Abstract
Sleep disturbances are a common problem among institutionalised older people. Studies have shown that this population experiences prolonged sleep latency, increased fragmentation and wake after sleep onset, more disturbed circadian rhythms and night-day reversal. However, studies have not examined the extent to which this is because of individual factors known to influence sleep (such as age) or because of the institutional environment. This paper compares actigraphic data collected for 14 days from 122 non demented institutional care residents (across 10 care facilities) with 52 community dwelling poor sleepers aged over 65. Four dependent variables are analysed: i) the ‘interdaily stability’ (IS); ii) the ‘intradaily variability’ (IV); iii) the relative amplitude (RA) of the activity rhythm; and iv) the mean level of activity during the 24 hour day. Data are analysed using a fixed effects, single level, model (using MLwiN). This model enables comparisons between community and institutional care groups to be made whilst conditioning out possible ‘individual’ effects of ‘age’, ‘gender’, ‘level of dependency’, ‘level of incontinence care’, and ‘number of regular daily/prescribed medications’. After controlling for the effects of a range of individual level factors, and after controlling for unequal variance across groups (heteroscedascity), there is little difference between community dwelling older adults and institutional care residents in IS score; suggesting that the stability of day to day patterns (such as bed times, get up times, lunch times etc) is similar within these two resident groups. However, institutional care residents do experience more fragmented rest/wake patterns (having significantly higher IV scores and significantly lower mean activity values). Our findings strongly suggest that the institutional care environment itself has a negative association with older people’s rest/wake patterns; although longitudinal studies are required to fully understand any causal relationships.