Adolescent Diabetes Management and Mismanagement

N A L A R T I C L E

Adolescent DiabetesManagement andMismanagementJILL WEISSBERG-BENCHELL, PHDALLEN M. GLASGOW, MDW. DOUGLAS TYNAN, PHD

PHILIP WIRTZ, PHDJANE TUREK, MSN, CDEJOSEPH WARD, RN, CDE

OBJECTIVE — To document the existence and prevalence of adolescent-generateddiabetes management techniques.

RESEARCH DESIGN A N D METHODS— One hundred forty-four adoles-cents completed the confidential questionnaire developed for this study. Glycohemo-globin was also obtained for each individual.

RESULTS — Within the 10 days before their clinic visit, many adolescents admittedto engaging in various mismanagement behaviors, with 25% admitting to missingshots. Parents tend to underestimate adolescent mismanagement. Missing shots wassignificantly related to poor control (P < 0.01). Older adolescents engaged in moremismanagement than their younger cohorts (P < 0.001). The questionnaire factoredinto two subscales: blatant mismanagement and faking.

CONCLUSIONS — This study shows the importance of recognizing the preva-lence of mismanagement among adolescents.

A major impediment to maintainingthe health of individuals with dia-betes is poor adherence to the pre-

scribed regimen. The complex, multifac-eted regimen makes nonadherence highlyprobable. Adolescents with insulin-de-pendent diabetes mellitus (IDDM) areless likely to adhere to their prescribedmedical regimens than either theiryounger or older cohorts (1-4). Studiessuggest that readmissions for diabetic ke-

toacidosis are primarily due to major de-viations from recommended therapy,such as missing insulin shots (5-9). Nev-ertheless, the prevalence of poor adher-ence among adolescents and its contribu-tion to poor control remain unclear. Thismay be true, in part, because none of theexisting studies directly asked aboutmissing shots or other major deviationsfrom the regimen (3,10-16).

The first aim of the present study

From Children's National Medical Center and George Washington University (P.W.), Washing-ton, DC.

Address correspondence and reprint requests to Jill Weissberg-Benchell, Ph.D., Children'sNational Medical Center, Department of Endocrinology, 111 Michigan Avenue NW, Washing-ton, DC 20010.

Received for publication 22 June 1992 and accepted in revised form 18 August 1994.IDDM, insulin-dependent diabetes mellitus.

was to document the existence and prev-alence of adolescent-generated manage-ment techniques or diabetes mismanage-ment. We were particularly interested intwo major behaviors: missing shots andmaking up results of/missing blood tests,although other regimen behaviors weredeemed vital to the study. The second aimof the study was to evaluate the reasonsgiven by adolescents for their mismanage-ment. Finally, the study was designed toassess the relationship between misman-agement and glycohemoglobin. Furtherexplorations of these issues are of para-mount importance, since adolescentsmay believe they are doing somethingadaptive, when in reality, mismanage-ment can have dangerous physiologicalimplications.

RESEARCH DESIGN ANDMETHODS

SubjectsAll IDDM patients between the ages of11 and 19 at least 1 year post-diagnosis,literate, and without a major psychiatricdiagnosis were considered eligible to par-ticipate in the study. Asked to participatewere 156 adolescents (cared for by thesame multidisciplinary diabetes team inan American Diabetes Association-ap-proved education program) attendingconsecutive diabetes clinic outpatient ap-pointments at our hospital; 144 adoles-cents agreed (mean age 14.6 years, SD =2.3). The group was equally divided be-tween males and females. Of those pro-viding race information, 94 were white,41 were black, and 5 were "other." Therange for duration of diabetes was 1-16years (mean = 5.34, SD = 3.62).

