CRITICAL ANALYSIS REPORT for 2 Articles

CriticalAnalysisArticles1.pdf

Home >Business & Finance homework help >Accounting homework help >CRITICAL ANALYSIS REPORT for 2 Articles

Hospital Ownership and Community Benefit: Looking Beyond Uncompensated Care Paula H. Song, PhD, assistant professor, health services management and policy. College of Public Health, Ohio State University, Columbus; Shoou-Yih D. Lee, PhD, associate professor, health management and policy. University of Michigan, Ann Arbor; Jeffrey A. Alexander, PhD, professor emeritus, health management and policy. University of Michigan; and Eric E. Seiher, PhD, assistant professor, health services management and policy, Ohio State University

E X E C U T I V E S U M M A R Y Not-for-profit (NFP) hospitals have come under increased public scrutiny for man- agement practices that are inconsistent with their charitable focus. Of particular concern is the amount of community benefit provided by NFP hospitals compared to for-profit (FP) hospitals given the substantial tax benefits afforded to NFP hospitals.

This study examines hospital ownership and community benefit provision beyond the traditional uncompensated care comparison by using broader measures of community benefit that capture charitable services, community assessment and partnership, and community-oriented health services. The study sample includes 3,317 nongovernment, general, acute care, community hospitals that were in opera- tion in 2006. Data for this study came from the 2006 American Hospital Association Hospital Survey and the 2006 Area Resource File.

We used multivariate regression analyses to examine the relationship between hospital ownership and five indicators of community benefit, controlling for hospi- tal characteristics, market demand, hospital competition, and state regulations for community benefit. We found that NFP hospitals report more community benefit activities than do FP hospitals that extend beyond uncompensated care. Our findings underscore the importance of defining and including activities beyond uncompen- sated care when evaluating community benefit provided by NFP hospitals.

For more information about the concepts in this article, contact Dr. Song at [email protected].

126

HOSPITAL OWNERSHIP AND COMMUNITY BENEFIT

I N T R O D U C T I O N Not-for-profit (NFP) organizations maintain a position of high public trust compared to their for-profit (FP) coun- terparts. However, NFP hospitals have come under increased public scrutiny for practices that appear to be incon- sistent with their not-for-profit ethos, which traditionally focuses on serv- ing their communities in a charitable manner (Panel on the Nonprofit Sector, 2005). Recent investigations into the activities of NFP hospitals have pointed to weak management and governance and noted a variety of misdeeds, such as excessive compensation for executives, inadequate provision of charity care, and deceptive pricing. Adding to this negative publicity has been a raft of law- suits against NFP hospitals alleging that these organizations have aggressive bill- ing and collection practices that violate their charitable obligations to federal, state, and local governments (Lagnado, 2003; Fox & Schaeffer, 1991). These and related concerns have given rise to ques- tions as to whether sufficient difference exists between NFP and FP hospitals to justify NFPs' tax-exempt status (Intemal Revenue Service, 2007).

Of particular concern is the amount of community benefit provided by NFP hospitals compared to FP hospitals, given the estimated $12.6 billion in tax benefits afforded to NFP hospitals each year due to their tax-exempt status (Internal Revenue Service, 2007; Gov- emment Accountability Office, 2008). Although this debate is not new, the resolution largely depends on how com- munity benefit is defined and measured (Schlesinger and Gray, 2006). This issue has gained renewed attention with the

Intemal Revenue Service (1RS) commu- nity benefit reporting requirements and the recent move by the state of Illinois to challenge property tax exemption for three local hospitals (Japsen, 2011). Following these actions, many hospi- tal advocacy groups, researchers, and policy makers are calling for broader definitions of community benefit for tax-exemption justification (Japsen, 2011; Byrd & Landry, 2012, Schlesinger & Gray 2006).

In part to address these concerns, the 1RS has taken steps to clarify the definition of community benefit and established additional reporting requirements for NFP hospitals. Effec- tive for 2010 filings, tax-exempt NFP hospitals are required to report in the revised Schedule H of 1RS Form 990 the organization's annual return; charity care policies; costs of providing charity care; and community-oriented activities, such as community health improve- ment services and community-building initiatives. The inclusion of these community-oriented activities implies that NFP hospitals are, or should be, engaged in these activities at levels that differentiate them from FP hospitals to justify their tax-exempt status. Yet little evidence in the literature demon- strates whether NFP hospitals provide more community-oriented activities and services than do FP hospitals. The majority of recent studies have focused on hospital services and activities that can be measured by expenditure levels, such as charity care, uncompensated care, shortfalls from Medicare and Med- ieaid, and costs of medical education. While this approach estimates a portion of hospitals' direct costs in providing

127

JOURNAL OF HEALTHCARE MANAGEMENT 58:2 MARCH/APRIL 2013

community benefit, it does not capture other community-oriented activities that are considered community benefit in 1RS Form 990 (Government Account- ability Office, 2008).

This study examines hospital owner- ship and community benefit provision beyond the traditional uncompensated care comparison by testing hospital ownership differences in broader mea- sures of community benefit. Our com- parison includes three main domains of community benefit—charitable care, community assessment and partner- ship, and community-oriented health services—that are consistent with the activities in 1RS reporting guidelines. The study results will enhance our understanding of how hospital owner- ship influences the scope and delivery of community benefit and will help inform policy on NFP hospitals and community benefit expectations.

