Recent reimbursement policies, as well as anticipated future reforms to the U.S. health care system, address unnecessary health care spending and gaps in quality that have been well documented (Obama, 2016). The Affordable Care Act and the Medicare and CHIP Reauthorization Act addressed these issues through new reimbursement initiatives that aim to incentivize more effective care across a continuum of providers (Goldsmith, 2011). Consistent with these incentives, payers are moving away from fee-for-service reimbursement structures and toward capitated and bundled payments (Shay & Mick, 2013). Many industry experts predict that health care providers will consolidate in response to these environmental changes (Berenson, Ginsburg, Christianson, & Yee, 2012 ; Goldsmith, 2011). The aging population, new payment methods, and the introduction of value-based payment (VBP) methods make sub-acute care (SAC) a potentially attractive market for hospitals. SAC refers to a type of care patients receive when they are no longer sick enough to be in an acute care facility but are not healthy enough to return home and still require 24-hour nurse care. The role of SAC has increased within the U.S. health system as a result of payers becoming more stringent on acute care length of stay associated with various diseases and health care episodes. Facilities included in the term SAC include inpatient rehabilitation facilities (IRFs) and skilled nursing facilities (SNFs). In the current environment, hospitals may have an incentive to buy or partner with SAC providers in order to gain control of services that exist across the continuum of care and better manage quality outcomes throughout the continuum of care.
Vertical integration in health care is defined as “the provision of a continuum of office-based primary care, acute care, and postacute care services within a single organizational or joint ownership structure, allowing for a coordinated progression of services across the patient care spectrum” (Shay & Mick, 2013, p. 16). Hospitals that vertically integrate SAC can do so by bringing IRFs and SNFs under their governance structure. Researchers have examined vertical integration in a variety of health care settings as a predictor of financial and organizational performance (Forbes & Lederman, 2010 ; Rothaermel, Hitt, & Jobe, 2006).
By vertically integrating into SAC, hospitals may gain a competitive advantage, improve the coordination of care across the care spectrum, discharge patients faster, and capture revenue streams associated with the SAC portion of the episode of care. Vertical integration has disadvantages that should be noted; it can create large internal costs, can perpetuate obsolete or inefficient synergies or processes, and comes with opportunity costs as it is capital and resource intense to adopt and implement (Harrigan, 1985). Despite the potential benefits of vertically integrating these services, not all hospitals are adopting a vertical integration SAC strategy. Little is known about factors that may facilitate or impede such efforts. Two previous studies examined hospital integration into SAC. Wheeler, Burkhardt, Alexander, and Magnus (1999) examined a multitude of financial, organizational, and market characteristics that determined what types of hospitals diversify into SAC. Wheeler et al. found that investor-owned organizations are associated with less vertically integrated SAC and that risk and financial performance are predictors of SAC vertical integration. Wang, Wan, Clement, and Begun (2001) suggest that vertical integration into SAC is associated with an increase in managed care. Both studies were completed using data sets that are now over two decades old and may not be generalizable to current market conditions.
Therefore, the purpose of this research is to utilize data collected between 2008 and 2012 of a national sample of short-term general hospitals to investigate relationships among organizational and market factors and hospital vertical integration into SAC services. Using a longitudinal database, we examine relationships between market and organizational factors and the likelihood that a hospital will be vertically integrated into SAC between the years 2008–2012. The results of this study will contribute to the understanding of how market and organizational factors influence hospital strategy, specifically vertical integration behavior. Furthermore, our findings may inform public policymakers about how hospitals are responding to new payment reforms and could advise hospital and health system leaders on potential future SAC strategies.
The SAC industry provides a broad range of services to patients as part of an acute care episode. Patients admitted to SAC facilities have complex care needs for services, such as rehabilitation, supportive care, and palliative care management (Mor, Intrator, Feng, & Grabowski, 2010). Medicaid and Medicare beneficiaries are the largest groups of SAC utilizers. The role of SAC is to aid in the recovery and rehabilitation of patients when they are no longer eligible for acute care services but still require 24-hour care. Facilities included in the term SAC include IRFs and SNFs. One of the biggest issues associated with SAC services is hospital readmissions, which could potentially be preventable (Goldfield et al., 2008). In addition, improved transitions in care represent an area that can potentially result in significant cost savings (Averill et al., 2010). Improved transitions of care that support care coordination, communication across providers, and continuity have been associated with reductions in hospital readmissions (Coleman, Parry, Chalmers, & Min, 2006).
