OBJECTIVES: To compare the efficacy of topical ocular hypotensives for the treatment of intraocular hypertension and glaucoma. Adverse events that may affect patients’ willingness to comply with treatment were also assessed. METHODS: A systematic review was conducted to identify randomized controlled trials in patients with glaucoma or ocular hypertension with a prostaglandin analogue treatment arm. A total of 181 eligible articles were identified with 73 suitable for meta-analysis. A mixed treatment comparison (MTC) was performed to estimate the relative efficacy of the treatments. Studies connected to latanoprost and reporting the mean and standard deviation in absolute intraocular pressure (IOP) at three months were used in the MTC since this maximised the analysis dataset. Baseline IOP was included as a covariate in the MTC. Random effects models were used, as study variance indicated some degree of heterogeneity. A second MTC was conducted to estimate the rate of hyperaemia-type adverse events. RESULTS: The mean IOP at three months for latanoprost is 17.42 mmHg from a baseline of 23.5 mmHg.  Latanoprost is statistically significantly better at lowering IOP versus timolol, and latanoprost and timolol is statistically significantly more efficacious versus latanoprost alone. There were no further statistically significant differences in mean IOP for latanoprost versus any other treatments. The hyperaemia event rate for latanoprost is 25.18%. Using the odds ratio results compared to latanoprost, timolol has a statistically significantly lower event rate. Travoprost, bimatoprost, travoprost and timolol, bimatoprost and timolol, latanoprost and brimonidine, latanoprost and dorzolamide have statistically significantly higher event rates. CONCLUSIONS: The results indicate that there is no clinically relevant difference in IOP lowering between treatments except for timolol. However there are significant differences in adverse events. In actual practice such events affect a patients’ willingness to comply with treatment and therefore the effectiveness of treatment.