Symptom Effects on Employment in a Structural Model of Mental Illness and Treatment: Analysis of Patients with Schizophrenia.

BACKGROUND: There is a long tradition in the health and mental health economics literatures of estimating the impacts of disorders on employment and earnings. Several analyses have associated mental illness with poorer labor market outcomes, often using indicators of disorders to measure mental illness, but it is unclear to what extent unobserved medical treatment biases the estimated impacts of disorders on labor market outcomes. In this study we argue that in order to judge the true employment costs of mental illness and the potential benefits of treatment it is necessary to account for the structural relationship between treatment, symptoms, and employment outcomes. AIMS OF THE STUDY: The study proposes a structural model for understanding mental illness impacts on employment and empirically estimates one element of this structural model that links symptoms of schizophrenia to patients' employment status. In addition, we use our empirical estimates to simulate employment consequences of more effective treatment and reductions in symptom levels. EMPIRICAL
METHODS: Our empirical analyses use a sample of 1,643 adults with a schizophrenia diagnosis. We predict the likelihood of three outcomes - not employed, employed in a sheltered or supported job, and employed in a non-supported job. Analyses include measures of demographic characteristics, illness history, location differences, and detailed symptom measures.
RESULTS: We find that negative symptoms have a substantial adverse impact on participation in both non-supported jobs and in sheltered or supported jobs. The impacts on employment of other symptoms of schizophrenia are not as large, but significant effects are also found for symptoms of depression. Simulations suggest, however, that only one-third of consumers would be employed in any type of job even given a large reduction in symptom levels. DISCUSSION: Negative symptoms are particularly important for role functioning and employment. The marginal effect on employment of a reduction in negative symptoms is several times greater than the effect of a comparable reduction in positive symptoms. Moreover, the effect of an improvement in symptoms on employment is stronger for non-supported employment than for working in sheltered or supported employment. Although commonly measured symptoms of schizophrenia impact employment, greater control of symptoms alone is unlikely to lead to large increases in employment for persons with schizophrenia in the near term. IMPLICATIONS FOR HEALTH CARE PROVISION AND USE: These findings suggest that improved treatment that results in reduced symptom levels will increase rates of employment among persons with schizophrenia, but that large employment impacts probably also require more effective rehabilitative therapies that target improvement in functioning. IMPLICATION FOR POLICY: Expansions of supported employment opportunities and removal of work disincentives in public income-support programs are two additional measures that may help to increase employment participation.