Approximately 25% of all Medicare beneficiaries discharged to a skilled nursing facility (SNF) are readmitted to a hospital within 30 days, costing an estimated $4.3 billion annually.1 With the implementation of value-based reimbursement, including the Hospital Readmissions Reduction Program (HRRP), many hospitals have implemented strategies to better manage patients who use SNF services after inpatient discharges. Recent evidence suggests that hospitals may be able to reduce readmissions through SNF networks, which describes when hospitals develop loose relationships with certain SNFs in their communities and target patient referrals to such facilities.2 Meanwhile, emerging evidence also shows that hospital-led SNF networks are not the only strategy hospitals adopt to deal with unnecessary readmissions of patients who use SNFs, and that hospitals may also choose to vertically integrate.3 Vertical integration, defined as “an arrangement whereby the organization offers, either directly or through others, a broad range of patient care and support services operated in a functionally unified manner,” potentially including services such as preacute, acute, and postacute care, organized around a hospital.4 Recent analyses shows that following the Affordable Care Act (ACA), more hospitals vertically integrated into SNF.5 Hospitals which vertically integrate into SNFs may do so by acquiring an SNF that enables them to develop a hospital-based SNF.
Research examining the intersection between hospital vertical integration into SNF and outcomes has focused on individual measures of patient quality outcomes.2,6–10 Although such research focusing on individual-level patient outcomes provides valuable insight into understanding hospital vertical integration into SNF and how individual patients may be managed, we believe this only tells part of the story of the impact of hospital vertical integration and a readmissions rates, which have become an ever important organizational level clinical outcome measure as a result of the ACA. To date, no study has specifically focused on the relationship between hospital vertical integration into SNF and hospital readmission rates, which reflects quality of care outcomes for an entire population of patients. Examining organizational-level hospital readmission rates enables us to better understand the managerial implications of vertical integration. Previous studies examining patient-level outcomes are testing the relationship between hospital vertical integration into SNF and clinical patient outcomes. By examining the hospital readmission rate, we are also able to understand whether vertical integration may have implications for how organizations manage the entire population of patients that transition to SNFs to reduce readmission rates experienced. In light of the financial penalties associated with 30-day readmission rates, it is critical to examine this outcome when understanding hospital vertical integration into SNF.
We examined the impact of hospital vertical integration into SNF on 30-day pneumonia readmission rates and 30-day heart failure (HF) readmission rates among acute care hospitals in the United States. We used a panel data of all US hospitals that included merged data from the American Hospital Association (AHA) Annual Survey, Medicare Hospital Compare data, and Rural Urban Commuting Codes (RUCA) between 2008 and 2011. Using hospital and year fixed-effects models, we examined the relationship between vertical integration into SNF and 30-day pneumonia and HF readmission rates and then explored whether vertical integration is differentially related to readmissions in rural/urban, for-profit/nonprofit, and system-affiliated versus stand-alone hospitals. This study provides critical knowledge for hospital administrators by informing them on the organizational-level impacts associated with hospital vertical integration into SNF.
Retrospective, longitudinal panel data on vertical integration was obtained from the AHA annual survey, combined with CMS' Hospital Compare and the RUCA codes. We modeled the impact of vertical integration on hospital readmissions using hospital and year fixed effects for all general acute care hospitals in the United States between 2008 and 2011. The study sample included hospitals that were or became vertically integrated during the study period and hospitals that were not vertically integrated during any years of study period.
The dependent variables of interest were hospital 30-day pneumonia readmission rates and 30-day heart failure readmission rates, which were available on the CMS Hospital Compare website. CMS calculates a case-mix adjusted hospital-wide 30-day readmissions for pneumonia and HF. Our primary analyses modeled the impact of hospital vertical integration into SNF on 30-day readmissions for both pneumonia and HF using hospital and year fixed effects. Our primary analyses included running two models, one for each independent variable. Specifically, our models took the following functional form:
, where yit is the dependent variable (30-day readmission rate) where i = hospital and t = time; β1 is the coefficient for hospital vertical integration into subacute care of the independent; variable (xit1); xit1 is hospital vertical integration into subacute care and is the unknown intercepts for a vector of hospitals; uit is the error term).