One quarter of all Medicare beneficiaries who utilize SAC are readmitted to a hospital within 30 days (Mor et al., 2010). VBP mechanisms (such as Accountable Care Organizations and the Hospital Readmission and Reduction Program) link hospital reimbursements to patient quality outcomes for entire episodes of care. This may incentivize hospitals to focus on strategies that may enable them to focus more on transitions in care and SAC outcomes. In light of these changes, hospitals may adopt new strategies toward SAC, of which can include vertical integration, joint ventures, and partnerships. Vertical integration of SAC defines a relationship in which a hospital owns a SAC facility. Theory has long argued that, through vertical integration, hospitals may be able to gain control of the SAC care part of the continuum and which has the potential to reduce adverse quality outcomes that impact the ability for a hospital to maximize revenue through VBP (Shay & Mick, 2013).
Resource dependence theory (RDT) is one of the most widely used theories in explaining vertical integration (Hillman, Withers, & Collins, 2009) and suggests that hospital strategic decision-makers will adopt a vertical integration strategy based on the resources available in their organization’s environment and within the organization itself. RDT encompasses three environmental constructs: munificence, dynamism, and complexity. Munificence is the availability of necessary resources in a firm’s particular environment (Pfeffer & Salancik, 2003). The availability of resources can change over time from scarce to abundant, and RDT predicts that successful organizations will take advantage of resource munificence. Dynamism and complexity relate to the level of uncertainty in an organization’s market. Dynamic environments are constantly changing, creating uncertainty for organizations (Yeager, Menachemi, Savage, Ginter, Sen, & Beitsch, 2014). Complexity refers to the amount of heterogeneity or diversity in a firm’s environment, which also creates uncertainty for decision-makers. RDT suggests that munificence, dynamism, and complexity influence the strategy a firm will adopt. Firms respond to these three components of their environment through strategic behaviors. These strategic decisions are made based on the resources available in an organization’s environment and within the organization itself. We therefore hypothesize that a hospital’s vertical integration into SAC is associated with the level of dynamism, complexity, munificence, and organizational resources.
Dynamism in the environment is directly associated with a firm’s decision to diversify (Harrigan, 1985). When relevant external factors are changing, hospitals may face information uncertainty, which can cause decision-makers to struggle to predict the future. Decision-makers do not possess all the information needed to make optimal strategic decisions. As a result, the strategic decision-making processes are driven by perceptions of what is occurring in an organization’s environment (Keats & Hitt, 1988). As an organization’s environment becomes more dynamic, decision-makers face more uncertainty.
The uncertainty created from dynamic environments impacts the strategic decisions hospital leaders are able to make. Previous research suggests that increased dynamism is associated with the adoption of less risky organizational strategies (Birnbaum, 1984 ; Mahon & Murray, 1981). Vertical integration into SAC strategies tend to be capital intensive and take significant time to adopt and implement, making them risky. Thus, decision-makers carefully plan and consider all potential scenarios when considering adopting such a strategy. Dynamic environments may make the consideration of different scenarios more difficult. It may create uncertainty when completing potential strategic implementation plans and add complexity to the decision-making process. As an organization’s environment becomes more dynamic, hospitals may be less likely to adopt a capital-intensive, risky strategy, such as vertical integration into SAC.
Hypothesis 1: Higher levels of dynamism will be negatively associated with hospital vertical integration into SAC.
Complexity in an organization’s environment may also be associated with vertical integration (Harrigan, 1985). Previous studies suggest that, when an organization’s environment becomes more complex, they are more likely to adopt a less risky strategy (Birnbaum, 1984 ; Mahon & Murray, 1981). Strategic decision-makers operating in more complex environments have to consider more factors when making decisions. Previous research indicates that, in these instances, organizations are less likely to adopt new strategies and also take less strategic risks (Menachemi, Mazurenko, Kazley, Diana, & Ford, 2012). As environments become more complex, hospital strategic decision-makers face information uncertainty regarding the future. For example, a hospital’s environment becomes more complex when a competitor opens a new service line that directly competes with their hospital. As a result, hospital leaders may be unwilling to make major strategic decisions or implement major organizational changes until they are sure their organization will be not be negatively impacted by the new competition. The decision to vertically integrate into SAC carries a significant amount of risk, particularly due to the associated cost of vertically integrating. When hospital decision-makers face increased complexity and uncertainty in their future, they will be less likely to adopt a risky strategy that utilizes a significant amount of capital, such as vertical integration into SAC.
Hypothesis 2: Higher levels of market complexity will be negatively associated with hospital vertical integration into SAC.
