Advertisement

Predictors of Comorbid Eating Disorders and Diabetes in People with Type 1 and Type 2 Diabetes

Published:September 07, 2016DOI:https://doi.org/10.1016/j.jcjd.2016.06.005

      Abstract

      Objectives

      The objective of this study was to identify psychosocial predictors of comorbid eating disorders (EDs) in individuals with type 1 and type 2 diabetes.

      Methods

      In this cross-sectional study, 140 people with diabetes answered an online survey covering sociodemographic information, body esteem, restrictive eating, medication omission, coping styles and depressive symptoms. Participants were recruited through advertisements on more than 100 websites, including forums, community organizations and Facebook groups focusing on either diabetes or EDs. Recruitment took place in Canada, Europe, Australia and the United States.

      Results

      On average, EDs developed after diabetes diagnoses in participants with type 1 diabetes but prior to diabetes diagnosis in participants with type 2 diabetes. In type 1 diabetes, avoidance coping styles and depressive symptoms predicted an additional diagnosis of EDs. Co-occurring EDs and type 2 diabetes were predicted by body mass indexes and task-oriented coping strategies.

      Conclusions

      Variables potentially influencing the development of EDs in people with diabetes differ according to the type of diabetes, body mass indexes, coping styles and depressive symptoms; they should be more systematically evaluated and closely monitored. In the context of diabetes management, prevention strategies for ED onset based on increased knowledge of the risk factors associated with EDs are necessary and could help decrease the risk for the health complications of diabetes.

      Résumé

      Objectifs

      L'objectif de la présente étude était de déterminer les prédicteurs psychosociaux de la cooccurrence des troubles des conduites alimentaires (TCA) chez les individus souffrant du diabète de type 1 ou de type 2.

      Méthodes

      Dans cette étude transversale, 140 personnes diabétiques ont répondu à une enquête en ligne couvrant les informations sociodémographiques, l'estime corporelle, les restrictions alimentaires, l'omission des médicaments, les stratégies d'adaptation et les symptômes de dépression. Les participants ont été recrutés par des annonces publicitaires sur plus de 100 sites Web, dont les forums, les organismes communautaires et les groupes Facebook portant sur le diabète ou les TCA. Le recrutement a eu lieu au Canada, en Europe, en Australie et aux États-Unis.

      Résultats

      En moyenne, les TCA se sont développés après le diagnostic de diabète chez les participants atteints du diabète de type 1, mais avant le diagnostic de diabète chez les participants atteints du diabète de type 2. Dans le diabète de type 1, les stratégies d'adaptation de type évitement et les symptômes de dépression ont prédit un diagnostic supplémentaire de TCA. La cooccurrence des TCA et du diabète de type 2 était prédite par les indices de masse corporelle et les stratégies d'adaptation centrées sur la tâche.

      Conclusions

      Les variables influençant potentiellement le développement des TCA chez les personnes diabétiques diffèrent selon le type de diabète, les indices de masse corporelle, les stratégies d'adaptation et les symptômes de dépression; elles devraient être évaluées de manière plus systématique et surveiller étroitement. Dans le contexte de la prise en charge du diabète, les stratégies de prévention de la manifestation de TCA fondées sur la connaissance des facteurs de risque associés aux TCA sont nécessaires et pourraient aider à diminuer le risque de complications liées au diabète.

      Keywords

      Mots clés

      Introduction

      Compared to individuals without diabetes, those with diabetes are at heightened risk for developing comorbid eating disorders (EDs) (
      • Grylli V.
      • Hafferl-Gattermayer A.
      • Schober E.
      • Karwautz A.
      Prevalence and clinical manifestations of eating disorders in Austrian adolescents with type-1 diabetes.
      ,
      • Nielsen S.
      Eating disorders in females with type 1 diabetes: An update of a meta-analysis.
      ,
      • Rodin G.
      • Olmsted M.P.
      • Rydall A.C.
      • et al.
      Eating disorders in young women with type 1 diabetes mellitus.
      ,
      • Smith F.M.
      • Latchford G.J.
      • Hall R.M.
      • Dickson R.A.
      Do chronic medical conditions increase the risk of eating disorder? A cross-sectional investigation of eating pathology in adolescent females with scoliosis and diabetes.
      ,
      • Starkey K.
      • Wade T.
      Disorder eating in girls with type 1 diabetes: Examining directions for prevention.
      ). Although such comorbidity is well documented in adolescents, very few studies have focused on its underlying mechanims in adults (
      • Herpertz S.
      • Nielson S.
      Comorbidity of diabetes mellitus and eating disorders.
      ). The nature and complexity of the interaction between diabetes and EDs make it necessary to treat type 1 diabetes and type 2 diabetes differently; each type has a specific relationship with EDs. Lack of knowledge regarding the predictors of concomitant EDs and diabetes hinders the development of appropriate preventive strategies and psychological treatments for individuals with diabetes.

