Problematic Debts in Health Insurance: effects on mental health care use

Background: Prior studies have consistently found that the incidence and persistence of mental health issues is higher among persons with personal debts, but the (causal) mechanisms for this relationship are not well understood. The literature has posited three alternative explanations for the association of personal debts and health. The first is the social causation hypothesis which posits that problematic debts cause worse mental health status. The second is the social selection hypothesis which states that worse mental health status increases the risk of having problematic debts. The third hypothesis is that some third variables, like major life events, cause both worse mental health status and problematic debts. So far, the literature has not been able to establish the (causal) pathway. Prior literature mostly uses cross-sectional survey data, different measures of personal -but not necessarily problematic- debts, and relies on self-rated health. Prior mental healthcare use and life events are usually not controlled for.

Method: Our nationwide individual-level data from the Netherlands includes detailed information on the timing and amount of problematic debts in health insurance, timing of (negative) life events (e.g. unemployment, divorce, being suspected of committing a criminal offence, widowing) and mental healthcare diagnosis and use in 2011-2015. The data also includes information on other socio-economic variables (e.g. income, other personal debts, education, job status), personal information (e.g. gender, age, marital status, ethnicity), household information and other healthcare expenditures. We first estimate how prior mental healthcare use and major life events affect the probability of going into problematic debts, while controlling for socio-economic variables and other healthcare expenditures. Next, we test whether having a problematic debt impacts mental health status, by comparing the mental health status changes of two groups of people (with and without problematic debts in health insurance). We match both groups on prior mental health status and other relevant characteristics by propensity score matching. We measure mental health status as overall mental disorder, adjustment disorders, burnout and depression. We also study whether the size of problematic debts matters.

Results: Like prior research, we find that a multiplicity of variables, such as negative life events (e.g. unemployment, divorce, or widowing), mental health care use in the prior year, socio-economic status and personal characteristics like gender, age and ethnicity affect the propensity of going into problematic debt. Our research is ongoing, and we are currently estimating the before and after models comparing two groups of people in which we study whether going into problematic debts impacts mental health status. Our preliminary results indicate a positive relationship.

Conclusion: This is the first research that uses nationwide individual-level data on timing and amount of problematic debts, life events and mental health status to estimate the relationship between problematic debt and mental health. Reduction of poverty and problematic debts is high on the political agenda, and given the consistently found relationship between mental health issues and problematic debts, it is of crucial importance to understand the diversity of mechanisms that drives the relationship between problematic debts and mental health status.