A comparison of the community benefit of nonprofit and for-profit hospitals

Controversy over the tax-exemption of nonprofit hospitals has been repeatedly discussed over the past decade. Many question whether the exemption is justified by hospitals' benefit to communities (Al-Agba, 2017). Past studies have analyzed this issue by focusing on charitable care, or uncompensated care, as a measure of community benefit (Bai, et al., 2021; Bruch and Bellamy, 2020). Most findings conclude that nonprofit hospitals do not provide enough charity care to justify the tax-exemption. In this paper, I instead focus on other metrics of a hospital's benefit to its community, aside from charity care. These metrics include the populations a hospital serves, the quality outcomes of the hospital, and the impact of the hospital's efforts detailed in its Community Health Needs Assessment. The results indicate that, for most metrics analyzed, nonprofit hospitals do not provide a significantly greater community benefit than forprofit hospitals. Nonprofit hospitals serve higher-income populations and do not serve more racially diverse counties, nor counties with lower rates of insurance. Additionally, nonprofit hospitals do not necessarily take on a greater portion of their revenue from Medicaid and Medicare patients. My analysis found no clear impact of hospital's efforts to resolve certain community issues such as substance abuse. However, these hospitals have more positive quality outcomes in terms of readmission rates. These findings help answer the question of if nonprofit hospitals benefit their communities to a greater extent than their for-profit counterparts.

ABSTRACT Controversy over the tax-exemption of nonprofit hospitals has been repeatedly discussed over the past decade. Many question whether the exemption is justified by hospitals’ benefit to communities (Al-Agba, 2017). Past studies have analyzed this issue by focusing on charitable care, or uncompensated care, as a measure of community benefit (Bai, et al., 2021; Bruch and Bellamy, 2020). Most findings conclude that nonprofit hospitals do not provide enough charity care to justify the tax-exemption. In this paper, I instead focus on other metrics of a hospital’s benefit to its community, aside from charity care. These metrics include the populations a hospital serves, the quality outcomes of the hospital, and the impact of the hospital’s efforts detailed in its Community Health Needs Assessment. The results indicate that, for most metrics analyzed, nonprofit hospitals do not provide a significantly greater community benefit than forprofit hospitals. Nonprofit hospitals serve higher-income populations and do not serve more racially diverse counties, nor counties with lower rates of insurance. Additionally, nonprofit hospitals do not necessarily take on a greater portion of their revenue from Medicaid and Medicare patients. My analysis found no clear impact of hospital’s efforts to resolve certain community issues such as substance abuse. However, these hospitals have more positive quality outcomes in terms of readmission rates. These findings help answer the question of if nonprofit hospitals benefit their communities to a greater extent than their for-profit counterparts. 2 TABLE OF CONTENTS ABSTRACT ................................................................................................................................ 2 1. INTRODUCTION .............................................................................................................. 4 2. LITERATURE REVIEW ................................................................................................... 9 3. DATA ............................................................................................................................... 11 4. ANALYSIS ....................................................................................................................... 14 4.1 Median Income of County ........................................................................................ 15 4.2 Minority Populations in County................................................................................ 16 4.3 County Insurance Rates ............................................................................................ 18 4.4 Revenue Mix ............................................................................................................. 19 4.5 Quality Outcomes ..................................................................................................... 21 4.6 CHNA Impact ........................................................................................................... 24 4.7 Hospital Revenue ...................................................................................................... 27 5. CONCLUSION ................................................................................................................. 29 6. APPENDIX ....................................................................................................................... 30 7. REFERENCES ................................................................................................................. 