Sedative load and mortality among residents of long-term care facilities: a prospective cohort study

Abstract

Background: Older people are often prescribed multiple drugs with sedative properties. Most research has focused on specific classes of sedative and psychotropic drugs. The cumulative effect of taking multiple drugs with sedative properties has been termed 'sedative load'. Few previous studies have investigated the sedative load among residents of long-term care facilities. No previous studies have assessed the possible association between sedative load and mortality.

Objective: The objective of this study was to describe the sedative load among residents of long-term care facilities, and to investigate a possible association between sedative load and mortality.

Methods: This was a prospective cohort study. The study population comprised all 1444 residents of 53 long-term care wards in seven hospitals in Helsinki during September 2003. Of the eligible residents, 1087 residents or their proxies provided written informed consent to participate. Medical, medication and follow-up mortality data were available for 1004 residents. The main outcome measures were sedative load and all-cause mortality.

Results: The mean age of the residents was 81.3 (SD 10.9) years, and the mean number of regularly used drugs per resident was 7.1 (SD 3.4). Fifteen percent of residents were categorized as non-users of sedative drugs, 32% as users of some drugs with sedative properties and 53% as residents with a high sedative load. There was a bivariate association between having a higher sedative load and younger age (p < 0.001), male sex (p = 0.006), not being widowed (p = 0.001), diagnosis of depression (p < 0.001), diagnosis of psychiatric illness other than depression (p < 0.001), not being diagnosed with dementia (p = 0.009) and a shorter duration of institutional care (p = 0.02). Unadjusted analysis revealed that having a higher sedative load was associated with increased survival (p = 0.04, log rank test). However, in the adjusted Cox proportional hazard model, only poor nutritional status (hazard ratio [HR] 1.55; 95% CI 1.32, 1.82), male sex (HR 1.37; 95% CI 1.12, 1.69), increasing age (HR 1.04; 95% CI 1.03, 1.05) and co-morbidity (HR 1.07; 95% CI 1.02, 1.13) were significantly associated with risk of death.

Conclusions: There is a very high rate of sedative and psychotropic drug use among residents of long-term care facilities in Helsinki. However, having a high sedative load was not associated with an increased risk of death. Further research is needed to investigate the possible association between sedative load and mortality using alternative models and methods, and in different resident populations.