The level of munificence in an organization’s environment may also impact an organization’s strategy and response to its environment. Munificence refers to the availability of resources necessary for an organization to be successful and may impact a hospital’s ability to acquire and manage the resources necessary to its revenue stream and associated provision of services. Health care strategy research has shown that organizations in more munificent environments are more productive and adopt more resource intense strategies (Balotsky, 2005 ; Trinh & O’Connor, 2000). For example, decision-makers may be more willing to adopt strategies that cater to affluent customers or payer groups that are most financially beneficial. By catering to these types of stakeholders, hospitals may be able to ensure that affluent patients will utilize their facilities instead of going to a competitor. In doing this, the organization secures revenue streams, which in turn provide greater overall financial stability. This may enable hospital decision-makers to adopt more capital-intense strategies such as vertical integration (Menachemi, Shin, Ford, & Yu, 2011). Greater levels of munificence may also mean that hospitals have access to the services in the community that they are interested in acquiring. SAC facilities cannot be vertically integrated (usually through a merger or acquisition) unless there is a facility in the marketplace to acquire. For example, hospitals in highly munificent (resource abundant) environments may be more likely to adopt a vertical integration SAC strategy because the resources to be successful may be more available within the environment. Lastly, hospitals located in more munificent areas may cater to more resourceful payers, which in turn may provide the financial resources needed to pursue resource-intensive strategies such as vertical integration into SAC.
Hypothesis 3: Higher levels of munificence will be positively associated with hospital vertical integration into SAC.
An organization’s ability to respond to its environment is also driven by the availability of internal resources. Organizations with greater availability of internal resources are able to respond more effectively to changing environmental threats that create uncertainties. Vertical integration may potentially reduce future uncertainties by bringing control of cash flows into the organization. Organizational resources may be important factors in strategic adoption. An organization’s existing resources may restrain how an organization is able to respond to the pressures brought on by the environment.
For example, larger hospitals or hospitals that are part of a system may have greater internal resources (e.g., administrative staff, clinicians) and therefore may be able to more easily utilize internal resources to accommodate the demands of the environment (Banaszak-Holl, Zinn, & Mor, 1996).
Hypothesis 4: Hospitals with greater organizational resources will be more likely to be vertically integrated into SAC.
Data Source and Study Sample
This study used longitudinal analysis (2008–2012) of all nonfederal, general acute care hospitals operating in the United States using secondary data from four sources. The American Hospital Association’s (AHA) Annual Survey of Hospitals was used to identify whether or not hospitals were vertically integrated into SAC. The AHA Annual Survey also collects information on hospital characteristics. The Area Health Resource File, the Centers for Medicare and Medicaid’s (CMS) Cost Report, and the Rural Urban Commuting Area (RUCA) data were also used. The Area Health Resource File provides county-level measures of health care markets. The CMS Cost Report provides hospital financial data. Lastly, the RUCA is a census tract-based classification of zip codes that characterizes rural and urban status. As a result of this process, our sample was composed of 3,775 hospitals, for a total of 16,269 hospital observations over the 5-year study period. Lastly, it should be noted that hospitals that closed during the study period or opened during the study period were not included in the sample. A summary of all variables and the associated data source can be found in Table 1.
The dependent variable was “vertical integration into SAC.” To operationalize this variable, we reviewed the AHA Annual Survey results from 2008 to 2012, specifically hospital responses to whether or not they had an SNF at the hospital. We used this variable as an indicator of whether or not they had a vertically integrated SNF. The variable was binary: 1 being vertically integrated and 0 being not vertically integrated in a given year.
To measure our constructs (dynamism, complexity, munificence, and organizational resources), we used variables from the AHA Annual Hospital Survey, the Area Resource File (ARF), CMS Cost Report, and RUCA codes.
Dynamism. Dynamism was operationalized using two measures: change in the county population at or below the federal poverty level between 2008 and 2011 (ARF; Mazurenko, Hearld, & Menachemi, 2017 ; Menachemi et al., 2012) and change in population between 2008 and 2011 (ARF; Menachemi et al., 2012 ; Tarver & Menachemi, 2017). Counties with an increase in the federal poverty level are more resource-scarce over time, as individuals experiencing poverty may be less able to consume health care and utilize an SNF. Next, change in population levels reflect potential changes in the demand and utilization for health care services.
Complexity. Complexity was operationalized using two measures: Herfindahl–Hirshman Index (HHI; Hsieh, Clement, & Bazzoli, 2010 ; Zinn et al., 1998) and market SNF availability. Based on hospital admissions, the HHI is defined as the sum of squares of hospital admissions of a hospital as a percentage of total admissions within a county (Hsieh et al., 2010). The HHI is a common measure of the market concentration, the size of a hospital in relation to its market, and thus the level of competition (Morrisey, Sloan, & Valvona, 1988). For the numerator, a hospital’s market size was measured based on its admissions in a given year. The denominator was the county hospital admissions in a given year. SNF availability in a county was defined as the number of SNFs in a county divided by the census population per 1 million people. This measure provided a representation of the complexity of the marketplace for SAC services in a county.