40 3 1. INTRODUCTION Background In 2010, the Illinois Supreme Court ruled that Provena Medical Center, a nonprofit hospital in Urbana County, Illinois, did not provide enough benefit to the community to justify its property tax exemption (State of Illinois Supreme Court, 2010). The ruling was based on the finding that the value of the hospital’s tax exemption was $268,276 more than the value of the charity care provided by the hospital to the community. More recently in 2017, the Internal Revenue Service revoked the tax-exemption of an undisclosed nonprofit hospital. The IRS cited the hospital’s inability to prove significant community benefit to justify the exemption (Ellison, 2017). These two incidences are reflective of the controversy prevalent in the last few decades regarding nonprofit hospital tax-exemptions. Many question whether nonprofit hospitals provide enough additional benefit to their communities compared to for-profits to justify a tax-exemption (Al-Agba, 2017). This paper analyzes several metrics of community benefit provided by hospitals in the United States and seeks to find a relationship between these metrics and the ownership status of hospitals. Of the more than 6,000 hospitals in the nation, over 60% of them are listed as nonprofit (Cheney, 2017). Despite the nonprofit title, many of these institutions report large sums of profit. These profits, representing the financial sum generated when revenues exceed business expenses, are used differently between nonprofit and for-profit hospitals. For-profit hospitals issue returns to investors. Their nonprofit counterparts instead reinvest the excess returns back into hospital operations, often used for expansions and improvements. The other difference between the types of ownership is that for-profit hospitals pay property, income, and sales taxes. All nonprofit hospitals have a tax-exemption from the IRS, and do not pay these taxes. Both designations of 4 hospitals serve the healthcare needs of their communities, participate in medical research, and provide teaching opportunities for medical students and professionals. Both must provide charity care, or treatment subsidized for those in need. Both designations are obligated to provide the same standard of care and aim for the same quality outcomes defined by the federal government (Centers for Medicare and Medicaid Services, 2020). In 1956, the United States Internal Revenue Service (IRS) issued Revenue Ruling 56-185 that specifically allowed nonprofit hospitals to be exempt from income and property taxes (Maseo, 2019). The IRS continues to uphold the exemption as long as hospitals claim a defined charitable purpose. These regulations reflect an expectation that hospitals exist to serve the needy, not to make profits. Though many organizations do charitable work or have a positive impact on society, the tax-exemption comes when the purpose of the organization is charitable. Therefore, in order to justify a tax-exemption, nonprofit hospitals must have a charitable purpose of healing and not a purpose of making profit. However, the IRS began to change regulatory language to allow for more profitable practices in nonprofit hospitals while still holding the exemption. The 1956 Revenue Ruling 56185 that required hospitals to offer care to everyone was replaced by Ruling 69-545 in 1969 that permitted some restriction of care based on ability to pay and insurance coverage. In 1983, Revenue Ruling 83-157 was passed, allowing states to discontinue emergency services in select hospitals, further allowing for low-income and uninsured patients to be denied care. One by one, these revenue rulings allowed for hospitals to pull away from the previous expectation of caring for everyone to becoming a payment-oriented and profitable organization. As the profits of the medical industry grew exponentially, questions arose regarding the responsibilities of hospitals. 5 Were they responsible for the issues in their communities? Should they be required to commit resources to impactful programs and outreach? Today, these questions surrounding the community benefit that nonprofit hospitals provide, or lack thereof, have led to much controversy in the health care field. Many argue that nonprofit hospitals are too profitable and do not provide enough community benefit to justify the tax exemption. Others respond that without the tax exemption, nonprofit hospitals would not be financially stable enough to remain in operation and many who rely on these hospitals in underserved communities would lose access to care. In regard to the level at which nonprofit hospitals benefit their communities, many hospitals point to the amount of charity care, or uncompensated care, provided in a year. This financial metric describes how much care, in US dollars, was provided to patients that had no means of payment. These treatments are covered by the hospital, hence the name “uncompensated” care. Many studies over the past decade have sought to measure the difference in uncompensated care provided by nonprofit hospitals and for-profit hospitals. Discussion of Analysis In this paper, I analyze other key metrics of community benefit provided by hospitals in the United States. An important aspect of a hospital’s community benefit lies in the demographics of the community served. If nonprofit hospitals serve communities with lower median incomes, greater racial and ethnic diversity, and lower rates of health insurance, they are benefitting the community, often at a greater expense to the hospital. However, my findings indicate that nonprofit hospitals serve counties with higher median incomes than for-profit hospitals, and they do not serve larger populations of racial minorities. 