933-P: Impact of Sleep and Activity on Glycemic Control and Quality of Life in Haitian Youth with Type 1 Diabetes

Abstract

Background: Sleep and exercise affect overall health. In youth with type 1 diabetes (T1DM), physical activity and adequate sleep reportedly improve glycemic control. Data from low-income countries is lacking. Objective: To describe sleep and activity in Haitian youth with T1DM and examine their impact on glycemic control, health-related quality of life (HRQL) and Life Satisfaction (LS). Methods: Single-center cross-sectional study in Haiti, including youth diagnosed with T1DM between 0-25 years old. Wristbands tracked activity and sleep for 10 days. Diabetes Quality of Life in Youth (DQoLY) questionnaire was used to evaluate HRQL and LS. Point-of-care hemoglobin A1c (A1C) values (max. measure 14%) were recorded. Linear regression was used to assess sleep and activity as predictors of A1C, HRQL and LS. Results: We included 66 participants (59% female, mean age 17.8±4.8 years, mean diabetes duration 3.7±3.4 years, mean BMI Z-score -0.86±1.1). Mean HRQL was 63/100 and mean LS 65/100. Mean A1C was 11.3%, and 22 patients (34%) had A1C >14%. Mean daily step count was 9448±8431 and mean sleep duration 7h31±1h17. Higher A1C was predicted by younger age (p=0.005) and lower BMI Z-score (p=0.0016), but not sex or diabetes duration. Better HRQL was predicted by younger age (p=0.029), and marginally by male sex (p=0.07). None predicted LS. Sleep duration and step count did not predict either outcome. When excluding patients with A1C>14% (to eliminate a possible ceiling effect), shorter sleep duration was a significant predictor of higher A1C (p=0.03) in uni- and multivariate models adjusted for age, sex, BMI and diabetes duration. Conclusions: Youth with T1DM in Haiti have poor glycemic control, and low HRQL and LS. Their average sleep duration and daily activity is similar to reported values in peers. While activity level did not affect A1C, HRQL or LS, shorter sleep duration predicted higher A1C in youth with A1C<14%. Prospective studies with larger sample sizes are needed to validate our findings. Disclosure M. Babinski: None. R. Duperval: None. J. E. Von oettingen: None. K. Altenor: None.