Quality in Nursing Homes
Summary and Keywords
In developed countries, the role of public authorities as financing bodies and regulators of the long-term care sector is pervasive and calls for well-planned and informed policy actions. Poor quality in nursing homes has been a recurrent concern at least since the 1980s and has triggered a heated policy and scholarly debate. The economic literature on nursing home quality has thoroughly investigated the impact of regulatory interventions and of market characteristics on an array of input-, process-, and outcome-based quality measures. Most existing studies refer to the U.S. context, even though important insights can be drawn also from the smaller set of works that covers European countries.
The major contribution of health economics to the empirical analysis of the nursing home industry is represented by the introduction of important methodological advances applying rigorous policy evaluation techniques with the purpose of properly identifying the causal effects of interest. In addition, the increased availability of rich datasets covering either process or outcome measures has allowed to investigate changes in nursing home quality properly accounting for its multidimensional features.
The use of up-to-date econometric methods that, in most cases, exploit policy shocks and longitudinal data has given researchers the possibility to achieve a causal identification and an accurate quantification of the impact of a wide range of policy initiatives, including the introduction of nurse staffing thresholds, price regulation, and public reporting of quality indicators. This has helped to counteract part of the contradictory evidence highlighted by the strand of works based on more descriptive evidence. Possible lines for future research can be identified in further exploration of the consequences of policy interventions in terms of equity and accessibility to nursing home care.
Concerns about quality in nursing homes (NHs) have been longstanding and widespread. General dissatisfaction with quality has fueled intense discussions among experts, policymakers, and the general public, due to the vulnerability of NH residents and to the role of public agencies in regulating NHs in most health systems.1 Moreover, in the United States, competition in NH markets is primarily based on quality, because of extensive price regulation by Medicare and Medicaid. Still, competition over quality remains a key factor also in systems different from the United States, where providers can set prices for a wider range of residents (Bardey & Siciliani, 2018; Gaynor & Town, 2012). As for the academic debate, scholars from different fields have drafted a rich research agenda that has contributed to pin down a general framework for the definition and measurement of NH quality and to evaluate the impact of the major policy initiatives.
This article critically discusses the most recent contributions of the health economics literature to this interdisciplinary debate. Although a few studies have developed theoretical analyses, we concentrate on the bulk of the literature that is empirically focused.2 Following Norton’s (2000) wide-ranging coverage of the economics of long-term care (LTC), Grabowski and Norton (2012) and Mukamel, Haeder, and Weimer (2014) have insightfully reviewed the empirical research on NH quality from a standpoint similar to ours. Our aim is to update and extend their work from a two-fold perspective. On the one side, by examining the influence on NH quality of key economic factors (intensity and mix of inputs, ownership status, price regulation, information asymmetry), the present contribution assesses the policy implications of recent empirical evidence. On the other side, it discusses the empirical challenges and illustrates how methodological advances have contributed to improve the rigor of policy evaluation.
Institutional circumstances are central to the NH industry. Favored also by the availability of rich data sets, most existing studies focus on the United States and are the core of this review. However, in recent years, research has also looked at European countries, and we present part of such evidence as well. The facilities assisting elderly residents are referred to by using various denominations depending on the institutional setting and the level of specialization (e.g., skilled nursing facilities, assisted living facilities, continuing care retirement communities, etc.). For ease of exposition, they are all referred to here as belonging to the general category of NHs, unless differently specified.
