Chain Reaction

This article was downloaded by: [Chongqing University] On: 24 March 2014, At: 15:34 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Aging & Social Policy Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/wasp20

Chain Reaction a

b

Martin Kitchener PhD , Ciaran O'Neill PhD & Charlene Harrington PhD

c

a

Department of Social and Behavioral Sciences , University of California , San Francisco, USA b

University of Ulster , U.K.

c

Department of Social and Behavioral Sciences , University of California , San Francisco, USA Published online: 11 Oct 2008.

To cite this article: Martin Kitchener PhD , Ciaran O'Neill PhD & Charlene Harrington PhD (2005) Chain Reaction, Journal of Aging & Social Policy, 17:4, 19-35, DOI: 10.1300/J031v17n04_02 To link to this article: http://dx.doi.org/10.1300/J031v17n04_02

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.

Downloaded by [Chongqing University] at 15:34 24 March 2014

This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions


GENERAL ARTICLE

Chain Reaction: An Exploratory Study of Nursing Home Bankruptcy in California Martin Kitchener, PhD University of California, San Francisco

Ciaran O’Neill, PhD University of Ulster, U.K.

Charlene Harrington, PhD University of California, San Francisco Martin Kitchener is Associate Adjunct Professor, Department of Social and Behavioral Sciences, University of California, San Francisco. His research interests include health care organization, policy, and long-term care. Ciaran O’Neill is Professor, University of Ulster, U.K. His research concentrates on health economics and policy. Charlene Harrington is Professor, Department of Social and Behavioral Sciences, University of California, San Francisco. Her research interests include health policy and long-term care for the elderly and disabled. Address correspondance to: Dr. Martin Kitchener, Department of Social and Behavioral Sciences, University of California, San Francisco, 3333 California Street, Suite 455, San Francisco, CA 94118 (E-mail: [email protected]). This research was funded by the California Research Bureau’s Contract Research Fund, Grant #L-1821, at the request of Senate Rules Committee. The views expressed in the paper are those of the authors and do not necessarily reflect those of the California Research Bureau or the California State Senate. Journal of Aging & Social Policy, Vol. 17(4) 2005 Available online at http://www.haworthpress.com/web/JASP  2005 by The Haworth Press, Inc. All rights reserved. doi:10.1300/J031v17n04_02

19


20

JOURNAL OF AGING & SOCIAL POLICY

ABSTRACT. This paper reports on an exploratory study of nursing home bankruptcy. From state and industry data regarding nearly 1,000 California facilities, it was possible to identify 155 homes in five chains (multi-facility organizations) that were operating in bankruptcy in 2000. When compared with facilities in non-bankrupt chains, while the bankrupt chain facilities had significantly worse financial liquidity, higher administrative costs, and higher payables to related parties, they also had more Medicare residents, fewer Medicaid residents, better solvency, and were located in less competitive county markets and in areas with higher Medicaid reimbursement rates. These findings indicate that, rather than facility characteristics and local market factors, strategic decisions taken at the corporate (chain) level are the major determinants of nursing facility bankruptcy status. [Article copies available for a fee from The Haworth Document Delivery Service: 1-800-HAWORTH. E-mail address: Website: © 2005 by The Haworth Press, Inc. All rights reserved.]

KEYWORDS. Nursing homes, chains, bankruptcy, policy

INTRODUCTION In 2001, there was mounting concern over reports that 1,800 of the nation’s 16,500 nursing homes were operating in bankruptcy while restructuring under protection of Chapter 11 of the U.S. Bankruptcy Code (American Health Care Association [AHCA], 2001). In contrast to the knowledge of hospital failures derived from national datasets and sophisticated models (Lee & Alexander, 1999), fewer data and limited research presented policymakers with an inadequate basis from which to assess the nature or scope of nursing home bankruptcy (Centers for Medicare and Medicaid Services [CMS], 2002a). At the national level, the nursing home industry depicted a “rising tide” of bankruptcy, attributed it to the 1998 introduction of the Medicare Nursing Facility Prospective Payment System (PPS) and demanded higher reimbursement rates to ensure facility survival and quality (Appleby, 1999; AHCA, 2001). By contrast, the U.S. General Accounting Office (GAO, 2000) ascribed the bankruptcy of five of the largest chains (multi-facility organizations) to factors including poor business decisions. In California, as in other states, a lack of research evidence left perceptions to be shaped by providers’ framing of the issue and media coverage of a small number of sudden facility closures involving: the