B A C K G R O U N D A N D L I T E R A T U R E R E V I E W Approximately two thirds of the hospi- tals in the United States are NFP (AHA, 2012). Under the current tax code, NFP hospitals are classified as charitable organizations and are exempt from all federal income taxes and most state and local taxes, and they have access to the tax-exempt debt markets. In return, NFP hospitals are expected to provide benefit to the communities in which they oper- ate by meeting the community benefit standard—a set of criteria established by the 1RS in 1969 to determine on a case- by-case basis whether an NFP hospital qualifies for tax exemption (Internal Revenue Service, 1969). FP hospitals are not tax exempt and thus are not

held to the same community benefit expectations as NFP hospitals are. NFP hospitals may have reasons other than tax exemption to provide community benefit. According to economic theory, NFP hospitals, though subject to a breakeven constraint, seek to maximize quantity and quality of services (New- house, 1970). In contrast, FP hospitals may limit community benefit provi- sion outside of tax liabilities because it is inconsistent with their objective to maximize shareholder value. When FP hospitals do provide community benefit outside of tax liabilities, their purpose may be to maintain or improve their competitive position and public per- ception in the local market (Clement, White, and Vladmanis, 2002).

However, studies that examine the relationship between hospital own- ership and community benefit have produced inconsistent results. Sev- eral studies found that NFP hospitals provide more charity care or uncom- pensated care than do FP hospitals (Thorpe and Phelps, 1991; Mann, Melnick, Bamezai, & Zwanzinger, 1997). Other studies found either that NFP and FP hospitals provide compa- rable levels of uncompensated care or that the difference between the two is relatively small (Government Account- ability Office, 2005; Congressional Budget Office, 2006).

Studies evaluating the adequacy of NFP hospitals' community benefit provision relative to tax benefits have generally found that NFP hospitals fall short of their obligations (Morrisey, Wedig, & Hassan, 1996; Nicholson, Pauly, Burns, Baumritter, & Asch, 2000). For example, Nicholson et al. (2000)

128

HOSPITAL OWNERSHIP AND COMMUNITY BENEFIT

used a public goods approach to estab- lish benchmarks for the expected level of community benefit provided by an NFP hospital compared to an equiva- lent FP hospital. Defining community benefit to include uncompensated care, cost of other unbilled public good services, losses sustained on medical research, and taxes for FP hospitals, they concluded that NFP hospitals were not meeting their expected level of com- munity benefit provision. In a study of California hospitals, Schneider (2007) found that the assessment of whether NFP hospitals provided equivalent lev- els of community benefit to those of FP hospitals depended on the definition of community benefit. When community benefit was defined as uncompensated care only, NFP hospitals' community benefit levels were found to be lower than those of FP hospitals. However, if it was defined more broadly to include price discounts, government shortfalls, and taxes paid by FP hospitals, NFP hospitals provided more community benefit than FP hospitals did. Simi- larly, Bazzoli, Clement, & Hsieh (2010) found that only when bad debt and shortfalls from Medicare were included in the 1RS Schedule H's definition of community benefit were private NFP hospitals found to provide adequate community benefit compared to FP hospitals.

Few studies have focused on hospi- tal ownership and community benefit using a broader definition of commu- nity benefit that includes community- oriented activities in addition to charity care or uncompensated care. Collec- tively, these studies suggest that when comparisons used a more inclusive

definition of community-oriented ser- vices, NFP hospitals did in fact provide more benefit to the community than did their FP counterparts (Arrington & Haddock, 1990; Boscarino & Chang, 2000). However, these studies were based on hospital survey data that are now outdated.

Our study contributes to the exist- ing literature in several ways. First, we used data from 2006 to evaluate hospital ownership and community benefit. These are the most recent data we have access to that contain propri- etary information on charitable care. Second, we used an inclusive definition of community benefit that captures a range of hospital community benefit activities, beyond uncompensated care, as suggested by the literature and the 1RS. Third, we evaluated the types of programs and services hospitals provide, focusing on the number of services. This focus complements the standard cost focus by allowing comparisons between hospitals that use varying accounting practices for calculating uncompensated care and community benefit.

Given that NFP hospitals are expected to provide community benefit in exchange for tax exemption and that their objective is to maximize quantity and quality of services over profit maxi- mization, we expected to find that NFP hospitals provide more community ben- efits than do FP hospitals. Specifically, we expected that NFP hospitals provide more community benefit than do FP hospitals along the three domains of community benefit we evaluated: chari- table care, community assessment and partnership, and community-oriented health services.

