RESUMO
BACKGROUND: Falls are considered as a predictive marker of poorer outcomes for people living with HIV (PLWHIV). However, the available evidences on the predictive value of falls are controversial. Our aim is to summarize the existing data about falls in PLWHIV. METHODS: A literature search was conducted using electronic databases (MEDLINE, Embase, and LILACS) for original observational studies. The primary outcome was any and recurrent falls' frequency in PLWHIV, and secondary outcomes were factors associated with falls. We conducted a random-effects meta-analysis with meta-regression to obtain a summary frequency of falls and recurrent falls. RESULTS: The pooled frequency for any fall was 26% [95% confidence interval (CI): 19% to 34%], compared with 14% for recurrent falls (95% CI: 9% to 22%). In studies comparing PLWHIV and people without HIV, we found no difference for any (pooled odds ratio 1.03, 95% CI: 0.90 to 1.17) or recurrent falls (pooled odds ratio 1.08, 95% CI: 0.92 to 1.27) between groups, but falls in middle-aged PLWHIV might be more associated with subjacent clinical conditions such as cognitive impairment, polypharmacy, use of medications with action in the central nervous system, and frailty, classic risk factors for falls in the elderly. CONCLUSIONS: The overall frequency of falls in PLWHIV seems to be lower than that presented by some initial studies, and several factors associated with falls are shared with elderly adults. Although both PLWHIV and people without HIV presented similar frequency of falls, we found that these events might be qualitatively different; therefore, an appropriate method to evaluate falls in this population to prevent adverse outcomes is warranted.