      Comorbid Eating Disorders and Type 1 Diabetes

      The association between EDs and type 1 diabetes has been studied far more than that between EDs and type 2 diabetes (
      • Crow S.
      • Kendall D.
      • Praus B.
      • Thuras P.
      Binge eating and other psychopathology in patients with type II diabetes mellitus.
      ). Available research on EDs and type 1 diabetes suggests that diabetes is a risk factor in the development of EDs (
      • Colton P.
      • Olmsted M.
      • Daneman D.
      • et al.
      Disturbed eating behavior and eating disorders in preteen and early teenage girls with type 1 diabetes: A case-controlled study.
      ,
      • Colton P.
      • Rodin G.
      • Bergenstal R.
      • Parkin C.
      Eating disorders and diabetes: Introduction and overview.
      ,
      • Davidson M.K.
      Eating disorders and diabetes: Current perspectives.
      ,
      • Mannucci E.
      • Tesi F.
      • Ricca V.
      • et al.
      Eating behavior in obese patients with and without type 2 diabetes mellitus.
      ). Characteristics such as elevated body mass indexes (BMIs), weight fluctuations resulting from repeated dieting, body dissatisfaction, negative emotions, depressive symptoms and poor coping strategies have been associated with the co-occurrence of EDs and type 1 diabetes (
      • Goebel-Fabbri A.E.
      Disturbed eating behaviors and eating disorders in type 1 diabetes: Clinical significance and treatment recommendations.
      ,
      • Mannucci E.
      • Rotella F.
      • Ricca V.
      • et al.
      Eating disorders in patients with type 1 diabetes: A meta-analysis.
      ,
      • Olmsted M.P.
      • Colton P.A.
      • Daneman D.
      • et al.
      Prediction of the onset of disturbed eating behavior in adolescent girls with type 1 diabetes.
      ). Contrary to people without diabetes, adolescents diagnosed with type 1 diabetes tend to see their weight increase substantially while transitioning from adolescence to early adulthood (
      • Davidson M.K.
      Eating disorders and diabetes: Current perspectives.
      ,
      • Delhaye M.
      • Robert J.J.
      • Vila G.
      Diabète insulino-dépendant et troubles des conduites alimentaires: Quels progrès? Insulin-dependent diabetes mellitus and eating disorders: A review.
      ). This weight gain often leads to the development of problematic eating behaviours, usually aimed at losing weight. Weight changes combined with elevated BMIs in patients with type 1 diabetes can exacerbate body dissatisfaction, which is known to be a strong predictor of problematic eating and EDs (
      • Fairburn C.G.
      Cognitive behavior therapy and eating disorders.
      ,
      • Jacobi C.
      • Hayward C.
      • de Zwaan M.
      • et al.
      Coming to terms with risk factors for eating disorders: Application of risk terminology and suggestions for a general taxonomy.
      ). Weight cycling in patients with type 1 diabetes can also precede the onset of problematic eating behaviours through dietary restrictions and fasting followed by binge-eating behaviours (
      • Fairburn C.G.
      Cognitive behavior therapy and eating disorders.
      ).
      Past studies have shown that the emphasis on diet associated with diabetes management accentuates the risk for developing EDs (
      • Rodin G.
      • Olmsted M.P.
      • Rydall A.C.
      • et al.
      Eating disorders in young women with type 1 diabetes mellitus.
      ,
      • Smith F.M.
      • Latchford G.J.
      • Hall R.M.
      • Dickson R.A.
      Do chronic medical conditions increase the risk of eating disorder? A cross-sectional investigation of eating pathology in adolescent females with scoliosis and diabetes.
      ,
      • Mannucci E.
      • Tesi F.
      • Ricca V.
      • et al.
      Eating behavior in obese patients with and without type 2 diabetes mellitus.
      ,
      • Colton P.A.
      • Rodin G.M.
      • Olmsted M.P.
      • Daneman D.
      Eating disturbances in young women with type I diabetes mellitus: Mechanisms and consequences.
      ,
      • Peveler R.C.
      • Bryden K.S.
      • Neil H.A.W.
      • et al.
      The relationship of disordered eating habits and attitudes to clinical outcomes in young adult females with type 1 diabetes.
      ,
      • Rydall A.C.
      • Rodin G.M.
      • Olmsted M.P.
      • et al.
      Disordered eating behavior and microvascular complications in young women with insulin-dependent diabetes mellitus.
      ). Despite a call for greater dietary flexibility in treating type 1 diabetes, patients are still regularly advised to focus on the quantity and nutritional value of the food they eat. Such practices may induce rigid, controlling and perfectionist attitudes toward food (
      • Goebel-Fabbri A.E.
      • Fikkan J.
      • Connell A.
      • et al.
      Identification and treatment of eating disorders in women with type 1 diabetes mellitus.
      ), which increase the risk for losing control over food intake and having binge-eating episodes (
      • Rodin G.
      • Olmsted M.P.
      • Rydall A.C.
      • et al.
      Eating disorders in young women with type 1 diabetes mellitus.
      ,
      • Delhaye M.
      • Robert J.J.
      • Vila G.
      Diabète insulino-dépendant et troubles des conduites alimentaires: Quels progrès? Insulin-dependent diabetes mellitus and eating disorders: A review.
      ). Moreover, to compensate for their binges, these patients may begin to voluntarily skip their insulin (
      • Colton P.
      • Olmsted M.
      • Daneman D.
      • et al.
      Disturbed eating behavior and eating disorders in preteen and early teenage girls with type 1 diabetes: A case-controlled study.
      ,
      • Colton P.A.
      • Rodin G.M.
      • Olmsted M.P.
      • Daneman D.
      Eating disturbances in young women with type I diabetes mellitus: Mechanisms and consequences.
      ), thereby discovering an efficient but harmful way to control their weight.
      Negative emotions and depressive moods in people with type 1 diabetes can increase appetite and trigger binge eating (
      • Fairburn C.G.
      Cognitive behavior therapy and eating disorders.
      ,
      • Goebel-Fabbri A.E.
      • Fikkan J.
      • Connell A.
      • et al.
      Identification and treatment of eating disorders in women with type 1 diabetes mellitus.
      ,
      • Safer D.L.
      • Telch C.F.
      • Chen E.Y.
      Dialectical behavior therapy for binge eating and bulimia.
      ,
      • Stein R.I.
      • Kenardy J.
      • Wiseman C.V.
      • et al.
      What's driving the binge in binge eating disorder? A prospective examination of precursors and consequences.
      ). It has been shown that, confronted with stressful situations and negative emotions, adolescents with type 1 diabetes and comorbid EDs are more likely to resort to negative coping strategies (i.e. self-criticism, magical thinking) than their peers without comorbid EDs (
      • Grylli V.
      • Wagner G.
      • Hafferl-Gattermayer A.
      • et al.
      Disturbed eating attitudes, coping styles, and subjective quality of life in adolescents with type 1 diabetes.
      ). These negative coping strategies increase the risk for depressive symptoms and are linked to poorer glycemic control (
      • Yi J.P.
      • Yi J.C.
      • Vitaliano P.P.
      • Weinger K.
      How does anger coping style affect glycemic control in diabetes patients?.
      ). In patients with type 1 diabetes, insulin manipulation can represent another dysfunctional coping strategy aimed at managing stressful situations and reducing negative emotions following a binge-eating episode (
      • Yan L.
      Diabulimia: A growing problem among diabetic girls.
      ).
      Goebel-Fabbri et al (
      • Goebel-Fabbri A.E.
      • Fikkan J.
      • Connell A.
      • et al.
      Identification and treatment of eating disorders in women with type 1 diabetes mellitus.
      ) suggested a developmental model of eating disorders in individuals with type 1 diabetes that emphasizes the eating restrictions inherent in diabetes management. For these patients, dietary constraints may represent the first step toward EDs. Three evolutionary pathways to EDs are possible from there. In the first, psychological and physical sensations of food deprivation enhance the risk for binge eating which, in turn, can increase negative feelings about one's weight and body, give rise to fears of gaining weight and lead to the omission of insulin as a compensatory behaviour. In the second, patients with type 1 diabetes gain weight despite restricting their eating, most often once insulin treatment begins. The second pathway is associated with feelings of guilt regarding weight and shape, fear of additional weight gain and depressive symptoms. Finally, in the third, restrictive dietary behaviours can generate perfectionist attitudes in patients with type 1 diabetes. In a context of rigidity and privation, depressive symptoms, hyperglycemia and insulin manipulation to control weight or compensate for overeating are more likely to occur.