129

JOURNAL OF HEALTHCARE MANAGEMENT 58:2 MARCH/APRIL 2013

M E T H O O S This study includes nongovernment, general, acute care, community hospi- tals that were in operation in 2006. The sample consists of 2,660 private NFP and 657 FP hospitals. The data for this study came ftom the 2006 American Hospital Association (AHA) Annual Hospital Survey and the 2006 Area Resource File. The AHA hospital survey is a national survey conducted annually of all U.S. hospitals and has an average response rate of 85 percent. Missing data are imputed on the basis of the previ- ous year's responses and comparisons to hospitals of similar size and orientation. The hospital survey captures informa- tion on organizational characteristics, service patterns, community assessment and partnership activities, utilization, staffing, payers, and finances. The Area Resource File is produced annually by the U.S. Department of Health and Human Services. It draws on a number of national databases, including that of the Census Bureau, to produce county- level data on population, sociodemo- graphic status, and health resource, which can be used to construct measures of local market characteristics.

Dependent Variables We evaluated the relationship between hospital ownership and community benefit in the three broad domains of community benefit being studied. Table 1 presents a detailed description of each community benefit indicator.

The charitable care domain is defined as healthcare provided to needy, underserved, and disadvantaged patients, including those who are unin- sured or underinsured and those who

are covered by means-tested insurance programs. The two measures are (1) uncompensated care and (2) Medicaid inpatient load. Uncompensated care is the conventional measure of charity care used in the literature because variation in hospital accounting practices makes it difficult to distinguish between char- ity care, defined as services performed without expectation of payment, and bad debt, referred to as the balance of services provided with expectation of reimbursement but ultimately unable to collect on (Morrisey et al., 1996; Schnei- der, 2007; Thorpe, Florence, & Seiber, 2001). We calculated uncompensated care costs by taking the sum of charity care and bad debt, adjusted by the hos- pital's cost-to-charge ratio, as a percent- age of total expenses. We used Medicaid inpatient load as a measure of charitable care because Medicaid enrollees face substantial barriers to accessing health providers and services due to compara- tively low Medicaid reimbursement rates (Asplin et al., 2005; Medicaid Access Croup, 1994). Accordingly, higher Medicaid inpatient loads indicate that hospitals provide a greater proportion ' of care to underserved populations. We measure Medicaid inpatient load by the ratio of the number of Medicaid inpa- tient days to total inpatient da}^.

The second domain of community benefit, community assessment and partnership, is defined as "organization- wide generation, dissemination, and use of community intelligence to address the health needs of local communi- ties" (Lee, Alexander, & Bazzoli, 2003, p. 168; see also Lee, Chen, & Weiner, 2004; Proenca, Rosko, & Zinn, 2000). The 2006 AHA Annual Survey contained

130

HOSPITAL OWNERSHIP AND COMMUNITY BENEFIT

T A B L E 1 Dependent Variable Description by Community Benefit Domain

Domain and indicator Variable Description

Charitable Care Domain

Uncompensated care [(Charity care + bad debt) x cost-to-charge ratio]/total expenses

Medicaid inpatient load Total Medicaid inpatient days/total inpatient days

Community Assessment and Partnership Domain

Community assessment Sum of 8 Questions Related to Community Assessment and and partnership Partnership (0 = No, 1 = Yes)

1. Does your hospital have a long-term plan for improving the health of its community?

2. Does your hospital have resources for its community benefit activities?

3. Does your hospital work with other providers, public agencies, or community representatives to condua a health status assessment of the community?

4. Does your hospital use health status indicators (such as rates of health problems or surveys of self-reported health) for defined populations to design new services or modify existing services?

5. Does your hospital work with other local providers, public agencies, or community representatives to develop a written assessment of the appropriate capacity for health services in the community?

6. If yes, have you used the assessment to identify unmet health needs, excess capacity, or duplicative services in the community?

7. Does your hospital work with other providers to collect, track, and communicate clinical and health information across cooperating organizations?

8. Does your hospital, either by itself or in conjunction with others, disseminate reports to the community on the quality

costs of healthcare services?

[continued)

131

JOURNAL OF HEALTHCARE MANAGEMENT 58:2 MARCH/APRIL 2013

TABLE 1 continued

Community-Oriented Health Services Domain

Community health

services provision

Emergency department services

Sum of 16 Community Health Services Provided (0 = No, 1 = Yes)

1. Community outreach 2. Crisis prevention

3. Child Wellness

4. Fitness center

5. Health fair

9. Occupational health

10. Patient education

11. Patient representative services

12. Social work

13. Support groups

6. Health information center 14. Teen outreach services

7. Health screening 15. Transportation services

8. Nutrition program 16. Volunteer services

Binary variahle (0 = No, 1 = Yes) for emergency department

services provision

9 items that assessed hospitals' engage- ment in these activities, such as whether hospitals have community benefit resources, long-term plans to improve community health through collabora- tion with community partners, and formal mechanisms to report on com- munity health needs. Previous research showed consistently that all but one of those items loaded on a common factor (i.e., explained the majority of the vari- ance of a common factor) and had good intemal reliability (Lee et al., 2003, 2004). Thus, we used the sum of the eight dichotomous items to represent the level of community assessment and partnership (ranging from 0 to 8).

