Antipsychotics and the risk of death in the elderly: an instrumental variable analysis using two preference based...

Observational studies have investigated the comparative safety of antipsychotics with varying results. Instrumental variable analysis has been suggested as a possible alternative to conventional analyses when there is concern about the effect of unmeasured confounding in observational studies. Using the example of the risk of death with typical compared to atypical antipsychotics, we aimed to explore the performance of two different instruments. We used the doctor prescribing preference instrument, which has been used in previous studies, to investigate further the assumptions of this instrument in the Australian population. We also propose an alternative instrument, nursing home facility preference.

With the Australian Department of Veterans' Affairs administrative claims database, we used an instrumental variable analysis to compare the risk of death after 12 months between the two antipsychotic classes.

Using the doctor prescribing preference instrument we estimated that typical antipsychotics were associated with an extra 24 (95% Confidence Interval (CI) 18-30) deaths per 100 patients per year compared to atypical antipsychotics, and an extra 10 (95% CI 7-14) deaths per 100 patients per year among nursing home residents. Facility prescribing preference was a stronger instrument (OR=19.2 95% CI 17.1-21.6) and provided a better balance of covariates than doctor prescribing preference.

Our study has shown that valid instruments in one population may not be directly applicable to other health care settings and testing of assumptions is crucial when performing IV analyses. Facility prescribing preference appears to be a potentially valid instrument for further work in this area.