      Comorbid Eating Disorders and Type 2 Diabetes

      The mechanisms involved in the causes of type 2 diabetes and EDs are unclear (
      • Crow S.
      • Kendall D.
      • Praus B.
      • Thuras P.
      Binge eating and other psychopathology in patients with type II diabetes mellitus.
      ). A sequence of onset opposite from that of comorbid EDs with type 1 diabetes has been suggested, with EDs—and more precisely, binge eating—being thought to predispose individuals to type 2 diabetes (
      • Colton P.
      • Rodin G.
      • Bergenstal R.
      • Parkin C.
      Eating disorders and diabetes: Introduction and overview.
      ,
      • Davidson M.K.
      Eating disorders and diabetes: Current perspectives.
      ,
      • Mannucci E.
      • Tesi F.
      • Ricca V.
      • et al.
      Eating behavior in obese patients with and without type 2 diabetes mellitus.
      ). Risk factors having been linked to the etiology of type 2 diabetes and the development of comorbid EDs include elevated BMIs (
      • Crow S.
      • Kendall D.
      • Praus B.
      • Thuras P.
      Binge eating and other psychopathology in patients with type II diabetes mellitus.
      ), weight gain (
      • Thomas J.G.
      • Butryn M.L.
      • Stice E.
      • Lowe M.R.
      A prospective test of the relation between weight change and risk for bulimia nervosa.
      ), body dissatisfaction and drive for thinness (
      • Herpertz S.
      • Albus C.
      • Kielmann R.
      • et al.
      Comorbidity of diabetes mellitus and eating disorders: A follow-up study.
      ), as well as depressive mood (
      • Carroll P.
      • Tiggemann M.
      • Wade T.
      The role of body dissatisfaction and bingeing in the self-esteem of women with type II diabetes.
      ).
      Patients with type 2 diabetes tend to be overweight or obese (
      • Carroll P.
      • Tiggemann M.
      • Wade T.
      The role of body dissatisfaction and bingeing in the self-esteem of women with type II diabetes.
      ,
      • Papelbaum M.
      • Appolinário J.C.
      • Moreira Rde O.
      • et al.
      Prevalence of eating disorders and psychiatric comorbidity in a clinical sample of type 2 diabetes mellitus patients.
      ), and their BMIs are higher when they have comorbid EDs (
      • Crow S.
      • Kendall D.
      • Praus B.
      • Thuras P.
      Binge eating and other psychopathology in patients with type II diabetes mellitus.
      ). Thus, they are particularly at risk for trying to consume less food to lose weight and for feeling dissatisfied with their bodies. The high levels of body dissatisfaction observed in overweight and obese individuals with type 2 diabetes tend to increase their drive for thinness (
      • Herpertz S.
      • Albus C.
      • Kielmann R.
      • et al.
      Comorbidity of diabetes mellitus and eating disorders: A follow-up study.
      ) and to be linked to more severe depressive symptoms (
      • Carroll P.
      • Tiggemann M.
      • Wade T.
      The role of body dissatisfaction and bingeing in the self-esteem of women with type II diabetes.
      ). They may also play key roles in the onset of EDs in patients with type 2 diabetes.
      In cases of EDs and type 2 diabetes, it has been suggested that a restrictive-eating and binge-eating pattern would lead to weight gain, which would then be associated with the onset of type 2 diabetes (
      • Mannucci E.
      • Tesi F.
      • Ricca V.
      • et al.
      Eating behavior in obese patients with and without type 2 diabetes mellitus.
      ,
      • Rodin G.M.
      • Daneman D.
      Eating disorders and ID diabetes: A problematic association.
      ). However, a sequence of onset in which EDs and obesity would develop before type 2 diabetes has yet to be demonstrated.

      Methods

      Objectives and hypotheses

      The objective of this research was two-fold. First, it wanted to shed light on the developmental sequence of diabetes and EDs. A different sequence of ED onset is expected for each type of diabetes: while EDs should develop after type 1 diabetes, they may precede type 2 diabetes. Second, this study aimed to identify key factors in the co-occurrence of EDs and diabetes while considering type 1 diabetes and type 2 diabetes separately. Relevant factors in the prediction models are BMI, body esteem, restrictive eating, medication omission, coping strategies and depressive symptoms. According to past research, these six factors could contribute in varying degrees to predicting comorbid EDs with type 1 diabetes and type 2 diabetes.

