Longitudinal trends in late-life insomnia: implications for prescribing.

Abstract

To assess trends in insomnia and hypnotic drug use in a representative sample of elderly general practice patients. Longitudinal study with three interview waves--1985, 1989 and 1993. Urban and suburban Nottingham. 1042 patients originally aged 65 and over randomly sampled from general practitioner lists. Point prevalence estimates, status (case/non-case/died) at 4-year follow-up, episode incidence and survival functions. At baseline (1985) 221 respondents met the survey criteria for insomnia. Of these, 36.1% continued to report severely disrupted sleep in 1989. Within this period 84 new cases of insomnia were identified (an incidence rate of 3.1% per person-year at risk). Controlling for age and sex, insomnia was unrelated to survival among prevalent cases, but significantly related to survival among incident cases (odds ratio = 1.7; 95% confidence interval = 1.1-2.5). Of 166 respondents using prescription hypnotics in 1985, 31.7% continued to report usage in 1989. Similarly, out of 41 new hypnotic drug users identified in 1989, 29.3% continued to report usage in 1993. Important clinical differences in the natural history of insomnia are evident when incident and prevalent cases are compared. Nevertheless, outcomes at 4-year follow-up suggest that, for the majority of surviving cases identified in a prevalence screen and for a substantial minority of incident cases, late-life insomnia shows a level of chronicity incompatible with hypnotic drug therapy as currently recommended.