The third domain in our definition of community benefit, community- oriented health services, includes ser- vices that are valued by the community but typically do not generate a profit for the hospital. The first indicator for this domain is the provision of com- munity health services, a composite

index constructed using a subset of the AHA data, which report on 16 distinc- tive community health services aimed at promoting community health through preventive and educational programs (Lee et al., 2003; BazzoU et al., 1997). The second indicator of community- oriented health services is whether the hospital provides emergency department services. Although the 1RS has loosened the 1969 guidelines by allowing NFP hospitals to waive the requirement of operating an emergency department, the emergency room is often considered the cornerstone of a hospital's provision of community care (Albrecht, Slobodkin, & Rydman, 1996; Griswold, 2000). Emer- gency departments at NFP hospitals remain an important point of access to care, particularly for Medieaid beneficia- ries, who often are unable to find a phy- sician accepting new Medieaid patients or unable to see a physician outside of the traditional 9:00 a.m. to 5:00 p.m. work schedule. Indeed, studies have

132

HOSPITAL OWNERSHIP AND COMMUNITY BENEFIT

found that Medicaid beneficiaries tend to be the highest utilizers of the emer- gency department (Mortensen & Song, 2008). Emergency rooms are also the last resort and safety net for many of the nation's uninsured, and a significant proportion of visits to the emergency room are for nonurgent care (Baker, Stevens, & Brook, 1995).

Independent Variables The key variable of interest is hospi- tal ownership as reported on the AHA survey. NFP hospitals include church- ovwied, independent, or corporate- owned NFP hospitals. FP hospitals include all other nongovernment, for- profit hospitals.

To account for those hospitals that may have more resources or whose organizational structure may influence their community benefit provision, we included variables for hospital size, urban or rural designation, teach- ing status, system affiliation, network affiliation, and operating performance. Hospital size was measured by the number of inpatient beds. Dummy indicators were created and set equal to 1 for urban, teaching, system-affiliated, and network-affiliated hospitals and 0 otherwise. Occupancy rate and operat- ing margin were included as indicators of operating performance. Occupancy rate is measured by the percentage of admissions, adjusted for outpatient and inpatient activity, over total available bed days. We took the lagged three-year average of operating margin to account for year-to-year fluctuations in profit- ability and to take into account that the current year's community benefit provision may be influenced by the

organization's financial performance in previous years.

We also included five variables to control for county-level market condi- tions, which may affect the level of demand for community benefit. These conditions include unemployment rate, education, per capita income, man- aged care penetration, and hospital competition. The first three variables represent the demand for community benefit services by the uninsured and underinsured. Unemployment rate was measured by the percentage of those in the labor force who were unemployed. Education was measured by the percent- age of the population over age 25 with at least a high school diploma or an equivalent degree. Per capita income is the average of individual annual income in the county. Managed care penetra- tion and hospital competition represent market forces that could influence a hospital's community benefit provision. Managed care penetration was measured by the percentage of Medicare patients enrolled in a managed care plan. Hospital competition was measured by 1 minus the Herfindahl-Hirschman Index, or the sum of the squared market share of all community hospitals in the county on the basis of adjusted inpa- tient admissions. This variable ranges from 0 to 1, with higher values indicat- ing a more competitive market.

Finally, to control for state-level regulation, we included a dummy indica- tor for states that have community ben- efit laws. Sixteen states have passed laws or guidelines specifying NFP hospital community benefit reporting require- ments or provision levels. We noted that, according to one study, FP hospitals

133

JOURNAL OF HEALTHCARE MANAGEMENT 5 8 : 2 MARCH/APRIL 2013

decreased their community assessment and partnership and health promotion activities in states that had community benefit laws, widening the gap between FP and NFP hospitals (Ginn, Shen, & Moseley, 2009). To control for this factor, we induded a variable that accounted for the interaction between state community benefit law and hospital ownership.

Analysis We used multivariate regression analyses to test the relationship between hospital ownership and the five community ben- efit indicators, controlling for hospital characteristics, market demand, hospi- tal competition, and state regulations for community benefit. We assumed nonlinearity in our analysis given the structure of our dependent variables. We used a generalized linear model with logit link to test percentage of uncom- pensated care and percentage of Med- icaid inpatient load (Models 1 and 2) and hospital ownership. For the count models, we used a negative binomial regression to test community assess- ment and partnership (Model 3) and community-oriented services (Model 4) and hospital ownership. Finally, we used logistic regression to evaluate emergency department services (Model 5) and hospital ownership. All models were adjusted to reflect robust standard errors to correct for geographic clustering in the data. Marginal effects were tested and are reported for significance at the p < .05 level. All analyses were performed in Stata 10.0 (StataGorp 2007).

RESULTS Table 2 presents descriptive statistics for dependent and independent variables

in our study sample by ownership status (n = 3,317). NFP hospitals represent 80 percent of the hospitals in our sample. On average, FP and NFP hospitals provide comparable levels of uncom- pensated care. However, NFP hospitals provide higher levels for all other indica- tors of community benefit compared to FP hospitals.

Table 3 reports the marginal effects from the regression models. These marginal effects represent the increase in the dependent variable for one unit increase of each independent variable. In the case of hospital ownership, our key variable of interest—the marginal effect—reveals the difference between NFP and FP hospitals. A positive mar- ginal effect indicates that NFP hospitals provide more community benefit than do their FP counterparts.