Consistent with previous work examining hospital vertical integration into SNF, vertical integration into SNF was obtained from the AHA annual survey.5,11 In each year study year, we used the variable where hospitals were asked to indicate whether their hospital had an SNF and hospitals that indicated yes and were deemed to be vertically integrated in a given year. The variable was binary; one being vertically integrated and 0 being not vertically. In secondary analyses, we examined whether hospital vertical integration into SNF was associated with a change in readmissions rates among different types of hospitals. Previous research examining hospital quality outcomes reported that organizational factors such as ownership, geographic location, and system membership may be associated with variation in quality outcomes. As a result, we ran stratified analyses by hospital characteristics including system membership, hospital geographic location (rural and urban), and hospital ownership type. Hospital ownership type was defined as investor-owned, not-for-profit and nonfederal, and governmental (all state and local municipality owned hospitals such as county hospitals). Federal hospitals, which include Veterans Affairs hospitals, US department of defense run hospitals were not included because such hospitals do not provide care for the general population, and the data for such institutions are not available.
We reported results as significant if the p values were less than .1, .05, or .001. All analyses were computed in the statistical package STATA, version 13.0, and in all cases, error terms were clustered within hospitals to account for repeated observations. The study was granted exemption status from the institutional review board of the second author's university.
When examining hospital vertical integration into SNF and 30-day HF readmission rates, there were 3,626 unique hospitals representing 11,328 hospital-year observations (Table 1). The mean hospital 30-day HF readmission rate was 22%. Among rural hospitals, the mean HF readmission rate was 21.85%, and pneumonia readmission rate was 17.56%. When examining hospital vertical integration into SNF and 30-day pneumonia readmission rates, there were 3,361 unique hospitals representing 12,007 hospital-year observations. The mean 30-day pneumonia readmission rate was 18%.
Urban hospitals reported a mean HF readmission rate of 22.04% and a mean pneumonia readmission rate of 17.88%. System-affiliated hospitals reported a mean HF readmission rate of 21.08% and a mean pneumonia readmission rate of 18.13%. NFP hospitals reported a mean HF readmission rate of 21.98% and a mean pneumonia readmission rate of 17.76%. Investor-owned hospitals report a mean HF readmission rate of 22.08% and a mean pneumonia readmission rate of 18.00%. Nonfederal governmental hospitals reported a mean HF readmission rate of 21.77 and a mean pneumonia readmission rate of 17.56%.
When examining the relationship between hospital vertical integration into SNF and readmission rates (Table 2), we found that vertical integration was associated with a reduction in hospital 30-day pneumonia readmission rates (β = −0.233, p = .039). Vertical integration into SNF was not significantly associated HF 30-day readmissions (β = 0.107, p = .553).
Next, we examined whether hospital vertical integration into SNF was associated with a change in readmissions rates among different types of hospitals. Rural hospitals that became vertically integrated into SNF were associated with a reduction in 30-day pneumonia readmission rates (β = −0.318, p = .012) but not associated with a reduction in 30-day HF readmissions (β = 0.241, p = .454). Vertical integration among urban hospitals was associated with a reduction in 30-day HF readmission rates (β = −0.576, p = .107). Similarly, when comparing investor-owned, not-for-profit, and nonfederal, governmental hospitals, not-for-profit hospitals that became vertical integrated into SNF were associated with a reduction in pneumonia 30-day readmission rates (β = −0.417, p = .004). Investor-owned and nonfederal governmental hospitals that became vertically integrated were not associated with pneumonia 30-day readmissions. Among investor-owned hospitals that vertically integrated into SNF were associated with an increase in HF 30-day readmissions (β = 0.974, p = .096). Not-for-profit and nonfederal governmental hospitals that vertically integrated were not associated with HF 30-day readmissions. When comparing system-affiliated hospitals and free-standing hospitals, neither hospital type was associated with pneumonia 30-day readmission rates or HF readmission rates.
Our study has several limitations. First, we used data from Hospital Compare for 30-day readmissions. Hospital Compare does not report readmissions data for hospitals if the reported outcome is below a certain volume threshold and instead provides a response of “too small to calculate,” eliminating some facilities in some years. We anticipate that this may impact certain hospital types (example, small hospitals) more than others, making it harder to understand the impact of vertical integration on these types of hospitals due to the absence of data. Next, we used vertical integration data from the AHA survey that relies on self-reported information that may be biased or inconsistent. As a result, there is a chance that organizations not represented in our data because they had missing data for certain years could provide a different perspective. Next, the independent variable only conveys the information related to the presence or absence of a vertically integrated SNF. It does not capture detailed information regarding how an SNF is vertically integrated and the nature of the coordination (for example, specific programs, protocols, services, location, etc.), which may impact outcomes associated with our dependent variable. Finally, our study was not able to identify the organizational differences that may be attributed to such differences in outcomes, and further research should focus on understanding the processes for each diseases that may be able to influence such outcomes.
