100. CROSS-SECTIONAL ASSOCIATIONS BETWEEN SLEEP and METABOLIC HEALTH: ROLES OF SCHIZOPHRENIA ILLNESS AND MEXICAN ETHNICITY

Abstract

Introduction

People with schizophrenia (PwS) have a higher risk of developing coronary heart disease (CHD) and sleep disturbances are an important risk factor for CHD and poor metabolic health. Studies have demonstrated that PwS have a higher prevalence of metabolic syndrome (MetS) and sleep disturbances. One study observed poor subjective sleep quality and longer sleep latency were associated with having metabolic disorder in drug-naïve, Chinese PwS.

Mexican ethnicity may be a modifier between sleep and metabolic health, which has been understudied among PwS. Short sleep duration measured by wrist-worn actigraphy has been linked with poor metabolic outcomes in Hispanic individuals. Furthermore, one study observed that people from Mexico are more likely to be obese and have diabetes compared to people from South America or Cuba, rendering individuals from Mexico a particularly vulnerable population for worse metabolic outcomes. Despite this, research on sleep and metabolic health in PwS, especially among Mexican PwS, is limited.

This study aims to explore the associations between sleep quality and metabolic health in PwS compared to non-psychiatric comparison participants (NCs), with a focus on people of Mexican ethnicity compared to people of non-Hispanic white (NHW) backgrounds. We hypothesize that (1) PwS will exhibit worse sleep quality, objective sleep measures and metabolic health compared to NCs, and (2) worse sleep quality and objective sleep measures will be associated with poorer metabolic health in PwS and NCs. An exploratory analysis will investigate the modifying relationship of Mexican ethnicity on sleep quality and metabolic health.

Methods

The sample included 199 English-speaking community-dwelling participants aged 30-70 from San Diego, including 83 adults with schizophrenia or schizoaffective disorder and 116 non-psychiatric comparison participants. We divided the analyses into Study 1 and Study 2 to include subsets of participants with the relevant sleep data. Study 1 includes 199 individuals with subjective sleep quality measures, assessed with the Pittsburgh Sleep Quality Index (PSQI). The PSQI includes overall sleep quality and component scores such as sleep quality, sleep latency, sleep duration, sleep efficiency, sleep disturbances, use of sleep medication, and daytime dysfunction. Study 2 includes 61 individuals who had objective sleep quality, assessed with 5-day Fitbit activity trackers. Objective sleep measures included mean and variability of total sleep time, bedtime, and wake after sleep onset (WASO). For both Study 1 and Study 2, metabolic health measures included body mass index (BMI), hemoglobin A1c (HbA1c), insulin resistance (HOMA-IR), and number of MetS criteria. For MetS criteria, participants were required to meet at least three of the following: elevated waist circumference, triglycerides, HDL levels, or blood pressure.

Statistical analyses involved chi-square tests and independent samples t-tests to compare diagnostic group. General linear models were used to examine associations between sleep quality and metabolic health, adjusting for covariates such as ethnicity, age, education, and gender. All analyses were conducted using SPSS, with significance set at p < 0.05.

Results

The total sample included 83 PwS and 116 NCs, of which 50.8% were female and mean age was 54.6 years. There were 27 Mexican PwS (57.4%) and 20 Mexican NCs (42.6%). PwS had worse overall subjective sleep quality than NCs (t=-5.045, p= LESS THAN 0.001, d=-0.725). PwS also had worse PSQI component scores (greater sleep latency, more sleep disturbances, increased use of sleep medication, and poorer daytime dysfunction) compared to NCs. For objective sleep measures, PwS had higher mean and variability of total sleep time as well as higher WASO compared to NCs. PwS had an earlier mean bedtime compared to NCs. Additionally, PwS had higher BMI, HbA1c levels, HOMA-IR, and greater number of MetS criteria than NCs.

Study 1 findings: Lower subjective sleep duration was significantly associated with higher HbA1c levels in NCs (B=.019, SE=.007, p=.006, η2=.067), but not in PwS. Other subjective sleep measures showed no significant associations with BMI, HbA1c, HOMA-IR, and MetS criteria.

Study 2 findings: Greater variability of total sleep time was associated with higher BMI in PwS (B=.038, SE=.018, p=.05, η2=.233). In NCs, decreased mean total sleep time was significantly associated with higher BMI (B=-.058, SE=.022, p=.015, η2=.183). Increased variability in bedtime (B=6.671e-6, SE=1.434e-6, p= LESS THAN .001, η2=.396) and later mean bedtime were also associated with higher HbA1c levels in NCs. (B=2.497e-6, SE=1.097e-6, p=.029, η2=.136). Due to the small sample size, HOMA-IR and MetS criteria outcomes were not able to be analyzed for these hypotheses.

For the models that included subjective sleep latency, sleep duration, sleep efficiency, sleep disturbances, use of sleep medication, and daytime dysfunction, Mexican ethnicity was significantly associated with higher HbA1c levels in PwS, but not in the NCs.

Conclusions

In PwS, the significant association between greater variability of mean total sleep time and higher BMI highlights the need for interventions focused on promoting consistent sleep schedules to manage metabolic risk. Additionally, the results suggest that Mexican ethnicity plays a key role in metabolic health in PwS. This finding highlights the importance of targeted screening and personalized interventions for Mexican PwS to attenuate the risk for poor metabolic health.