Expanding Physician Owned Hospitals (POH) Creates Problems

Abstract

I was engaged by Catholic Health Initiative (CHI) beginning March 9, 2010 to examine the impact of increasing the number of physician-owned hospitals (POHs) in Nebraska. This “White Paper” represents my work to-date on this task. In the study that follows, I summarize past research investigating cost issues related to POHs, and I undertake a cursory analysis of the impact of Nebraska’s POH, the Nebraska Heart Hospital (NHH), on the cost of heart operations for the period July 2008 to June 2009. I found that the cost per patient day for all heart procedures was 39.9 percent higher at the NHH compared to all other hospitals in the region and was 45 percent higher than the average for all Nebraska hospitals. Had NHH charged the same price per patient day as other hospitals in the region, savings to patients would have been $15.9 million for the period July 2008 to June 2009.


Introduction
In economics it is theorized that an increase in supply, with no change in demand, results in lower prices and higher output. However, this is often not the case for hospitals. That is, an increase in hospital availability does not necessarily produce lower prices. In fact the opposite outcome is often observed with the addition of a hospital, or hospital beds, actually increasing hospital prices and health care costs.

It has been found that in the hospital sector, Roemer’s Law applies. That is, "in an insured population, a hospital bed built is a bed filled." This rule was produced by researcher Milton Roemer (1959), working at the UCLA School of Public Health. Roemer and colleagues found a positive correlation between per capita short-term hospital beds available and the number of hospital days used. Thus, the addition of hospital beds was found to drive up medical prices in a community.

Supporting this nexus, a 2008 study (Trinh, et al., 2008) concluded that hospital competition was associated with higher levels of duplication of inpatient, ancillary, and high-tech services. Furthermore, the researchers found that this duplication of inpatient services resulted in higher costs and higher hospital operating margins. The positive link between hospital utilization and hospital cost has been especially acute for physician-owned hospitals (POH). That is, past research has concluded that the addition of physician-owned facilities to an area, contrary to economic theory, generates higher prices.

Physician-Owned Hospitals (POHs)
In a recently published study, it was found that the entrance of POHs and limited-service hospitals to communities is associated with significant growth in total hospital volumes and total hospital spending (Lewin Group, 2004). In a related study, Mitchell (2007) found that the entry of a physician-owned orthopedic hospital between 1999 and 2004 drove up market area utilization of complex spinal fusion procedures by 121 percent. By the end of the period, Mitchell concluded that 91 percent of orthopedic procedures were performed in POHs with the residual nine percent being completed by full-service community hospitals. In addition to reducing community hospital utilization, it has been concluded that POHs generate higher costs for health care in an area. For example, an analysis by the Medicare Payment Advisory Commission (MedPAC) found that heart, orthopedic and surgical specialty hospitals had higher inpatient costs per discharge than community hospitals (Lewin Group, 2004).

Several factors account for the positive relationship between the addition of POHs and health care costs, and the negative correlation between the addition of a POH and lower utilization at full-service community hospitals.

Consumer not decision maker. Economic theory requires that for effective market operations, informed consumer decision making characterize both the input and output markets. In the market for hospital care, this is not the case since physicians have a significant amount of control over the patient’s choice of hospitals. For POHs this is especially problematic since physicians have a motivation to refer patients to facilities in which they have a stake or ownership interest. For example, Greenwald (2006) found that 80 percent of physicians were prone to refer Medicare patients to orthopedic facilities for which they had more than a five percent stake. Federal law generally prohibits physician self-referral, but grants an exception known as the “whole-hospital exemption” which allows physicians, who have an ownership interest in an entire hospital, to refer patients to hospitals where they are authorized to perform services. According to the GAO (2003), 96 percent of physician-owned limited-service hospitals that opened between 1990 and 2003 were located in states without Certificates of Need (CON) laws.