Munificence. Munificence was operationalized using two measures: The percentage of population eligible for Medicare in the hospital’s county (Menachemi et al., 2012 ; Tarver & Menachemi, 2017) and urban/rural location (Menachemi et al., 2011 ; Yeager et al., 2014). Medicare is the largest payer of SAC services in the United States, and Medicare beneficiaries utilize SAC services more than any other population. Organizations rely on Medicare payments as a financial resource. Larger Medicare-eligible populations represent greater levels of munificence in a community from the perspective of the hospital. Rural and urban location measures the availability or resources in a community. Urban areas are frequently associated with greater availability of health care resources in a community (Horev, Pesis-Katz, Mukamel, 2004).
Organizational resources. Organizational resources were measured with five variables. Hospital size was defined as the number of hospital beds. This was a continuous variable. Financial performance was defined as operating margin and is a continuous variable. Operating margin is a financial measure that reflects the proportion of a hospital’s revenue that remains once all wages, overhead, material costs, depreciation, and interest expense have been paid. Ownership status was measured as a series of dummy variables reflecting “not for profit,” “investor owned,” or “nonfederal governmental” categories. Ownership status was chosen to operationalize organizational resources because it can influence a hospital’s access to capital to make strategic investments. Health system affiliation was measured as a dummy variable reflecting whether or not a hospital was a member of a health system or not. This information is self-reported by each hospital. Finally, swing beds could be considered a substitute for vertically integrating into SNFs. They are beds that CMS has allowed an organization to use as an acute care bed and later a skilled nursing bed as patients moves between types of care. The swing beds variable was operationalized to consider whether or not an organization was licensed to have swing beds in a given year.
To understand if organizations were more likely to vertically integrate over time, a dummy variable was included for time. A summary of all constructs can be seen in Table 1.
Descriptive statistics for the independent variables and dependent variables were analyzed to determine the variability of each, to test the assumptions of the regression model, and to test for outliers in the data. Because the dependent variables is a binomial indicator, we conducted a pooled cross-sectional analysis (2008–2012) using a logit regression model to assess the relationship between market and organizational factors and hospitals being vertically integrated into SAC. Two models were run: For our first model, we tested the relationship between dynamism, complexity, and munificence variables and hospital vertical integration into SAC, controlling for states. We ran this to build a baseline in which our full model could be assessed. For the second model, we ran the full model, testing the relationship between dynamism, complexity, munificence, and organizational factors, controlling for states. Marginal effects were calculated using the margins, dydx command, in STATA for both models. Standard errors were clustered within hospitals for both models. All analysis was completed using STATA 13.0.
There were 16,269 hospital observations in the sample representing 3,775 individual hospitals. Within the sample, 52% of hospitals were located in urban areas (see Table 2). Sixty-one percent of hospitals were not for profit, 21% were nonfederal governmental, and 18% were investor-owned. Fifty percent were associated with a health system, and 31% had swing beds. On average, hospitals in the sample had 164 staffed beds, and the mean HHI was .674. In addition, on average, 16% of county populations were eligible for Medicare.
A cross tabulation was conducted of hospitals that were vertically integrated into SNFs over the period of the study (2008–2012). In 2008, 25.3% of hospitals in the sample had a vertically integrated SNF at the hospital level (see Table 2). This number increased each year, to 26.1% in 2009, 26.7% in 2010, and 27.9% in 2011. By 2012, 29.4% of hospitals in the sample had a vertically integrated SNF at the hospital level.
It was hypothesized that hospitals in more dynamic environments would be less likely to be vertically integrated into SAC (H1). Findings did not support this hypothesis; when operationalizing dynamism as the change in population in a county (2008–2011), an increase in the population by 1,000 was associated with a very small (1.19e−06%) increase in the likelihood that hospitals would be vertically integrated into SAC. The coefficient on the change in poverty level between 2008 and 2011 was not statistically significant. Second, it was hypothesized that hospitals practicing in environments that are more complex would be less likely to be vertically integrated into SAC (H2). Our analysis produced equivocal results for this hypothesis. A percentage point increase in the number of SNF in a county as a ratio of the county population was associated with a 1.8% decrease in the likelihood of being vertically integrated into SAC (p < .001). When using the HHI, the findings were not in the direction predicted, although they were not statistically significant either.
In Hypothesis 3, it was hypothesized that hospitals in more highly munificent environments would be more likely to be vertically integrated into SAC. This hypothesis produced equivocal results. When operationalizing munificence as hospital location (urban/rural), hospitals located in rural areas had a 6.9% greater likelihood of being vertically integrated into SAC, relative to hospitals in urban areas (p < .001). When operationalizing munificence as the percentage of population eligible for Medicare in a county, a 1 point increase was associated to 1.8% greater likelihood of being vertically integrated into SAC (p < .001).