MeasuresThe diabetes mismanagement question-naire is a 10-item multiple-choice test(see APPENDIX). Items were derived fromself-reports by patients seen in our clinic,in educational groups, and in therapy ses-sions. The questions addressed three ba-

DIABETES CARE, VOLUME 18, NUMBER 1, JANUARY 1995 77

Diabetes mismanagement

Table 1—Admitted and perceived mismanagement in 10 days before clinic visit

Missed insulin injectionMissed injection and then took extra

to coverMade up blood test results because

did not do testTook extra insulin to cover

inappropriate foodAte inappropriate foodMissed meals and snacksFixed blood glucose monitor to give

lower numberChanged test strip to give lower

numberMade up blood test results because

real ones were too highFaked illness

144Youths

(%)

2513

29

34

815610

11

29

13

55Parents

(%)

117

24

24

87627

20

18

26

22Doctors

(%)

107

26

17

664320

14

23

12

Range ofDrs.' responses

0-400-30

5-50

0-60

4-1004-1000-50

5-50

5-50

2-50

Youths refers to teens who admit to engaging in that behavior at least once. Parents refers to those whobelieve their teens engage in that behavior. Doctors refers to mean and range of physicians' estimates of thepercentage of youths in their practice who engage in that behavior.

sic areas of diabetes care: blood testing,insulin shots, and diet.

Subjects were asked how manytimes (within the past 10 days) they hadengaged in the various behaviors. Thequestionnaire was scored on a likert-typescale, with answers ranging from 1 (nev-er) to 5 (frequently). In keeping with ourspecific interest in missing shots andblood tests, two additional questionswere posed. One asked for reasons whyshots were missed, and the other askedfor reasons why blood tests were missed.A list of possible answers were provided,as were blank spaces for adolescents to fillin additional answers.

Glycohemoglobin was measuredby Quickstep column (Bio-Rad, Rich-mond, CA). This test was excluded inthree patients known to have abnormalhemoglobin.

ProcedureThe study's purpose was reviewed witheach participant. Anonymity was ex-

plained and assured to each subject. Afteradolescents completed the questionnaire,a blood sample was obtained for anHbAlc.

We were interested in comparingthe admitted mismanagement practices ofour adolescents with the expectations formismanagement held by both parentsand pediatric endocrinologists. Towardthe end of our study, 55 consecutive par-ents who accompanied their adolescentsto clinic were asked to respond to eachitem in the way they believed their teen-ager would. No one refused. In addition,56 physicians who work with adolescentsand are members of the Lawson-WilkinsPediatric Endocrine Society were asked toestimate the frequency of each misman-agement behavior among the adolescentsin their practice. These physicians do notwork with the patients who participatedin the present study. Twenty-two physi-cians completed and returned this ques-tionnaire.

RESULTS— The mean HbAlc for thispopulation was 8.0, SD = 1.55 (normalrange 4.9-6.1). The median was 7.6 andthe mode was 7.4.

Adolescent mismanagementTable 1 displays the admitted misman-agement practices of our adolescentswithin the 10 days before their most re-cent clinic visit. A relatively large numberof adolescents admit to some mismanage-ment behaviors, with 25% admitting tomissing their insulin shots at least oncewithin the 10 days before their clinic visit.

Parent estimates of mismanagementThe agreement between parents and teenswas assessed by comparing the answers ofthe 55 parents and their teenagers on eachof the 10 questionnaire items. Initial anal-yses of the questionnaire data revealedthat the distribution is skewed. Therefore,Fisher's exact probability tests were per-formed (Table 2). These analyses revealedsignificant differences between parents'perceptions and their teenagers' re-sponses for missing shots and for takingextra insulin to cover eating inappropri-ate foods. Two other items approachedsignificance: missing injections and thentaking extra insulin to cover and makingup blood test results because the real oneswere too high. Parents underestimatedthe frequency with which teenagers en-gage in a variety of mismanagement be-haviors.

Physician estimates ofmismanagementOnly 22 of the 56 physicians surveyedcompleted and returned the question-naire. Therefore, the following data mustbe viewed with caution (Table 1).