6 Similarly, there are many patients in the United States who are unable to pay for their care because of a lack of insurance. In many cases, hospitals treat these patients, but they are not compensated for it. Hospitals that are located in counties with lower rates of insurance are at a greater financial risk for not being compensated yet are providing critical benefit to these communities. My findings suggest that nonprofit hospitals do not serve communities with significantly lower rates of insurance than for-profit hospitals, and therefore, they do not provide additional benefit to communities in need in terms of insurance rates. Hospitals can also benefit their community by treating a greater portion of patients on public insurance. For the purposes of this paper, we look at percent of revenue that comes from Medicare and Medicaid reimbursements. Due to the lower reimbursement rates of these plans, serving these patients also negatively impacts the hospital’s revenue. However, often these patients need a greater level of care and treating them benefits the community. I find that nonprofit hospitals do not attribute a significantly different portion of their revenue to patients with Medicare, and they actually report a smaller portion of revenue from Medicaid than their for-profit counterparts. In the case of serving patients with public insurance, these findings suggest that nonprofit hospitals do not provide additional community benefit. A hospital’s ability to provide quality care to its patients is a large factor in how well it serves the community. Centers for Medicare and Medicaid Service penalize hospitals with excessively negative outcomes, including high readmission rates, when patients are treated and then return soon after for a related cause, and hospital acquired conditions, when a patient contracts a condition due to conditions at the hospital. Hospitals with higher quality and lower penalty rates are of greater benefit to their community. My analysis finds that nonprofit hospitals have fewer penalties for readmissions, which suggests a higher quality of care. However, the 7 difference between nonprofit and for-profit hospitals in terms of penalties for hospital acquired conditions is not statistically significant. Nonprofit hospitals are required to submit Community Health Needs Assessments in order to maintain a tax exemption. These assessments contain an evaluation of the issues within a hospital’s community and are submitted along with an implementation plan regarding how the hospital plans to address these issues. The ability of a hospital to resolve issues or improve conditions in their community is a measure of their community benefit. My analysis to evaluate the impact of these assessments differs from the rest in terms of methodology, which is detailed in the analysis section. My findings find no statistically significant impact of hospitals efforts in the communities they serve as planned in these Community Health Needs Assessments. Finally, though hospital revenue is not a direct measure of community benefit, I did include revenue as a separate analysis. My intent was to learn more about nonprofit hospitals’ ability to allocate resources toward community benefit in comparison to that of for-profit hospitals. Though hospital profitability would be a more ideal measure, revenue was a more reliable reported metric. I found that, holding all else constant, nonprofit hospitals receive higher revenues per bed than for-profit hospitals. They also have higher total revenues on average. This may indicate that nonprofit hospitals have the financial means to allocate equal or greater funds toward community benefit than for-profit hospitals. These metrics of community benefit are an important addition to the conversation of the tax-exemption justification and help one to understand if nonprofit and for-profit hospitals differ in their benefit to their community. 8 2. LITERATURE REVIEW One metric used to determine the level of community benefit provided is charity care. Charity care is an expense that occurs when a patient is underinsured and has no means to pay for necessary treatments. In 2019 alone, hospitals in the United States reported $41.6 in charity care (American Hospital Association, 2021). The institutions are not compensated for this treatment but are incentivized to provide it in order to prove a community benefit. At times, this value may be inflated when bad debt, an expense occurring when hospitals expect a payment that never comes, is included. Both nonprofit and for-profit hospitals provide this charity care. A recent study found that nonprofit hospitals did not provide more charity care than their public and for-profit counterparts. For every $100 in total expenses incurred, nonprofit hospitals spent $2.30 on charity care, compared to $4.10 in public hospitals and $3.80 in for-profit hospitals. The authors state that this level of charity care provided by nonprofit hospitals does not align with their favorable tax exemption (Bai, et al. 2021). A 2018 study also found that charity care, or uncompensated care, does not differ significantly between nonprofit and for-profit hospitals. In 2018, nonprofit hospitals spend on average 2.95% of their