Objectives. We took advantage of a 2-intervention natural experiment to investigate the impacts of neighborhood demolition and housing improvement on adult residents’ mental and physical health. Methods. We identified a longitudinal cohort (n = 1041, including intervention and control participants) by matching participants in 2 randomly sampled cross-sectional surveys conducted in 2006 and 2008 in 14 disadvantaged neighborhoods of Glasgow, United Kingdom. We measured residents’ self-reported health with Medical Outcomes Study Short Form Health Survey version 2 mean scores. Results. After adjustment for potential confounders and baseline health, mean mental and physical health scores for residents living in partly demolished neighborhoods were similar to the control group (mental health, b = 2.49; 95% confidence interval [CI] = −1.25, 6.23; P = .185; physical health, b = −0.24; 95% CI = −2.96, 2.48; P = .859). Mean mental health scores for residents experiencing housing improvement were higher than in the control group (b = 2.41; 95% CI = 0.03, 4.80; P = .047); physical health scores were similar between groups (b = −0.66; 95% CI = −2.57, 1.25; P = .486). Conclusions. Our findings suggest that housing improvement may lead to small, short-term mental health benefits. Physical deterioration and demolition of neighborhoods do not appear to adversely affect residents’ health.