Vertical integration into SNFs was associated with reduced pneumonia 30-day readmission rates among acute care hospitals, suggesting that when hospitals vertically integrate into SNF, they may be better able to manage pneumonia patients across the care continuum. Understanding the effects of hospital vertical integration of SNF on outcomes is complicated.10 Fragmentation of delivery between acute care and SNFs has potentially contributed to poor-quality outcomes such as readmissions, and vertical integration of different providers along the care continuum may reduce such fragmentation. Previous research suggests that through hospital vertical integration of SNFs, care may improve through enhanced patient hand-offs and more consistent quality of care pathways.12,13 Previous research shows that patients may experience better discharge outcomes when discharged to a vertically integrated postacute care setting.7,10,14 Vertical integration into SNFs was associated with reduced pneumonia 30-day readmission rates among acute care hospitals, suggesting that when hospitals vertically integrate into SNF, they may be better able to manage pneumonia patients across the care continuum. Our findings may suggest that through hospital vertical integration of SNF services, hospitals may be able to better support patient care processes that can improve quality outcomes of pneumonia patients transitioning to an SNF.
We also show that the effects of vertical integration varied by type of hospitals. Among rural hospitals, but not urban hospitals, vertical integration was associated with a reduction in 30-day readmissions rates for pneumonia patients. The most common reason for readmission for pneumonia patients is subsequent pneumonia.15 Previous research has found only a weak association between hospital location (urban vs. rural) and 30-day pneumonia readmissions.16 Rural hospitals face significant barriers to responding to the demands of the health care marketplace because they are often smaller, have a limited work force, and constrained financial resources.17–19 Although these factors could make it challenging to implement processes to reduce unplanned pneumonia readmissions, our study suggests that among rural hospitals, vertical integration may be able to facilitate environments that support processes that improve the likelihood of an unplanned readmission. Specifically, vertical integration may improve intrafacility communication, better manage care pathways, and training for managing patients at risk of being readmitted through follow-up procedures and continuity of care.
By contrast, among urban hospitals, vertical integration was associated with a reduction in HF 30-day readmissions (but not for pneumonia readmission). Research on HF disease management shows the complex nature of the disease and the need to use complex care management processes across the continuum. There is significant variation in strategies hospitals adopt to manage 30-day HF readmissions.20 Previous research identified 6 hospital strategies associated with lower 30-day HF readmission rates, which include (1) partnering with physicians, (2) partnering with other acute care hospitals, (3) having nurses responsible for medication reconciliation, (4) scheduling follow-up appointments before discharge, (5) establishing processes to provide discharge papers or electronic summaries to patient's primary physician, (6) assigning staff to follow-up with an laboratory test results after patients are discharged.21 One explanation for our findings is that vertical integration may contribute to these or other strategies being adopted by urban hospitals.
We also found that, among not-for-profit hospitals, vertical integration into SNF was associated with a reduction in the 30-day pneumonia readmission rates, but no relationship for investor-owned facilities. Previous research examining the impact of hospital ownership on patient safety outcomes has found inconsistent results.22 The findings of our study suggest that not-for-profit hospitals are able to take advantage of being vertically integrated. For example, upon vertically integrating a SNF facility, not-for-profit hospitals may not be constrained by a need to return profits to investors, enabling them to make investments necessary to most effectively integrate care for pneumonia, but not HF. These investments could include training staff members on processes and patient safety, integrating electronic medical records systems between the two facilities and creating a new policies and procedures for patients transferred between acute care and SNF. Recently published research has found that hospital ownership is consistently associated with the cost of care for postacute care.23 Another potential explanation for the variation among hospital ownership types may be based in the fact that not-for-profit hospitals more consistently participate in value-based purchasing programs, which for the most part are voluntary. Therefore, this participation may explain some of the variation we identified in our results.
One question from our results is why does vertical integration matter for pneumonia readmission rates and not for HF readmission rates? One possible explanation is that readmissions and the processes attributed to reducing the risk of being readmitted are different for HF and pneumonia, and such processes may not be impacted through a vertical integration strategy. Pneumonia and HF have different care pathways and operating procedures for managing patients that transition from acute care to SNF. One obvious difference between these two diseases is that pneumonia is episodic while HF is a chronic disease. Vertical integration may more easily facilitate the adoption of care strategies that improve outcomes for pneumonia episodes, but be less effective for mitigating unplanned readmissions for patients with chronic conditions.
