P19 Detection of eating disorders in type 1 diabetic patients through the DEPSR survey

Abstract

Introduction: Eating disorders (ED) are more prevalent among adolescents with type 1 diabetes compared to healthy adolescents. This association is worrying since it generates poor metabolic control and therefore increases the risk of developing the complications of diabetes, increasing the risk of mortality by three times (4). This is why early detection and proper management ofED in this population at risk represents a relevant aspect in their care.

Objective: We examined the clinical utility of the DEPSR (diabetes eating problem surveyrevised) survey, a short selfreport questionnaire developed for patients with type 1 diabetes, in order to identify adolescents at risk. We also analysed whether other variables including clinical, psychosocial and conflict characteristics in relation to diabetes management are associated with a positive DEPS-R screening survey.

Methods: Transversal, observational, analytical design study. Thirtythree adolescents with type 1 diabetes aged between 10 and 18 years old participated in an outpatient control of the Diabetes Service of the Hospital Pediátrico Dr Humberto Notti during the month of September 2019. All participants completed the DEPSR screening survey as well as a questionnaire regarding psychosocial characteristics and conflict around diabetes management (Diabetes Family Conflict Scale Revised DFCSR). The clinical variables were obtained from the participants' medical records. Differences were examined between the groups with positive (score greater than 20) and negative DEPR. Adolescents who submitted a positive screening survey were offered a referral to a specialized eating team for further evaluation.

Results: Of 33 participants (mean age +/- SD= 13. 78 years +/- 2), 39 (13/33) DEPSR were positive. A similar number of female patients in relation to male patients obtained positive DEPSR (54% vs 46%, p= 0. 27). BMI z score was higher in the positive DEPSR group (mean= 0. 99 SD= 0. 44) than the negative DEPSR group (mean= 0. 3 SD=0.15,
p 0. 04). Those with positive DEPSR had higher Hb A1c % values (mean=10.2 +/- 2. 7 vs 8. 3 +/- 1, p= 0. 006). The group of patients with positive DEPSR had higher scores in relation to conflict linked to diabetes management (p 0. 002). Those adolescents with separated/divorced parents scored higher in DEPSR (p 0. 03). In regression analysis, reporting by adolescents of greater conflict in relation to diabetes management
(OR= 6. 4, p=0.02) and % of Hb A1 c greater than (OR= 4. 8, p= 0. 04) were predictors of a positive score in DEPS-R.

Discussion: This study strengthens the idea that BMIz score is an important predictor for eating disorders among adolescents with type 1 diabetes so attention to body weight management and adolescents perception of body weight should be increased. With respect to diabetes control, several studies have looked for correlation between ACT and HbA1c values as a surrogate for diabetes control, with inconsistent results. Our results confirm a significant association between ACTs and suboptimal glycemic control. Interestingly, our findings showed a strong correlation between the degree of conflict regarding parent-adolescent diabetes management and the positive DEPS-R survey.

Conclusion: The DEPSR survey is a useful clinical tool for identifying adolescents with type 1 diabetes at risk for developing disordered eating behaviors. Suboptimal diabetes control and increased conflict in diabetes management are associated with a positive DEPSR score. It remains to be established in future work whether patients with positive DEPSR scores, after being evaluated by a multidisciplinary team specialized in eating
disorders (ED), meet DSM V diagnostic criteria for these pathologies and can determine the positive predictive value of DEPS-R.