Send to

Choose Destination
Aust N Z J Public Health. 2010 Dec;34(6):589-93. doi: 10.1111/j.1753-6405.2010.00613.x. Epub 2010 Oct 7.

Factors associated with choice of antipsychotic treatment in elderly veterans: potential confounders for observational studies.

Author information

Data Management and Analysis Centre, Discipline of Public Health, University of Adelaide, South Australia.



Antipsychotics are commonly used in the elderly despite limited efficacy and safety data from randomised controlled trials. Observational comparative safety studies of antipsychotics vary, which may be due to confounding.


To compare the characteristics of typical and atypical antipsychotic initiators.


Using the Australian Government Department of Veterans' Affairs claims dataset, we compared patient and prescribing physician characteristics and health care utilisation between atypical and typical antipsychotic initiators. Significant independent predictors of use were calculated using a multivariate log-binomial model.


Compared to patients initiated on typical antipsychotics (n=10,966), patients initiated on atypical antipsychotics (n=9,239) were less likely to be male (Relative Risk (RR)=0.91, 95% CI 0.89-0.94) and have prior dispensing of morphine (RR=0.53, 95% CI 0.49-0.57) and oral corticosteroids (RR=0.86, 95% CI 0.81-0.91) and to have been hospitalised for myocardial infarction or pneumonia. Patients initiated on atypical antipsychotics were more likely to be in aged care (RR=1.08, 95% CI 1.05-1.12), to be prescribed the medicine by their usual doctor (RR=1.12, 95% CI 1.09-1.16) and have prior dispensing of anticholinesterases (RR=1.19, 95% CI 1.15-1.23), antidepressants (RR=1.18 95% CI 1.15-1.22) and anti-parkinson medications (RR=1.30, 95% CI 1.25-1.36).


Differences between typical and atypical antipsychotic initiators indicate the potential for confounding in observational studies. Future pharmacoepidemiogical research in Australia, investigating the adverse events of antipsychotics, should consider the variables identified in this study to control for confounding.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Wiley
Loading ...
Support Center