Abstract
BACKGROUND: The independent effects of depressive symptoms and sleep problems for future physical illness risk have yet to be studied systematically across a variety of disease endpoints. METHODS: We analysed data from 7395 participants (65.81 ± 9.39 years; 54.8% female) from the English Longitudinal Study of Ageing (ELSA). Baseline was wave 4 and participants were followed up for 6 years until wave 7. Sleep was measured using an adapted version of the Jenkins Sleep Problems questionnaire and depressive symptoms using the Centre for Epidemiological Studies Depression scale. Participants with the illness of interest at baseline [coronary heart disease (CHD), cancer, diabetes/high blood glucose, arthritis] were excluded from models predicting the onset of that illness at follow-up. Logistic regression was used, entering depressive symptoms and sleep problems simultaneously into models controlling for a wide range of covariates. RESULTS: In fully adjusted models depressive symptoms predicted incident CHD (OR 1.11, 95% CI 1.04-1.20, p = 0.004) and diabetes/high blood glucose (OR 1.13, 95% CI 1.04-1.22, p = 0.002) independent of sleep problems; both depressive symptoms (OR 1.10, 95% CI 1.04-1.16, p = 0.002) and sleep problems (OR 1.14, 95% CI 1.02-1.26, p = 0.019) predicted incident arthritis. CONCLUSIONS: Sleep problems and depressive symptoms, and a combination of both, were differentially associated with physical illness onset 6 years later. Our findings highlight the importance of taking into account somatic and affective experiences when looking across a variety of different physical illnesses.