Another potential explanation that may explain the lack of a statistically significant association between vertical integration into SNFs and HF readmissions is that hospital integration into SNFs may not be appropriate interventions to reduce 30-day HF readmissions.24 A literature review of interventions to reduce readmissions and identified predischarge interventions, postdischarge interventions, and interventions bridging the transition in which SNFs were not identified as a significant component associated with reducing readmissions. Hospital strategies they are associated with reducing HF readmissions that include partnering with community physicians, local hospitals, assigning nurses, and health care workers to be responsible for medication adherence, arranging follow-up appointments, and providing patients with test results upon discharge, and ensuring processes send health information to patient's primary physician.20 Although this study examining hospital strategies to reduce HF readmissions has not explicitly identified SNFs as a critical component, HF is prevalent in an estimated 20%–37.4% of individuals in SNFs and FH patients discharged to an SNF are at greater risk for being readmitted to the hospitals compared with patients who are discharged to their home.25–27 There remains a critical need to better understand how HF SNF patients are treated and the factors which may impact quality outcomes, which make it difficult to evaluate the role hospital ownership may play in influencing certain care processes.25
In conclusion, this study examined the relationship between hospital vertical integration into SNF and 30-day readmission rates for pneumonia and HF. We found that vertical integration was associated with a reduction in hospital 30-day pneumonia readmission rates. We found no statistically significant reduction in hospital readmission rates when examining the relationship between hospital vertical integration into SNFs and HF readmission rates. We found variation in the impact of vertical integration on readmission rates among different hospital organizational types. Vertical integration among urban hospitals was associated with a reduction in 30-day HF readmission rates. Vertical integration among IO hospitals was associated with an increase in 30-day HF readmissions. Finally, vertical integration among rural hospitals and NFP hospitals was associated with a reduction in 30-day pneumonia readmissions rates.
Vertical integration continues to be a potential strategy for hospitals to adopt in response to the pressures to integrate care. Hospital administrators considering expanding into SNFs should consider our findings when trying to determine whether vertical integration into SNFs is a strategy that may help them achieve reduction in certain readmissions types. The relationship between hospitals and SNFs is complex, and our findings may help hospital leaders develop realistic expectations when choosing to adopt a SNF vertical integration strategy. Leaders may also choose to more closely examine ways in which their organizations have developed a vertically integrated SNF. It is critical to evaluate ways in which such organizational structures can more effectively be managed to ensure hospitals are able to reduce readmissions and provide high-quality patient care. Finally, there is a need to evaluate and characterize different ways that hospitals vertically integrate.
Tory H. Hogan, PhD, is an assistant professor in the Division of Health Services Management and Policy at the Ohio State University College of Public Health in Columbus, OH.
Christy Harris Lemak, PhD, MPH, MBA, FACHE, is a professor and Department Chair in the Department of Health Services Administration at the University of Alabama at Birmingham School of Health Professions in Birmingham, Alabama.
Larry R. Hearld, PhD, MBA, MSA, is an associate professor in the Department of Health Services Administration at the University of Alabama at Birmingham School of Health Professions in Birmingham, Alabama.
Jack Wheeler, PhD, MA, is a professor emeritus in the Department of Health Management and Policy at the University of Michigan School of Public Health in Ann Arbor, Michigan.
Bisakha (Pia) Sen, PhD, is a professor in the Department of Health Care Organization and Policy at the University of Alabama at Birmingham School of Public Health in Birmingham, Alabama.
Nir Menachemi, PhD, MPH, is a professor and department Chair in the Department of Health Policy and Management at the Indiana University Fairbanks School of Public Health in Indianapolis, Indiana.
1. Mor V, Intrator O, Feng Z, Grabowski DC. The revolving door of rehospitalization from skilled nursing facilities. Health Aff. 2010;29(1):57–64.
2. McHugh JP, Foster A, Mor V, et al. Reducing hospital readmissions
through preferred networks of skilled nursing facilities. Health Aff. 2017;36(9):1591–1598.
3. Hogan TH, Lemak CH, Ivankova N, Hearld LR, Wheeler J, Menachemi N. Hospital vertical integration
into subacute care as a strategic response to value-based payment incentives, market factors, and organizational factors: A multiple-case study. Inquiry. 2018;55:46958018781364.
4. Conrad DA, Dowling WL. Vertical integration
in health services: Theory and managerial implications. Health Care Manage Rev. 1990;15(4):9–22.