      Sample and procedure

      Participants were recruited through online forums, Facebook, e-mail and community organization websites concerning diabetes and EDs in Canada, Australia, Europe and the United States. More than 100 contacts and postings were made to recruit participants. Between May and June 2011, 140 participants with diabetes completed online questionnaires. Four groups of participants were formed according to their responses concerning the problematic eating behaviour items in the Eating Disorders Examination Questionnaire-6.0 (EDEQ-6): type 1 diabetes (n=54); ED-type 1 diabetes (n=39); type 2 diabetes (n=24) and ED-type 2 diabetes (n=22). Based on selected EDEQ-6 items, the first two authors (CG and AA) independently determined whether the participants had current EDs or not and, if so, which one. Afterwards, the evaluators compared their classifications (ED or not) and ED diagnoses (when applicable) for the first 35 participants. A satisfactory inter-raters agreement of .95 was obtained, suggesting that the classifications and ED diagnoses are reliable. In the few cases where discrepancies were found between the two raters, they discussed it until a consensus was reached.

      Instruments

      Sociodemographic questionnaire

      Participants were asked for their ages, weights, heights and socioeconomic and marital statuses. They were also questioned about how old they were when they received diabetes and ED diagnoses, if applicable, as well as about medication omission related to fear of gaining weight. BMIs were calculated based on participants' self-reported weights and heights (BMI=kg/m2).

      Eating Disorder Examination Questionnaire-6.0

      The Eating Disorder Examination Questionnaire (EDEQ-6) (
      • Fairburn C.G.
      Cognitive behavior therapy and eating disorders.
      ) includes 28 items concerning behaviours and attitudes typically associated with eating problems. In this study, ED diagnoses were based on the presence and frequency of specific eating behaviours (e.g. binge eating, fasting, vomiting, exercising intensively, abusing laxatives and diuretics) as per the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) criteria (
      ). The EDEQ-6 has high reliability, with an internal consistency of 0.93, and the items assessing eating symptomatology show a high degree of temporal stability (
      • Fairburn C.G.
      Cognitive behavior therapy and eating disorders.
      ,
      • Mond J.M.
      • Hay P.J.
      • Rodgers B.
      • et al.
      Validity of the Eating Disorder Examination Questionnaire EDEQ in screening for eating disorders in community samples.
      ).

      Body Esteem Scale

      The Body Esteem Scale (BES) (
      • Mendelson B.
      • White D.
      • Mendelson M.
      Manual for the Body-Esteem Scale for adolescents and adults.
      ) contains 23 items divided into 3 subscales: weight, appearance and attribution. The BES subscales have an internal consistency of 0.94, 0.92 and 0.81, respectively (
      • Mendelson B.
      • White D.
      • Mendelson M.
      Manual for the Body-Esteem Scale for adolescents and adults.
      ).

      Beck Depression Inventory II

      Depressive symptomatology was assessed using the Beck Depression Inventory-II (BDI-II) (
      • Beck A.T.
      • Steer R.A.
      • Brown G.K.
      Manual for the beck depression inventory-II.
      ). This 21-item self-reported questionnaire measures the presence and severity of depressive symptoms. The BDI-II has an internal consistency of 0.92 and good stability over time (test-retest reliability of 0.93).

      Coping inventory for stressful situations

      Participants' coping strategies were assessed using the Coping Inventory for Stressful Situations (CISS) (
      • Endler N.S.
      • Parker J.D.A.
      Coping inventory for stressful situations (CISS) manual.
      ), which contains 48 items divided into 3 main subscales: task, emotion and avoidance. The internal consistency of the subscales ranged from 0.87 to 0.92 for task, 0.82 to 0.90 for emotion and 0.76 to 0.85 for avoidance (
      • Endler N.S.
      • Parker J.D.A.
      Coping inventory for stressful situations (CISS) manual.
      ).

      Statistical analyses

      Self-reported ages of diabetes and ED onset were used to determine which disorder (diabetes or EDs) developed first. 2-way ANOVAs, chi-square and z tests with contrasts were used to compare the 4 groups that were formed (EDs-type 1 diabetes; EDs-type 2 diabetes; type 1 diabetes; type 2 diabetes). Predictive models of comorbid EDs-type 1 diabetes and of EDs-type 2 diabetes were tested using stepwise logistic regressions with backward elimination.

      Results

      Sociodemographic and eating characteristics

      The 4 groups differed in mean age, ages of diabetes onset and BMIs. A 2-way ANOVA indicated significantly different effects, depending on the type of diabetes and the comorbid conditions: respondents with type 1 diabetes were younger (x¯=34 years old) than those with type 2 diabetes (x¯=52 years old) (F[1136]=58.692; p<.001, η2=.27), and respondents with diabetes only were older (x¯=46 years old) than those with ED-diabetes (x¯=40 years old) (F[1136]=7.733; p<.006, η2=.04). Regarding age of diabetes onset, significantly different effects by diabetes type and comorbidity effects were observed, but no interactional effects were seen (Table 1). Participants with type 1 diabetes were diagnosed with diabetes at younger ages (x¯=15 years old) than those with type 2 diabetes (x¯=42 years old) (F[1136]=123.942; p<.001, η2=.42). Moreover, participants with diabetes and comorbid EDs developed their diabetes when they were younger (x¯=25 years old) than people with diabetes only (x¯=32 years old) (F[1136]=8.790; p=.004, η2=.03). Table 1 illustrates separated means for the 4 groups.
      Table 1Sociodemographic information
      Age
      Separated means for the 4 groups but no interactional effects.
      Age of diabetes onset
      Separated means for the 4 groups but no interactional effects.
      Age of ED onset
      MeanSDMeanSDMeanSD
      Type 1 diabetes38.64015.07617.74016.770
      Type 2 diabetes53.87112.63346.83911.332
      ED-type 1 diabetes29.13610.32013.1368.75419.8007.722
      ED-type 2 diabetes50.2008.63737.2678.54024.36412.682
      ED, eating disorder; SD, standard deviation.
      a Separated means for the 4 groups but no interactional effects.
      Another 2-way ANOVA showed different effects by diabetes type, a comorbidity effect and an interactional effect for BMIs. Participants with type 1 diabetes had lower BMIs (x¯=27) than those with type 2 diabetes (x¯=37) (F[1135]=43.386; p<.001, η2=.24). Participants in the ED-type 1 diabetes group had higher BMIs (x¯=28) than those in the type 1 diabetes group (x¯=26), and people in the ED-type 2 diabetes group had higher BMIs (x¯=42) than those in the type 2 diabetes group (x¯=32) (F[1135]=6.488; p=.012, η2=.04).
      All 4 groups were additionally compared in terms of restrictive eating and medication omission. Chi-square results showed no statistically significant differences between ED-type 2 diabetes and the 3 other groups in both behaviours. However, participants with ED-type 1 diabetes presented significantly higher restrictive eating behaviours (χ2(n=140)=17.27; p<.001) and omitted their medications in order to lose weight significantly more frequently (χ2[n=140]=37.42; p<.01) than did people with type 1 diabetes and type 2 diabetes only. More precisely, 95.5% of patients with ED-type 1 diabetes restricted their eating (z=3.4), while 66% (z=−2.5) of people with type 1 diabetes and 64.5% (z=−2.0) of people with type 2 diabetes reported restrictive eating. With regard to medication omission, 50% (z=6.0) of patients with ED-type 1 diabetes, 8% (z=−2.6) of patients with type 1 diabetes and 0 (z=−3.2) patients with type 2 diabetes reported this behaviour.

