Sleep duration and mortality among older adults in a 22-year follow-up study: an analysis of possible effect modifiers.

The aim of this study was to assess the relationship of sleep duration and all-cause mortality among 2,449 Polish community-dwelling older citizens of Krakow observed during 22 years of follow-up. In particular, the role of some demographic, psychosocial and health-related conditions were investigated in terms of modification effect. In the prospective study, background information was gathered by face-to-face interview. Vital data were obtained from the population registry. Cox regression models were used to assess the role of sleep duration in mortality, in the analyses of potential effect modifiers and the shape of the relationship. Sleep duration was observed to be a significant predictor of all-cause mortality. Life-weariness, functional activity, total number of chronic diseases and age (65-79, 80+) were found to be effect modifiers for the relationship between sleep duration and mortality. Further investigation showed a U-shaped mortality risk associated with the duration of sleep among individuals with a high level of life-weariness, high functional activity and in individuals aged 80 and over. On the other hand, a linear relationship between longer sleep duration and mortality was observed among older people with no experience of life-weariness, without chronic diseases, with medium functional activity and aged 65-79, but also among those who reported three and more chronic conditions. Results of our study support available evidence showing the relationship between sleep duration and mortality among older adults and suggest that any public health intervention in this area should consider also other coexisting modifiable psychosocial and functional determinants.