Objective: To investigate use of prescription and non-prescription drugs, predictors of drug use and symptoms of postprandial and postural hypotension were examined in 765 men and 846 women. Method: A questionnaire mailed to 6324 subjects randomly chosen from the electoral roll. Results: Drug use, both prescription and non-prescription, was greater ( P< 0·0001) in women (mean 4·1 total drugs, CI 3·9, 4·3) than in men (3·5 drugs, CI 3·3, 3·7); 41% of women and 36% of men used at least four drugs. Total drugs used were significantly greater in women (standardized regression coefficient (b)=0·101, P< 0·0001), and increased with the number of diagnoses (b=0·461, P< 0·0001) and taking antidepressants (b=0·160, P< 0·0001). Use of non-prescription drugs was also greater in women (b=0·094, P< 0·0001) and in association with arthritis (b=0·053, P=0·036) but decreased with age (b=−0·013, P< 0·0001). Potentially important drug interactions occurred in 10% of subjects and included hypoglycaemic agents with b-blockers and asthma with b-blockers. Symptoms of postprandial hypotension, reported in 2·6% of subjects, increased with the number of drugs used (OR 1·17, CI 1·05, 1·31) and with diabetes (OR 2·27, CI 1·05, 4·89). Symptomatic postural hypotension, recorded by 13% of subjects, related significantly to age (OR 1·04, CI 1·02, 1·06), tranquillisers (OR 1·98, CI 1·11, 3·54), antidepressants (OR 2·23, CI 1·30, 3·84) and ACE inhibitors (OR 2·14, CI 1·42, 3·22). Conclusion: Reduction in the number and frequency of drug doses and better knowledge about drugs may improve compliance. Observed associations with polypharmacy or hypotension suggest possible approaches to decreasing drug-related illness and limiting the adverse effects of hypotension in the elderly.