      Onset of EDs

      Although 89.5% (z=4.7) of participants with ED-type 1 diabetes reported having developed EDs after the onset of diabetes, 81.8% (z=4.7) of those with ED-type 2 diabetes had EDs before their diabetes diagnoses (χ2[n=49]=22.244; p<.001). On average, individuals in the ED-type 1 diabetes group developed EDs 7 years after their initial diabetes diagnoses, and individuals in the ED-type 2 diabetes received ED diagnoses 13 years prior to their diabetes diagnoses.
      No statistically significant difference was observed for the age of ED onset between the groups with ED-type 1 diabetes (x¯=20) and ED-type 2 diabetes (x¯=24) (t[1.44]=−1.454; p>.05). Also, no statistically significant difference in ED types was found between the 2 comorbid groups. In the ED-type 1 diabetes group, 9% (z=1.2) of respondents had anorexia nervosa, 30% (z=.7) had bulimia nervosa, 25% (z=−2.0) had binge eating disorders and 36% (z=.7) had unspecified eating disorders. No diagnoses of anorexia nervosa (z=−1.2) were found in the ED-type 2 diabetes group; 20% (z=−.7) had bulimia nervosa; 53% (z=2.0) had binge eating disorders and 27% (z=−.7) had unspecified eating disorders (χ2[n=59]=4.838; p>.05).

      Predictors of comorbid conditions

      Two stepwise logistic regressions with backward elimination were performed to identify predictors of EDs in people with diabetes. The analysis was broken down by diabetes typology and included BMI, body esteem, restrictive eating, medication omission, coping styles and depressive symptoms. Table 2 presents the results, which show that body esteem, restrictive eating and medication omission because of fear of gaining weight did not contribute significantly to predicting EDs in people with either typeof diabetes. In the case of participants with type 1 diabetes, avoidance coping style (χ2 [1, n=72]=6.023; p=.014) and depressive symptoms (χ2 [1, n=72]=18.152; p<.001) were identified as significant predictors of comorbid EDs. Together, these two variables account for 36.40% of the variance (Cox and Snell R2). The odds ratio value for avoidance coping style and depressive symptomatology indicated that if their levels increased by 1 point, the risk for having comorbid EDs increases by 1.080 and 1.124, respectively.
      Table 2Logistic regression results for type 1 and type 2 diabetes
      Included in the modelB(SE)
      Bias of the standard error p<.05.
      Odds ratioLowerUpper
      95% CL for odds ratio
      Type 1 diabetes
       Avoidance0.077(.031)1.0801.0161.149
       Global BDI0.117(.028)1.1241.0651.187
      Type 2 diabetes
       BMI2.238(.848)9.3741.77849.412
       Task−0.149(0.063).862.761.975
      BDI, Beck Depression Inventory-II; BMI, body mass index; B(SE).
      a Bias of the standard error p<.05.
      In participants with type 2 diabetes, BMIs (χ2 (1, n=38)=6.964; p=.008) and task-oriented coping strategies (χ2 [1, n=38]=5.542; p=.019) were significant predictors of EDs. They explained 51% of the variance (Cox and Snell R2). The odds ratio value for BMI indicated that a 1-point increase is associated with a 9.374-point increase. For task-oriented coping strategies, a 1-point increase was associated with a 1.160 (1/0.862=1.160)-point decrease. Table 3 presents the descriptive results (z test and contrast) for each group.
      Table 3Descriptive results
      BMI
      Descriptive values represent means that are not transformed. The F test and contrast were performed on the transformed variable, and it respected the postulates.
      Body esteem appearanceBody esteem attributionBody esteem weightTask-oriented copingAvoidance copingEmotion copingBDI
      Mean (SD)Mean (SD)Mean (SD)Mean (SD)Mean (SD)Mean (SD)Mean (SD)Mean (SD)
      Type 1 diabetes25.7322.3414.1018.6853.8339.9740.2833.67
      (5.475)
      Differences among all groups.
      (6.263)
      No statistical significant differences.
      (3.734)
      Differences between ED-type 1 diabetes and ED-type 2 diabetes.
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      (6.374)
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      (10.316)
      Differences between ED-type 1 diabetes and ED-type 2 diabetes.
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      (10.630)
      No statistical significant differences.
      (14.382)
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      (10.813)
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      Type 2 diabetes31.6522.9313.7816.4455.6839.5237.1633.68
      (7.740)
      Differences among all groups.
      (5.491)
      No statistical significant differences.
      (3.389)
      Differences between ED-type 1 diabetes and ED-type 2 diabetes.
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      (6.483)
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      (13.284)
      Differences between ED-type 1 diabetes and ED-type 2 diabetes.
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      (10.067)
      No statistical significant differences.
      (12.449)
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      (9.848)
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      ED-Type 1 diabetes27.6922.8513.0516.0049.2542.2852.5350.57
      (8.880)
      Differences among all groups.
      (8.245)
      No statistical significant differences.
      (4.151)
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      (5.159)
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      (11.941)
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      (10.985)
      No statistical significant differences.
      (11.065)
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      (16.384)
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      ED-Type 2 diabetes41.7024.389.6213.84640.3841.3152.0853.92
      (8.460)
      Differences among all groups.
      (8.382)
      No statistical significant differences.
      (2.931)
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      (4.120
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      (9.596)
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      (9.911)
      No statistical significant differences.
      (11.679)
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      (16.358)
      Differences between comorbid groups, both diabetes types and diabetes groups only.
      ED, eating disorder; SD, standard deviation.
      a Differences among all groups.
      b Differences between ED-type 1 diabetes and ED-type 2 diabetes.
      c Differences between comorbid groups, both diabetes types and diabetes groups only.
      d No statistical significant differences.
      e Descriptive values represent means that are not transformed. The F test and contrast were performed on the transformed variable, and it respected the postulates.

