A New Analysis of Spending Growth Rate Declines

In recent years researchers at the Bureau of Economic Analysis have begun to study health spending more carefully and to create new data sources and analytic approaches to understand trends in this spending.   (HA Article)   An article in the current Health Affairs describes the Bureau’s findings in regard to the health spending growth slowdown that occurred during the period 2000 to 2010.  The core of BEA’s work is the creation of the Health Care Satellite Accounts which are designed to facilitate health spending by medical condition.  From a national perspective, total health spending, and its year-to-year changes, are largely driven by the number of people, the number of medical conditions these people need treated, and the unit costs of the various services and goods utilized in providing that treatment.  Unfortunately, in our lifetimes, national health spending has never gone down, either in aggregate or on a per capita basis. From 2000 to 2005 the annual per capita growth rate was 6.8% and from 2005 to 2010 it was 4.5%, so there was a slowing in spending.  Medical care in the BEA method is assigned to one of 263 Clinical Classification Software groups.  Claims data and survey data was then blended to examine trends in prevalence and costs per case for each condition.  BEA researchers found that the spending growth decline was exclusively due to changes in cost per case and not to a lower prevalence of disease; i.e., fewer cases.  In fact, annual prevalence rates actually grew slightly in the second half of the study period, particularly for genitourinary conditions, infectious diseases, skin conditions, and nervous system conditions  Somewhat dismayingly, prevalence for routine care declined by 2.7%.

They also found that the rate and cause of spending increases varied significantly across conditions.  For example, the circulatory conditions disease category, which is quite large as it includes hypertension and heart disease, experienced a 3.2% decline in growth rate between the two time periods, which represents over half the total slowdown of 2.3%.  Endocrine systems diseases, such as diabetes and high cholesterol, saw a 6.5% lower growth rate in the second time period and routine care growth fell by 5.2%.  But these major contributors to slower growth were offset to some extent by other categories, like nervous system conditions which actually had spending accelerations in the second half of the decade.  The five largest categories in 2010 were circulatory conditions, routine care, musculoskeletal, respiratory and endocrine systems, which collectively accounted for about 50% of total spending.  It is notable that in a number of these categories the change from branded to generic drugs probably contributed significantly to lower costs per case.  The cost per case growth rate decreased most significantly for circulatory conditions, endocrine system diseases, and genitourinary conditions, all of which have medications as a significant treatment component.  None of the most common condition categories experienced an increase in cost per case in the second half of the treatment period.  The researchers suggest that the lower spending growth per case is largely due to less utilization per case and switching to lower cost treatments in the case, not absolute declines in unit prices for the same treatment.  This study confirms that utilization is less the cause of our health spending issues than unit costs are.