Reasons offered for missingshots/testsForgetting to take insulin shots or bloodtests was the most commonly cited reasonfor missing these tasks (Table 3). With theexception of forgetting, however, the is-

78 DIABETES CARE, VOLUME 18, NUMBER 1, JANUARY 1995

Weissberg-Benchell and Associates

Table 2—Numbers and percentages of youth-parent pairs admitting mismanagement

Missed insulin injectionn%

Missed injection and then tookextra to covern%

Made up blood test resultsbecause did not do testn%

Took extra insulin to coverinappropriate foodn%

Ate inappropriate foodn%

Missed meals and snacksn%

Fixed blood glucose monitorto give lower numbern%

Changed test strip to givelower numbern%

Made up blood test resultsbecause real ones were toohighn%

Faked illnessn%

Youth: NoParent:

No

4095

4796

3280

3585

327

1450

4692

3981

3489

3472

Yes

25

24

820

615

873

1450

48

919

411

1328

Youth:YesParent:

No

969

467

1067

750

410

726

5100

467

1165

686

Yes

431

233

533

750

3890

2074

00

233

635

114

Fisher's exactprobability

0.02

0.06

0.31

0.01

0.15

0.10

1.00

0.59

0.06

0.66

sues around missing shots and missingblood tests appear to be different fromeach other. Adolescents missed shots be-cause they did not believe they wereneeded, or to prevent lows. Missing bloodtests seemed instead to be primarily dueto a desire to appear to others to be ingood control.

Relationship between HbAlc andmissing shots/testsGiven the skewed distribution of ourdata, a series of logistic regression analy-ses were performed in which the variablesof missing shots and missing blood testswere regressed on the basis of patient age,sex, duration of diabetes, and HbAlc level

(both individually and as a set). Results ofthese analyses reveal that individuals whoadmitted to missing shots were older andhad higher HbAlc levels than did thosewho denied ever missing shots (Table 4).No significant differences were found formissing blood tests.

Factor analysisThe 10 questionnaire items were sub-jected to a principal components analysisto determine if the individual itemsformed reliable subscales. Two factorswere formed (Table 5). Conceptually,these factors were J) blatant mismanage-ment and covering up and 2) faking ill-ness and faking test results. The eigen-value for blatant mismanagement was2.09 and for faking was 1.32. This re-sulted in 25 and 18% of the total itemvariance explained by each of the two fac-tors, respectively. The resulting two fac-tors have acceptable internal consistency(a = 0.74 and 0.60, respectively).

Relationship amongmismanagement, control, anddemographicsA multiple regression was calculated forthe relationship between glycohemoglo-bin and the demographic variables. Cau-casian teens had lower HbAlc levels thandid non-Caucasians (r2 = 0.08, F[l,127]= 10.64, P = 0.001). The associationsamong glycohemoglobin and demo-graphics with the factor scores were alsoevaluated. The blatant mismanagementfactor was significantly related to age (r2

= 0.05, F[l,125] = 7.18, P < 0.01).Older adolescents engage in mismanage-ment more than their younger cohorts.No relationship between the factor scoresand glycohemoglobin was found.

CONCLUSION— The purpose ofour study was to document 1) the exis-tence and prevalence of mismanagement,2) the reasons for engaging in misman-agement, and 3) the impact of misman-agement on control.



Table 3—Reasons offered for missing shots and blood tests

Shots1 just forgotI was away from home and forgot to bring it with meI didn't think the evening shot was necessary1 didn't plan on eating, so 1 missed the shot1 wanted to prevent a low reactionI wanted to see what would happenI wanted to lose weightI didn't want to give a shot in front of friends1 wanted to get sick to get out of something1 never miss my shots

Blood testsI just forgotDidn't test, wanted complete record, so made upI wanted to show good results to doctorI wanted to show good results to familyI'm fussed at when high, so made up lower numberI was away from home and forgot to bring it with meIt hurts to testI didn't want to test in front of friendsI never made up blood tests

4024121086551

41

3528242222206322

The percentages add up to more than 100 because respondents were permitted to choose more than onereason.

PrevalenceAdolescents admit to engaging in a widevariety of mismanagement behaviors,primarily around food. Not followingthe prescribed diet is one of the easiestand most tempting mismanagementbehaviors. Food is involved in manypeer-related activities and has social im-plications. Similarly, diet-related mis-management behaviors were the onesphysicians and parents cited as the mostprevalent.

Of greater concern, however, isthe prevalence of missing blood tests(29%), making up blood test results togive lower numbers (29%), and missinginsulin shots (25%). Parents underesti-mated the prevalence of a variety of mis-management behaviors among their ado-lescents.