total expenses on charity care, compared to 2.62% of for-profit hospital total expenses. However, this difference is not statistically significant. It is important to note that charity care varies by institution and geography (Bruch and Bellamy, 2020). An earlier study in 2002 found that a gap of over $800 million exists between the value of the tax-exemption and the value of charity care provided by private, nonprofit hospitals. Most hospitals provide a value of charity care less than the value of the tax-exemption. This held true even when 50 percent of the hospital’s bad debt was included in the charity care value. However, this gap is not consistent across all communities. In areas with lower incomes and greater 9 poverty rates, hospitals provided more charitable care than the tax-exemption. The opposite is true of hospitals in wealthy areas that provide much less charity care than their tax-exemption (Kane and Wubbenhorst, 2002). Researchers have also quantified the efforts made by nonprofit hospitals in their community, including hospital services, access to emergency care, involvement in health professionals’ education, and price discounts, among other measures. They found that there is approximately a 20 percent deficit in community benefits by nonprofit hospitals compared to the value of the tax-exemption. Even in hospitals displaying a larger contribution of uncompensated care, it is often apparent after a deeper inspection that the prices or services provided are exorbitantly marked up, making the dollar amount of uncompensated care inflated. They state that without more regulation, nonprofit hospitals will continue to fall short in helping their community enough to justify a tax exemption (Nicholson et al., 2000). These studies conclude that nonprofit hospitals are not supplying enough charity care to their communities to match the value of the tax-exemption. Though several of these studies were conducted before the Affordable Care Act was passed in 2009, adding more requirements for these hospitals, it can be reasonably assumed that the general conclusions still stand as more recent studies align with their findings. These findings present a barrier to nonprofit hospitals justifying their tax-exemption based solely on their level of charity care. However, proponents of the tax-exemption point out that charity care is not the only means in which nonprofit hospitals serve their communities. The findings of these studies are important to consider when evaluating the issue of nonprofit hospitals’ community benefit in comparison to their for-profit counterparts, and how this evaluation contributes to understanding if the tax exemption is justified. However, research 10 is lacking when trying to understand the overall benefit to a community provided by a hospital, aside from the narrow metric of charity care or even community benefit expenditures. Additionally, the literature lacks an analysis of the direct impact of CHNAs and their effectiveness in improving the community. This paper seeks to address those gaps and strengthen the research surrounding this subject. 3. DATA I collected demographic data at the county level from Torch Insight, an analytics tool and database operated by Milliman. This data includes racial demographics, such as percent of the county population that is black or non-white, non-black (effectively, “other”). It also includes insurance breakdown information such as percent of the county population with no insurance or public insurance. The county demographic data of median income and population was downloaded from the US Census Bureau. Hospital data, including financial metrics and ownership designations, was collected from Torch Insight. Outcome data, including penalties for negative outcomes at the hospital level, was collected from the Kaiser Family Foundation. Data on overdose death rates was collected from the Centers for Disease Control and Prevention’s WONDER database. I conducted the analysis on nonprofit hospitals’ Community Health Needs Assessments (CHNAs) using CHNA documents found on individual hospital websites. All data collected in this paper reflect the period of 2013 to 2019. 5,246 hospitals across 50 states were included in the regressions. 1,117 of these hospitals are public hospitals operated by the state, 2,646 of these hospitals are nonprofit hospitals, and 1,483 of these hospitals are for-profit hospitals. As the purpose of this research is to compare the benefits of private nonprofit and private for-profit hospitals, public hospitals are primarily used 11 4. ANALYSIS As discussed in Section 1, my analysis seeks to understand the relationship between hospital ownership (primarily nonprofit in comparison to for-profit) and key metrics of community benefit. I evaluate this benefit across seven metrics: median county income, racial demographics, revenue mix (primarily rates of public insurance), county health insurance rates, quality outcome measures, CHNA impacts, and hospital revenue. To conduct these analyses, I run logistic and linear regressions using the statistical software Stata. Logistic regressions are used when the dependent variable is binary. This occurs in my analyses on median income, racial demographics, and insurance rates. Because I seek to determine if these variables play a role in determining if nonprofit hospitals elect to be located in a county, they