Quality Measures in Nursing Homes
Defining quality in the NH industry is a challenging task in itself because of the multidimensional nature of health outcomes.3 The prevailing framework finds its foundation in the classic Donabedian’s view that links structural characteristics to processes and to health outcomes (Donabedian, 1980; Harrington, 2005). Structural indicators are meant to capture institutional features in terms of internal organization (e.g., ownership status) and resource availability (e.g., infrastructure, capital and labor inputs). Process indicators describe how care is delivered (e.g., use of physical restraints, tube feeding, catheters, antipsychotics, etc.), while outcome indicators measure the endpoint of assistance. The latter are typically based either on clinical outcomes (e.g., mortality, hospitalizations, falls, urinary tract infections, ulcer pressure sores, etc.) or on administrative performance measures (e.g., deficiencies or sanctions issued by regulatory authorities, liability claims; Hillmer, Wodchis, Gill, Anderson, & Rochon, 2005). Even though favorable structural characteristics and well-managed processes are thought to positively contribute to patients’ well-being, the final outcome depends also on how they interact with other factors including patients’ initial health status. To control for the confounding factors that may prevent a proper identification of the relationship between the variables of interest, accurate risk-adjustment statistical methods are required (Dranove, 2012).
A key step in these studies is the choice of the quality indicators (Castle & Ferguson, 2010). Because of limited information on health outcomes, researchers sometimes rely on input-based measures, such as volume and composition of nursing staff. This is justified by the prominent role of nursing in the production function and by the purported positive association between staff and quality. This notwithstanding, the approach has been criticized on the grounds that high staffing might reflect inefficiency rather than high quality. By contrast, process indicators are more directly linked to care provision, but the mode of treatment may heavily depend on patients’ initial conditions. The adoption of certain processes may thus capture the acuity level of residents, instead of true differences in practice style. In response to such concerns, outcome-based quality measures have been advocated, spanning from extreme events, such as mortality and hospitalization, to less serious conditions, such as falls and pressure sores. The perspective has been broadened further by prompting the use of administrative indicators, such as the incidence of litigation procedures or of deficiency citations by oversight bodies (e.g., Konetzka, Norton, & Kilpatrick, 2004; Konetzka, Park, Ellis, & Abbo, 2013). These measures capture failure to reach preestablished standards of care, whereas clinical outcomes might be affected by patient initial health status. Yet, comparability across different contexts may be limited, as the indicators reflect the views of a certain regulatory authority or judiciary.
In most cases, NHs are subject to a legal obligation of nondiscrimination, and the literature almost unanimously assumes uniform quality within the same facility. This notwithstanding, worries about heterogeneity of treatment have emerged especially in the United States, where NHs receive different revenues across payer types. The primary purchaser of long-stay NH services is Medicaid, a joint state-federal program for the poor and disabled, which accounts for roughly 50% of all NH expenditure. Postacute NH care is covered by Medicare, a federal program providing insurance to almost all persons aged 65 or over, and accounting for about 12% of total NH expenditure. The remainder of NH care is financed primarily by private out-of-pocket payments (Grabowski, Feng, Hirth, Rahman, & Norton, 2013). Coverage and payment for Medicaid and Medicare enrolees are administrated at the federal or state levels, in most cases by employing a prospective payment system (PPS) where reimbursement rates are predetermined and based on a prospective, risk-adjusted, cost-based reimbursement methodology. By contrast, private-paying residents are charged the price chosen by the NHs, subject to market competition. Because of the different revenues across provider types, with the lowest reimbursement rates generally paid by Medicaid (Konetzka, Grabowski, Perraillon, & Werner, 2015), NHs may be expected to have an economic incentive to differentiate the quality of care based on the type of payer served (Hackmann, 2019). To investigate this concern, Grabowski, Gruber, and Angelelli (2008) examined whether NHs discriminate between payer types in terms of care quality. Using longitudinal individual-level data, they found no evidence of patients receiving different quality of care by payer type. This result held also after accounting for potential endogeneity, due to unobserved heterogeneity across NHs that treat different shares of Medicaid/Medicare patients, or to unobserved patient characteristics associated with Medicaid enrollment.4 By lending support to the commonality assumption, such findings point to substantial cross-subsidies within NHs and alleviate the concern that NHs are not perfectly price regulated. The evidence is also reassuring for the possibility of using facility-based quality indicators when individual information is not available.
