Background Medication use is a potentially modifiable risk factor for falling; psychotropic and cardiovascular drugs have been indicated as main drug groups that increase fall risk. However, evidence is mainly based on studies that recorded falls retrospectively and/or did not determine medication use at the time of the fall. Therefore, we investigated the associations indicated in the literature between medication use and falls, using prospectively recorded falls and medication use determined at the time of the fall. Methods Data from the B-PROOF (B-vitamins for the prevention of osteoporotic fractures) study were used, concerning community-dwelling elderly aged >= 65 years. We included 2,407 participants with pharmacy dispensing records. During the 2- to 3-year follow-up, participants recorded falls using a fall calendar. Cox proportional hazard models were applied, adjusting for potential confounders including age, sex, health status variables and concomitant medication use. Results During follow-up, 1,147 participants experienced at least one fall. Users of anti-arrhythmic medication had an increased fall risk (hazard ratio [HR] 1.61; 95 % confidence interval [CI] 1.12-2.32) compared with non-users. Similarly, non-selective beta-blocker use was associated with an increased fall risk (HR 1.41 [95 % CI 1.12-1.78]), while statin use was associated with a lower risk (HR 0.81 [95 % CI 0.71-0.94]). Benzodiazepine use (HR 1.32 [95 % CI 1.02-1.71]), and antidepressant use (HR 1.40 [95 % CI 1.07-1.82]) were associated with an increased fall risk. Use of other cardiovascular and psychotropic medication was not associated with fall risk. Conclusion Our results strengthen the evidence for an increased fall risk in community-dwelling elderly during the use of anti-arrhythmics, non-selective beta-blockers, benzodiazepines, and antidepressant medication. Clinicians should prescribe these drugs cautiously and if possible choose safer alternatives for older patients.