are used as the independent variables. The dependent variable, nonprofit (indicating nonprofit hospital ownership) is binary, so logistic regressions are used in the analysis. In these models, the log odds of the outcomes are modeled as a linear combination of the predictor variables. Logistic models fit these analyses better, because they do not violate homoskedasticity nor Ordinary Least Squares (OLS) assumptions. In regressions where the outcome variable is continuous, and not binary, multiple linear regressions are used. This occurs in my analyses of revenue mix, CHNA impact, and hospital revenue. I use these models to determine the value of dependent variables based on the values of independent variables. For all regressions, the statistical null hypothesis is that nonprofit hospitals do not differ significantly from for-profit hospitals across the various metrics. This is different only for the CHNA Impact analysis, in which the hypothesis is that CHNAs with a focus on substance abuse have an impact that does not differ significantly for CHNAs without this focus. I drew conclusions on the community benefit provided by nonprofit hospitals by determining if this 14 hypothesis was rejected in the regression or held true. Hypotheses were rejected if the p-value was lower than 0.10, 0.05, or 0.01 and statistical significance was then found at the confidence levels of 10%, 5%, or 1%, respectively. In efforts to limit bias, I ensured that my dataset represented hospitals from all 50 states in both urban and rural areas. Few observations were dropped from the dataset consisting of all hospitals due to a lack of data. 4.1 Median Income of County I begin by analyzing the relation between county-level income measures and the organizational form of hospitals. The intention behind this analysis is determining if nonprofit hospitals serve communities with greater low-income populations in need. If nonprofit hospitals serve lower-income populations than for-profit hospitals, that indicates a greater level of benefit provided. I ran several regressions: 1) a logit regression with only the dependent binary variable of nonprofit ownership and the independent variable of median income, 2) a logit regression with added control variables for racial demographics, and 3) a logit regression with absorbed state fixed effects. This is particularly important to control for factors that differ by state, such as the status of Medicaid expansion. The null hypothesis is that nonprofit hospitals do not differ significantly from for-profit hospitals in median income of counties served. Summary statistics indicate that the median county income in which a nonprofit hospital is located is, on average, $60,293. The median county income in which a for-profit hospital is located is, on average, $59,519. Nonprofit hospitals are located in counties with an average median income of $774 higher than for-profit hospitals. 15 Summary statistics indicate that nonprofit hospitals serve counties with, on average, 12% of the population identifying as black and 5% identifying as non-black and non-white (“other”). For-profit hospitals serve counties with, on average, 10% of the population identifying as black and 5% identifying as non-black and non-white. Logit regressions displayed in Tables 2 and 3 were conducted to determine the difference in racial diversity within the communities that nonprofit hospitals serve in comparison to forprofit hospitals. A simple regression (2a) with percent black as the independent variable and nonprofit as the dependent variable, indicates that a 1% increase in the percent of the county identifying as black is associated with a 0.364 increase in the log odds that a nonprofit hospital is located in the county. This finding is significant at the 5% level. As county demographics were added to the regression (2b) to control for median income and other racial factors, this effect remains statistically significant and increases slightly to an increase of 0.384. However, as state fixed effects are added (2c), this increase remains unchanged, but the impact becomes statistically insignificant. Therefore, holding all else equal, nonprofit hospitals do not serve counties with a statistically different percent of the population identifying as black compared to for-profit hospitals. Regressions with percent non-white, non-black as the independent variable show us that a 1% increase in the percent of the county identifying their race as “other”, or non-white, nonblack, is associated with an increase in log odds of 0.571 that a nonprofit hospital is located in the county (3a). This finding is significant at the 10% level. As county demographics were added to the regression (3b) to control for median income and other racial factors, this effect remains the same. As state fixed effects are added (3c), the impact becomes statistically insignificant at 17 that nonprofit hospitals do not differ significantly from for-profit hospitals in percent of revenue from Medicaid and Medicare. On average, 21% of a nonprofit hospital’s revenue is from Medicare, and 12% is from Medicaid. This is compared to 21% of a for-profit hospital’s revenue is from Medicare and 13% is from Medicaid. Linear regressions, displayed in Tables 5 and 6, were conducted to determine the difference that a hospitals nonprofit status has on revenue mix. A simple regression (5a) with percent revenue Medicare as the dependent variable and nonprofit as the independent variable, with public as a control, indicates that nonprofit hospitals are associated with an insignificant decrease of 0.1% in percent of revenue coming from Medicare. As more variables were added to the regression (5b) to control for hospital size, county insurance rates, number of alternative hospitals in the county, and other county demographics, this effect remains insignificant and decreases to a minimal impact. As state fixed effects are added (5c), the insignificant impact increases to a decrease of 3%. The percent of the county’s population with public insurance, the median income, and the number of public hospitals in the county are also significant variables impacting the percent of revenue from Medicare. These results allow us to conclude that nonprofit hospitals do not serve a significantly greater portion of Medicare patients than forprofit hospitals. A simple regression (6a) with percent revenue Medicaid as the dependent variable and nonprofit as the independent variable, with public as a control, indicates that nonprofit hospitals also do not serve a greater portion of Medicaid patients. Nonprofit hospitals are associated with an insignificant decrease of 0.5% in percent of revenue coming from Medicaid. As more variables were added to the regression (6b) to control for hospital size (measured by number of 20 determine if nonprofit hospitals were penalized less than their nonprofit counterparts, indicating higher quality outcomes. If so, this would demonstrate nonprofit hospitals’ added benefit provided. The null hypothesis is that nonprofit hospitals do not differ significantly from forprofit hospitals in penalties for readmissions and hospital acquired conditions. On average, nonprofit hospitals in the United States were penalized 0.58% for readmissions in 2018, compared to 0.72% of for-profit hospitals. 14% of nonprofit hospitals were penalized for hospital acquired conditions in 2018, compared to only 9% of for-profit hospitals. Regressions, displayed in Tables 7 and 8, were conducted to determine the difference in outcomes between nonprofit and for-profit hospitals. The variable measuring readmission penalties is a continuous variable indicating by what percent payments from CMS were reduced due to readmissions. Because the variable is continuous, linear regressions were used. A simple regression (7a) includes 2018 readmission penalties as a dummy dependent variable and nonprofit as the independent variable, with public as a control along with previous years penalty variables. Penalty variables from previous years were included to ensure that the impact of hospital ownership on penalties was isolated from any effects of previous year performance. This regression indicates that nonprofit hospitals are associated with a decrease of 5% in the amount of penalization, significant at the 5% level. As more variables were added to the regression (7b) to control for public insurance factors, this effect decreases to 4% and remains significant at the 10% level. As state fixed effects are added (7c), the significant effect remains the same. Previous year penalties and percent of revenue from Medicaid are significant variables as well. My findings indicate that nonprofit hospitals are penalized less often than for-profit hospitals for readmissions and therefore have better outcomes in this area. 22 4.6 CHNA Impact For nonprofit hospitals to obtain their tax-exemption afforded to them by the IRS, they must conduct an analysis of the needs of their community, structure a plan to address identified issues, and submit a Community Health Needs Assessment (CHNA). A CHNA evaluates specific needs of a community and then holds hospitals accountable to acting on them. The hospital must describe what resources they will allocate to resolve the issue. Failure to file a CHNA will result in a nonprofit hospital losing their tax-exempt status. An effective CHNA would be followed by a change in the community over the next few years according to the plan set in place by the hospital. These CHNAs may be a valuable resource in determining a hospital’s community benefit. However, if there is not a demonstrated impact of the hospital’s efforts as detailed in the CHNAs, these reports may not be an effective way of measuring or regulating community benefit provided by nonprofit hospitals. The intention of this analysis is to determine if there is an impact of the CHNAs. In order to conduct this analysis, I identified substance abuse as an issue commonly addressed in CHNAs. I then sought to measure the change in substance abuse overdoses in a community five years after the CHNA was filed for the hospital in the county. The null hypothesis is that counties with a hospital addressing substance abuse in its CHNA do not differ significantly from counties not addressing the issue in terms of substance abuse overdoses. To determine the effectiveness of Community Health Needs Assessments (CHNAs), I compiled 335 of the reports filed in 2013 from hospitals around the community. The remaining hospitals filed their reports in a different year, had not begun the process of submitting CHNAs as of 2013, or their report was not readily available. CHNAs were limited to hospitals located in 24 “one hospital counties” in that there are no competing hospitals in