Nursing Staff and Quality of Care
Since inadequate staff levels have often been held responsible for the unsatisfactory quality of assistance, the relationship between nurse staffing and quality of NHs has been subject to scrutiny, and the hypothesized positive association between staff volume and residents’ well-being has fostered the introduction of minimum staffing standards (Harrington, Schnelle, McGregor, & Simmons, 2016).5 Despite a generally favorable view, these measures have also received some critiques because constraints on the input mix may negatively affect financial performances, by increasing costs and diverting resources away from alternative uses (e.g., investments in infrastructures). Furthermore, greater demand for nursing services due to mandated staffing standards may increase nursing wages, thus inducing facilities above the minimum threshold to reduce their staff, especially if quality is poorly observable (Chen & Grabowski, 2015).
A rich stream of studies, mainly in the area of health services research, has attempted to assess the impact of nurse staffing on patients’ outcomes in order to provide a guidance about how to set minimum thresholds. Systematic reviews have led to fairly different conclusions.6 In most cases, the empirical findings point toward a positive association between staffing and NH quality, even though some studies do not find any significant impact or, more rarely, display a negative association. Moreover, the available evidence pinpoints a stronger association between patient outcomes and per-patient hours of registered nurses (RNs), the more specialized group, while the link appears weak for less skilled workers such as licenced nurses and almost vanishes for nurse aids (NAs; Konetzka, Sterns, & Park, 2008).
The central role of nursing staff emerges also when considering subjective rather than objective measures of NH quality. In the former approach, judgments about quality are elicited from the main actors of the NH market. In most cases, surveys are administered to NHs’ residents to measure satisfaction with the service and to assess the relative importance they attribute to a range of quality domains (e.g., safety, comfort, autonomy, etc.). For example, Barsanti, Walker, Seghieri, Rosa, and Wodchis (2017) compared residents’ attitudes using two large surveys from Ontario (Canada) and Tuscany (Italy). They found that, according to the residents’ view, the areas of prior intervention should be improvement in staff’s professional skills and competences (Tuscany) and the ability to understand residents’ feelings and emotional needs (Ontario). Buljac-Samardzic and van Woerkom (2018) took a different perspective and examined the workers’ view in order to identify the factors affecting quality of care, focusing in particular on whether perceived workload influences care provision. They stress the potential mitigating effect of team support, which allows caregivers to concentrate on the professional skills where they feel stronger (strength use). According to the authors’ results, workload does not seem to be negatively associated to care quality and patient safety, as perceived by workers. The study also found that workers who are supported by their team and have the opportunity to exert their professional strengths avoid compromising quality of care even when subject to intense workload and stress. The authors conclude that leveraging workers strength is an important management tool for ensuring high (perceived) quality of NH care under increasing pressure from the demand side.
Policy Shocks and Causal Effects
The conflicting findings of the literature that studies the link between nurse staffing and quality partly reflect the specificities of the experiences analyzed. Nevertheless, serious empirical challenges have hindered the causal identification of the effect of interest. One of the main obstacles arises from potential endogeneity of staff hiring decisions, usually taken jointly with average resident acuity. Unobserved case-mix differences may yield downward biased estimates, since more complex patients are typically associated with less favorable outcomes and higher staffing at the same time. Consequently, omitted variable bias and reverse causality likely affect contributions based on cross-sectional data that fail to account for unobserved heterogeneity.
A new generation of studies has made systematic use of longitudinal data and exploited policy shocks to causally identify how changes in nursing volume and skill mix affect quality. Konetzka et al. (2008) first pointed out the methodological shortcomings of the literature. In their study, they included facility-level fixed effects to control for time-invariant unobserved heterogeneity. In addition, they relied on instrumental variables (IVs) to address potential endogeneity caused by the joint decisions regarding nursing staff and patients’ acuity. IVs for NH staff exploit the adoption of a PPS for Medicare beneficiaries. Such change was associated to a reduction in average reimbursement that produced an exogenous financial shock possibly affecting decisions concerning nursing staff. Parameters identification was achieved thanks to the variation in time and intensity of the shock across NHs. In the IV estimates, the quality-enhancing role of registered nurses, reducing both pressure sores and urinary tract infections, was larger in magnitude compared to the case that treated staff as exogenous. Such evidence confirms the importance of accounting for endogeneity of nursing staff to avoid downward-biased estimates.