Kitchener, O’Neill, and Harrington

21

traumatic transfer of frail residents, the state’s incurring temporary administration costs of $2m (California Department of Health Services [CDHS], 2002a), and a review of state oversight procedures (Bonnet, 2001). The primary aim of this study is to provide a foundation of evidence about nursing home bankruptcy for policymakers, the public, and future researchers. In the absence of comparable national data on facility bankruptcy, this exploratory study had two main goals: (1) to identify the scale and nature of nursing home bankruptcy in California, and (2) to use the state’s rich facility-level dataset to identify characteristics of bankrupt homes that might provide early warning signals to regulators, and which could be further examined (e.g., within predictive models) if national data become available. The study is reported in the four main sections of this paper: (1) conceptual framework, (2) research design and methods, (3) findings, and (4) discussion of findings with reference to national developments, areas for future research, and policy implications. CONCEPTUAL FRAMEWORK Because of the dearth of research into nursing home bankruptcy, the conceptual model for this study was derived from nursing home cost studies (Aaronson et al., 1995) and analyses of hospital bankruptcies and closures (Wertheim & Lynn, 1993). While some variables employed in studies of hospital failure (e.g., leadership tenure) are not available for nursing homes, two key findings from these literatures were incorporated into our framework. First, facility characteristics (e.g., size) and market factors were incorporated (Lee & Alexander, 1999). Second, because much of the variation contained within facility financial data is explained by single ratios of leverage, profitability, and liquidity (Zeller et al., 1997), a single ratio of each characteristic was included in the framework. Facility Characteristics Following nursing home cost studies and hospital failure research, facility characteristics were predicted to be associated with facility costs and hence, with the likelihood of bankruptcy (Wertheim & Lynn, 1993).


22


Number of Beds (Size). Hospital studies suggest that as the relative number of resources increases, facilities are better placed to avoid bankruptcy (Cleverly, 1985). Similarly, nursing home analyses report associations between size and outcomes, including: higher net patient revenues (HCIA & Arthur Andersen, 1999), better facility financial status (Cohen & Spector, l996), and fewer closures (Angelelli et al., 2003). California nursing home cost reports for 1999 show that facilities with 1-59 beds had an average loss of income per patient day of $4.86, compared with earnings of $1.82 per patient day for facilities with 60-99 beds, and earnings of $2.55 per patient day for facilities with 100 or more beds (California Office of Statewide Health Planning and Development [COSHPD], 2002). Occupancy Rates. Nursing home studies report that facilities with lower occupancy rates have higher average patient costs (Ullmann, 1984) and are more likely to close because of poor quality of care (Angelelli et al., 2003). Hospital studies also suggest that low occupancy affects facility financial position and hence, the likelihood of bankruptcy (Wertheim & Lynn, 1993). Chain Membership. Cost studies of hospitals and nursing homes report that chain facilities have generally lower operating costs (Arling et al., 1991; McKay, 1991) that could lead to better financial status, and hence reduce the likelihood of bankruptcy. Ownership Type. While nursing home ownership type (e.g., forprofit, not-for-profit) is associated with facility outcomes including licensing termination (Angelelli et al., 2003) and closure (Kitchener et al., 2004), non-profit and government facilities are omitted from this study because they are not eligible to declare Chapter 11 bankruptcy. Revenue and Cost Factors Medicare and Private Pay Residents. Because Medicare skilled nursing facility (SNF) reimbursement rates are set at the federal level and are higher than most state Medicaid rates (Swan et al., 2001), facilities with higher percentages of Medicare residents may have higher revenues and net incomes. While higher percentages of Medicare residents might once have reduced the likelihood of bankruptcy, following Medicare PPS in 1998, the average Medicare reimbursement rate for freestanding facilities declined from $305 in 1997 to $240 in 1999. This may have increased the risk of bankruptcy among facilities with higher percentages of Medicare residents (AHCA, 2001).