      Conclusions

      The results of this study suggest that, depending on the type of diabetes, the comorbid onset of EDs differs. As suggested by Goebel-Fabbri et al (
      • Goebel-Fabbri A.E.
      • Fikkan J.
      • Connell A.
      • et al.
      Identification and treatment of eating disorders in women with type 1 diabetes mellitus.
      ), the dietary and weight-management preoccupations inherent in type 1 diabetes appear to precede, maybe even predispose to, ED development. In contrast, EDs appear first in most cases of comorbid EDs with type 2 diabetes. This latest result is particularly interesting because the sequence of ED onset in adults with type 2 diabetes had not yet been tested. Moreover, the results highlight that patients with ED-type 1 diabetes are more likely to present with restrictive eating and medication omission in order to lose weight than are patients with type 1 diabetes and type 2 diabetes. Factors associated with co-occurrence of an ED and type 1 diabetes as well as type 2 diabetes also differ. As theoretically suggested, negative coping strategies and depressive symptoms were predictive of EDs in patients with type 1 diabetes (
      • Fairburn C.G.
      Cognitive behavior therapy and eating disorders.
      ,
      • Goebel-Fabbri A.E.
      • Fikkan J.
      • Connell A.
      • et al.
      Identification and treatment of eating disorders in women with type 1 diabetes mellitus.
      ,
      • Grylli V.
      • Wagner G.
      • Hafferl-Gattermayer A.
      • et al.
      Disturbed eating attitudes, coping styles, and subjective quality of life in adolescents with type 1 diabetes.
      ,
      • Yi J.P.
      • Yi J.C.
      • Vitaliano P.P.
      • Weinger K.
      How does anger coping style affect glycemic control in diabetes patients?.
      ); and BMIs (
      • Crow S.
      • Kendall D.
      • Praus B.
      • Thuras P.
      Binge eating and other psychopathology in patients with type II diabetes mellitus.
      ) and low task-coping styles were relevant factors in predicting EDs with type 2 diabetes.
      Most patients with type 1 diabetes received their diabetes diagnoses during adolescence, a particularly risky period for ED onset (
      • Starkey K.
      • Wade T.
      Disorder eating in girls with type 1 diabetes: Examining directions for prevention.
      ). Takii et al (
      • Takii M.
      • Uchigata Y.
      • Kishimoto J.
      • et al.
      The relationship between the age of onset of type 1 diabetes and the subsequent development of a severe eating disorder by female patients.
      ) suggested that being diagnosed with type 1 diabetes between the ages of 7 and 18 years is associated with a higher risk for developing severe EDs compared with being diagnosed in early childhood or adulthood. They also suggested that developing diabetes during puberty reduces the ability to manage stress. Insulin manipulation and its impact on weight loss could also precipitate the development of EDs in adolescent patients with type 1 diabetes (
      • Takii M.
      • Uchigata Y.
      • Kishimoto J.
      • et al.
      The relationship between the age of onset of type 1 diabetes and the subsequent development of a severe eating disorder by female patients.
      ). In patients diagnosed with type 2 diabetes, it seems that the developmental course of this comorbid condition is inversed; EDs precede type 2 diabetes. This may be related to a weight gain effect, likely attributable to food-restriction behaviours in people who binge eat. In fact, restrictive eating increases the risk for binge eating (
      • Fairburn C.G.
      Cognitive behavior therapy and eating disorders.
      ) which, combined with sedentary behaviours and in the absence of compensatory behaviours, has the potential to increase BMIs and, subsequently, to accentuate the risk for developing type 2 diabetes over time. This explanation could correspond to a large proportion of the ED-type 2 diabetes sample, given that most of them reported bingeing and that over half (53%) of them received diagnoses of binge-eating disorder. Such diagnoses imply that they eat excessive amounts of food at least once a week, with a sense of losing control over what they eat and without adopting compensatory behaviours (e.g. no fasting, exercising, vomiting or insulin omission) (
      ). In this context, they are likely to gain weight and, over time, continuing to do so can lead to the onset of type 2 diabetes.
      The results of this study highlight relevant factors associated with the co-occurrence of EDs and diabetes. In patients with bothtype 1 and type 2 diabetes, coping strategies appear to be a significant determinant of whether or not people are at risk for having comorbid EDs. People with ED-diabetes may feel overwhelmed by all the tasks associated with diabetes management (e.g. nutrition, medication, exercises) (
      • Goebel-Fabbri A.E.
      Disturbed eating behaviors and eating disorders in type 1 diabetes: Clinical significance and treatment recommendations.
      ). In trying to cope with this challenging condition, they may use less adaptive coping styles, such as denying or avoiding thinking about the medical risks associated with bad diabetes management. In such contexts, for patients with diabetes, developing EDs may even seem to be attempts to gain further control over their eating and weight through eating restrictions and, in some cases, insulin omission. Also, individuals with ED-diabetes may be more likely than those with only diabetes to use binge eating for purposes of emotional regulation (
      • Fairburn C.G.
      Cognitive behavior therapy and eating disorders.
      ).
      Depressive symptoms, for their part, were identified as predictors of comorbid conditions in patients with ED-type 1 diabetes. Mood disturbances have previously been identified as significant predictors of disturbed eating behaviours (
      • Starkey K.
      • Wade T.
      Disorder eating in girls with type 1 diabetes: Examining directions for prevention.
      ) and, according to the present findings, when found in patients with type 1 diabetes, they could increase their risk for subsequently developing comordid EDs.
      In this study, BMIs stand out as an important factor associated with the development of EDs in patients with type 2 diabetes. As suggested, the high mean BMI observed in participants with ED-type 2 diabetes could be a consequence of binge eating; binge eating leads to higher BMIs which, in turn, further increase the risk for the onset of type 2 diabetes. The importance of BMI in the prediction of ED-type 2 diabetes can also be understood by the strong emphasis patients with EDs put on their abilities to control their weight and shape. In fact, they tend to base their self-worth almost exclusively on this control (
      • Fairburn C.G.
      Cognitive behavior therapy and eating disorders.
      ). When type 2 diabetes is co-occurring, they must, in addition, closely manage their glycemia through their eating habits. Indeed, diabetes management in patients with type 2 diabetes can accentuate the felt pressure to lose weight, given that it involves increased control of the types of foods consumed, especially carbohydrate-rich foods, as well as strict BMI control and regular exercise.
      Although this study provides much-needed knowledge concerning the developmental sequence of EDs and diabetes onset in individuals with type 1 and type 2 diabetes, generalization of its results is limited by the use of a questionnaire to infer the participants' ED diagnoses, its small number of participants and the sensitive nature of the subject of eating for individuals with diabetes. With more than 100 posts published in ED and diabetes forums and websites, we can conclude that recruitment among the online community with diabetes was difficult. Patients with type 2 diabetes were particularly reluctant to participate and, in some cases, overtly expressed mistrust and reacted very emotionally toward being asked about eating problems. Results of this study may also suffer from a retrospective recall bias; participants had to specify the age at which they had received their ED and diabetes diagnoses and were not followed longitudinally. Therefore, these findings need replication.