Tackling the same problem, Tong (2011) took advantage of a legislation change that raised minimum staffing standards in California and showed that total employment increased by 10% in low staffing NHs, while no change was found in high staffing ones. Then, the causal effect of staff availability on residents’ mortality was identified using IVs. Post-policy staffing levels were instrumented considering whether facilities were subject to constraints because of the legal change. The results from the IV estimates confirm that higher staffing levels significantly reduce on-site mortality. Again, failing to account for higher employment levels in centers that assist more needy residents would have biased downward the estimated impact of staffing variables on outcomes. In the same vein, Lin (2014) exploited exogenous changes in NH regulation. The data covered all U.S. states that either strengthened existing staffing requirements or imposed new ones, and quality was measured by deficiencies assigned by regulatory agencies. The analysis accounted for both time-invariant and time-varying unobserved heterogeneity using panel IV methods. IVs were computed as the distance between prepolicy staffing levels and the new thresholds, assuming that the policy shock affects NH quality through changes in nurse staffing only. The distinction between high-skilled RNs and low-skilled NAs was found to be central: RNs ensure better performances in terms of both lower frequency and severity of the violations of care standards, whereas quality is not affected by the level of NAs. Formal testing found evidence of endogeneity and the ordinary least squares results largely underestimated the magnitude of the impact of nursing staff on quality compared to IVs.
The fact that staff requirements are often set in terms of total direct hours of care paves the way to possible strategic responses through a change in nurse skill mix. The policy implicitly incentivizes input substitution in favor of less specialized personnel (NAs), at the expenses of the more qualified and costly ones (RNs). By using a greater proportion of low-paid unskilled nurses, NHs may partially offset the higher cost of the workforce due to the more stringent occupational standards. This may have unintended adverse effects on quality, given the contribution of skilled nursing to high-quality care. Chen and Grabowski (2015) estimated the causal effect of a change in minimum staffing requirements using a difference-in-differences (DID) design. NHs in California and Ohio were the treated observations, as these states underwent important reforms in minimum staffing in the early 2000s. Facilities located in states not subject to minimum staff regulation were the control group. Quality was measured by indicators including deficiencies, process measures, and preventable episodes (e.g., pressure ulcers and contractures). The study evaluated short-, medium-, and long-run effects of the policy change using a 10-year panel data set. The findings lend support to the hypothesis of strategic behavior since a positive total net effect in the supply of hours of nursing care was associated with a reduction in the use of specialized workers. The positive net effect persisted even three or more years after policy implementation and can be ascribed to the increase in working hours of low-skilled NAs that outweighs the reduction of high-skilled RNs. Regarding quality, there were improvements in terms of lower frequency and severity of deficiencies and contractures, even though other outcome and process indicators displayed no significant post-policy change.
Considering a system outside the United States, Friedrich and Hackman (2017) studied the impact on health outcomes of a large and exogenous shock affecting nurse employment in Denmark. They exploited a parental leave program that gave the opportunity to take up to one-year absence per child aged zero to eight: a policy that led to a sudden and persistent 12% reduction in the stock of working nurses. Implementing a DID analysis, the authors used the exogenous variation in nurse employment at the county-health-sector-year level to estimate the effect of exposure to the reform. The results displayed a 13% increase in mortality among the elderly aged 85 and older and, consistent with theoretical predictions, indicated a decrease in the hospitalization rate among sicker residents.