In the charitable care domain, hos- pital ownership is significantly related to uncompensated care but shows no relationship with Medicaid inpatient load. The marginal effects presented in Table 3 indicate that after control- ling for hospital characteristics, market characteristics, and state regulation, uncompensated care as a share of total expenses at NFP hospitals averages 0.8 percentage points higher than at similar FP hospitals (Model 1). While statistically significant, the relatively small magnitude of the marginal effect is consistent with previous studies that found modest differences between levels of uncompensated care provided by NFP and FP hospitals. In contrast. Model 2 revealed no statistically sig- nificant relationship between hospital ownership and Medicaid inpatient load. While NFP hospitals do show a 0.7

134

HOSPITAL OWNERSHIP AND COMMUNITY BENEFIT

TABLE 2 Descriptive Statistics by Ownership, Nongovernment

Investor Owned

(n =

Mean

Community Benefit Indicators

Uncompensated care %

Medicaid inpatient load O/n , 0

Community assessment and partnership (0-8)

Community health ser-

vices (0-16)

Emergency department (l=Yes, 0 = No)

Hospital Characteristics

Size (number of beds)

System afifiliation ( 1 =

Yes, 0 = No)

Network affiliation (1 = Yes, 0 = No)

Urban (1 =Yes, 0 = No)

Teaching (1 = Yes, 0 =

No)

Occupancy rate %

Operating margin %

Market Conditions

Unemployment rate %

Education (% > high school education)

Per capita income ($ per 1000)

Medicare HMO %

Herfindahl-Hirschman Index (0-1)

State Regulation

State c o m m u n i t y benefit

l a w ( l =Yes, 0 = No)

0.045

0.160

4.294

5.619

0.699

152.4

0.878

0.198

0.691

0.015

0.527

0.059

0.054

0.76

28.99

0.13

0.545

0.416

= 657)

0.001

0.005

0.130

0.175

0.018

4.903

0.013

0.016

0.018

0.005

0.007

0.006

0.001

0.003

0.318

0.006

0.014

0.019

General

Not-

{n.

Mean

0.047

0.188

6.035

8.909

0.874

198.8

0.561

0.366

0.612

0.074

0.607

0.015

0.052

0.805

30.86

0.106

0.472

0.422

Acute Care

for-Profit

= 2,660)

0.001

0.003

0.054

0.089

0.006

3.886

0.010

0.009

0.005

0.004

0.002

0.000

0.001

0.172

0.002

0.007

0.010

Hospitals, 2006

All Hospitals

(n = 3,1

Mean

0.047

0.182

5.690

8.257

0.839

189.7

0.624

0.333

0.627

0.063

0.591

0.023

0.053

0.796

30.49

0.11

0.486

0.421

$17)

0.000

0.003

0.052

0.082

0.006

3.282

0.008

0.004

0.003

0.002

0.000

0.001

0.152

0.002

0.006

0.009

135

JOURNAL OF HEALTHCARE MANAGEMENT 58:2 MARCH/APRIL 2013

» o

£ o o

M m

S Bl 09

.5 >

m I

i I

i .1

11- I

s d

fS 00o

o o r N o O O r i O t N C N r O 0 0 0 0 0 0 0

Ö Ö Ö Ö Ö Ö ó

< 7 ^ \ r ) O c ^ l l — I ( N O O p O Ö Ö dl ó ó

O O t N o o o o t N i — i t n i / i o o t N o o o o

ö ö ö ö öö ö ö ö ö öII ö ö ö ö ó

I I I

O ^ ^ O "-1 "-1 CN (N r}> i-H

ö ö ö ö ö ö '-i

c N * r s N o 0 0 0 , - H O O C O O O t N p c O i n p p i - i ( N

n o o r O O C O O O O U - )

p d

oq d

00 o o

oc-, i j O O O V O < N O O

P P P P p N

* r J > o o , - H ^ ^ ^ , - ^ Í O i n O C N t N

o ^ .—I t̂ o ĉ in o o o o o o (Np p p p p p p d d ó d d d d

I I I I

o in o tN o rs 00 o CO op p p p p d d d d d I I I I I

•a

c •a

•2 a;

PP do"

ó ó

9 0 0 o

0 0 0

(0 0 5 2

- 0

10 8]

(0 3 8 5

08 8

(0 2 9 2

1 3 9

(0 35 1

18 6

(Ti <N

r^ ^^ en i n

o o

3 9 0

(N (N

3 4 8

9 4 2

00 6

00 8

18 3

4 7 8

2 2 7

20 6

9 3 3

3 3 5

i n

o 1

3 2 6

# «- 00 00

0 0 0

00 5

1—1

00 5

00 7

) - 0 .

00 8

0 3 0

9 3 0

16 8

04 0

T O

S o

g o

0 1 2

01 3

c o o C N t N

;ooo

-0 .

o o o

o 1

o o o

00 3;

o 1

00 5

03 4

-0 .

00 7;

00 0

00 4

) - 0 .