23

Medicaid Residents. Studies have identified a positive relationship between percentage of Medicaid residents and factors such as costs per day (Harrington et al., 1998) and the likelihood of closure (Angelelli et al., 2003). California facilities with higher levels of Medicaid residents may be at greater risk of bankruptcy because, as noted earlier, Medicaid reimbursement rates are lower than Medicare and private pay rates (COSHPD, 2002). Reimbursement Rates and Geographic Regions. While urban nursing homes have been found to have higher costs (Ullmann, 1984) and be more likely to close (Angelelli et al., 2003), rural facilities are reported to have more financial problems (Smith et al., 1992). To take geographical differences in costs of living into account, California sets its Medicaid reimbursement rate to vary by size and geographic region. Patient day rates for homes with 59 beds or less are: $87.26 in the Los Angeles (L.A.) Region, $100.28 in the (San Francisco) Bay Area counties, and $93.31 in all other counties (COSHPD, 2002). While facilities in regions with higher Medicaid reimbursement rates may be less likely to experience bankruptcy, this would depend on the extent to which Medicaid rates cover the regional costs. Expenditure Factors Administrative Costs. Administrative costs may influence the financial viability of nursing facilities in two ways (Wertheim & Lynn, 1993). On one hand, paying sufficient wages and benefits to administrators may help attract and retain individuals who may have the capacity to identify and deal with financial problems early. Against this, administration represents an overhead, which if allowed to become unnecessarily high, could endanger the financial position of a facility. Maintenance Costs. High maintenance costs could place a facility in financial jeopardy as well as indicating that the building is in need of remodeling or rebuilding. This would drive up maintenance costs and probably other costs as well, for example, administration dealing with contractors, etc. A case study of nursing home closures in Michigan identified high maintenance costs among older homes as an important factor (Hirschel, 2002). Funds Paid to Related Parties. As more nursing facilities join chains, an increasing number have financial relationships with parent organizations (related parties). From parents, facilities may receive funds that can help with operating costs, but many also pay administrative


24


fees to parents. It could be expected that “better” facility net flows (funds received minus funds paid) could improve the stability of a facility and hence be negatively associated with bankruptcy. Nurse Staffing Levels. Nurse staffing levels vary widely; they are a highly significant positive factor in average operating costs, and they may have negative effects on facility financial outcomes (Bliesmer et al., 1998). On the other hand, where facilities compete on staffing and quality, higher staffing could lead to higher revenues and improve the financial status of the facility. Financial Status Indicators While financial status (as indicated by measures of liquidity, leverage, and solvency) is central within bankruptcy studies, Zeller et al. (1997) demonstrated that most variance within measures can be explained with the use of a single ratio for each. In general, higher liquidity (acid test ratio) indicates a stronger financial position, higher leverage (liabilities to asset ratio) indicates financial problems, and higher profitability (net income ratio) indicates better financial health. Thus, facility bankruptcy should be negatively associated with profitability and liquidity, and be positively associated with leverage. Resident Characteristics Socio-Demographic Factors. Higher facility percentages of the aged 85 and over population should increase the casemix of residents (and thus the per-patient cost of care) and could weaken the financial position of the facility (Ullmann, 1990). Higher disability rates among African Americans may have a negative effect on the net income of facilities with higher proportions of such residents (Headen, 1992). While the evidence is mixed on this issue (Ullmann, 1990), homes with higher proportions of minorities may also provide a proxy for low-income areas, which in turn could negatively impact on facility viability. Resident Acuity (Casemix). Nursing facility studies have shown a strong positive relationship between acuity (casemix) and nurse staffing time and hence, cost (Fries et al., 1994). As staff hours increase, the costs for a facility should increase, and, if not adequately reimbursed, higher facility casemix could increase the likelihood of bankruptcy.


25


Market Competition Nursing home and hospital studies suggest that in more competitive county markets, private pay rates and facility income may be lower (Zinn, 1994; Banaszak-Holl et al., 1996) and that facilities are more likely to fail (Angelelli et al., 2003). Thus, more competitive county markets were expected to be associated with facility bankruptcy. RESEARCH DESIGN AND METHODS Sample All free-standing, licensed, and/or certified nursing facilities operating in California in 2000 were considered for inclusion in the study (n = 1,156). Hospital-based nursing homes (211) were excluded because none was reported to be in bankruptcy, and cost report data for these facilities are not comparable with those of freestanding facilities. All government and other non-profit facilities (160) were omitted because they are not eligible to file for Chapter 11 bankruptcy status. Forty-one freestanding homes were omitted because they either did not file cost reports for the study period and/or had extensive missing data. Thus, the final study sample comprised the 955 freestanding California facilities that were comparable in terms of data availability and eligibility to operate under Chapter 11 bankruptcy status. Variables and Data Sources The primary sources of data for this study were the cost and utilization reports compiled from the uniform reports that all California facilities submit for all payers (COSHPD, 2002). Table 1 displays the variables selected to operate our conceptual framework, their definitions, data sources, and the hypotheses for facility bankruptcy. All facility and market characteristics (Herfindahl Index) were taken, or calculated, from COSHPD (2001) cost reports. Consistent with standard practice in bankruptcy studies, revenue and cost variables were standardized by resident day, and all financial data were examined for the year prior to bankruptcy (1998-99). Resident characteristics were taken from facility utilization reports (COSHPD, 2001). Because other acuity measures were not available for this study, facility average need for assistance with activities of daily living (ADL) data were taken from

26



TABLE 1. Variables, Measures, Data Sources, and Hypotheses

Variables

Measures

Data Sources Bankruptcy Hypotheses

Facility Characteristics Number of Beds

Total number of licensed beds.

