In recent years there has been a great deal of attention on the variation in service volume across providers and regions. Our suite of tools, collectively entitled Potentially Preventable Events (PPE), has led our research group to engage with a variety of stakeholders in their efforts to minimize volume variation. This variation typically results from inefficiencies, poor quality of care leading to the use of otherwise unnecessary services or the overuse of services resulting from practice pattern. Volume is a sensible target for cost reduction efforts and, when detailed as variation across peer providers or regions, is hard to justify. Price (transaction price) comparisons are more complex but, arguably, have greater bearing on total U.S. healthcare cost. At least this is what we are told each year by the policy folks at the Organisation for Economic Co-operation and Development (OECD).
The U.S. appears to have made headway recently in stabilizing growth in healthcare spending, but there is real concern that price increases are once again looming on the horizon. Constraints placed upon Medicare and Medicaid spending have been cited by the American Hospital Association (AHA) as resulting in underpayments from both Medicare (88% paid) and Medicaid (90% paid)—gaps that will require productivity gains or cost shifting to private payers, the latter being a pattern identified by some researchers¹.
The data tell us that national health expenditures (NHE) stabilized as a percentage (17.4%) of GDP in 2009-2013, a recessionary period in which both GDP and NHE growth fell. This is an unsurprising correlation since NHE is both limited by GDP and contributes to its level. Recently the Center for Sustainable Health Spending provided analysis of 2014 spending and concluded that NHE inflation is gaining momentum at a rate of 5% (from 3.6% in 2013), climbing to an estimated 17.8% of GDP in December 2014.
What is of particular interest when we look at NHE is the percentage attributed to hospital spending (see graph). While below its peak from the early 1980s, the share of NHE attributed to hospitals is now at the highest level since 1996 and rising. This runs counter to the perception of falling inpatient admissions, efforts to contain cost through bundling and episodes, and initiatives to reduce readmissions and complications. It does however fit with two other trends being observed in the hospital sector. The first is ongoing consolidation of hospitals, which leads to the leveraging of higher unit prices in negotiation. This issue has been of increasing concern to the FTC and helps explain why hospital revenue is increasing while admissions are decreasing. A second is that outpatient hospital revenue has grown to offset reduced inpatient revenue.
Source: AHA Statistics 2013 Table 3.
There is increasing evidence of growth in hospital-based outpatient revenue. For example, visits have risen and “gross” revenue has grown as a share of total revenue (see table). Gross revenue relates to charges rather than reimbursement received—it should be cautioned that charges are particularly susceptible to manipulation in order to maximize percent of charge contract terms favored in many commercial insurance outpatient contracts².
Contributing to the relative rise of outpatient revenue is the ongoing shift of inpatient procedures to the outpatient domain with attendant increases in the use of observation status. Aspects of this shift, such as the push for 1-day stays to be classified outpatient under the two-midnight rule, have received considerable attention. Another driver is the acquisition of community-based providers by hospitals leading to increases in reimbursement due to site of service differentials. These differentials are the subject of heated debate, with MedPAC recommending—and the AHA opposing–equalization (site neutrality). A number of physician groups support the concept of neutrality and are members of the Alliance for Site Neutral Payment Reform. This issue is only likely to grow as patients are increasingly caught between paying higher out-of-pocket costs for services that can be delivered in more cost-effective settings or failing to meet eligibility requirements for classification as an inpatient admission despite being retained over an extended period in a hospital bed.
When price variation results from differences in local input prices there is not a lot to be done, at least in the short term, and we might as well move on. Price variation resulting from differences in mission- related cost should be the subject of reform as we discussed in a recent blog; but reform would likely generate longer term rationality in payment as opposed to short term gains in value. What we should be able to address more competently are price differences induced by site of service alone.
Site-of-service differentials will not be fixed by volume constraints – nobody is arguing that the service is unnecessary – but requires that more attention be placed on, for example, defining when cases are too complex for treatment in a physician’s office or ambulatory surgical center. Or conversely when a case is suitable for treatment in an outpatient hospital unit rather than requiring the additional overhead associated with inpatient stays. These determinations hold the potential for vast savings through lowering structural cost and payment over time. However, right now, we are only able to make general claims as to the suitability of treatment site.
To help reduce price variation induced by service site we need to improve the classification of patient complexity, particularly in ambulatory care where increasingly complex surgery is being done that had, only a few years ago, been performed strictly on an inpatient basis. Improved case mix measurement needs to be matched with commensurate reform in payment policy that diminishes the payment differential between inpatient and outpatient services, particularly surgical interventions. Lastly, payment reform needs to recognize that, at least in the short term, improvement in case mix will not be sufficient to obviate the need for low and high outlier policies. If we are ever going to control healthcare costs we need to stop dreaming that volume reduction in isolation will save us and recognize that site of service and unit prices are a major part of the solution.
Richard Fuller, MS, is an economist with 3M Clinical and Economic Research.
¹Franzini L, Taychakhoonavudh S, Parikh R, White C. Medicare and private spending trends from 2008 to 2012 diverge in Texas. Med Care Res Rev. 2015;72(1):96–112. doi:10.1177/1077558714563174.
²Fuller RL. An analysis of real price effects resulting from charge setting practices in the US hospital sector.; 2014. Available at: http://jktgfoundation.org/data/price setting practices in the US hospital sector_FINAL.pdf.