring provided me the ideal research factory, where he patiently

gave me the opportunity to grow and make me fall in love with carrying out

research. Everyone should have a teacher, coach as good and as generous as

David. David wrote long and extraordinary critiques of the early drafts of the

manuscript. I am thankful that I had a chance to know him, work with him, and

learn from him.

A big thank you to my prof. Niek van Dijk, who deftly and thoughtfully

guided me and shared his academic and orthopaedic expertise. I always looked

forward to our meetings in Amsterdam to discuss the progress of this thesis.

I am also very grateful for our elaborate conversations about my future and your

guidance in making the right decisions.

Dr. Daniel Haverkamp, Prof. Gino Kerkhoffs, Dr. Rudolf Poolman and

Dr. Ton Vervest thank you for your support in carrying out the first decision aids

study. We have been able to achieve a lot in a short time frame.

Many thanks also to the talented and supportive colleagues at the

Orthopaedic Hand and Upper Extremity Service: Johann Blauth, Arjan Bot,

Stephanie Becker and Valentin Neuhaus who taught me statistics, helped

136

Dec

isio

n-m

akin

g in

Ort

hop

aed

ic S

urg

ery

improve the various manuscripts and supported the long afternoons at the back

office. We had a great time. I owe an enormous debt to my many co-authors

and colleagues (Jade Anderson, Stijn Bekkers, Robin Blok, Jeroen Bosen, Jan Paul

Briët, Pim van Dijk, Anne Caroline Döring, Saroj Golay, Jillian Gruber, Charlotte

Hoogstins, Stein Janssen, Prakash Jayakumar, Ahmet Kinaci, John King, Jasmijn

du Long, Dennis Makrawung, Mariano Menendez, Jos Mellema, Dirk ter Meulen,

Sjoerd Nota, Thijs Oosterhoff, Celeste Overbeek, Rajesh Reddy, Alexander Rakic,

Silke Spit, Joost Strooker, Mark van Suchtelen, Christiaan Swellengrebel, Teun

Teunis, Emily Thornton, Timothy Voskuijl, Dick Vuijck and Nick Wickramasinghe),

whose great ideas fill this manuscript, and to all the kind people who have taken

the time to teach me what I know about carrying out scientific research.

I am forever thankful to Stefan Breugem, Job Doornberg and Gerard

Schaap, who gave me my first insights in carrying out research, shared their

enthusiasm for orthopaedic surgery and recommended to start my PhD-

program as research assistant at the Hand and Upper Extremity Service. I also

owe a debt to the many people who were generous in sharing their time to

improve my academic and orthopaedic endeavors, like Jakob van Oldenrijk and

Inger Sierevelt.

My “paranimfen” Frederick Mansell en Christophe Wijffels are wonderful

friends and sources of inspiration. I am also very grateful for the Danes: Sjoerd

Nota, Jan Paul Briët, Peter Paul Zwetsloot and Olvert Berkhemer for having such

a great time in Boston, Somerville, throwing BBQ parties and your friendship.

I would like to thank one person in particular: Nicoline, whose love,

support, guidance, critical view, intelligence and most of all friendship make

every day a joy.

I am blessed to have family, friends and team mates who contributed

indirectly to this thesis, by supporting me while I was abroad, who showed

me it was okay to make sacrifices and made my return to the Netherlands feel

as if had never been away. My parents, Gerard and Merel, and sisters, Lois en

Annemijn, encouraged and supported me to achieve my dreams.

Curriculum Vitae

Michiel Hageman was born in Al Jubail, Saudi Arabia on April 7th, 1985. After a

short interlude in The Netherlands, Michiel lived with his family in Malaysia until

1992. Back in the Netherlands, after graduating from high school (VWO, Den Haag)

in 2004, he studied at the medical school of the University of Amsterdam. During

his study Michiel worked for the Bio-Implant Service (BIS) the Netherlands as

orthopaedic tissue-donation surgeon. In his final year of his bachelor he conducted

a research internship at the department of orthopaedic surgery of the Academic

Medical Center Amsterdam (prof. dr. C.N. van Dijk). The experiences at BIS, his

research internship and clinical internship at the AMC made him enthusiastic to

continue working in the medical field of orthopaedic surgery. After obtaining the

medical doctor’s degree in 2011, he worked as PhD student at the department of

Orthopaedic Hand and Upper Extremity of the Massachusetts General Hospital,

Boston – United States as well as the Slotervaart Ziekenhuis in Amsterdam, which

finally resulted in this thesis. During his time in Boston Michiel developed a special

interest in “Shared Decision Making” and “Decision Aids” to facilitate the decision-

making. Together with his friend and colleague Teun Teunis, Michiel launched

PATIENT+, dedicated to support shared decision-making with digital decision

aids. Subsequently Michiel and Teun wrote the book SAMEN Beslissen: waarom

moeilijk doen als het SAMEN kan? and were awarded the best value best health care

initatieve of 2017 (Doelmatigheidsprijs 2017).

In 2014, Michiel started his training for orthopaedic surgery at the department

of general surgery at the Onze Lieve Vrouwe Gasthuis (dr. M.Gerhards).

He continued his residency at the department of orthopaedic surgery at the AMC

(prof. dr. C.N. van Dijk) and Slotervaart Ziekenhuis (dr. H. van der Vis). During his

clinical work, Michiels’ interests to innovate and develop products to improve

health care further increased. At the end of 2017 he decided to focus solely on

PATIENT+. Michiel will lead and support the team of PATIENT+ to develop, integrate

and evaluate decision aids into innovative health care delivery systems.

ISBN 978 94 91549 88 5

omsl.proefschrift.Hageman.indd 1 26-02-18 14:06