In many countries, pain is the most common indication for use of antidepressants in older adults. We reviewed the evidence from randomized controlled trials on the efficacy and safety of antidepressants, compared to all alternatives for pain in older adults (aged ≥65 years). Trials published from inception to 1 February 2024, were retrieved from 13 databases. Two independent reviewers extracted data on study and participant characteristics, primary efficacy (pain scores, converted to 0-100 scale) and harms. Estimates for efficacy were pooled using a random effects model and reported as difference in means and 95% CI. Quality of included trials was assessed using the Cochrane risk of bias tool. Fifteen studies (n = 1369 participants) met the inclusion criteria. The most frequently studied antidepressants were duloxetine and amitriptyline (6/15 studies each). Pain related to knee osteoarthritis was the most studied (6/15 studies). For knee osteoarthritis, antidepressants did not provide a statistically significant effect for the immediate term (0-2weeks), (-5.6, 95% confidence interval [CI]: -11.5 to 0.3), but duloxetine provided a statistically significant, albeit a very small effect in the intermediate term, (≥6weeks and <12 months), (-9.1, 95% CI: -11.8 to -6.4). Almost half (7/15) of the studies reported increased withdrawal of participants in the antidepressant treatment group vs. the comparator group due to adverse events. For most chronic painful conditions, the benefits and harms of antidepressant medicines are unclear. This evidence is predominantly from trials with sample sizes of <100, have disclosed industry ties and classified as having unclear or high risk of bias.