Sammendrag på engelsk (abstract)
Objectives: To assess the long-term mortality risk associated with antipsychotic drug (AP) use in nursing homes.
Design: A longitudinal study with 5 assessments over a 75-month follow-up period.
Setting: A representative sample of nursing home patients in 4 Norwegian counties.
Participants: At baseline, 1163 patients were included. At the last follow-up, 98 patients were still alive.
Measurements: Prevalent drug use at each assessment was registered. Level of dementia, neuropsychiatric symptoms, level of functioning, medical health, and use of restraints were recorded at each assessment. A Cox regression model with time-dependent psychotropic drug use as the main predictor was estimated and adjusted for confounders.
Results: In unadjusted Cox regression, a lower mortality risk was associated with the use of other psychotropic drugs, but not APs, compared with nonusers. In the adjusted analysis, neither use of APs nor other psychiatric drugs was associated with increased mortality risk. Higher age, male gender, not being married, medical disease burden, lower level of functioning, more severe degree of dementia, and a higher number of drugs were all associated with increased mortality risk.
Conclusion: In this long-term study of nursing home patients, AP drug use was not associated with increased risk of mortality. This is in line with results from earlier studies of clinical samples, but contrasts with results from randomized controlled trials and registry-based studies. The findings should be interpreted with caution. Taking into account the modest benefit and high risk of adverse effects of AP drug use, nonpharmacological treatment remains the first-line treatment approach
Journal of the American Medical Directors Association , 2016