      Clinical considerations

      Despite some limitations, results from this study suggest that healthcare professionals working with individuals with diabetes should be vigilant about problematic eating behaviours in their patients, especially restrictive eating and medication omission. Such behaviours may hide other problematic eating behaviours that patients with diabetes may be reluctant to disclose, such as binge eating, using laxatives or fasting. Consequently, careful evaluation and monitoring of ED behaviours is advisable, and a questionnaire like the Diabetes Eating Survey Revised could be useful (
      • Markowitz J.T.
      • Butler D.A.
      • Volkening L.K.
      • et al.
      Brief screening tool for disordered eating in diabetes: Internal consistency and external validity in a contemporary sample of pediatric patients with type 1 diabetes.
      ). From a preventive stance, a deeper knowledge of the risk factors associated with EDs, such as BMIs, depressive symptoms and coping styles, seems necessary in order to limit the likelihood of ED onset and to decrease the risk for the health and medical complications associated with diabetes and comorbid EDs.
      When EDs are found in patients with diabetes, treatment should prioritize eating regulation and binge-eating reduction (
      • Gagnon C.
      • Aimé A.
      • Bélanger C.
      • Markowitz J.T.
      Comorbid diabetes and eating disorders in adult patients: Assessment and considerations for treatment.
      ). Emphasis on dietary restrictions and weight loss must be proscribed because they can exacerbate the ED symptomatology and the medical problems associated with both EDs and diabetes. A less rigid approach to eating, weight and shape appears more likely to be effective. While paying attention to possible depressive symptoms, treatment should also focus on developing healthful coping styles. As suggested by Starkey and Wade (
      • Starkey K.
      • Wade T.
      Disorder eating in girls with type 1 diabetes: Examining directions for prevention.
      ), when it comes to the management of stress and depressive symptoms in patients with diabetes, it may be particularly relevant to work on improving problem solving.

      Author Contributions

      CG researched data, wrote and edited the manuscript; AA wrote, edited and reviewed the manuscript; CB contributed to and reviewed the manuscript.

