Online Program

Using propensity scores and difference-in-differences methods to estimate the effects of mental health parity

Colleen Barry, Johns Hopkins Bloomberg School of Public Health 
Susan Busch, Yale School of Public Health 
Howard Goldman, University of Maryland School of Medicine 
Haiden Huskamp, Harvard Medical School 
*Elizabeth Stuart, Johns Hopkins University 

Keywords: causal inference, health care reform

This presentation describes a study design to estimate the effect of mental health parity laws. Mental illness is a major public health concern in the United States, and is associated with significant morbidity and mortality. Parity laws require that the same level of benefits provided for medical/surgical benefits also be provided for mental health benefits, which before parity had some additional financial requirements and treatment limits. This study looks at the effects of the federal parity program on mental health spending, use, and care patterns. The study uses claims data from a large national health plan and takes advantage of a change in the federal law, as well as the fact that some states already had state parity laws and so were unaffected by the federal law change. We combine two statistical methods for estimating causal effects in non-experimental studies, difference-in-differences estimation and propensity score matching. Broader potential use of the methods to investigate the effects of state or local policies will also be discussed. The newly-enacted federal insurance parity law has the potential to reduce unmet need and improve financial protection for those with mental illness; however, specific provisions of the law may hinder these goals. This study will produce new information about how key facets of this law affect utilization and spending and will help inform the implementation of federal health care reform.