COSHPD Cost Report

-

Occupancy Rate

Resident days divided by bed days.

COSHPD Cost Report

-

Percent Medicare Resident Days

Days paid by Medicare as percent of total days care per facility.

COSHPD Cost Report

+

Percent Medicaid Resident Days

Days paid by Medicaid as percent of total days care per facility.

COSHPD Cost Report

-

Los Angeles Region

Facilities receiving lower Medicaid reimbursement rate for L.A. counties.

Medicaid Rate Areas

+

Bay Area Region

Facilities receiving higher Medicaid reimbursement rate for Bay Area Counties.

Medicaid Rate Areas

_

Administrative Costs per Resident Day

Total administrative costs standardized by resident days.

COSHPD Cost Report

+

Maintenance Costs per Resident Day

Total maintenance costs standardized by resident days.

COSHPD Cost Report

+

Net Related Party Payables Per Resident Day

Total Receivables minus Payables from related party companies standardized per resident day.

COSHPD Cost Report

_

Nurse Staffing Hours per Resident Day

Total productive hours (excluding vacations, sick days, mealtimes) for: full-time, part-time, and contract staff; directors of nursing; supervisory and registered nurses (RN), licensed practical/vocational (LVN/LPN) for the year, standardized by resident days.

COSHPD Cost Report

+

Profitability

Net Income Margin: ratio of net income to total healthcare revenue. Higher ratio shows a stronger position.

COSHPD Cost Report

-

Liquidity

Acid Test Ratio: cash plus marketable securities divided by total current liabilities. Higher ratios indicate a stronger financial position.

COSHPD Cost Report

-

Solvency

Liability to Assets Ratio: total liabilities to total assets. Higher ratios show a weaker position.

COSHPD Cost Report

+

Percent Aged Over 85

Annual percent of facility residents aged over 85 years.

COSHPD Utilization Rpt

+

Percent Black

Annual percent of African American facility residents.


+

Revenue Factors

Expenditure Factors

Financial Indicators

Resident Characteristics



Variables

Measures

Percent Hispanic

Annual percent of Latino facility residents.

Acuity

The average percentage of residents that are totally dependent in three activities of daily living (ADLs) eating, toileting, transfer from bed/chair.

27 Data Sources

Bankruptcy Hypotheses


+

OSCAR Data

+

Market Concentration Herfindahl for Days of Care

The Herfindahl Index (HI) for each county was calcu- COSHPD lated using total days of care for each facility in Cost county in 1999. Days of care per facility were divided Report by days of care in its county. For each county, facility proportions were squared and summed to create HI. Index range 0-1, with higher values representing less competition/more concentration.

-

the federal On-Line Survey Certification and Reporting (OSCAR) system for 1999 (CMS, 2002b). The region variable was derived from the three California Medicaid reimbursement rate areas (COSHPD, 2002). Identification of Bankrupt Facilities Two stages of fieldwork were required to identify facilities operating in bankruptcy in California. First, using multiple federal and industry information sources and a press search conducted by the California Research Bureau (CRB), a list was compiled of eight bankrupt chains operating facilities in California in 2000. No bankrupt independent (nonchain) nursing homes were identified. Second, the California nursing home licensing information system (CDHS, 2002b) was used to identify 155 Californian facilities in the eight bankrupt chains. Analysis Given the exploratory nature of the project, the available data, and the sample characteristics (all identified bankrupt facilities being chain members), the dataset was analyzed descriptively in three main steps. First, for each nursing home in the study, a record file was created to comprise bankruptcy status in 2000 (yes/no) plus facility variables for the previous year. Second, descriptive statistics were computed on all variables for four categories of facility: (a) all freestanding (955); (b) non-bankrupt independent (188); (c) bankrupt chain mem-


