IMPORTANCE: When a patient with neovascular age-related macular degeneration or diabetic macular edema does not respond to an initial anti-vascular endothelial growth factor agent, usually after several injections, ophthalmologists may switch to another anti-vascular endothelial growth factor agent. Authors of case series have suggested beneficial effects from switching. However, to our knowledge, there are no studies with an appropriate control group to evaluate how such patients would do without switching agents. OBJECTIVE: To assess outcomes in patients who have a poor initial response but continue treatment without switching agents. DESIGN, SETTING, AND PARTICIPANTS: We obtained data from 2 multicenter clinical trials, the Comparison of Age-Related Macular Degeneration Treatments Trials (CATT) and the Diabetic Retinopathy Clinical Research Network (DRCR.net). Based on typical clinical reasons for switching agents, we developed "switching rules" at both 3 and 6 months after initiation of treatment. Using these switching rules, we identified a 3-month and a 6-month cohort of "treatment failures" from both CATT and DRCR.net studies. INTERVENTIONS: Although the cohorts from each study met criteria for switching, they were treated with the initial agent throughout the study (bevacizumab or ranibizumab in CATT and ranibizumab in DRCR.net). MAIN OUTCOMES AND MEASURES: Primary outcomes were change in visual acuity and change in central retinal thickness on optical coherence tomography from the 3- or 6-month visit at which switching rules were met. RESULTS: The 126 patients from CATT and the 59 patients from DRCR.net who were selected for the switching analysis were similar in age, sex and race/ethnicity to the overall study populations. Among the participants who met the criteria for switching, the CATT participants were a mean (SD) of 79.7 (7.8) years of age, 65.9% women, and 97.6% white, while the DRCR.net participants were a mean (SD) of 65.5 (9.3) years of age, 44.1% women, and 76.3% white In all 4 cohorts, there was a 3- to 5-letter improvement in mean visual acuity over the 3 months after the switching rules were met, although all patients continued on their originally assigned treatment. Mean central retinal thickness also improved by 40 to 70 μM. CONCLUSIONS AND RELEVANCE: These results demonstrate the importance of having a comparison group to evaluate the effect of switching anti-vascular endothelial growth factor agents for treatment of neovascular age-related macular degeneration or diabetic macular edema. Without a comparison group, it is impossible to know whether any improvement observed after switching was related to the new treatment or was related to regression to the mean and time effects as observed in the 4 cohorts presented here. Randomization to switching or not switching drugs would provide a basis for valid conclusions about the effects of switching.