Medicare's Bending Cost Curve. July 28, 2014. V. Summary of Service Category Findings


While these finding are not definitive, they offer suggestive information about the factors that may have affected recent Medicare spending trends. As noted below, in many cases, multiple factors may have contributed to trends for individual service categories.

Drugs and Imaging: This has taken the form of patent expirations for existing blockbuster drugs and accompanying increased generic uptake, and fewer blockbuster drugs being introduced into the market in recent years.13 It is also possible that the anticipation or actual implementation of payment and delivery models that hold providers accountable for costs has resulted in more cost effective use of medical technology. Medicare spending on imaging may have also been influenced by nationwide factors leading to less use of imaging services, including coverage trends in the private sector that had spillover effects in physician practice behavior related to treating Medicare patients.14 The decline in imaging rates may in part be due to a provision in the Deficit Reduction Act that capped payment rates for imaging studies conducted at freestanding imaging centers and physician offices at the hospital outpatient rate.

Table 3

Percentage of Reduction in Medicare Spending Growth between 2000-2008 and 2009-2012

Attributable to Price and Utilization, by Service Category

  Percent Attributable to:
Service Category Price* Utilization
Part D --- ---
HOPD 0% 100%
Acute Inpatient 0% 100%
Other Inpatient 88% 12%
SNF 21% 79%
Hospice 10% 90%
Home Health --- ---
DME 37% 63%
Part B Drug 19% 81%
Physician E&M 0% 100%
Other Part B 100% 0%
Anesthesia^ --- ---
Dialysis 100% 0%
Testing 27% 73%
Other Procedures 0% 100%
Imaging 22% 78%
ASC 0% 100%

* Includes both changes in fee schedule amounts and case mix.

† The percentages for Part D drugs are not calculated, since Part D was introduced in 2006 with a ramp up in enrollment (and therefore utilization) in 2007 making it difficult to identify non-enrollment-related utilization growth trends.

‡ The percentages for home health are not calculated, since the Summary File

does not include episodes of care, which serves as the basis for payment.

^ The contribution of price and utilization for anesthesia services is not calculated, since the service category experienced an increase in its growth rate during this period.

Data Source: CMS Medicare Beneficiary Summary File

Inpatient Hospital and Post-Acute: The trends in inpatient and post-acute care services may reflect a number of factors. The rate of hospital admissions per beneficiary decreased towards the end of the decade (data not shown in table). CBO found that the average annual rate of surgical discharges declined in the latter part of the 2000s after having grown slightly in earlier years.15 There has been an ongoing shift in the site of surgical care from inpatient to lower cost outpatient settings, such as Ambulatory Surgical Centers, reflecting innovations in medical procedures and payment incentives.16 This shift has also affected post-acute care, reducing the need to use skilled nursing facilities. Other types of discharges also declined. Part of this was due to the rising share of beneficiaries who are young and do not require as much inpatient hospital care. Finally, the rate of readmissions for beneficiaries 65 years old or greater began to decline in 2012. The Hospital Readmission Reduction Program was implemented in October of 2012, and a number of other programs have been implemented with a focus on improving care coordination and reducing readmissions. The readmission rate has declined significantly, leading to the avoidance of 150,000 readmissions in 2012 and 2013.17

Home Health: Trends in home health spending were likely directly affected by specific Medicare policy initiatives. Although strongly suggestive, these maps are only provided for illustrative purposes and are not intended to demonstrative fully causul connections. Maps 1 and 2 display the ratio of 2009-2012 to 2000-2008 Medicare per enrollee spending growth rates by hospital referral region for Home Health and DME spending. Map 1 (Home Health) shows the locations of Medicare Fraud Strike Force teams, which have been in existence since 2007. These teams are placed within, or adjacent to, key locations with historically high rates of Medicare fraud. As the map illustrates, these locations have witnessed some of the largest declines in Home Health spending growth in recent years as the Medicare program continues to reinforce its program integrity efforts. These teams represent one of a number of program integrity initiatives the Department has undertaken in recent years, which were further strengthened by the Affordable Care Act. Hence, although certain regions exhibiting large drops in spending growth on home health services such as Nevada, Utah, Oklahoma, and New Mexico do not have Fraud Strike Force teams, they have benefitted from implementaton of other program integrity tools such as enhanced provider screening.

Durable Medical Equipment (DME): Trends in durable medical equipment (DME) spending were also directly affected by specific Medicare policy initiatives. The DME map shows the locations where the first round of competitive bidding for durable medical equipment took place, starting in 2011. Again, given the declines in DME spending growth seen in recent years in all but one of these locations, these data suggest that the competitive bidding program was successful in reducing prices and inappropriate utilization. Note that spending for these services was also affected by provisions of the Deficit Reduction Act of 2005 that reduced payment rates, and DME services were also a focus of Medicare Fraud Strike Force activities described above.

Hospice: A number of factors also likely affected hospice spending. There was substantial growth in this industry during the early 2000s attributable to both the number of beneficiaries choosing hospice for end of life care and the number of patients remaining in hospice care for more than 6 months. However, the percent of hospice episodes greater than 180 days, which increased from 6.5 percent in 1999 to 12.7 percent in 2009, has remained relatively constant from 2010-2012. The leveling of the trend toward long stays probably contributed to the slowdown in spending growth as well as other factors including the face-to-face encounter required by the ACA for extending hospice election.

13 Katie Thomas, ‘Use of Generics Produces an Unusual Drop in Drug Spending’, The New York Times, 18 March 2013, section Business Day < [accessed 15 November 2013].

14 David W. Lee and Frank Levy, ‘The Sharp Slowdown In Growth Of Medical Imaging: An Early Analysis Suggests Combination Of Policies Was The Cause’, Health Affairs, 31 (2012), 1876–84 <>.

15 Levine and Buntin.

16 Elizabeth L. Munnich and Stephen T. Parente, ‘Procedures Take Less Time At Ambulatory Surgery Centers, Keeping Costs Down And Ability To Meet Demand Up’, Health Affairs, 33 (2014), 764–69 <>.

17New HHS Data Shows Major Strides Made in Patient Safety, Leading to Improved Care and Savings (Baltimore, MD: Centers for Medicare & Medicaid Services, 7 May 2014) <> [accessed 27 July 2014].

View full report


"ib_medicost.pdf" (pdf, 674.63Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®