Greenspace quality and proximity are stronger predictors of mental health than quantity or access

• Compared models of poor mental health with distinct urban greenspace characteristics. • Proxies for greenspace quality and proximity emerged as the strongest predictors. • Greenspace quantity and access were weaker in explaining variation in mental health. • Findings highlight the need to prioritise urban greenspace quality and proximity. As evidence grows for the mental health benefits of urban greenspaces, understanding which specific characteristics matter most is critical for guiding effective urban design and policy. This study compares multiple measures of greenspace exposure–focusing on proximity, quantity, access and quality–to determine which best predict changes in mental health. Individual-level mental health data were obtained from a nationally representative longitudinal survey in Great Britain (British Household Panel Survey, 1991–2008). The sample was restricted to individuals who moved neighbourhoods ( n = 492), and a Before-After-Control-Intervention (BACI) study design was applied. Individual repeated measures from before and after the move were used to quantify the effects of changes in greenspace exposure on the probability of poor mental health, here defined using GHQ-12 with a score ≥ 3. Generalised linear mixed models were developed and their relative performance compared to identify the best-performing models. Among seven models tested, models incorporating local bird species richness (proxy for greenspace quality) and proximity to local public greenspace provided the best fit, outperforming models based on quantity and access. To our knowledge, this is the first study to compare the effect of multiple greenspace measures on mental health using a BACI design specifically within an urban context. Our findings suggest the potential importance of urban greenspace quality and proximity in shaping mental health outcomes compared to measures of quantity and access. However, these findings are caveated by a temporal disconnect between exposure and outcome data; we therefore recommend that future studies validate these findings using temporally-aligned data. Based on this study’s findings, we recommend that future environmental planning and public health strategies prioritise these characteristics to maximise the mental health benefits of urban greenspaces.