Cherry Picking. POHs treat a smaller share of Medicaid patients, or patients with high needs, but low financial resources. For example, it was found that 13 percent of community hospital admissions were Medicaid patients. However, only 3 percent of admissions to POHs were Medicaid patients (TrendWatch, p. 5). According to Cram (2005), POH patients were less likely to have co-existing conditions and were less likely to have had a heart attack prior to surgery. Mitchell (2005) discovered that POH were more apt to treat low severity cases and cases with lower comorbidities. In an analysis of eleven studies examining POHs, it was concluded by each of the studies that POHs “cherry-pick the most profitable patients.”[1]

Limited emergency rooms. By limiting the share of patients entering via the emergency room, POHs reduce their cost by avoiding patients with more severe or complicating health factors. In a report by the Office of the Inspector General, less than one-third of POHs have physicians on site at all times. Furthermore, they found that 25 percent of POHs did not have adequate written policies for managing medical emergencies and only about half of POHs have an emergency room. The study also discovered that POHs with emergency rooms were more likely to have only one bed.

Higher Margins at POHs. An analysis of Medicare cost reports in 2006 found that 57 percent of physician-owned facilities had margins above 9 percent. In contrast, only 17 percent of acute care general hospitals had margins greater than 9 percent (MedPac, 2005). It was found that steering patients to the higher margin hospitals resulted in higher medical costs. Mitchell (2005) found that physician owners of cardiac hospitals treated patients with more generous insurance and these patients incurred more profitable surgical diagnosis-related groups (DRGs).

Nebraska Examples of POHs
POH Heart Facilities. In a national study, Nallamothu (2007) discovered that the opening of a POH focusing on heart health resulted in a significant increase in usage and a doubling of the rate of coronary revascularization of Medicare beneficiaries. Confirming past research, it was concluded in an examination of a Lincoln, Nebraska POH that physicians steered patients to the physician-owned facility (McManis,2005). Additionally, it was concluded that this physician-owned facility focused on services that were reimbursed at higher rates and on patients that were the best payers. Through the referral process, it was concluded that this hospital avoided, or limited, emergency cases and tended to treat healthier patients (Nallamothu, 2007).

Table 1 (available on request) compares the hospital experience for heart procedure patients at NHH to that of all hospitals in the region and to all Nebraska hospitals for the period July 2008 to June 2009.[2] Data indicate that charges for heart procedures at the NHH were significantly higher than that at all hospitals in the region and at all Nebraska hospitals. Average cost per patient day were $14,304 for NHH, $10,228 for all hospitals in the region, and $9,867 for all Nebraska hospitals. Thus, costs at NHH were 39.8 percent higher than the average for all hospitals in the region, and 45.0 percent higher than the average for all Nebraska hospitals. If NHH had charged the average price for all hospitals in the region, patient savings for 2008-09 would have been $15,873,279. If NHH had charged the average price for all Nebraska hospital for 2008-09, patient savings for 2008-09 would have been $17,279,288.

In Table 2 (available on request) is listed the average length of hospitalization for NHH, for all hospitals in the region, and for all Nebraska hospitals. As presented, NHH’s patients were hospitalized for shorter periods of time than patients in the comparison groups. This difference may have been produced by: 1) NHH admitting and treating healthier patients that were required to remain in the hospital for a shorter period of time, and/or 2) NHH physicians discharging patients after a shorter stay. Since hospitals are generally compensated by DRG, rather than length of stay, keeping patients in the hospital for a longer period of time reduces hospital profitability. However, while differences in Table 2 support this hypothesis, they are not statistically significant at the 95 percent level of confidence but merit additional analysis to be definitive.


In addition to charging higher prices, data suggest that the NHH pulled patients away from Bryan LGH. According to the Lewin Group, full-service hospitals, such as Bryan LGH are less able to support essential but money-losing care as they lose more profitable patients to POH (Lewin, 2004). Not surprisingly, according to Bryan LGH Administration, the number of heart surgeries at their hospital dropped from 482, the year before NHH opened, to 60, the year after NHH opened. Data from Table 1 also show higher utilization at NHH.