Our 10-item scale factored intotwo subscales: blatant mismanagementand faking. These items did not fall intotechnique-related subscales such as shots,

Table 4—X* analysis of missed shots and missed blood tests in relation to demographicvariables, duration, and HbAlc

AgeSexRaceDurationHbAlc

*P<0.05.

Shots

Single variablemodel

5.26*0.672.912.784.39*

Multivariablemodel

4.04*0.021.020.875.17*

Blood tests

Single variable Multivariablemodel model

0.000.001.140.051.35

0.510.193.270.312.20

blood tests, or diet. Instead, more con-ceptually based subscales were found.Perhaps youths who are uninterested incaring for their diabetes will not selectspecific diabetes care techniques to avoid,but instead will avoid most, if not all, ofthem. Similarly, youths who do not wantothers to know how well or poorly theycontrol their diabetes will tend to misrep-resent their behavior in all aspects of dia-betes care.

Reasons for mismanagementUnderstanding the reasons for misman-agement is a complex issue, and a ques-tionnaire such as ours can only offer asimple first look. Our results suggest thatforgetting shots and tests is common. Inaddition, many teens missed shots be-cause they did not think the evening shotwas necessary. Teens fabricated bloodtests because of perceived pressure fromfamily and physicians to produce goodresults.

Relationship betweenmismanagement and controlFinding a statistically significant relation-ship between mismanagement and con-trol is often an elusive goal (17). Never-

Table 5—Factor loadings

Factor a

MismanagementMissed insulin shots (0.70)Made up results for (0.70)

tests never doneTook extra insulin to (0.63)

cover foodMissed shots, took extra (0.56)

to coverAte inappropriate foods (0.52)Missed meals or snacks (0.45)

FakingFaked test to read lower (0.70)Altered strips to lower (0.45)

readingAltered monitor to (0.45)

lower readingFaked illness (0.45)


Weissberg-Benchell and Associates

theless, we did find that the specificbehavior of missing shots is related tocontrol.

SummaryThe results of the present study show thateven in a diabetes program that offers acomprehensive multidisciplinary team,patients do struggle with adherence totheir regimen. It is vital that health careprofessionals recognize the existence andprevalence of mismanagement so we canhonestly discuss these behaviors with ourpatients and their families. Without suchdiscussions, our ability to make effectivedecisions about treatment plans becomesmoot. Concrete strategies for minimizingor avoiding mismanagement should be-come a standard part of diabetes educa-tion. Finally, parents' underestimationsregarding the prevalence of a variety ofmismanagement behaviors (missingshots, in particular) suggest that they arenot monitoring their adolescents' diabetesregimens as closely as may be needed.The idea of encouraging parents to in-crease supervision of their adolescentshas caused some controversy amonghealth care professionals. Nevertheless,

increased parental supervision has beenquite effective in decreasing general non-compliance (17) and has also been shownto be effective for teens with IDDM (18).Therefore, health care providers mayneed to encourage families to play a moreactive role in their adolescents' diabetescare than they have in the past.

Acknowledgments—This work was sup-ported by a grant from the Gilbert and JayleeMead Family Foundation in Greenbelt, Mary-land.

APPENDIX: DIABETESQUESTIONNAIRE— Most peoplehave difficulty taking care of their diabe-tes in the precise way their physician toldthem to. This questionnaire asks aboutdifferent parts of your diabetes care. Donot spend much time on any one item.There are no right or wrong answers, aseveryone takes care of their diabetes indifferent ways. We are interested in whatyou do. Please circle the answer that bestfits the way things are for you.

1 = Never 2 = Once or Twice 3 = Three or Four Times 4 = More than Four Times 5 = Frequently

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1. In the past 10 days, how often did you fix your glucose monitor togive lower readings?