28


bers (155); and (d) non-bankrupt chain members (612). Because some of the variable distributions were slightly skewed, the data were normalized using a series of Box Cox transformations (Greene, 1997). For variables whose range covered non-positive values (e.g., net income), the amount necessary to make the minimum value positive was added to the distribution. Third, differences in means (Z) tests were conducted to identify significant differences between facilities in bankrupt chains and facilities in non-bankrupt chains. To account for the multiple comparisons made, results are reported with Bonferroni corrections (Bland & Altman, 1995). FINDINGS This study of nursing home bankruptcy in California produced three sets of unexpected findings. First, the California state agency responsible for nursing home industry oversight did not monitor bankruptcy in 2002 and could not identify bankrupt facilities. Second, all cases of nursing home bankruptcy identified in this study were members of eight chains operating in the state. Unlike hospital failure studies (Wertheim & Lynn, 1993), no independent facilities were identified as operating in bankruptcy. Four of the identified bankrupt chains (113 Californian facilities) filed in 1999 and remained in bankruptcy through 2000 (Lenox, Sun, Aspen & Vencor). The other four chains (42 Californian facilities) entered bankruptcy in 2000 (IHS, Mariner, Hermitage, and TLC). Five of the bankrupt chains were national, and three (Lenox, Aspen, and TLC) were smaller, regional chains. With 155 bankrupt homes in 2000 (just over 16,000 beds), California’s 16% rate of bankrupt facilities was higher than the 11% national average (AHCA, 2001). The third surprising finding of this study was that, as shown in Table 2, few facility or market factors were associated with the bankruptcy status of individual facilities. Against expectations from hospital research and nursing home cost studies, bankrupt chain facilities were no different in terms of occupancy rates and size when compared with homes in non-bankrupt chains. In terms of revenues, facilities in bankrupt chains had higher percentages of Medicare days of care. Contrary to expectations, facilities in bankrupt chains had lower percentages of Medicaid days of care than facilities in non-bankrupt chains. In terms of expenditure factors, as expected, when compared with homes in non-bankrupt chains, facilities in bankrupt chains had higher administration costs and had less advantageous net flows of funds to re-


29


TABLE 2. Comparison of Free-Standing For-Profit Bankrupt and Non-Bankrupt Chain Facilities in California, 1999-2000 All Facilities N = 955

Non-Bankrupt, Independent Facilities N = 188

Bankrupt Chain 1 Facilities N = 155

Non-Bankrupt Chain Facilities N = 612

Mean (Median)

Mean (Median)

Mean (Median)

Mean (Median)

87.29% (89.92)

86.91% (89.61)

88.56% (91.14)

87.08% (89.72)

Facility Characteristics Occupancy Rate

Mean (Median)

SD

Mean (Median)

SD

Mean (Median)

SD

Mean (Median)

SD

103.92 (99.00)

49.59

94.10 (92.50)

51.64

105.76 (99.00)

41.38

106.46 (99.00)

50.55

Percent Medicare Resident Days

6.47 (5.90)

5.20

5.11 (3.80)

6.40

8.89** (8.10)

5.07

6.27 (5.70)

4.59

Percent Medicaid Resident Days

66.37 (74.00)

25.30

63.97 (75.25)

30.25

61.43* (67.80)

21.45

68.35 (75.90)

24.33

Number of Beds Revenue Factors

Los Angeles region (lower Medicaid reimbursement)

32.25%

48.02%

20.00%**

34.31%

Bay Area region (higher Medicaid reimbursement)

17.49%

42.11%

25.16%**

13.89%

Expenditure Factors Administrative Costs per Resident day

18.66 (17.52)

7.92

18.12 (16.13)

8.92

20.84** (20.19)

8.73

18.27 (17.38)

7.28

Maintenance Costs per Resident Day

11.96 (11.42)

3.05

12.68 (11.86)

3.64

11.82 (11.45)

2.32

11.78 (11.29)

2.99

Net Related Party Payables per Resident Day

⫺0.09 (0.00)

1.43

0.05 (0.00)

0.19

⫺0.53* (⫺0.04)

3.18

⫺0.02 (0.00)

0.75

3.03 (2.90)

0.71

3.14 (3.00)

0.91

3.05 (3.00)

0.49

2.99 (2.90)

0.68

Net Income Margin Profitability

1.47 (2.26)

10.87

1.57 (2.13)

10.04

2.96 (3.71)

7.42

1.06 (1.79)

11.80

Acid Test Ratio Liquidity

0.30 (0.02)

0.87

0.71 (0.22)

1.42

0.15** (0.01)

0.70

0.22 (0.01)

0.61

Liability to Assets Ratio Solvency

0.87 (0.67)

1.15

0.85 (0.68)

0.94

0.62** (0.43)

0.57

0.94 (0.72)

1.30

38.97 (40.83)

17.33

41.48 (45.22)

19.32

40.25 (41.05)

15.07

37.87 (38.92)

17.4

Nurse Staffing Hours Per Resident Day Financial Indicators

Resident Characteristics Percent Aged Over 85

30



TABLE 2 (continued) All Facilities N = 955

Non-Bankrupt, Independent Facilities N = 188

Bankrupt Chain Facilities1 N = 155

Non-Bankrupt Chain Facilities N = 612

Mean (Median)

Mean (Median)

Mean (Median)

Mean (Median)

Percent Black

10.21 (4.30)

15.49

10.26 (3.89)

16.57

8.64 (3.51)

13.52

10.61 (4.76)