On the whole, the more recent evidence consistently points to the importance of controlling for endogeneity when assessing the benefits of higher and skilled nursing staff on NH quality care. Empirical strategies that exploit exogenous policy shocks confirm that failure to account for potential endogeneity would lead, in most cases, to underestimate the impact of interest. However, despite these improvements, the overall picture is far from homogeneous. Even the contributions that explicitly tackle causal inference in policy evaluation are subject to restrictive assumptions. In particular, the IV approach requires that the policy shock used as instrument does not affect quality through channels other than the change in nurse staffing. Such an assumption is not easily testable, and, in some circumstances, it may threaten identification if NHs implement wide-ranging responses to policy shocks. Moreover, the available evidence shows that, even when quality improves for some indicators, others may display no significant change.
Minimum Staffing Standards: Financial Consequences and Market Exit
More compelling staffing standards may have negative financial implications because of the rise in operating costs. Particularly if private-pay residents account only for a residual share of the market as in the United States, NHs have limited opportunities to compensate higher staffing costs by increasing revenues. Because of this, NHs may decide to change the workforce skill mix, reducing the share of more costly, specialized workers. However, substitution of skilled with unskilled workers can have adverse effects on quality, thus offsetting the expected benefits of increased staff intensity. Hence, a comprehensive evaluation of the gains and losses of enacting minimum staff requirements should also account for its impact on NHs’ financial performances.
Bowblis (2015) examined financial indicators encompassing per-bed patient costs, revenues, and profit margins, as well as total operating margins. After controlling for differences in regulations across U.S. states and for changes in the stringency of requirements over time, the study did not detect a significant impact of minimum staffing regulation on per-bed costs and revenues. On the contrary, per-bed profits and overall NH profits dropped significantly in states with more stringent regulation. Financial losses were larger in NHs relying on Medicaid patients relatively more, supporting the hypothesis that such providers have relatively lower opportunities to increase revenues. Even if the impact on cost-per-bed is negligible on average, regulators should carefully evaluate the financial implications of raising staffing thresholds, as the number of NHs incurring in financial losses might increase. The consequences of an uncompensated regulation-induced rise in operating costs can be exacerbated considering that many facilities operate with small financial margins and may be driven out of the market because of more stringent staffing requirements.
NHs’ market exit in the United States is the focus of the study by Bowblis and Ghattas (2017). They examined the differential responses to newly enacted minimum staffing requirements between NHs that were largely above the threshold before policy implementation, those that were around the threshold, and those that were significantly below it. They found that staff adjustment was heterogeneous and depended upon the initial distance from the threshold. Convergence in total staffing levels was mainly due to the rise of nonskilled nurses, while high staff providers reduced their personnel. The incidence of NH exits appears insignificant and not associated to any detectable pattern, giving no support to the concern that compelling staffing standards may cause shortages in the supply of NH services. The determinants of market exit have been investigated also for the United Kingdom. Allan and Forder (2015) showed that NHs operating in more competitive local contexts as well as facilities delivering lower quality care have a higher probability of being subject to closure. The evidence is consistent with the hypothesis that NHs displaying poor performances in terms of quality are more likely to suffer from regulatory actions due to errors and mismanagement, which may ultimately drive these providers out of the market.
Labor and Technology Substitution
Despite the attention devoted to the relationship between a skilled and an unskilled labor force, very little is known about the substitutability/complementarity between innovative technologies and labor in NHs. This industry is labor intensive, and the use of advanced technologies and robotics is still in its infancy. Moreover, most potential applications pertain to the development of technological aids and supports aimed at facilitating home care (e.g., devices for communication and home health monitoring, physical aids, etc.). Their diffusion will presumably favor the choice of assisting the elderly dependent patients at their domicile, thus postponing NH admission. However, the lack of familiarity of elderly people with information technology (IT) has hindered a generalized adoption of IT devices to assist dependent individuals, though this is expected to change with the aging of generations more familiar with technological devices (Stone et al., 2016).
