Role of utilization and price in spending variation in a privately insured Texas population: 2008-2011

Background: A great deal of research has gone into understanding the inefficiencies inherent in the US health care system and policy makers have been seeking ways to control costs and improve quality. Central to these efforts has been the research on health care cost variation. Current research on spending variation has predominantly focused on Medicare and this is a limitation in its potential to affect policies. More insight on the amount of inappropriate variation may be gained by exploring spending variation levels in the commercial sector. We examine and contrast the drivers of trends and geographic variation in spending, price, and utilization, by health care market areas - Hospital Referral Regions (HRRs) or Hospital Service Areas (HSAs,) - for the populations covered by Blue Cross Blue Shield of Texas (BCBSTX), the state's largest private insurer. Methods: The study uses claims and enrollment data from 2008 through 2011 for BCBSTX members living in Texas, aged under 65, and members of PPO plans. The outcome variables are: spending per member per year, utilization metrics, and price (allowed amount per service). Spending, utilization and price are decomposed into hospital inpatient services, hospital outpatient services, professional services, and prescription drugs. We report the range, mean, median, standard deviation and coefficients of variation (standard deviation/mean) for each outcome variable. To indicate how each geographic area differs from the overall Texas mean we use maps and scatterplots reporting the standard deviation from the mean and the index of variation. Spending trends and variation in BCBSTX are decomposed into price and utilization indices. The price index is further decomposed into a cost-of-doing-business index and an ‘excess price' index. Similarly, the utilization index is decomposed into an age-sex index, a health-risk index, and an ‘excess utilization' index. Results: We found considerable variation in spending, utilization, and price at the HRR level, and even higher variation at the HSA level. Overall, growth in spending declined from 2008/09 to 2010/11. This trend masks increased price growth in 2009/10 accompanied by significant reduction in utilization growth. In 2010/11, we saw continued low spending growth, with a reduction in price growth, but a constant or increasing utilization growth. Conclusions: There is evidence of considerable inefficiencies in health care delivery in this Texas population as reflected in large spending variations. Trend analyses indicate a slowing of spending growth