WASHINGTON, Dec. 17 /PRNewswire/ -- The cost to Medicare of a patient receiving treatment at a physician owned and operated hospitals is "4.6 to 6.00% lower" than hospitals not run by physicians, according to a new study by Oxford Outcomes, an internationally respected research agency.
Medicare savings projected by the Congressional Budget Office (CBO) has been given as the primary reason for efforts to limit growth and eventually force physician owned hospitals to close.
"Every other accusation against over 200 of America's best and safest hospitals has either been withdrawn or discredited," said Molly Sandvig, JD, executive director of Physician Hospitals of America (PHA). "All that is left now putting tens of thousands of people out of work and further injuring local economies if Congress does not remove the attack on physician owned hospitals contained in the health care bill."
The new study and analysis was directed by Dr. John Schneider and shows that while moving against physician owned hospitals puts billions of dollars and thousands of jobs at risk, there will be no savings to Medicare, despite earlier CBO estimates. In fact, Medicare pays less to physician hospitals than to non-physician owned hospitals for the same treatments.
According to Schneider, "The estimated savings from limiting physician owned hospitals (POH) growth are overstated for a number of reasons, including the fact that the costs of providing care in a POH is less than the costs of providing care in non-POHs. "
Among Dr. Schneider's findings:
Both the U.S. House passed health care reform bill and the bill under consideration in the U.S. Senate already spell economic trouble for nearly all of the state and local economies where hospitals owned by physicians are located. The economic impact is placed at a loss of well over $5 billion and could eventually cost up to 100,000 jobs.
"In allowing the attack on physician owned hospitals to remain in the health care bill, nothing is gained, the only results are incredibly negative economic impact, and loss of healthcare access," said Sandvig.
PHA is asking House and Senate members to take a second look at the effort to undermine physician owned hospitals since the core economic argument, projected Medicare cost savings, has vanished. "When members of the United States Congress understand the senseless chaos these provisions will cause in many communities in 37 states, with absolutely no benefit to Medicare, we are confident they will be changed," added Sandvig.
Physician Hospitals of America (PHA) is the national organization representing more than 220 hospitals owned and operated by reform-minded physicians themselves. PHA is on the web at: www.physicianhospitals.org.
The Oxford Outcome study by Dr. John Schneider is available for download as a .pdf at: http://www.physicianhospitals.com/var/files/member_updates/member_updates188289.pdf
He may be reached at:
ISSUE BRIEF | December 15, 2009
Restrictions on Physician-Owned Hospital Growth:
Comparing Medicare Expenditures on POHs and Non-POHs
The health reform bill which has passed in the U.S. House and is currently being debated by the U.S. Senate, includes language that places limits on the growth of physician-owned hospitals (POHs).(1) CBO analyses of the various health reform proposals have assumed that these limitations will result in overall Medicare expenditure savings over a 10-year period.
In this Issue Brief we argue that the estimated savings from limiting POH growth are overstated because the costs of providing care in POHs is less than the costs of providing care in non-POHs. The study reaches the following conclusions:
In an earlier study we have shown that Medicare expenditures do not vary according to whether a community has a POH.(2) The study examined the effect of POHs on Medicare per enrollee expenditures at the metropolitan area (MSA) level nationwide, spanning the eight-year time period from 1998 to 2005. The study used fixed-effects panel data estimation with instrumental variables to account for the bias introduced by endogenous POH market entry (i.e., POHs may be more likely to open in high-growth / high-demand markets with high levels of Medicare per enrollee expenditures).
After controlling for other variables that are likely to affect expenditures (especially the age and sex distribution of the MSA), we find no association between POH presence and Medicare expenditures per enrollee at the MSA level. The results are robust to changes in model specification, estimation technique and definition of geographic market. These findings suggest that the "demand inducement" aspects of physician ownership of acute care hospitals (if any) has no meaningful impact on market-level Medicare expenditures per enrollee. We argued based on these findings that current policies based on an assumption that POHs are associated with significant increases in total expenditures may need to be reassessed.