Summary
This “White Paper” has compared past studies that analyzed the impact of POHs on health care costs. In general, research has concluded that the addition of POHs increases cost for patients in an area. Moreover, the addition of POHs to an area tends to undercut the economic viability of full-service community hospitals in the service area. By selecting patients with less severe illnesses, by limiting emergency room care, and by earning higher profit margins, POHs tend to increase medical costs in an area. The fundamental factor accounting for this relationship is that the decision of which medical facility to use is made, or influenced, by the physician, not the patient. Thus, physicians with a financial stake in the POH are motivated to send high profit margin patients to the POH. Cost data for one of Nebraska’s two POHs, NHH, support the conclusion that Nebraska’s POHs charge higher prices for medical procedures than community hospitals.[1]


References
American Hospital Association, “Self-referral to Physician-owned Hospitals,” April 17, 2008.

Cram P, et al. “Cardiac Revascularization in Specialty and General Hospitals,” New England Journal of Medicine, April 7, 2005, Vol. 352(14), pp. 1454-1462.

Government Accountability Office. (April 2003). Specialty Hospitals: Information on National Market Share, Physician Ownership, and Patients Served. Washington, DC.

Greenwald, Leslie, et al. 2006. “Specialty versus Community Hospitals: Referrals, Quality, and Community Benefits,”. Health Affairs, 25(1), 106-118.

Lewin Group (2004). “TrendWatch: Impact o Limited-service Providers on Communities and Full-service Hospitals, Sept. 2004, Vol. 6(2).

McManis Consulting, 2005. “Impact of Physician-owned Limited-service Hospitals,”

MedPac. 2005. “Report to the Congress: Physician-Owned Specialty Hospitals,” Medicare Payment Advisory Commission, Washington, DC.

Mitchell, J.M. (2005). “Effects of Physician-owned Limited-service Hospitals: Evidence from the Market for Cardiac Inpatient Care in Arizona,” Health Affairs Web Exclusive, Oct. 25, 2005.

Mitchell, J.M. (2007). “Utilization Changes Following Market Entry by Physician-owned Specialty Hospitals,” Medical Care Research and Review, 64(4), 395-415.

Nallamothu, BK, et al. “Opening of Specialty Cardiac Hospitals and Use of Coronary Revascularization in Medicare Beneficiaries,” Journal of the American Medical Association, March 7, 2007, Vol. 297(9), pp. 962-968.

Office of the Inspector General, “Physician-Owed Specialty Hospitals’ Ability to Manage Medical Emergencies,” http://oig.hhs.gov/oei/reports/oei-02-06-00310.pdf Department of Health and Human Services, January 2008.

Roemer, Milton. “Hospital costs relate to the supply of beds,” Modern Hospital. 1959 Apr;92(4):71-3.

TrendWatch, “Physician Ownership and Self-referral in Hospitals: Research on Negative Effects Grows,” American Hospital Association, April 2008.

Trinh, Hanh Q, Begun, James W and Luke, Roice D. “Hospital service duplication: Evidence on the medical arms race,” Health Care Management Review 33, no. 3 (Jul-Sep 2008): p. 192.

[1]An examination of spinal fusion operations at Nebraska’s other POH, the Nebraska Orthopedic Hospital, shows the same relationship as presented in Tables 1 and 2. That is, for the POH, cost- per-patient day were generally higher and the length of hospitalization was shorter than for community hospitals.

[1]The eleven studies were 1) MedPAC, 2) McManis: Black Hills, SD Case Study, 3) McManis: Lincoln, NE Case Study, 4) McManis: Oklahoma, OK Case Study, 5) McManis: Wichita, KS, Case Study, 6) TrendWatch, 7) Government Accountability Office, 8) Mitchell: Oklahoma City, 9) NEJM: Cran et al. , 10) Office of the Inspector General, 11) JAMA: Naltanothu et al.

Ernie Goss