the county. This was done to isolate the effect of a hospital’s CHNA efforts from any effects of competing hospitals’ efforts. Of these CHNAs, 203 of them mention “substance abuse” within the document, as counted by a PDF-to-text software. These 203 reports mention “substance abuse” an average of six times throughout the document. Assuming that CHNAs mentioning substance abuse a minimal number of times did not identify decreasing substance abuse as an area of focus within the community at that time, I generated a variable substance abuse focus designating CHNAs that mention “substance abuse” more than five times as focusing significantly on the issue. This threshold was determined after thoroughly reading a subset of the CHNAs and identifying key focus issues. I then used CDC overdose data to track the change in opioid overdoses in those counties from 2013 to 2018, the five years after the report was filed. Once this setup was complete, I ran a robust linear regression on the data as done with other metrics in this paper, using opioid focus as the independent variable and overdose change as the dependent variable. A variable describing overdose deaths per 100k population in 2013 was added to control for the level of substance abuse prevalence already in the county. The regressions are limited to a dataset with hospitals filing a CHNA report in 2013 with no competing hospitals for reasons stated above. On average, hospitals whose CHNAs focused on substance abuse saw an increase in overdose deaths from 2013 to 2018 of ten deaths per 100k population, compared to an increase of 7.5 deaths per 100k population for those that did not focus on substance abuse. However, this effect could occur due to the fact that hospitals focusing on substance abuse in their CHNAs are likely located in areas where substance abuse is a greater issue. These linear regressions, displayed in Table 9, were conducted to determine the impact that a hospitals’ efforts to decrease substance abuse has on the rate of overdose deaths within a 25 4.7 Hospital Revenue Though revenue is not a direct indication of a hospital’s community benefit, it is related to a hospital’s ability to direct resources to community benefit activities and efforts. Hospital profitability is an ideal metric to measure this ability as it accounts for expenses, but total revenue was a more accessible and consistent datapoint than general profitability. I sought to determine whether nonprofit hospitals had greater or less revenue than for-profit hospitals, in an effort to gauge nonprofit hospitals ability to allocate funds toward community benefit. The null hypothesis is that nonprofit hospitals do not differ significantly from for-profit hospitals in total revenue. Summary statistics indicate that nonprofit hospitals have on average, an annual revenue $96.3 million, compared to $47.6 million for for-profit hospitals. These large differences in revenue are likely due to a number of factors, including hospital size. To account for the varying size of hospitals across the United States, I generated a variable revenue per bed, representing the total hospital annual revenue divided by the number of beds within the hospital (an indication of size). Linear regressions, displayed in Table 10, were conducted to determine the difference in comparable revenue between nonprofit and for-profit hospitals. A simple regression (10a) with revenue per bed as the dependent variable and nonprofit as the independent variable, with public as a control, indicates that nonprofit hospitals have a decrease in revenue per bed of $37,315 annually compared to for-profit hospitals, statistically significant at the 5% level. As variables controlling for insurance mix are added to the regression (10b), this relationship flips to indicate an increase in revenue per bed of $92,066 compared to for-profit hospitals, statistically significant at the 1% level. As more variables are added (10c) to control for county 27 5. CONCLUSION Though previous studies have found no statistically significant difference in charitable care provided by nonprofit hospitals to that by for-profit hospitals, it is important to look holistically at additional factors leading to hospitals’ community benefit. However, my analysis supports the claims of past researchers that nonprofit hospitals do not benefit the community significantly more than their for-profit counterparts. The indications that nonprofit hospitals serve communities with higher incomes, and do not serve populations with a greater portion of uninsured or racially diversity support claims that they do not provide additional benefit to the communities in which they are located. Additionally, my analyses indicate that nonprofit hospitals do not treat more Medicare and Medicaid patients as a portion of revenue, and there is no significant evidence that their Community Health Needs Assessments are effective, at least in decreasing rates of substance abuse. However, it is also significant to consider that nonprofit hospitals are associated with fewer penalizations for readmissions than for-profit hospitals, benefitting the community by decreasing the number of visits to the hospital. It is not clear if nonprofit hospitals are penalized less often for readmissions because they do provide a higher quality of care, or because of other reasons. The finding that nonprofit hospital status is associated with more revenue generated per bed and higher total revenues on average than for-profit hospitals indicates that nonprofit hospitals do have potential resources to allocate to community benefit. This analysis does not take into account expenses but is interesting to consider along with the other findings in this paper. However, more research is needed to contribute to the discussion on the benefit of nonprofit hospitals and the justification of the tax-exemption. These results are not exhaustive of all critical factors in this issue but highlight areas in which nonprofit hospitals do and do not benefit the community in ways significantly greater than their for-profit counterparts. 