Increased income over time predicts better self-perceived mental health only at a population level but not for individual changes: An analysis of a longitudinal cohort using cross-lagged models

Background: The literature identifies a strong relationship between mental health and income, but there is little research that clarifies the directional association between household income and self-perceived mental health (SPMH) overtime either at between-perso+n or within-person levels. This study investigates whether higher income predicts better SPMH overtime and poor SPMH predicts lower income overtime both at between-person or within-person levels. Methods: Data analyzed was from the Montreal Southwest Social and Psychiatric Epidemiology Catchment Area study (ZEPSOM), a longitudinal community-based cohort. The baseline survey was conducted in 2007/8 with follow-up every two years. We traced a total of 3464 participants over a period of 8 years. To examine the associations between income and SPMH at both between-person or within-person levels, cross-lagged panel models (CLPMs) and random intercept cross-lagged panel models (RI-CLPMs) were used. Gender and age effects were examined using multiple group analyses. Complete case analyses evaluated the findings’ robustness. Results: At between-person levels, higher household income predicted higher SPMH, but not vice versa. These associations were stronger among men and older adults. At within-person levels, higher income did not predict higher SPMH. No significant gender- or age- group differences were observed. Complete case analyses supported the findings. Limitations: Loss to follow-up may affect the generalizability of the research findings. Conclusions: This study suggests that higher household income predicts higher SPMH at between-person levels. Policy and programs aiming at promoting mental health should focus on low-income individuals, especially men and older adults.