00 2

3 0 0

o 1

00 2

o o p p d d

I I 1I

i 00

-73 Q C

"ñ. -^ "ÏÎ is T3 ^ O

li II

2 Ç _ , c^ c/3 Z ¡13

-a pa

S ä ^ s s I I . C/3 c/3

5 „

l e

136

HOSPITAL OWNERSHIP AND COMMUNITY BENEFIT

percentage point higher share of Med- ieaid patients, the difference does not achieve significance at the p < .05 level. In this model, hospital size, occupancy rate, and county unemployment rate were related to higher Medieaid inpa- tient loads, and hospitals that were system affiliated, urban, teaching, more profitable, located in communities with higher per capita income, and operat- ing in competitive markets had lower Medieaid inpatient loads.

In contrast to their scores in the charitable care domain, NFP hospitals consistently scored higher in the commu- nity assessment and parmership domain and community-oriented health service domain. Within the community assess- ment and parmership domain, NFP own- ership is associated with a 1.74 point (on the 8-point scale) increase in a hospital's level of community assessment and part- nership relative to FP ownership (Model 3). This scale increase indicates that more NFP hospitals have plans to address the health of their community, work with other providers to track and communi- cate health information across organiza- tions, and engage in other community assessment and parmership activities comprising the community assessment and partnership indicator described in Table 1. To test whether this difference is attributable to particular elements of the scale, we ran separate regressions for each component in the community assessment and parmership composite score (results not presented). Rather than differing on a single element, NFP hospitals were 20 to 30 percentage points more likely than FP hospitals to respond affirmatively to each question included in the community assessment and partnership indicator.

As in the case of the community assessment and partnership domain, we found, after controlling for hospital characteristics, market characteristics, and state regulation, that NFP hospitals are more likely to provide community- oriented health services than are their FP counterparts. Our results indicate that NFP ownership is associated with an increase oí 1.76 (out of 16) community health-oriented services provided com- pared to FP hospitals. The disaggregated regressions indicate that NFP hospitals were significantly more likely to offer health fairs, nutrition programs, and the other services included in the commu- nity health services provision indica- tor than were the FP hospitals. Finally, we found that NFP hospitals were 8.2 percentage points more likely to provide emergency department services than were FP hospitals.

Across all models, the presence of state community benefit laws had no effect on community benefit provision, with the exception of the community assessment and partnership indicator (Model 3). In states with a community benefit law, hospitals provided more community benefit with respect to the community assessment and partner- ship indicator; however, the interaction between state community benefit laws and hospital ownership was negative, suggesting that ownership differences in community assessment and partner- ship are less pronounced in states with community benefit laws than in states without such laws.

D I S C U S S I O N We found that NFP hospitals reported more community benefit activities than

137

OF HEALTHCARE MANAGEMENT 58:2 MARCH/APRIL 2013

did FP hospitals in ways that extend well beyond uncompensated care, the con- ventional measure of community bene- fit. Overall, we found that NFP hospitals provided more community benefit than did FP hospitals with regard to chari- table care, community assessment and partnership, and community-oriented services. Specifically, NFP hospitals, on average, provide relatively small, but statistically significantly more, uncom- pensated care. In the domains of com- munity assessment and partnership and community-oriented services, we found that NFP hospitals participated more often in community assessment and partnership, provided more community health services, and were more likeiy to have an emergency department than were FP hospitals. However, we found no significant difference in the provision of Medicaid services.

The results of this study contribute updated evidence to the existing litera- ture that when community-oriented activities are included in the commu- nity benefit definition, NFP hospitals provide higher levels of community benefit than do FP hospitals (Arrington & Haddock, 1990; Boscarino & Chang, 2000). Our results also provide support for the underlying assumption of 1RS Schedule H that NFP hospitals already engage in community-oriented activities at levels that differentiate them from their FP counterparts. This finding is important not only because this under- lying assumption is the basis for Sched- ule H reporting requirements but also because similar comparative analyses on hospital ownership and commu- nity benefit activities using Schedule H reports will be limited, as the reporting

requirements apply only to tax-exempt hospitals.

Consistent with previous studies, our findings underscore the impor- tance of defining and including activi- ties beyond uncompensated care when evaluating the amount of community benefit provided by NFP hospitals (Schlesinger & Cray, 2006; Schneider, 2007; Bazzoli et al., 2010). Reporting requirements and policies that focus pri- marily on uncompensated care or only on easily monetized measures under- estimate community benefit provided by NFP hospitals and increase NFP hospitals' risk of jeopardizing their tax- exempt status. Therefore, the definition and measurement of community benefit are critical to properly evaluating the level of community benefit NFP hospi- tals provide in return for tax exemption.

The results of our study also have practice implications for NFP hospital managers. For many hospitals, the 1RS Schedule H reporting requirements have prompted the first opportunity to take inventory of their community benefit activities. To do so, managers must have information systems in place that can track both the activities and the expenses incurred to conduct those activities. In addition, the importance for hospitals to track and report com- munity benefit activities in addition to charity care will become more impor- tant under healthcare reform. Assuming the Affordable Care Act will be fully implemented, the number of uninsured will decrease drastically after 2014 as a result of health insurance expansion, which in tum will reduce the demand for charity care services. Therefore, NFP hospitals must be able to demonstrate

138

HOSPITAL OWNERSHIP AND COMMUNITY BENEFIT

the community benefit value they pro- vide beyond charity care in ways that distinguish them from FP hospitals and maintain their tax-exempt status. Prin- cipe, Adams, Maynard, & Becker (2012) suggest several ways in which the Afford- able Care Act may influence the compo- sition of community benefit provided by hospitals. For example, the reduced demand for charitable services may result in increased revenues for hospi- tals, enabling NFP hospitals to provide additional community benefit, such as preventive services or subsidized health services (Principe et al., 2012).