For these reasons, the scope for widespread applications of robotics IT to NH care is somewhat limited compared to other health providers (e.g., hospitals), although innovations such as platforms for improving clinical patient supervision and management finds a favorable environment especially in the most advanced and highly skilled NH facilities (Zhang et al., 2016). According to Lu, Rui, and Seidmann (2018), the introduction of innovative ITs, meant to help streamline nursing activity and reduce medical errors, improved clinical quality. Such evidence suggests that IT may lead to better health outcomes even in a labor-intensive sector such as NH care. As for input substitution, no significant difference in utilization of nursing staff was detected after the introduction of the innovation. However, even with no effect on average, there is evidence of a considerable heterogeneity according to where a certain NH locates in the vertical spectrum of quality.7 Complementarity between labor and IT prevailed in NHs whose quality was above the median, with an increase of 5.6% of the nurse-to-patient ratio following the adoption of the new technology. Conversely, NHs falling below the median registered a drop of 7.6% in nurse staffing after the new technology was introduced, pointing to substitutability between labor and IT among poorly staffed providers at baseline.
Ownership Status and Quality
For-Profit and Not-for-Profit Nursing Homes
Among the structural factors that influence NH quality, ownership status exerts a prominent role. In the United States, the main distinction is between private for-profit (FP) and not-for-profit (NFP) NHs, although in European countries public facilities are also relevant players. In more recent years, the consequences of NHs being either independently operated or part of multifacility chains and the role of chain size have attracted increasing interest (Harrington, Olney, Carrillo, & Kang, 2012; Kai et al., 2016; Lu & Wedge, 2013). Identifying systematic differences in health outcomes by NH type, along with insights on the possible channels that generate such differences, are central policy objectives.
Economic theory provides little backing for identifying the best-performing type of provider in terms of quality. It is difficult to predict a priori the effect of ownership status on quality, depending on multiple mechanisms at work (Brekke, Siciliani, & Straume, 2012). On the one hand, being subject to the nondistribution constraint, NFP NHs can invest extra resources in high-quality care since they need not compensate shareholders and pay profit taxes. Moreover, NFP status may be associated with distinctive attitudes, including altruistic mission of the organization and workers’ motivations. On the other hand, the profit motive strengthens incentives toward operational efficiency and can induce FP providers to raise quality to attract patients, thereby increasing revenues. Since economic theory does not come out with clear-cut predictions, establishing whether ownership status affects NH quality remains an empirical issue.
Despite extensive empirical research comparing NFP and FP NHs, no conclusive statement can be drawn on their relative capacity to deliver higher quality care. Systematic reviews and meta-analyses provide mixed evidence, with a great deal of cases showing no detectable quality differences by ownership type. When significant gaps emerge, NFP NHs tend to outperform FP ones, although a few studies also point in the opposite direction (Comondore et al., 2009; Hillmer et al., 2005). A major threat to the identification of the causal impact of ownership status comes from possible unobserved differences in patients’ characteristics. This notwithstanding, most of the existing literature does not account for possible sources of unobserved heterogeneity between types of facilities. To the extent that the two groups of providers serve patients who differ in acuity levels and payment type, the empirical findings remain poorly informative and potentially misleading regarding what would happen to quality should an NH change ownership status (Konetzka, 2009).
To respond to these concerns, a stream of recent studies has attempted to isolate the impact of NH ownership on quality relying on rigorous identification strategies. In two related works, Grabowski et al. (2013) and Hirth, Grabowski, Feng, Rahman, and Mor (2014) used the difference in distance between patient’s residence to the closest NFP NH and patient’s residence to the closest FP NH (“differential distance”) as exclusion restriction for ownership status. The IV captured patient’s relative exposure to NFP versus FP supply of NH services and addressed self-selection into different types of facilities according to the (unobserved) patient acuity. Failing to control for this would introduce endogeneity bias in the estimate of the ownership coefficient. The identifying assumption of the model is that individuals do not consider the relative weight of NFP NH services when making their residential choice, with the IV being likely correlated with the ownership status of the admitting NH but assumed to be uncorrelated with health status.