29 7. REFERENCES Al-Agba, N. (2017, April 25). The Fairy Tale of a Non-Profit Hospital. Retrieved from https://thehealthcareblog.com/blog/2017/04/25/the-fairy-tale-of-a-non-profit-hhospital/ Bai, G., et. al. (2021, April). Analysis Suggests Government and Nonprofit Hospitals’ Charity Care Is Not Aligned with Their Favorable Tax Treatment. Health Affairs. Retrieved from: https://www.healthaffairs.org/doi/abs/10.1377/hlthaff.2020.01627 Bakken, E., & Kindig, D. (n.d.). Could Hospital Community Benefit Enhance Community Health Improvement? WMJ. https://citeseerx.ist.psu.edu/viewdoc/download? doi=10.1.1.681.8245&rep=rep1&type=pdf Bruch, J. D., & Bellamy, D. (2020). Charity Care: Do Nonprofit Hospitals Give More than ForProfit Hospitals? Journal of General Internal Medicine. Carlton, E. L., & Singh, R. (2018, May). Joint Community Health Needs Assessments as a Path for Coordinating Community-Wide Health Improvement Efforts Between Hospitals and Local Health Departments. American Journal of Public Health. https://ajph.aphapublications.org/doi/full/10.2105/AJPH.2018.304339 Chaudhry. (n.d.). The Affordable Care Act and Community Benefit. Retrieved from Association of Maternal and Child Health Programs website: http://www.amchp.org/AboutAMCH P/Newsletters/Pulse/SeptOct2016/Pages/The-Affordable-Care-Act-and-Community-B enefit.aspx Cheney, C. (2017, June 20). Top 5 Differences Between NFPs and For-Profit Hospitals. Retrieved from https://www.healthleadersmedia.com/finance/top-5-differences- betweennfps-and-profit-hospitals 40 Evans, M. (2012, January 14). Juggling the Lineup. Retrieved from https://www.modernhealth care.com/article/20120114/MAGAZINE/301149993/juggling-the-lineup Exempt Purposes - Internal Revenue Code Section 501(c)(3). (n.d.). Retrieved from https://www.irs.gov/charities-non-profits/charitable-organizations/exempt-purposesinternal-revenue-code-section-501c3 Herring, B., Gaskin, D., Zare, H., & Anderson, G. (n.d.). Comparing the Value of Nonprofit Hospitals' Tax Exemption to Their Community Benefits. US National Library of Medicine: National Institutes of Health. History of Hospitals. (n.d.). Retrieved March 18, 2020, from Penn Nursing website: https://www. nursing.upenn.edu/nhhc/nurses-institutions-caring/history-of-hospitals/ Intermountain LDS Hospital. (n.d.). LDS Hospital Implementation Strategy Plan 20202022. Kane, N. M., and W. H. Wubbenhorst. 2000. Alternative Funding Policies for the Uninsured: Exploring the Value of Hospital Tax Exemption. Milbank Quarterly 78:185 – 212. Lee, B. Y. (2016, May 8). Very Profitable Nonprofit Hospitals...But Where Are the Profits Going? Retrieved from https://www.forbes.com/sites/brucelee/2016/05/08/veryprofitable-nonprofit-hospitals-but-where-are-the-profits-going/#f1cf2f836b22 Luthra, S. (n.d.). Obamacare Pushed Nonprofit Hospitals to Do Good Beyond Their Walls. Now What? Retrieved 2017, from Kaiser Health News Masseo, P. (n.d.). IRS rule changes helped create a payment-focused hospital system. Retrieved from https://www.statnews.com/2019/03/05/irs-rule-changes-payment-driven- hospitalsystem/ Memorial Sloan Kettering Cancer Center. (n.d.). 2018 Community Service Plan Update. 41 Nicholson, S., M. V. Pauly, L. R. Burns, A. Baumritter, and D. A. Ash. 2000. Measuring Community Benefits Provided by For-Profit and Nonprofit Hospitals. Health Affairs 19 Rosenbaum, S., & Margulies, R. (n.d.). Tax-Exempt Hospitals and the Patient Protection and Affordable Care Act: Implications for Public Health Policy and Practice. Rosenthal, E. (n.d.). An American Sickness: How Healthcare Became Big Business and How You Can Take It Back. Stanford Healthcare & Community Partnership Program. (n.d.). Partnering to Improve: FY2017-2019 Implementation Strategy. Valdovinos, E., Le, S., & Hsia, R. Y. (2015, August). In California, Not-For-Profit Hospitals Spent More Operating Expenses on Charity Care Than For-Profit Hospitals Spent. Health Affairs. https://www.healthaffairs.org/doi/full/10.1377/hlthaff.2014.1208 What Is a Community Health Needs Assessment (CHNA)? (n.d.). Retrieved from American Lung Association website: https://www.lung.org/our-initiatives/tobacco/cessation-andprevention/hospital-community-benefits/what-is-a-community-health-needs-assessment Young, G. J., Desai, K. R., & Lukas, C. V. (1997, February). Does the Sale of Nonprofit Hospitals Threaten Health Care for The Poor? Retrieved from https://www.health affairs.org/doi/full/10.1377/hlthaff.16.1.137 Young, G. J., Flaherty, S., Zepeda, E. D., Singh, S. R., & Cramer, G. R. (2018, January). Community Benefit Spending by Tax-Exempt Hospitals Changed Little After ACA. Health Affairs. https://www.healthaffairs.org/doi/10.1377/hlthaff.2017.1028 42