In Hypothesis 4, it was hypothesized that greater organizational resources would be positively associated with vertical integration into SAC. Once again, this hypothesis produced equivocal results. When operationalizing organizational resources as swing beds, organizations that has swing beds were associated with a 6.9% greater likelihood that they would be vertically integrated into SAC, which supports our hypothesis (p < .001). However, this hypothesis was not supported when organizational resources were operationalized as system affiliation and investor-owned (compared to not for profit and nonfederal governmental). Being affiliated with a system were 4.1 percentage points less likely to be engaged in vertical integration into SAC (p < .001). Investor owned hospitals were 5.2 percentage points less likely to be engaged in vertical integration into SAC (p < .1). Although not statistically significant, bed size was negatively associated with hospital vertical integration into SAC. Operating margin and not-for-profit ownership were not statistically significant either.
Lastly, we tested the relationship between vertical integration and time to see if hospitals were vertically integrating more over time. As compared to 2008, 2009 was positively associated with hospital vertical integration in SAC (marginal effects = 0.9%, p < .05). As compared to 2008, 2010 was positively associated with hospital vertical integration in SAC (marginal effects = 2.4%, p < .001). 2011 was positively associated with hospital vertical integration into SAC (marginal effects = 1.4%, p < .1). 2012 was also positively associated with hospital vertical integration in SAC (marginal effects = 4.3%, p < .001). A summary of all results can be found in Table 3.
The purpose of this study was to examine the correlates of hospital vertical integration into SAC. The overall findings of this study suggest that environmental factors may play a role in whether or not hospitals are vertically integrated into SAC. Measures of munificence, complexity, dynamism, and organizational resources were significantly associated with hospital vertical integration into not always in the direction that was expected. These findings provide mixed support of theory and may have practical implications for current national efforts aimed to improve quality outcomes through integrated delivery systems.
We found that hospital vertical integration into SNFs was associated with the degree of environmental munificence. Policymakers should note that hospitals in rural areas were more likely to be vertically integrated into SNFs compared to hospitals in urban areas. This finding did not support our hypothesis because rural areas are resource-poor environments. It is possible that rural hospital leader makers may be more willing to adopt risky strategies such as vertical integration in order to ensure their hospital can accomplish integrated delivery goals for patients with limited postacute care options. This finding is consistent with previous literature that has suggested that hospitals in rural areas respond differently to environmental pressures in comparison to urban hospitals (Mick et al., 1993).
Rural hospitals also face scarcity in the availability of services, physicians, and nurses (Davidson & Moscovice, 2003). The constrained environments of rural areas also create unique patterns of hospital utilization, readmissions rates, and utilization of SAC services (Coburn, Bolda, & Keith, 2003 ; Coburn, Keith, & Bolda, 2002). Our findings, however, are in contrast with previous studies that found that rural hospitals are less likely to integrate and merge with other health care organizations (Trinh & O’Connor, 2000). Perhaps in the current time period, rural hospitals may have already been struggling to adapt to multiple strategic demands. For example, they are less likely to adopt an electronic medical record (DesRoches et al., 2013), less likely to adopt imaging technology innovations (Nystrom, Ramamurthy, Wilson, 2002), and face significant barriers to creating and participating in an accountable care organization or other integrated delivery system (Ortiz, Bushy, Zhou, & Zhang, 2013). In light of previous research findings that identify many of the difficulties rural hospitals face, one explanation is that hospitals may be the only resource available to provide SAC in their community; therefore, they may feel pressure to vertically integrate. Vertical integration may be a fit for markets with fewer resources. Further research should examine more closely the ways that rural hospitals are achieving integration with limited local resources.
As predicted, vertical integration into SAC was associated with the size of the Medicare-eligible population in the hospital’s community. Conversely, Medicare is the largest payer of SNF services; it can be assumed that Medicare payers have a demand for SNF services, and thus, when the demand for SNF services increases in a community, hospitals may respond by vertically integrating these services. Hospitals in markets with a smaller percentage of Medicare-eligible individuals may not view vertical integration as a viable strategy when the demand for the service is not as high. Hospital leaders may want to consider the current and projected volume of Medicare patients in their market as vertical integration strategies are reviewed. Future studies could examine how the number of hospitals, SAC providers, and Medicare-eligible patients in a market are collectively associated into SAC strategies over time.