Based on these statistical findings, it seems unlikely that restrictions on POHs have the ability to generate Medicare savings. Underlying these findings, an important question is whether adjusted DRG payment rates are higher or lower for POHs relative to their non-POH counterparts. Given the structure of the current Medicare inpatient prospective payment system, the main ways in which DRG per-case net payments vary are by the relative magnitude of the adjustments.(3)
For example, the typical urban teaching hospital receives upward price adjustments for indirect medical education (IME), Medicaid disproportionate share status (DSH), and "outlier" cases (i.e., cases that are exceptionally complex and have a large number of comorbid conditions).(4) Medicare's rate setting process is described in the Appendix.
In the current study, we analyze a small number of high-volume DRGs common to POHs and non-POHs. We calculate mean allowed severity-adjusted Medicare charges for POHs and non-POHs in the same metropolitan area. Our results confirm the expected finding that mean allowed charges by POHs are consistently lower than allowed charges of non-POHs in the same community.
There are two parts to the analysis. In Part I, we calculate the median difference between all MS-DRGs(5) that we were performed in POHs and non-POHs in the same metropolitan area. There were 67 MS-DRGs that met these criteria.
In Part II, we selected six relatively high-volume MS-DRGs representing cardiac and orthopedic care-- services that are common to POHs and non-POHs. These focal MS-DRGs are shown in Table 1.
TABLE 1 - MS-DRGs Selected for Analysis
MS-DRG Description ------ ----------- 291 Heart failure & shock w/ MCC 292 Heart failure & shock w/ CC 293 Heart failure & shock w/o CC 460 Spinal fusion except cervical w/o MCC 470 Major joint replacement, lower ext. w/o CC 491 Back & neck proc w/o spinal fusion w/o MCC --- ------------------------------------------
For both parts, data on median allowed Medicare charges were retrieved from the dataset version of CMS's Hospital Compare data. (6) The use of MS-DRGs helps reduce the effects of case mix severity, which can impact the cost of hospital care. These data were retrieved for all hospitals in markets (defined as metropolitan statistical areas, or MSAs) with at least one POH. This enabled the direct comparison of allowed charges between POHs and non-POHs in the same market.
The results of Part I were that, for the 67 MS-DRGs, the median difference in allowed Medicare charges was $734.19 lower. Put differently, the "typical" patient treated in a POH will have allowed charges $734.19 lower than the same case treated in a non-POH.(7) This represents a 4.6% difference between POHs and non-POHs.
The results of Part II, which focuses on the six MS-DRGs, show similar differences between POHs and non-POHs. For all six MS-DRGs, median allowed charges for POHs were significantly less than median allowed charges by non-POHs in the same MSA. The magnitude of the significant differences ranged from 3.6% to 11.1% (Table 2). These differences were statistically significant for four of the six MS-DRGs (one of the cardiac codes and all three orthopedic codes).
TABLE 2 - Differences in Medicare Allowed Charges, POH v. Non-POH
MS-DRG Non-POH POH Diff. % Diff. ------ ------- --- ----- ------- 291 $7,875 $7,716 -$159 -2.0% 292 $6,148 $5,926 -$222 -3.6%* 293 $5,209 $5,193 -$16 -0.3% 460 $21,162 $18,892 -$2,270 -10.7%* 470 $11,751 $10,730 -$1,021 -8.7%* 491 $6,098 $5,424 -$674 -11.1%* --- ------ ------ ----- -------
Note: * = difference is statistically significant at p = .05
These results confirm that on a per-case basis, adjusted for severity, the Medicare program pays less for care provided in POHs as compared to payments to non-POHs in the same community. The average difference, across all six MS-DRGs, is approximately 6%.
This analysis is consistent with our earlier econometric study, and serves to further illustrate the importance of non-trivial differences in per-case allowed charges. Multiplied over the total national number of cases in each of these MS-DRGs, the cost implications of moving POH cases to non-POH settings would be substantial.
In this Issue Brief we argue that the estimated savings from limiting POH growth are overstated because the costs of providing care in POHs is less than the costs of providing care in non-POHs. The study reaches two conclusions. First, for the same types of inpatient cases receiving the same treatment, a patient treated in a POH will cost the Medicare program on average $734 (4.6%) less than a similar patient receiving the same treatment in a non-POH. Second, for a subset of high-volume cardiac and orthopedic procedures performed at POHs and non-POHs in the same community, allowed Medicare charges for POHs were 6% less than allowed charges for nearby non-POHs
Summary of Medicare Rate Setting Process(8)
Step 1 - Hospitals submit a bill for each Medicare patient they treat to their Medicare fiscal intermediary (a private insurance company that contracts with Medicare to carry out the operational functions of the Medicare program). Based on the information provided on the bill, the case is categorized into a Medicare Severity Diagnosis Related Group (MS-DRG), which determines how much payment the hospital receives.
