Effect of health insurance parity laws on healthcare coverage for alcohol-related hospitalizations in the United States

Background. A previous analysis of national hospital discharge data found substantially lower rates of healthcare coverage among patients hospitalized for alcohol-related mental and behavioral disorders (ARMBD) than those hospitalized for other conditions (Yi et al. 2003). The current study further examines the effect of state parity laws for alcohol-related treatment on reducing this gap in healthcare coverage.

Methods. State parity law data were drawn from the Alcohol Policy Information System (APIS), and hospital discharge data from the 2004 Nationwide Inpatient Sample (NIS). The NIS contains 8 million records of hospital discharges in 2004 from about 1,000 community hospitals in the U.S. ARMBD discharges were identified by ICD-9-CM diagnoses for alcoholic psychoses, alcohol dependence syndrome, and nondependent abuse of alcohol. Lack of healthcare coverage was defined by “self-pay” being the principal payment source indicated on the patient's record. State insurance parity mandates for alcohol-related treatment were coded along two dimensions: (1) must cover and (2) must offer. Logistic regression estimated the effect of parity laws on the likelihood of having to pay by self for MBDA hospitalizations adjusting for age, sex, neighborhood income and metropolitan status.

Results. Compared with discharges with non-ARMBD diagnoses, the risk for having no insurance coverage was five times as high among discharges with a principal diagnosis of ARMBD (adjusted OR=5.02, p<.01). However, this gap was reduced by about half for discharges in states with “must cover” parity provisions (adjusted OR=0.52, p<.01) and by 40% for discharges in states with “must offer” parity provisions (adjusted OR=0.59, p<.05).