Investor-owned hospitals were less likely to be vertically integrated into SAC, compared to not-for-profit hospitals and nonfederal governmental hospitals, a finding that is consistent with previous health care management studies on this topic (Wheeler et al., 1999). Previous research has found that investor-owned hospitals may be less likely to provide diversified services compared to not-for-profit hospitals (Shortell, Morrison, Hughes, Friedman, & Vitek, 1987) and that not-for-profit hospitals were more likely to vertically integrate into SAC than investor-owned hospitals (Wheeler et al., 1999). One potential explanation for our findings is that investor-owned hospitals adopt strategies that will return financial rewards to investors and investor-owned hospitals may perceive vertical integration into SAC as a strategy that will not be profitable. Future research is needed to better understand why investor-owned hospitals are less likely to adopt vertical integration into SAC and how these organizations pursue SAC strategies.
Hospitals with swing beds were more likely to be vertically integrated into SNFs. The labor and institutional knowledge associated with providing SAC care is a unique set of skills that differ from acute care. Organizations that have swing beds may have the workforce and institutional knowledge to operate a SAC facility successfully. In addition, hospitals that had swing beds may be more likely to vertically integrate because they understand the challenges and dynamics associated with this part of the care continuum.
Hospitals that are part of a health care system were less likely to be vertically integrated. This study measured vertical integration at the hospital level, so it is possible that hospitals that are part of a system were vertically integrated at a health system level as compared to the hospital level. In light of the growing presence of health systems in the U.S. health system, health care management research should explore the role of health systems in vertical integration into SAC strategy adoption.
Lastly, between 2009 and 2012 hospitals became more vertically integrated into SAC as compared to 2008. Hospital vertical integration for this part of the care continuum could be the result of an overall increase in the complexity and munificence of the health care market, brought on by the anticipation of the Affordable Care Act (ACA) and adoption of VBP mechanism. Further research is needed to understand whether this has occurred prior to 2008 and after 2012. It is also critical that researchers seek to understand this trend and what it means for patient care.
It is critical to consider the theoretical implications of our study. We utilized RDT to justify the hypotheses and justification of the variables we select to measure each construct. Consistent with other health care studies utilizing RDT as a theoretical framework to examine strategy adoption, we utilized multiple variables to represent the three environmental constructs or RDT. For each dimension, our results either yielded conflicting findings or inconsistent findings where some variables were not significant. One potential explanation is that the variables that did not support our hypothesis failed to accurately operationalize the dimensions we were measuring. We chose to operationalize dynamism, munificence, and complexity in ways that were consistent with previous studies that have examined hospital strategy with the RDT framework. Studies that have found a positive finding tend to be published; therefore, it is possible that research with null findings, which have operationalized RDT constructs, has not been published (Yeager, et al., 2014).
Another potential explanation is that RDT may not accurately explain market and organizational factors associated with hospital vertical integration into SAC strategy adoption. We chose RDT to explain the phenomena because of its emphasis on the resource environment and the premise that organizational decision-makers act upon both their perceptions of their external resource environment and their own organizational resource endowments to maintain power and minimize dependence. RDT is but one theory that may help to understand vertical integration, and when adopting this theory we made certain assumptions about how and why hospitals may pursue vertical integration into SAC strategies. In light of our findings, other theories may provide a better explanation of vertical integration into SAC behavior. For example, institutional theory may explain that vertical integration is a strategy adopted in an attempt to achieve or maintain legitimacy in a changing health care environment. Utilizing institutional theory, we could hypothesize that hospitals in environments that are changing are more likely to be vertically integrated into SAC, which is in-line with our finding that change in county population between 2008 and 2011 was positively associated with hospital vertical integration into SAC. The transaction cost economics theory may frame vertical integration in the context of a make or buy decision in an attempt to achieve optimum efficiency. Utilizing the transaction cost economics theory, we may potentially hypothesize that hospitals experiencing high transaction costs when transitioning patients to SAC are more likely to vertically integrate into SAC in an attempt to lower such transaction costs. Future research on this topic should explore other theoretical frameworks to potentially better explain this organizational behavior.
Despite the valuable contributions of this research, the study has several limitations. First, we used data from the AHA Annual Hospital Survey, and therefore, we relied on hospitals accurately reporting their SAC strategy. In some years, responses to questions were inconsistent with previous report. Unfortunately, this was the only source for data related to vertical integration into SAC. Next, because of the complex nature of our model, our statistical software (STATA 13.0) would not allow us to cluster our standard errors within systems. Although there is limited empirical understanding of the role health systems play in vertical integration strategy adoption among hospitals, we were not able to account for this in our model. In addition, because of limitations we also were not able to account for whether or not hospitals have a SAC facility in their system when running our model.