15.62

Percent Hispanic

10.98 (7.69)

11.67

10.13 (6.81)

10.35

8.19* (4.82)

10.35

11.95 (8.82)

12.24

32.24 (32.33)

14.22

34.54 (34.67)

15.92

29.65 (29.67)

13.39

32.20 (32.33)

13.77

0.05 (0.02)

0.09

0.05 (0.02)

0.09

0.07** (0.03)

0.12

0.05 (0.02)

0.09

Percent Dependent in ADLS Market Concentration Herfindahl for Days of Care

1Z test of the difference in transformed means was conducted with Bonferroni corrections for all comparisons

between bankrupt chain facilities and non-bankrupt chain facilities. *p < 0.05, **p < 0.01

lated parties (i.e., they paid more than they received). There were no differences in nursing hours per resident day or in maintenance costs per day. When compared with homes in non-bankrupt chains, facilities in bankrupt chains displayed significantly weaker financial liquidity but had superior solvency and no difference in profitability. Thus, weaker liquidity was the only indicator of financial jeopardy of bankrupt chain facilities. Other than having significantly fewer Hispanic residents, nursing homes in bankrupt chains had no significant difference in resident characteristics (percentages of residents over age 85, Black residents, and residents with higher acuity levels). Finally, and perhaps most surprisingly, higher percentages of facilities in bankrupt chains were located in less competitive county markets and in regions where the Medicaid reimbursement rates were higher (e.g., the Bay Area). DISCUSSION AND CONCLUSION Studies of nursing home costs and hospital failures suggested that facility and market characteristics would be predictors of nursing home bankruptcy. This exploratory study of nursing home bankruptcy in California reports, however, that all bankrupt facilities were members of



31

chains and while they had significantly worse financial liquidity, higher administrative costs, and higher payables to related parties, they also had more Medicare residents, fewer Medicaid residents, better solvency, and were located in less competitive county markets and in areas with higher Medicaid reimbursement rates. The finding that bankrupt chain facilities had higher rates of Medicare patients does offer some support to providers’ claims regarding the negative impact of Medicare PPS (AHCA, 2001). However, a recent GAO (2002) study found that the 10 largest chains had a median Medicare margin of 18.2% in 1999 and 25.2% in 2000. Moreover, while profit margins (for all payers) for all facilities were about 1.8% in 2000, the largest chains had margins of double this amount (3.8%). This suggests that although chains’ bankruptcy decisions may have been impacted by the change in the Medicare payment system, their Medicare operations continued to be profitable in 1999 and 2000. Overall, these study findings indicate that strategic decisions taken at the corporate (chain) level are the major determinant of nursing facility bankruptcy. Noting that some nursing home chain bankruptcy filings may have involved separate petitions for operating divisions (geographic or functional), it seems that corporations could have entered only some of their facilities into bankruptcy. While this point and our findings underscore the general contention of the GAO (2000) that some chain bankruptcies may have had little to do with facility viability, they raise two important questions about the nature of nursing home bankruptcy. First, because this study was unable to identify any bankrupt-independent facilities in California, national research is required to establish whether this is a state-level phenomenon or whether, unlike independent hospitals, independent nursing homes do not enter bankruptcy. Any revealed difference could be explained by the possibility that, when compared with nursing homes, independent hospitals enter bankruptcy because they are harder to sell or close. However, at this stage, it is not known whether any independent nursing homes have entered bankruptcy in the nation. Second, if national research established that independent nursing homes do not enter bankruptcy, the question arises as to what does happen to failing nursing homes? While the present study was not designed to assess the outcomes of nursing home bankruptcy, among our sample, bankruptcy was followed by facility closure in only two chains (eight homes) in California. Moreover, only 32 freestanding California nursing facilities closed during the five-year period 1997-2001 (Kitchener et al., 2004). It is also worth noting that one of the small bankrupt chains