      References

        • Grylli V.
        • Hafferl-Gattermayer A.
        • Schober E.
        • Karwautz A.
        Prevalence and clinical manifestations of eating disorders in Austrian adolescents with type-1 diabetes.
        Wien Klin Wochenschr. 2004; 116: 230-234
        • Nielsen S.
        Eating disorders in females with type 1 diabetes: An update of a meta-analysis.
        Eur Eat Disord Rev. 2002; 10: 241-254
        • Rodin G.
        • Olmsted M.P.
        • Rydall A.C.
        • et al.
        Eating disorders in young women with type 1 diabetes mellitus.
        J Psychosom Res. 2002; 53: 943-949
        • Smith F.M.
        • Latchford G.J.
        • Hall R.M.
        • Dickson R.A.
        Do chronic medical conditions increase the risk of eating disorder? A cross-sectional investigation of eating pathology in adolescent females with scoliosis and diabetes.
        J Adolesc Health. 2008; 42: 58-63
        • Starkey K.
        • Wade T.
        Disorder eating in girls with type 1 diabetes: Examining directions for prevention.
        Clin Psychol. 2010; 14: 2-9
        • Herpertz S.
        • Nielson S.
        Comorbidity of diabetes mellitus and eating disorders.
        in: Treasure J. Schmidt U. Van Furth E. The handbook of eating disorders. John Wiley, Chichester2003: 401-414
        • Crow S.
        • Kendall D.
        • Praus B.
        • Thuras P.
        Binge eating and other psychopathology in patients with type II diabetes mellitus.
        Int J Eat Disord. 2001; 30: 222-226
        • Colton P.
        • Olmsted M.
        • Daneman D.
        • et al.
        Disturbed eating behavior and eating disorders in preteen and early teenage girls with type 1 diabetes: A case-controlled study.
        Diabetes Care. 2004; 27: 1654-1659
        • Colton P.
        • Rodin G.
        • Bergenstal R.
        • Parkin C.
        Eating disorders and diabetes: Introduction and overview.
        Diabetes Spectrum. 2009; 22: 138-142
        • Davidson M.K.
        Eating disorders and diabetes: Current perspectives.
        Can J Diabetes. 2003; 27: 62-73
        • Mannucci E.
        • Tesi F.
        • Ricca V.
        • et al.
        Eating behavior in obese patients with and without type 2 diabetes mellitus.
        Int J Obes. 2002; 26: 848-853
        • Goebel-Fabbri A.E.
        Disturbed eating behaviors and eating disorders in type 1 diabetes: Clinical significance and treatment recommendations.
        Curr Diab Rep. 2009; 9: 133-139
        • Mannucci E.
        • Rotella F.
        • Ricca V.
        • et al.
        Eating disorders in patients with type 1 diabetes: A meta-analysis.
        J Endocrinol Invest. 2005; 28: 417-419
        • Olmsted M.P.
        • Colton P.A.
        • Daneman D.
        • et al.
        Prediction of the onset of disturbed eating behavior in adolescent girls with type 1 diabetes.
        Diabetes Care. 2008; 31: 1978-1982
        • Delhaye M.
        • Robert J.J.
        • Vila G.
        Diabète insulino-dépendant et troubles des conduites alimentaires: Quels progrès? Insulin-dependent diabetes mellitus and eating disorders: A review.
        Ann Med Psychol (Paris). 2002; 160: 565-573
        • Fairburn C.G.
        Cognitive behavior therapy and eating disorders.
        Guilford Press, New York2008
        • Jacobi C.
        • Hayward C.
        • de Zwaan M.
        • et al.
        Coming to terms with risk factors for eating disorders: Application of risk terminology and suggestions for a general taxonomy.
        Psychol Bull. 2004; 130: 19-65
        • Colton P.A.
        • Rodin G.M.
        • Olmsted M.P.
        • Daneman D.
        Eating disturbances in young women with type I diabetes mellitus: Mechanisms and consequences.
        Psychiatr Ann. 1999; 29: 213-218
        • Peveler R.C.
        • Bryden K.S.
        • Neil H.A.W.
        • et al.
        The relationship of disordered eating habits and attitudes to clinical outcomes in young adult females with type 1 diabetes.
        Diabetes Care. 2005; 28: 84-88
        • Rydall A.C.
        • Rodin G.M.
        • Olmsted M.P.
        • et al.
        Disordered eating behavior and microvascular complications in young women with insulin-dependent diabetes mellitus.
        N Engl J Med. 1997; 336: 1849-1854
        • Goebel-Fabbri A.E.
        • Fikkan J.
        • Connell A.
        • et al.
        Identification and treatment of eating disorders in women with type 1 diabetes mellitus.
        Treat Endocrinol. 2002; 1: 155-162
        • Safer D.L.
        • Telch C.F.
        • Chen E.Y.
        Dialectical behavior therapy for binge eating and bulimia.
        Guilford Press, New York (NY)2009
        • Stein R.I.
        • Kenardy J.
        • Wiseman C.V.
        • et al.
        What's driving the binge in binge eating disorder? A prospective examination of precursors and consequences.
        Int J Eat Disord. 2007; 40: 195-203
        • Grylli V.
        • Wagner G.
        • Hafferl-Gattermayer A.
        • et al.
        Disturbed eating attitudes, coping styles, and subjective quality of life in adolescents with type 1 diabetes.
        J Psychosom Res. 2005; 59: 65-72
        • Yi J.P.
        • Yi J.C.
        • Vitaliano P.P.
        • Weinger K.
        How does anger coping style affect glycemic control in diabetes patients?.
        Int J Behav Med. 2008; : 167-172
        • Yan L.
        Diabulimia: A growing problem among diabetic girls.
        Nephrol News Issues. 2007; 21: 38
        • Thomas J.G.
        • Butryn M.L.
        • Stice E.
        • Lowe M.R.
        A prospective test of the relation between weight change and risk for bulimia nervosa.
        Int J Eat Disord. 2011; 44: 295-303
        • Herpertz S.
        • Albus C.
        • Kielmann R.
        • et al.
        Comorbidity of diabetes mellitus and eating disorders: A follow-up study.
        J Psychosom Res. 2001; 51: 673-678
        • Carroll P.
        • Tiggemann M.
        • Wade T.
        The role of body dissatisfaction and bingeing in the self-esteem of women with type II diabetes.
        J Behav Med. 1999; 22: 59-74
        • Papelbaum M.
        • Appolinário J.C.
        • Moreira Rde O.
        • et al.
        Prevalence of eating disorders and psychiatric comorbidity in a clinical sample of type 2 diabetes mellitus patients.
        Rev Bras Psiquiatr. 2005; 27: 135-138
        • Rodin G.M.
        • Daneman D.
        Eating disorders and ID diabetes: A problematic association.
        Diabetes Care. 1992; 15: 1402-1412
      1. Diagnostic and statistical manual of mental disorders-5: DSM-5. American Psychiatric Association, Arlington (VA)2013
        • Mond J.M.
        • Hay P.J.
        • Rodgers B.
        • et al.
        Validity of the Eating Disorder Examination Questionnaire EDEQ in screening for eating disorders in community samples.
        Behav Res Ther. 2004; 42: 551-567
        • Mendelson B.
        • White D.
        • Mendelson M.
        Manual for the Body-Esteem Scale for adolescents and adults.
        Universite Concordia, Montreal1997
        • Beck A.T.
        • Steer R.A.
        • Brown G.K.
        Manual for the beck depression inventory-II.
        Psychological Corp, San Antonio1996
        • Endler N.S.
        • Parker J.D.A.
        Coping inventory for stressful situations (CISS) manual.
        Multi-Health Systems, Toronto1990
        • Takii M.
        • Uchigata Y.
        • Kishimoto J.
        • et al.
        The relationship between the age of onset of type 1 diabetes and the subsequent development of a severe eating disorder by female patients.
        Pediatr Diabetes. 2011; 12: 396-401
        • Markowitz J.T.
        • Butler D.A.
        • Volkening L.K.
        • et al.
        Brief screening tool for disordered eating in diabetes: Internal consistency and external validity in a contemporary sample of pediatric patients with type 1 diabetes.
        Diabetes Care. 2010; 33: 495-500
        • Gagnon C.
        • Aimé A.
        • Bélanger C.
        • Markowitz J.T.
        Comorbid diabetes and eating disorders in adult patients: Assessment and considerations for treatment.
        Diabetes Educ. 2012; 38: 537-542