Psychological Distress Mediates the Association between Poor Sleep Quality and Prediabetes

Abstract

Background: Evidence indicates that poor sleep quality increases risk for prediabetes. Psychological distress may be a potential mechanism through which poor sleep quality affects glucose metabolism. Sleep, prediabetes, and distress are linked to type 2 diabetes. We examined the relations between poor sleep quality and prediabetes, and tested whether psychological distress mediates their relation in a cross-sectional study. Methods: Subjects were 133 men (n=49) and women (n=84) aged 37.9±12.5 years with body mass index (BMI) between 30.3±6.9kg/m2. Subjective sleep quality was determined by the Pittsburg Sleep Questionnaire Index (PSQI). Subjects were stratified based on PSQI score into normal (NSQ; PSQI≤5, n=68) vs. poor sleep quality (PSQ; PSQI>5, n=65). Two hour-oral glucose tolerance test (OGTT) was conducted in each subject, and prediabetes was diagnosed using the ADA diagnostic criteria. Psychological distress was assessed using the Childhood Trauma Questionnaire, and depression was self-evaluated using Beck Depression Inventory. PROCESS Macros method was used to assess the direct relation between sleep quality and prediabetes, and the potential mediating effects of psychological distress on that relation. All data analysis was adjusted for age, gender, race, education, employment status, depression and BMI. Results: Experiencing poor sleep quality was associated with higher fasting glucose levels (95.9±7.6 vs. 98.0±10.2, p<0.01) and 2h-post prandial glucose levels (102.8±24.7 vs. 108.7±30.5, p<0.01) compared to NSQ group. PSQ group also had greater OGTT-derived QUICKI (p<0.05). Psychological distress mediated the relation between poor sleep quality and prediabetes with adjustment for co-variables. Conclusions: Poor sleep quality is associated with prediabetes through psychological distress. Efforts to improve sleep quality and decrease psychological distress may yield lower type 2 diabetes rate in this high-risk prediabetic population. Disclosure L. Li: None.