In addition, we used data from the ARF, which was not available for every year of the study. This is mitigated by the fact that many of the measures did not change within 1–2 years’ time. Lastly, we used financial data from the CMS Cost Report, and only hospitals that provide care to Medicare beneficiaries are included in this analysis. This limitation may impact the generalizability of the findings, although it was mitigated by the fact that most hospitals in the United States accept Medicare.
Our findings raise important issues regarding potential hospital responses to VBP approaches. Little attention has been given to how the environment may influence hospital strategies toward managing patients that move between acute care and SAC. One key significant finding is that, over the course of the study (2008–2012), each year more hospitals became vertically integrated into SAC. Very little is understood about what this change means for providers, patients, and payers. Health care delivery leaders, payers, and policymakers must focus efforts toward understanding what this change means for patients and how they transition from acute care to SAC providers. In addition, our study adds new information about how hospital vertical integration into SAC is related to environmental factors. Unfortunately, the current CMS reimbursement programs do not take these factors into consideration. Hospital leaders should consider how to best align their organization’s SAC strategy in their operating environment. This study may also provide valuable insight into how health care delivery leaders can engage and understand their competitors SAC strategy. Given the capital intensity and riskiness associated with a vertical integration strategy, hospital leaders should consider the conditions of their organization’s environment as part of their SAC strategy formulation process. Hospital leaders may also find additional value to utilize the variables of this study to more informatively scan the parts of their environment and also evaluate their own internal organization’s capacity when determining their own SAC strategy.
As hospitals continue to adapt to payment changes brought about by the ACA, it is important to understand how hospitals will integrate into SAC. The continuum of care continues to move patients to SAC services, and funding agencies are focusing efforts toward reducing payments to organizations that provide SAC services. The demand for better coordination of care between hospitals and SAC facilities will only grow as penalties for readmissions increase and bundled payments become larger. The ability of hospitals to adapt to the changing health care landscape through vertical integration is related to their market and organizational resources. As policymakers continue to implement different components of the ACA, it is critical that they are aware of how hospitals in less resource-abundant environments respond.
Averill R. F., Goldfield N. I., Vertrees J. C., McCullough E. C., Fuller R. L., Eisenhandler J. (2010). Achieving cost control, care coordination, and quality improvement through incremental payment system reform. The Journal of Ambulatory Care Management
, 33(1), 2–23.
Balotsky E. R. (2005). Is it resources, habit or both: Interpreting twenty years of hospital strategic response to prospective payment. Health Care Management Review
, 30(4), 337–346.
Banaszak-Holl J., Zinn J. S., Mor V. (1996). The impact of market and organizational characteristics on nursing care facility service innovation: A resource dependency perspective. Health Services Research
, 31(1), 97–117.
Berenson R. A., Ginsburg P. B., Christianson J. B., Yee T. (2012). The growing power of some providers to win steep payment increases from insurers suggests policy remedies may be needed. Health Affairs
, 31(5), 973–981.
Birnbaum P. H. (1984). The choice of strategic alternatives under increasing regulation in high technology companies. Academy of Management Journal
, 27(3), 489–510.
Coburn A. F., Bolda E. J., Keith R. G. (2003). Variations in nursing home discharge rates for urban and rural nursing facility residents with hip fracture. The Journal of Rural Health
, 19(2), 148–155.
Coburn A. F., Keith R. G., Bolda E. J. (2002). The impact of rural residence on multiple hospitalizations in nursing facility residents. The Gerontologist
, 42(5), 661–666.
Coleman E. A., Parry C., Chalmers S., Min S. J. (2006). The care transitions intervention: Results of a randomized controlled trial. Archives of internal medicine
, 166(17), 1822–1828.
Davidson G., Moscovice I. (2003). Rural hospitals: New millennium and new challenges
. Minneapolis, MN: Rural Health Research Center, University of Minnesota.
DesRoches C. M., Charles D., Furukawa M. F., Joshi M. S., Kralovec P., Mostashari F., Jha A. K. (2013). Adoption of electronic health records grows rapidly, but fewer than half of U.S. hospitals had at least a basic system in 2012. Health Affairs
, 32(8), 1478–1485.
Forbes S. J., Lederman M. (2010). Does vertical integration
affect firm performance? Evidence from the airline industry. The RAND Journal of Economics
, 41(4), 765–790.
Goldfield N. I., McCullough E. C., Hughes J. S., Tang A. M., Eastman B., Rawlins L. K., Averill R. F. (2008). Identifying potentially preventable readmissions. Health Care Financing Review
, 30(1), 75–91.
Goldsmith J. (2011). Accountable care organizations: The case for flexible partnerships between health plans and providers. Health Affairs
, 30(1), 32–40.