2. In the past 10 days, how often did you miss shots?3. In the past 10 days, how often did you take extra insulin to cover

something you should not have eaten?4. In the past 10 days, how often did you make up blood test results

because the real ones were too high?5. In past 10 days, how often did you make up blood test results

because you had not done them?6. In the past 10 days, how many snacks and meals did you miss?7. In the past 10 days, how often did you miss your shots and then

take extra insulin to make up for it?8. In the past 10 days, how often did you change blood sugar test strips

to give a different reading?9. In the past 10 days, how often did you fake a headache,

stomachache, or other signs of high blood sugars to get out of doingwhat you didn't want to do?In the past 10 days, how often have you eaten something you knowyou really should not have eaten?

10

References

1. Jacobson AM, Hauser ST, Wolsdorf JI,Houlihan J, Milley JE, Herskowitz RD,Wertlieb D, Watt E: Psychologic predic-tors of compliance in children with recentonset of diabetes mellitus. J Pediatr 110:805-811, 1987

2. Christensen NK, Terry RD, Wyatt S,Pichert JW, Lorenz RA: Quantitative as-sessment of dietary adherence in patientswith insulin-dependent diabetes mellitus.Diabetes Care 6:245-250, 1983

3. Johnson SJ: Diabetes mellitus in child-hood. In Handbook ofPediatric Psychology.

Routh DK, Ed. New York, Guildford,1988, p. 9-31

4. Johnson SJ: Assessing daily managementin childhood diabetes. Health Psychol5:545-564, 1986

5. Glasgow AM, Weissberg-Benchell JA,Tynan WD, Epstein SF, Driscoll C, TurekJ, Beliveau E: Readmissions of childrenwith diabetes mellitus to a children's hos-pital. Pediatrics 88:98-104, 1991

6. White K, Kolman ML, Wexler P, Polin G,Winter RJ: Unstable diabetes and unsta-ble families: a psychosocial evaluation ofdiabetic children with recurrent ketoaci-dosis. Pediatrics 73:749-755, 1984

7. Golden MP, Herrold AJ, Orr DP: An ap-proach to prevention of recurrent diabeticketoacidosis in the pediatric population. JPediatr 107:195-200, 1985

8. SchadeD, DrummD.Duckworth W,Pat-ton P: The etiology of incapacitating brit-tle diabetes. Diabetes Care 8:12-20, 1985

9. Gray DK, Marreno DG, Godfrey C, OrrDP, Golden MP: Chronic poor metaboliccontrol in the pediatric population: astepwise intervention program. DiabetesEducator 14:516-520, 1988

10. Schafer LC, Glasgow RE, McCaul KD,Dreher M: Adherence to IDDM regimens:relationship to psychosocial variables andmetabolic control. Diabetes Care 6:493-498,1983

11. Glasgow RE, McCaul KD, Schafer LC:Self-care behaviors and glycemic controlin type I diabetes. J Chronic Dis 40:399-412,1987

12. Hanson CL, Henggeler SW, Harris MA,Mitchell KA, Carle DL, Burghen GA: As-



sociations between family members' per-ceptions of the health care system and thehealth of youths with insulin-dependentdiabetes mellitus. J Pediatr Psychol 13:543-554, 1988

13. Meldman L: Diabetes as experienced byadolescents. Adolescence 22:433-444,1987

14. Chase PH, Rainwater NG: Missed insulininjections, a common syndrome. PractDiabetol 8:20-23, 1989

15. Cox DJ, Irvine A, Gonder-Frederick L,Nowacek G, Butterfield J: Fear of hypo-glycemia: quantification, validation, andutilization. Diabetes Care 10:617-621,1987

16. Wilson DP, Endres RK: Compliance withblood glucose monitoring in childrenwith type 1 diabetes mellitus. J Pediatr108:1022-1030,1986

17. Johnson SB, Kelly M, Henretta J, Cun-ningham W, Tomer A, Silverstein JH: A

longitudinal analysis of adherence andhealth status in childhood diabetes. J Pe-diatr Psychol 17:537-554, 1992

18. Patterson G: A Social Learning Approach:Coercive Family Process. Eugene, OR,Castallia, 1982

19. Snyder J: Behavioral analysis and treat-ment of poor diabetic self-care and anti-social behavior: a single subject experi-mental study. Behav Ther 18:251-263,1987


Documents

Adolescent Diabetes Management and Mismanagement