5. Hogan TH, Lemak CH, Hearld LR, Sen BP, Wheeler JR, Menachemi N. Market and organizational factors associated with hospital vertical integration
into sub-acute care. Health Care Manage Rev. 2019;44(2):137–147.
6. Combs J, Liu Y, Hall A, Ketchen D. How much do high-performance work practices matter? A meta-analysis of their effects on organizational performance. Personnel Psychol. 2006;59(3):501–528.
7. Rahman M, Zinn JS, Mor VJ. The impact of hospital-based skilled nursing facility closures on rehospitalizations. Health Serv Res. 2013;48(2 pt 1):499–518.
8. Liu K, Black KJ. Hospital-based and freestanding skilled nursing facilities: Any cause for differential Medicare payments? Injury. 2003;40(1):94–104.
9. Stearns SC, Dalton K, Holmes GM, Seagrave SM. Using propensity stratification to compare patient outcomes in hospital-based versus freestanding skilled-nursing facilities. Med Care Res Rev. 2006;63(5):599–622.
10. Rahman M, Norton EC, Grabowski DC. Do hospital-owned skilled nursing facilities provide better post-acute care quality? J Health Econ. 2016;50:36–46.
11. Wheeler J, Burkhardt J, Alexander JA, Magnus SA. Financial and organizational determinants of hospital diversification into subacute care. Health Serv Res. 1999;34(1 pt 1):61.
12. Dilwali PK. From acute care to home care: The evolution of hospital responsibility and rationale for increased vertical integration
. J Healthc Manag. 2013;58(4):267–276.
13. Wilding HJ. Integrating care: From horizontal to vertical integration
. Health Care Manag (Frederick). 2010;18(3):15–20.
14. Schoenfeld AJ, Zhang X, Grabowski DC, Mor V, Weissman JS, Rahman MJS. Hospital-skilled nursing facility referral linkage reduces readmission rates among medicare patients receiving major surgery. Surgery. 2016;159(5):1461–1468.
15. McHugh MD, Ma CJ. Hospital nursing and 30-day readmissions
among Medicare patients with heart failure, acute myocardial infarction, and pneumonia. Med Care. 2013;51(1):52.
16. Whittle J, Lin CJ, Lave JR, et al. Relationship of provider characteristics to outcomes, process, and costs of care for community-acquired pneumonia. Med Care. 1998:977–987.
17. DesRoches CM, Charles D, Furukawa MF, et al. Adoption of electronic health records grows rapidly, but fewer than half of US hospitals had at least a basic system in 2012. Health Aff (Millwood). 2013;32(8):1478–1485.
18. Ortiz J, Bushy A, Zhou Y, Zhang HJR. Accountable care organizations: Benefits and barriers as perceived by rural health clinic management. Rural Remote Health. 2013;13(2):2417.
19. Succi MJ, Lee SY, Alexander JA. Effects of market position and competition on rural hospital closures. Health Serv Res. 1997;31(6):679.
20. Bradley EH, Curry L, Horwitz LI, et al. Contemporary evidence about hospital strategies for reducing 30-day readmissions
: A national study. J Am Coll Cardiol. 2012;60(7):607–614.
21. Bradley EH, Curry L, Horwitz LI, et al. Hospital strategies associated with 30-day readmission rates for patients with heart failure. Circ Cardiovasc Qual Outcomes. 2013;6(4):444–450.
22. Romano PS, Geppert JJ, Davies S, Miller MR, Elixhauser A, McDonald KM. A national profile of patient safety in US hospitals. Health Aff (Millwood). 2003;22(2):154–166.
23. Carroll NW, Hearld LR, Joseph R. Hospital ownership of postacute care providers and the cost of care [published ahead of print February 21, 2019]. Health Care Manag Rev. 2019. doi:.
24. Hansen LO, Young RS, Hinami K, Leung A, Williams MV. Interventions to reduce 30-day rehospitalization: A systematic review. Ann Intern Med. 2011;155(8):520–528.
25. Jurgens CY, Goodlin S, Dolansky M, et al. Heart failure management in skilled nursing facilities: A scientific statement from the American heart association and the heart failure society of America. Circ Heart Fail. 2015;8(3):655–687.
26. Manemann SM, Chamberlain AM, Boyd CM, et al. Skilled nursing facility use and hospitalizations in heart failure: A community linkage study. Paper presented at: Mayo Clinic Proceedings. 2017.
27. Allen LA, Hernandez AF, Peterson ED, et al. Discharge to a skilled nursing facility and subsequent clinical outcomes among older patients hospitalized for heart failure. Circ Heart Fail. 2011;4(3):293–300.