Session: Mental Health
Room: Upson 117
Time: Mon 15:00-16:30

Presenter: Benjamin Cook (Harvard Medical School - Cambridge Health Alliance. Psychiatry)

Discussant: Noelia DuchovnyCongressional Budget Office

Abstract

Introduction: In 2001, the U.S. Surgeon General summarized findings of racial/ethnic disparities in mental health care among children and promoted a vision for reducing these disparities. This paper describes the progress made in reducing these disparities, recognizing that the U.S. Food and Drug Administration’s numerous warnings concerning SSRI-class antidepressants during this time period may have had a differential impact on racial/ethnic groups’ use of mental health care services.

Methodology: We use 2002-2007 Medical Expenditure panel Survey (MEPS) data to measure trends in disparities in receipt of mental health care. We implement the Institute of Medicine (IOM) definition of racial-ethnic disparities using a rank-and-replace methodology which adjusts for variables related to need for mental health care but allows differences in variables related to system characteristics to enter into disparity predictions. We use two-part models for visits (hurdle models measuring any visit and number visits conditional on >1 visit) and expenditures (logit and generalized linear model measuring any expenditure and expenditures conditional on >$0 expenditure). This allows us to identify whether disparities are attributable to accessing the mental health care system or to the care received after entry into the system. We also examine whether differences in expenditures are due to differences in price (perhaps due to low reimbursement under public coverage) or differences in quantity of services. In additional analyses we exploit the longitudinal nature of the MEPS and determine factors that predict which children access services following a decline in mental health.

Findings: Preliminary findings show that, after adjustment for need, Black-White disparities exist in both 2002-2003 and 2005-2006 (this will be updated to include 2007 MEPS data), and that there is no significant trend in these disparities. Hispanic-White disparities were significant in 2002-2003 but not in 2005-2006 and this reduction in disparities is significant. As expected, indicators of socioeconomic status are significant mediators of these racial/ethnic disparities. These disparity trends are largely driven by disparities in any mental health care use as opposed to quantity of care once care is initiated. Work to be completed includes the use of longitudinal data to assess disparities in initiation.

Conclusions: After adjustment for need, the mental health care system continues to provide less care to African American children. These overall disparities are predominantly driven by disparities in any use of care. These results suggest that policies should be targeted to reduce disparities in initiation of children’s mental health care. We recommend that future efforts to identify trends in disparities in mental health services should use methodologies that adjust for health status, yet allow socioeconomic factors to mediate differences.

Key Terms

Mental Health Care Disparities; Children's Mental Health; FDA warnings; Antidepressant Use

Authors:

Benjamin Cook (Harvard Medical School - Cambridge Health Alliance. Psychiatry) , Colleen Barry (Yale School of Public Health. Division of Health Policy and Administration) and Susan Busch (Yale School of Public Health. Division of Health Policy and Administration)