The Effect of Massachusetts Health Care Reform on Hospital Inpatient Use

Purpose and Background.   The Commonwealth of Massachusetts implemented the first phase of a broad health care reform program in April 2006. While this reform has been widely viewed as the model for the Patient Protection and Affordable Care Act (ACA), how the ACA will affect the hospital sector is currently undefined. Covering the uninsured will increase demand for health care services, but which services and by how much are unknown. The objective of this study is to estimate the effects of the Massachusetts health care reform on the use of inpatient hospital services in Massachusetts. 

 Data and Methods. We used Healthcare Cost and Utilization Project (HCUP) State Inpatient Databases (SID) for the years 2004-2011 in 37 states including Massachusetts. The SID are a set of all-payer inpatient care databases capture the universe hospital inpatient stays in a given state, which are translated into a uniform format to facilitate multi-state comparisons and analyses. We identified a control group of hospitals from other states for each year based on their characteristics that match the hospitals in Massachusetts. More specifically, we fit a logistic regression model to calculate the predicted probability of being a Massachusetts hospital based on the number of discharges; average length of stay; average charge per discharge; teaching status; urban or rural location; patient age distribution; and patient payer mix in 2005. For every Massachusetts hospital, there were 5 non-Massachusetts hospitals within the 20% caliper of the standard deviation of the propensity scores. Hospital-specific utilization was summarized by calendar quarter and difference-in-differences time series models were estimated based on the multi-year implementation of reform initiatives in Massachusetts. We identified a pre-reform period (Q1 2004-Q3 2006), during period (Q4 2006-Q2 2007), and two post-reform periods (Q3 2007-Q1 2009 and Q2 2009-Q4 2011). Dependent variables were the quarterly estimate for each hospital of the natural logarithm of total discharges, average length of stay, the coefficient of variation in length of stay, and cost per discharge. Independent variables included annual measures of the Herfindahl-Hirschman Index, county-level measures of population, household income, unemployment rate, labor force participation rate and a dichotomous variable indicating if the hospital was in Massachusetts and the stage of policy implementation in the state. The regression models also controlled for the differential effects of the dramatic changes that occurred across the country.

 Principal Findings. Our descriptive results indicate that the number of discharges grew somewhat more rapidly in Massachusetts than in the rest of the country. Our risk adjusted results show that the full implementation of the reform legislation led to 5.8% more discharges, 5.0% shorter lengths of stay, a 2.5% reduction in the variation in a hospital’s length of stay and no change in cost per discharge – all relative to control hospitals in other policy environments.

 Conclusions. Massachusetts health care reform had a modest impact on inpatient utilization and that impact became greater the longer the reform was in place.