32


that closed facilities in California had serious quality-of-care problems that may have been more important factors for closure than the facilities’ financial status. In light of this and other evidence that very few independent nursing homes close (Angelelli et al., 2004), it seems that failing independent nursing homes either continue to operate (possibly in a precarious state) or are sold. It has been reported that following bankruptcy restructuring, some large chains such as Sun have sold, or plan to sell, some of their Californian facilities. Further evidence on the empirical nature of nursing home failure will require national analyses of the competing risks (closure, bankruptcy, and sale) facing different types of nursing homes (e.g., large and small chain-members vs. independent homes) to match work conducted in fields of hospitals (Lee & Alexander, 1999). Further research is also needed to examine the implications of chain bankruptcy for quality of care, and for issues of health policy and governance. At the national level, since five of the largest nursing home chains declared bankruptcy in 1999-2000, one has changed its name, none has proceeded to corporate dissolution, and no evidence has emerged of widespread facility closures. Rather, most chains have reduced their number of beds and reported improving financial positions to Wall Street investors. By 2002, revenues among the top chains ranged from $1.3 billion to $2.5 billion, and earnings before interest, taxes, depreciation, amortization, and rent (EBITDAR) for the largest seven chains averaged 11% (CMS, 2003). The policy significance of nursing home chain bankruptcy came into focus when, in response to mounting concern rather than evidence, Congress used the Balanced Budget Refinement Act ([BBRA], 1999) and the Benefits Improvement and Patient Protection Act ([BIPA], 2000) to reinstate temporarily some of the per diem reimbursement that nursing homes lost under Medicare PPS. These two provisions were worth an estimated $1.7 billion to the nursing home industry in 2003 (CMS, 2003). The development of such policy in the absence of evidence underscores recommendations made in a recent Institute of Medicine (2001) report regarding the need to better monitor the financial and operational arrangements of the nursing home industry, particularly national chains. The information deficit concerning nursing facility bankruptcy is increasingly problematic in the light of growing concerns regarding corporate governance. As some of the larger chains emerge from bankruptcy with new leaders, there has been acknowledgement of (prior) managerial problems. For example, the new Chair and CEO of Sun ad-



33

mitted, “If you look at the companies in our business that didn’t go into Chapter 11, they didn’t grow in the same way, they didn’t accumulate debt in the same way. They took the same hit on reimbursement that others did, but it didn’t force them into Chapter 11” (Piotrowski, 2002). In light of this admission, the findings presented here, and because government provides 71% of revenues for the large corporate nursing home chains (CMS, 2002a), they could be required to provide federal and state governments with more and better information to demonstrate that vulnerable residents are being cared for in stable facilities. In the absence of a national effort such as the one recommended by the Institute of Medicine (2001), networks of information exchanges should be established between states. Data from these efforts could then be made available to consumers, policymakers, and regulators by posting them on nursing home websites maintained by CMS and roughly half of all states (Harrington et al., 2003). A model for development may emerge from California where the facility-level financial data and the bankruptcy data collected for this study are reported on a public website. This study underscores the need to collect and analyze such financial and bankruptcy data on a national basis to better inform the public and reduce the potential for nursing home policy to be driven by partial understandings of phenomena such as bankruptcy. RECEIVED: 09/03 REVISED: 06/04 ACCEPTED: 09/04 REFERENCES Aaronson, W. E., Zinn, J. S., & Rosko, M. D. (l995). Subacute care, medicare benefits, and nursing home behavior. Medical Care Research and Review, 52(3): 364-388. American Health Care Association (AHCA) (2001). Facts and Trends 2000: The Nursing Facility Sourcebook. Washington, DC: AHCA. Angelelli, J., Mor, V. Intrator, O., Feng, A., & Zinn, J. (2003). Oversight of nursing homes: Pruning the tree or just spotting bad apples. The Gerontologist. 43 (Special Issue II): 67-75. Appleby, J. (1999). Nursing homes face cutbacks, closures. USA Today, September 30, p. C1. Arling, G., Nordquist, R. H., & Capitman, J. A. (1991). Nursing home cost and ownership type: Evidence of interaction effects. Health Services Research, 22: 255. Balanced Budget Refinement Act (BBRA) (1999). Passed by 105th Congress and signed into law by President Clinton in 1999.