Several limitations in this study are important to note. First, the dependent variables do not represent an exhaustive list of services and activities that may be accepted as community benefit. How- ever, these indicators have been used extensively in the literature and rep- resent the major community-oriented activities and services provided by hos- pitals (Lee et al., 2003; Proenca et al., 2000; Bazzoli et al., 1997). Second, our study is cross-sectional, which allows us to provide supporting evidence that NFP hospitals offer a greater range of community benefit than do FP hospi- tals. Restricted availability of key data on uncompensated care in the AHA survey precluded a longitudinal analysis, which would have allowed us to evaluate these effects over time and better control for unobserved effects. In addition, our analysis provides marginal effects, which estimate the difference between NFP and FP hospitals when all other control variables are at their means. If hospitals' characteristics were stratified at differ- ent levels, the results might differ. While beyond the scope of this study, fiiture

research should decompose these aver- age effects to determine whether differ- ences in community benefit exist among NFP hospitals that vary along certain structural and financial characteristics. Finally, although this study focuses on the levels of community benefit activi- ties and services hospitals provide, our data and results do not allow us to draw inferences as to whether these services were offered in excess of the tax benefits NFP hospitals received.

Our findings suggest that hospital ownership matters in regard to com- munity benefit provision. While the literature shows that NFP hospitals provide modestly more uncompensated care, our results indicate that they also provide significantly more community services beyond uncompensated care. The challenge of appropriately assigning a monetary value to these community service activities relative to tax benefits remains. Future research should address these measurement issues before deter- mining whether NFP hospitals supply enough community benefit, inclusive of broader activities constituting commu- nity benefit, to warrant exemption from taxes.

REFERENCES Albrecht, G. L., Slobodkin, D., & Rydman, R.I.

(1996). The role of emergency depart- ments in American health care. Research in the Sociology of Health Care, 13B, 289-316.

American Hospital Association (AHA). (2012). Fast facts on US hospitals. Retrieved from http://www.aha.org/research/rc/stat -studies/fast-facts.shtml

Arrington, B., & Haddock, C. C. (1990). Who really profits from not-for-profits? Health Services Research, 25(2), 291-304.

Asplin, B. R., Rhodes, K. V., Levy, H., Lurie, N., Crain, A. L, Cariin, B. P., & Kellerman, A. L. (2005). Insurance status and access to

139

JOURNAL OF HEALTHCARE MANAGEMENT 58:2 MARCH/APRIL 2013

ui^ent ambulatory care follow-up appoint- ments. Journal of the American Medical Association, 294(10), 1248-1254.

Baker, D. W., Stevens, C. D., & Brook, R. H. (1995). Determinants of emergency department use by ambulatory patients at an urban public hospital. Anruls of Emer- gency Medicine, 25(2), 311-316.

Bazzoli, G. J., Clement, J. P. & Hsieh, H. (2010). Community benefit activities of private, nonprofit hospitals. Journal of Health Poli- tics, Policy and Law, 35(6), 999-1026.

Bazzoli, G. J., Stein, R., Alexander, J. A., Conrad, D. A., Sofaer, S., & Shortell, S. M. (1997). Public-private collaboration in health and human service delivery: Evidence from community partnerships. Milbank Quarterly, 75(4), 533-561.

Boscarino, J. A., & Chang, J. (2000). Nontra- ditional services provided by nonprofit and for-profit hospitals: Implications for community health. Journal of Healthcare Management, 45(2), 119-136.

Byrd, J. D., & Landry, A. (2012). Distinguish- ing community benefits: Tax exemption versus organizational legitimacy. Journal of Healthcare Management, 57(1), 6 6 - 7 8 .

Clement, J. P., White, K. R., & Vladmanis, V. (2002). Charity care: Do not-for-profits influence for-profits? Medical Care Research and Review, 59{l), 59-78.

Fox, D. M., & Schaeffer, D. C. (1991). Tax administration as health policy: Hospi- tals, the Internal Revenue Service, and the courts. Journal of Health Politics, Policy and Law, 1(5,251-279.

Ginn, G. O., Shen, J. J., & Moseley, C. B. (2009). Community benefit laws, hospital ownership, community orientation activi- ties, and health promotion services. Health Care Management Review, 34(2), 109-118.

Griswold, B. (2000). State of emergency. ER closings pose nationwide threat. Revolu- tion, 1(2): 2 4 - 2 7 , 2 9 .

Japsen, B. (2011, September 10). State chal- lenging hospitals' tax exemption. New York Times. Retrieved from http://www.nytimes .com

Lagnado, L. (2003, June 10). Taming hospi- tal billing: Lawmakers push legislation to curb aggressive collection against uninsured patients. Wall Street Journal. Retrieved from http://www.wsj.com

Lee, S. Y. D., Alexander, J. A., & Bazzoli, G. J. (2003). Who do they serve? Community

responsiveness among hospitals af&liated with health systems and networks. Medical Care, 41(1), 165-179.

