Linking Subjective and Objective Measures of Economic Hardship to Parent and Child Mental Health: A Longitudinal Analysis of a Low-Income Sample

Background and Purpose: Household poverty and other tangible indicators of economic hardship are linked to poorer mental health outcomes for children and adults. However, objective economic measures have limited utility in studies of low-income samples due to a lack of heterogeneity. Subjective measures of economic hardship may be especially useful in this context. Using prospective data from a study of economically disadvantaged women with children, this study examines the extent to which objective and subjective measures of economic hardship correspond with child and adult mental health outcomes.

Methods: Data were drawn from the Families and Children Thriving Study, a longitudinal investigation of low-income families who received services from a home visiting program in Wisconsin. The analysis included 1,282 households with valid data at three time points from the child’s birth through early childhood. Program records provided a baseline measure of objective economic hardship based on an income-to-needs ratio reflecting federal poverty level (FPL) guidelines. Households were categorized into one of three groups (sample proportions in parentheses): (1) < 50% FPL (32.8%); (2) 50-100% FPL (33.8%); (3) > 100% FPL (33.4%). Participants also completed a baseline survey that included a 5-item measure of subjective economic hardship (sample item: I have trouble affording the basic things that my family needs). Participant surveys included PROMIS® assessments of self-reported mental health at three time points and child mental health at time 3 (average age = 6.0).

Latent growth modeling (LGM) was used to assess the degree to which subjective and objective measures of economic hardship correspond with caregiver mental health trajectories. Structural equation modeling (SEM) was used to analyze relations between both economic hardship measures and a latent child mental health variable, with and without a control for caregiver mental health at baseline. Measures of caregiver age and race/ethnicity were included as covariates in both modeling approaches. All analyses were performed in Mplus 8.11.

Results: The LGM and SEM models fit the data well (e.g., CFI = .991-.996; TLI = .984-.987; RMSEA = .024-.041, SRMR = .013-.020). Measures of objective and subjective economic hardship were significantly associated with poorer adult mental health at baseline (β = .10 and .49, respectively), but only subjective hardship predicted a worse mental health trajectory (β = .28). Objective and subjective indicators of hardship were linked to an increase in negative child mental health symptoms (β = .07 and .16, respectively). However, the associated effects of objective and subjective economic hardship on child mental health were not statistically significant after accounting for caregiver mental health.

Conclusions and Implications: When compared to an income-to-needs measure, a subjective measure of economic hardship was associated with more robust effects on caregiver and child mental health. The findings have salient implications for understanding the relationship between economic hardship and mental health, especially within the context of research on low-income families. Future work should explore the extent to which self-reported measures of perceived hardship are influenced by the mental health of respondents.