34


Banaszak-Holl, J., Zinn, J. S., & Mor, V. (1996). The impact of market and organizational characteristics on nursing care facility innovation: A resource dependency perspective. Health Services Research, 31(1): 97-118. Benefits Improvement and Protection Act (BIPA). 2000. Amends Section 1919(b) 42 USC 13995I-3(b) of the Social Security Act. Passed by 106th Congress and signed into law by President Clinton, 12/21/2000. Bland, J. M., & Altman, D. G. (1995). Multiple significance tests: The Bonferroni method. British Medical Journal, 310: 170. Bliesmer, M. M, Smayling, M., Kane, R., & Shannon, I. (1998). The relationship between nursing staffing levels and nursing home outcomes. Journal of Aging & Health, 10(3): 351-371. Bonnet, J. P. (2001). Lodi nursing home to shut down at end of June. Lodi News-Sentinel, June 6, p. 1. California Department of Health Services (CDHS) (2002a). The cost of temporary management and receivership, 1993-2001. Data table supplied to authors. Sacramento, CA: DHSS. California Department of Health Services (CDHS), Certification and Licensing Division (2002b). Automated Certification and Licensing Administrative Information and Management System (ACLAIMS) Data. Sacramento, CA: CDHS. California Office of Statewide Health Planning and Development (COSHPD) (2002). Long-Term Care Facility Financial Data for Period 12/31/1998 through 12/30/ 2000. File prepared for this study. Sacramento, CA: OSHPD. Centers for Medicare and Medicaid Services (CMS) (2002a). Health Care Industry Market Update: Nursing Facilities. February 6. Washington, DC: CMS. Centers for Medicare and Medicaid Services (CMS) (2002b). On-Line Survey, Certification, and Reporting (OSCAR) system data. Baltimore, MD: DHSS. Centers for Medicare and Medicaid Services (CMS) Scully, T. (2003). Health Care Industry Market Update: Nursing Facilities. Washington, DC: CMS, May 20. Cleverly, W. O. (1985). Predicting Hospital Failure with the Financial Flexibility Index. Healthcare Financial Management, 39(5): 29-33. Cohen, J. W., & Spector, W. D. (1996). The effect of Medicaid reimbursement on quality of care in nursing facilities. Journal of Health Economics, 15: 23-28. Fries, B. E., Schneider, D., Foley, W., Gavazzi, M., Burke, R., & Cornelius, E. (1994). Refining a case-mix measure for nursing facilities: Resources utilization groups (RUGS-III). Medical Care, 32(7): 668-685. General Accounting Office (2000). Nursing Homes: Aggregate Medicare Payments Are Adequate Despite Bankruptcies. Testimony Before the Special Committee on Aging, U.S. Senate. GAO/T-HEHS-00-192. Washington, DC: General Accounting Office, September 5. General Accounting Office (2002). Skilled Nursing Facilities: Medicare Payments Exceed Costs for Most but Not All Facilities. Report to Congressional Requestors. GAO/ HEHS-03-183. Washington, DC: General Accounting Office, December. Greene, W. H. (1997). Econometric Analysis 3rd Edition. Upper Saddle River, NJ: Prentice-Hall. Harrington, C., Carrillo, H., Mullan, J., & Swan, J. H. (1998). Nursing facility staffing in the states in the 1991-95 period. Medical Care Research and Review, 55(3): 334-363.



35

Harrington, C., O’Meara, J., Kitchener, M., Simon, L. P., & Schnelle, J. F. (2003). Designing a report card for nursing facilities: What information is needed and why. The Gerontologist, 43(II): 47-57. HCIA, & Arthur Andersen (1999). The Guide to the Nursing Home Industry 2000. Washington DC: HCIA and Arthur Andersen. Headen, A. E. (1992). Time costs and informal social support services as determinants of differences between black and white families in the provision of long-term care. Inquiry, 29: 440-450. Hirschel, A. (2002). Personal communication, telephone conversation, February 6. Kitchener, M., Bostrom, A., & Harrington, C. (2004). Smoke without fire: Nursing facility closures, 1997-2001. Inquiry, 41(2): 189-202. Lee, S.-Y. D., & Alexander, J. A. (1999). Consequences of organizational change in U.S. hospitals. Medical Care Research and Review, 56(3): 227-277. McKay, N. (1991). The effect of chain ownership on nursing home costs. Health Services Research, 26(1):109-124. Piotrowski, J. (2002). Post-acute firms see slow recovery. Modern Healthcare. January 28: 40-41. Smith, H. L., Piland, N. F., & Fisher, N. (1992). A comparison of financial performance, organizational characteristics and management strategy among rural and urban nursing facilities. Journal of Rural Health, 8(1): 27-40. Swan, J., Bhagavatula, V., Algotar, A., Seirawan, M., Clemena, W., & Harrington, C. (2001). State Medicaid nursing home reimbursement rates: Adjusting for ancillaries. The Gerontologist, 41(5): 597-604. Ullmann, S. G. (1984). Cost Analysis and Facility Reimbursement in the Long-Term Health Care Industry. Health Services Research, 19(1): 83-102. Ullmann, S. G. (1990). An examination of total charges in the long-term care industry. Journal of Applied Gerontology, 9(2):157-171. Wertheim, P., & Lynn, M. (1993). Development of a prediction model for hospital closure using financial accounting data. Decision Sciences, 24(3): 529-546. Wunderlich, G., & Kohler, P. (Eds.), Institute of Medicine. (2001). Improving the Quality of Care of Long-Term Care. Washington, DC: National Academy Press. Zeller, T., Stanko, B., & Cleverly, W. (1997). A new perspective on hospital financial ratio analysis. Healthcare Financial Management, 52(11): 62-66. Zinn, J. (1994). Market competition and the quality of nursing home care. Journal of Health Politics, Policy, & Law, 19(3): 555-581.