Step 2 - The base payment rate is comprised of a standardized amount, which is divided into a labor-related and non-labor share. The labor-related share is adjusted by the wage index applicable to the area where the hospital is located and if the hospital is located in Alaska or Hawaii, the non-labor share is adjusted by a cost of living adjustment factor. This base payment rate is multiplied by the MS-DRG relative weight.
Step 3- If the hospital is recognized as serving a disproportionate share of low-income patients (DSH), it receives a percentage add-on for each case paid through the PPS. This percentage varies depending on several factors, including the percentage of low-income patients served. It is applied to the MS-DRG-adjusted base payment rate, plus any outlier payments received.
Step 4- If the hospital is an approved teaching hospital it receives a percentage add-on payment for indirect medical education (IME) for each case paid through the PPS. This percentage varies depending on the ratio of residents-to-beds.
Step 5- A hospital can receive an additional payment for those cases that include technologies that meet the new technology add-on payment criteria.
Step 6- Next, the costs incurred by the hospital for the case are evaluated to determine whether it is eligible for an additional payment as an outlier case. This additional payment is designed to protect the hospital from large financial losses due to unusually expensive cases. Any outlier payment due is added onto the MS-DRG-adjusted base payment rate.
ABOUT OXFORD OUTCOMES
Oxford Outcomes, Inc. is an international outcomes research agency specializing in health economics, health outcomes research, and policy analysis. Founded in 1997, Oxford Outcomes has provided consulting services to pharmaceuticals, medical devices, professional associations, hospitals and health systems, and all levels of government in the U.S., Canada, and the United Kingdom. The Health Economics practice consists of leading experts based in Morristown NJ, Vancouver BC, Toronto ON, and Oxford UK. Many of our experts are academic based, and the majority of Oxford Outcomes research projects result in publications in peer-reviewed medical, economics, and policy journals.
(1) Refer to TITLE IV--TRANSPARENCY AND PROGRAM INTEGRITY Subtitle A--Limitation on Medicare Exception to the Prohibition on Certain Physician Referrals for Hospitals
(2) J.E. Schneider et al., "The Effects of Physician-Owned Hospitals on Medicare Expenditures Per Capita: An Instrumental Variable Approach to Endogenous Market Entry," Contemporary Economic Policy In Press (2009).
(3) Payment levels are also adjusted for local wages (via the wage index) and overall patient acuity (via the case-mix index). For the purposes of this simple model, we ignore these because they are based on normalized indices as opposed to "add ons."
(4) There are a number of studies that have documented these differences. See generally GAO, "Comparative Analyses of Payments for Selected Hospital Services," in Report to the Subcommittee on Health, Committee on Ways and Means, House of Representatives (Washington, DC: U.S. General Accounting Office, 1990), S. Nicholson, "The Effects of Medicare Payment Subsidies to Teaching Hospitals," Leonard Davis Institute of Health Economics Issue Brief 7, no. 4 (2002).
(5) According to CMS, "Patients that have similar clinical characteristics and similar costs are assigned to an MS-DRG. The MS-DRG will be associated with a fixed payment amount based on the average cost of patients in the group. Patients are assigned to a MS-DRG based on diagnosis, surgical procedures, age and other information. Medicare uses this information that is provided by hospitals on their bill to decide how much they should be paid."
(6) Median Medicare payments for the same MS-DRG refers to the midpoint of all payments to the hospital for a particular MS-DRG (that is, half the payments were lower and half the payments were higher than the median payment).
(7) Calculation of median differences is weighted by caseload volume within MS-DRG. Differences are significant at p </= 0.05
(8) Adapted from www.cms.hhs.gov/acuteinpatientpps/02_stepspps.asp
Available Topic Expert(s): For information on the listed expert(s), click appropriate link.
Molly Sandvig, JD
SOURCE Physician Hospitals of America
|SOURCE Physician Hospitals of America|
Copyright©2009 PR Newswire.
All rights reserved