Grabowski et al. (2013) restricted the analysis to postacute patients newly admitted to an NH to avoid path dependency and measured quality as rehospitalisation within 30 days of discharge, death rates, activities of daily living (ADLs) functioning, pain, and mobility. Their findings are consistent with the hypothesis that NFP NHs attract patients in worse health conditions on average. When endogenous sorting of patients was not accounted for, no clear performance pattern according to ownership status emerged, with NFP institutions performing better in terms of reduced risk of rehospitalization but worse for pain incidence and ADLs. On the contrary, the overall picture changes once ownership status is instrumented using differential distance: IV estimates indicate that NFP NHs consistently deliver better health outcomes for all indicators.
Hirth et al. (2014) focused on hospitalization of long-stay residents. Previous literature consistently finds lower hospitalization rates among residents in NFP facilities. However, such a difference may not reflect a causal relationship but instead may depend on unobserved differences in the residents’ initial conditions or on higher quality of NFP care. Also, the two groups of providers may differ in terms of varying admission thresholds, if certain providers are better equipped to treat severe patients who would be otherwise referred to the hospital. Separating selection effects from those due to different hospitalization thresholds or different quality of care is crucial to identify the channel leading to diverse patterns in hospital referrals. With this aim, the authors implemented an IV strategy that accounted for the endogeneity of ownership status and NH quality and explored differences in hospitalizations splitting referrals based on whether or not they could be prevented. A disproportionate decline in preventable hospitalizations should be observed if the negative correlation between NFP NHs and hospitalization rates is driven by their higher quality of care. Findings support the existence of a causal relationship between NFP status and hospitalization rates, with IV estimates displaying a negative relationship that was larger in magnitude than the one estimated assuming ownership status is exogenous. Moreover, the decline in hospitalization was not concentrated on preventable conditions, thus reinforcing the hypothesis that the negative correlation between NFP status and hospitalization is most likely explained by different hospitalization thresholds, with the NFP facilities being more prone to manage critical patients “in-house.”
Although the link between ownership status and quality of NHs has attracted a great deal of attention in the United States, there is still limited research in other countries. Recent exceptions include Stolt, Blomqvist, and Winblad (2011) and Bergman, Johansson, Lundberg, and Spagnolo (2016) for Sweden and Barron and West (2017) for the United Kingdom. Stolt et al. present a cross-sectional analysis of Swedish elderly care. After adjusting for sociodemographic differences across municipalities, the results suggest that private-sector (mainly FP) facilities are associated with a significantly lower staff-to-resident ratio compared to public-sector facilities. On the other hand, private-sector providers seemed to perform better in other dimensions of quality related to aspects of personal service. These include choosing between different food alternatives, offering a more reasonable duration between evening meal and breakfast, and leaving residents to participate in the formulation of their care plan.
Bergman et al. (2016) examined the effect of opening the market for elderly care services to private provision in Sweden between 1990 and 2009. Assuming that the population below 70 years of age remains unaffected by the change in NH provision, they used a difference-in-differences-in-differences framework, which compares the gap in mortality rates between those older than 69 and those younger than 70 for the municipalities that had partially privatized provision and for the ones that did not (yet) reform the local market. They recorded an increase in quality associated with the shift from pure “in-house” to private provision, with a reduction of the mortality rate by 1.6%. This estimate, however, captures the joint effect of introducing private service provision and competition, and the empirical strategy does not allow the disentanglement of the impact of privatization per se from the associated increase in competition.
Barron and West (2017) contributed to the debate by conducting a U.K.-based research study. Controlling for a range of variables potentially associated with quality, they contrasted the quality of care provided by FP NHs to that enjoyed by residents in facilities operated by local authorities or NFP organizations. Quality measures consisted of indicators based on inspections provided by the Care Quality Commission over the domains of safety, effectiveness, respect, meeting needs, and leadership. The evidence showed that NFP NHs and those run by local authorities perform better in terms of all quality domains compared to FP NHs.
An important influence on the relationship between ownership status and quality is also exerted by asymmetric information. Because of limited observability, quality of NH services is largely noncontractible. Asymmetries of information confer providers strategic advantages over patients, but NFP organizations are expected to have fewer incentives than FP facilities to exploit such advantages to cut quality (Hansmann, 1980; Rose-Ackermann, 1996). Hence, when quality is poorly observable to consumers, the NFP status may act as a signaling device and mitigate market failures (Hirth, 1999).