Harrigan K. R. (1985). Vertical integration
and corporate strategy. Academy of Management Journal
, 28(2), 397–425.
Hillman A. J., Withers M. C., Collins B. J. (2009). Resource dependence theory: A review
. Journal of Management
Horev T., Pesis-Katz I., Mukamel D. B. (2004). Trends in geographic disparities in allocation of health care resources in the US. Health Policy
, 68(2), 223–232.
Hsieh H. M., Clement D. G., Bazzoli G. J. (2010). Impacts of market and organizational characteristics on hospital efficiency and uncompensated care. Health Care Management Review
, 35(1), 77–87.
Keats B. W., Hitt M. A. (1988). A causal model of linkages among environmental dimensions, macro organizational characteristics, and performance. Academy of Management Journal
, 31(3), 570–598.
Mahon J. F., Murray E. A. (1981). Strategic planning for regulated companies. Strategic Management Journal
, 2(3), 251–262.
Mazurenko O., Hearld L. R., Menachemi N. (2017). The relationship between the external environment and physician e-mail communication: The mediating role of health information technology availability. Health Care Management Review
, 42(2), 122–131.
Menachemi N., Mazurenko O., Kazley A. S., Diana M. L., Ford E. W. (2012). Market factors and electronic medical record adoption in medical practices. Health Care Management Review
, 37(1), 14–22.
Menachemi N., Shin D. Y., Ford E. W., Yu F. (2011). Environmental factors and health information technology management strategy. Health Care Management Review
, 36(3), 275–285.
Mick S. S., Morlock L. L., Salkever D., de Lissovoy G., Malitz F. E., Wise C. G., Jones A. (1993). Rural Hospital Administrators and Strategic Management Activities1. Journal of Healthcare Management
, 38(3), 329–351.
Mor V., Intrator O., Feng Z., Grabowski D. C. (2010). The revolving door of rehospitalization from skilled nursing facilities
. Health Affairs
, 29(1), 57–64.
Morrisey M. A., Sloan F. A., Valvona J. (1988). Defining geographic markets for hospital care. Law and Contemporary Problems
, 51(2), 165–194.
Nystrom P. C., Ramamurthy K., Wilson A. L. (2002). Organizational context, climate and innovativeness: Adoption of imaging technology. Journal of Engineering and Technology Management
, 19(3-4), 221–247.
Obama B. (2016). United States health care reform: Progress to date and next steps. JAMA
, 316(5), 525–532.
Ortiz J., Bushy A., Zhou Y., Zhang H. (2013). Accountable care organizations: Benefits and barriers as perceived by Rural Health Clinic management. Rural and Remote Health
, 13(2), 2417.
Pfeffer J., Salancik G. R. (2003). The external control of organizations: A resource dependence perspective
. Stanford, CA: Stanford University Press.
Rothaermel F. T., Hitt M. A., Jobe L. A. (2006). Balancing vertical integration
and strategic outsourcing: Effects on product portfolio, product success, and firm performance. Strategic Management Journal
, 27(11), 1033–1056.
Shay P. D., Mick S. S. (2013). Post-acute care and vertical integration
after the Patient Protection and Affordable Care Act. Journal of Healthcare Management
, 58(1), 15–27; discussion 27-8.
Shortell S. M., Morrison E. M., Hughes S. L., Friedman B. S., Vitek J. L. (1987). Diversification of health care services: The effects of ownership, environment, and strategy. Advances in Health Economics and Health Services Research
, 7, 3–40; 111-4.
Tarver W. L., Menachemi N. (2017). Environmental market factors associated with electronic health record adoption among cancer hospitals. In Health Care Management Review
. Epub ahead of print.
Trinh H. Q., O’Connor S. J. (2000). The strategic behavior of U.S. rural hospitals: A longitudinal and path model examination. Health Care Management Review
, 25(4), 48–64.
Wang B. B., Wan T. T., Clement J., Begun J. (2001). Managed care, vertical integration
strategies and hospital performance. Health Care Management Science
, 4(3), 181–191.
Wheeler J. R., Burkhardt J., Alexander J. A., Magnus S. A. (1999). Financial and organizational determinants of hospital diversification into subacute care. Health Services Research
, 34(1 Pt. 1), 61–81.
Yeager V. A., Menachemi N., Savage G. T., Ginter P. M., Sen B. P., Beitsch L. M. (2014). Using resource dependency theory to measure the environment in health care organizational studies: A systematic review of the literature. Health Care Management Review
, 39(1), 50–65.
Zinn J. S., Weech R. J., Brannon D. (1998). Resource dependence and institutional elements in nursing home TQM adoption. Health Services Research
, 33(2 Pt 1), 261.