A large epidemiological study has found that prescription of an antipsychotic drug to an elderly patient with dementia is associated with increased risk of death subsequently; the risk possibly being greater with conventional as opposed to atypical agents.
The study authors note that elderly people with dementia may develop a range of challenging behavioural and psychological symptoms, and antipsychotic drugs may be used to control these. Trials suggest that any improvements from the drugs may be outweighed by adverse effects. However, evidence from various studies has suggested that they may be associated with an increased risk of mortality.
This study aimed to determine the risk of all-cause mortality in elderly people with dementia who were prescribed these drugs, using population-based data from healthcare databases, in Ontario, Canada, to generate the study cohort. Ontario has a population of about 12 million, most of whom are covered by a universally-funded healthcare programme, and about 1.4 million are aged 65 or over. From these, the authors identified all those aged 66 and over with a diagnosis of dementia, and excluded all who had a history of psychotic illness, were receiving palliative care, or who had been prescribed any antipsychotic medication in the year before entering the cohort.
All new antipsychotic medication dispensed after cohort entry was identified and divided into conventional and atypical. Primary outcome was all-cause mortality 30, 60, 90 and 180 days after initial dispensing, and mortality in those prescribed conventional and atypical antipsychotics compared with that of those prescribed none. Potential confounding factors were controlled for by developing propensity scores to match users and non-users.
A total of 27,259 propensity score-matched pairs were identified: 9,100 community-dwelling and 4,036 hospitalised patients for atypical antipsychotics compared to non-use, and 6,888 and 7,235 for atypical compared to conventional antipsychotic use.
Analysis indicated that new use of atypical antipsychotics was associated with a statistically significant increase in the risk for death at 30 days compared with non-use. This applied in both the community-dwelling cohort, with an adjusted hazard ratio (HR) of 1.31 (95% CI, 1.02–1.70; absolute risk difference, 0.2 percentage point) and the long-term care cohort (HR 1.55; 95% CI, 1.15–2.07; absolute risk difference, 1.2 percentage points).
Use of conventional antipsychotics was associated with a higher risk of death at all time points. The effect was sensitive to possible unmeasured confounding factors, especially for atypical agents: a factor (or combination of factors) that increased the risk of death in those prescribed the drugs by 1.5 would make the association no longer statistically significant.
The authors conclude that their analysis supports previous reports of an increased risk of death in elderly people with dementia who are prescribed antipsychotic drugs, and that the risk is greater with conventional than atypical drugs. They discuss potential mechanisms for the effect, but also caution that the study has limitations.
The risk estimates are relatively small and could be confounded by other variables that could not be included. Nevertheless, they highlight the need to balance the risks and benefits carefully when considering the use of these drugs in this patient population, and restrict them to situations where non-pharmacological measures have been ineffective.
Ann Intern Med 2007;146:775-86