Lee, S. Y. D., Chen, W. L., & Weiner, B. J. (2004). Communities and hospitals: Social capital, community accountability, and service provision in U.S. community hospitals. Health Services Research, 39(5), 1487-1508.

Mann, J. M., Melnick, G. A., Bamezai, A., & Zwanzinger, J. A. (1997). Profile of uncompensated hospital care, 1983-1995. Health Affairs, 16(4), 223-231.

Medicaid Access Group. (1994). Access of Medicaid recipients to outpatient care. New England Journal of Medicine, 330(20), 1426-1430.

Morrisey, M. A., Wedig, G., & Hassan, M. (1996). Do nonprofit hospitals pay their way? Health Affairs, J5(4), 132-144.

Mortensen, K. M. & Song, P H. (2008). Minding the gap: A decomposition of emergency department use by Medicaid enrollees and the uninsured. Medical Care, 46(10), 1099-1107.

Newhouse, J. P. (1970). Toward a theory of nonprofit institutions: An economic model of a hospital. American Economic Review, 60(1), 64-74.

Nicholson, S., Pauly, M. V., Burns, L. R., Baum- ritter. A., & Asdi, D. A. (2000). Measuring community benefits provided by for-profit and nonprofit hospitals. Health Affairs, 19(6), 168-177.

Panel on the Nonprofit Sector. (2005). Strengthening transparency, governance, and accountability of charitable organiza- tions: A final report to Congress and the nonprofit sector. Retrieved from http:// www.nonprofitpanel.org/Report/final /Index.html

Principe, K., Adams, E. K., Maynard, J., & Becker, E. R. (2012). The impact of the individual mandate and Intemal Revenue Service Form 990 Schedule H on com- munity benefit fi"oni nonprofit hospitals. American Journal of Public Health, 102(2), 220-237.

Proenca, E. J., Rosko, M. D., & Zinn, J. S. (2000). Community orientation in hospi- tals: An institutional and resource depen- dence perspective. Health Services Research, 35(5), 1011-1035.

Schlesinger, M., & Cray, B. H. (2006). How nonprofits matter in American medicine.

140

HOSPITAL OWNERSHIP AND COMMUNITY BENEFIT

and what to do about it. Health Affairs, 25(4), W287-.W303.

Schneider, H. (2007). Paying their way? Do nonprofit hospitals justify their favorable tax treatment? Inijuiry, 44(2), 187-199.

StataCorp. 2007. Stata Statistical Software: Release 10. College Station, TX: StataCorp LP.

Thorpe, K. E., Florence, C. S., & Seiber, E. E. (2001). Hospital conversions, margins, and the provision of uncompensated care. Health Affairs, 19(6), 187-194.

Thorpe, K. E., & Phelps, C. E. (1991). The social role of not-for-profit organizations: Hos- pital provision of charity care. Economic Inquiry, 29(3), 472-484.

U.S. Congressional Budget Office. (2006, December 6). Nonprofit hospitab and the provision of community benefits (Publication No. 2707). Retrieved from http://www .cbo.gov/ftpdocs/76xx/doc7695/12 -06-Nonprofit.pdf

U.S. Government Accountability Office. (2005). Nonprofit, for-profit, and govern- ment hospitals (Report No. CAO-05-743T). Retrieved from http://www.gao.gov/new .items/dO5743t.pdf

U.S. Government Accountability Office. (2008). Variation in standards and guidance limits. Comparison of how hospitals meet community benefit requirements (Report No. GAO-08 880). Retrieved from http://www .gao.gov/products/GAO-08-880

U.S. Internal Revenue Service. (1969). Revenue ruling 68-545. Washington, DG: Internal Revenue Service.

U.S. Internal Revenue Service. (2007, June 14). Draft Form 990 Redesign Project—Sched- ule H. Retrieved from http://www.irs.gov /pub/irs-tege/draftform990redesign_schh _instr.pdf

P R A C T A P P L I C A T

Ellen Zaman, FACHE, director, Community Affairs, Children's Hospital Los Angeles, California

H ealthcare organizations have endless opportunities for innovation in providingcommunity benefit. The authors of this article guide us through a study of more than 3,317 nongovernment, acute care hospitals to examine hospital ownership and the provision of community henefit. Their finding that not-for-profit (NFP) hospitals report more community benefit activities beyond uncompensated care than do for- profit (FP) hospitals raises important issues for organizations whose work is touched by the debate on community benefit and tax exemption. Key to this study is examin- ing the various domains: charitable care, community assessment and partnership, and community-oriented health services.

While often difficult—but not impossible—^to quantify, these community- oriented health services and partnership activities offer practitioners an opportunity to demonstrate how programs and interventions contribute to the health of popu- lations. Examples abound of innovative programs in hospitals across the country. Although some of these may have been developed in response to geographic and demographic needs, they can often be replicated at other organizations. After all, most of our organizations were founded in response to community needs; we can learn from these examples.

141

and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright

holder's express written permission. However, users may print, download, or email articles for individual use.