Chou (2002) tested this hypothesis using visits from spouse or adult children as a proxy for (absence of) asymmetry of information. The underlying assumption was that family members acquire information about quality during their visits. Therefore, patients receiving visits were assumed not to suffer from asymmetric information, while those who did not have regular contacts with relatives were hypothesized to be poorly informed. The results corroborated the research hypothesis. The study found no significant differences in clinical outcomes (mortality, ulcers, urinary tract infections, and dehydration) between FP and NFP NHs in patients who received visits from relatives. On the contrary, NFP NHs performed better than FP NHs among patients receiving no visits. Similar results emerged taking cognitively unaware patients as the uninformed parts.
Moving from the same conjecture, Jones, Propper, and Smith (2017) found that the launch of public reporting initiatives based on quality report cards led to a significant drop in the share of NFP NHs in the U.S. market, as the reputational benefits enjoyed by the NFP status lost part of its relevance when quality information became common knowledge.8 Moreover, high exit rates concentrated among low-quality NFP NHs, which were more clearly identified as poor performers in the new context. Such findings confirm that the role of NFP status as a signal for high quality depends on the degree of information asymmetry.
Additional Ownership Characteristics
The influence of more specific ownership-related features on quality has been investigated as well. Bowblis and McHone (2013) used an identification strategy based on the differential distance between patient residence and types of facilities to compare quality performance between continuing care retirement communities (CCRCs) and traditional NHs. CCRCs are characterized by high costs of admissions and target the richer fraction of private-pay patients. Besides the core LTC services, CCRCs also deliver postacute rehabilitation care, typically serving local patients. The study compared quality provided to Medicare patients that received postacute rehabilitation in centers affiliated with CCRCs with the same type of patients assisted in traditional NHs.9 Also, in this case, estimates that did not account for potential endogeneity of patient case-mix produced contradicting results, with centers affiliated with a CCRC performing either better or worse than traditional NHs, depending on the outcome considered. Conversely, more consistent evidence emerged when choice was estimated using the distance between the patient’s residence and the closest CCRC NH relative to the distance between the patient’s prior residence and the nearest traditional NH as an exclusion restriction. CCRCs affiliated centers do not seem to perform better than standard NHs in terms of clinical quality, measured by the incidence of pain, delirium, and pressure ulcers. While no difference was detected for the first two indicators, CCRS performed significantly worse for the third. Such findings contrast with conventional wisdom according to which CCRS deliver better quality care than standard NHs.
With the development of more articulated ownership structures, the distinction between investor-owned NHs and those run by salaried managers also gains more relevance. Managers owning equity shares of the enterprise face incentives different from those who are salaried and might be willing to contain expenditures on unobservable components of quality to raise profits. However, they may also be keener to develop initiatives that increase NH attraction capacity in the long run. The net effect on quality of care of these diverging incentives is ambiguous. To achieve causal identification of the impact of manager-owned status, Huang and Bowblis (2018) accounted for self-selection of patients with different underlying characteristics by means of a control function approach. They first estimated the probability of choosing a manager-owned NH, and then quality was regressed against a set of controls, including the endogenous variable and the first stage residuals. The exclusion restriction was the differential distance of prior residence to the closest manager-owned NH, relative to the distance from prior residence to the closest NH run by salaried managers. The main findings indicate that manager-owned facilities provide higher quality for most indicators.
The empirical evidence points to the importance of carefully evaluating the distribution of information and of accounting for the multidimensional nature of the quality of care when analyzing the differential performance between FP and NFP providers. NFP status seems to have an edge over FP status, at least in terms of selected measures of quality when patients do not observe quality. However, when asymmetries of information become smaller, differences in some aspects of performance fade away. Last, the relation between privatization and quality is likely to vary with different market structures, depending also on the potential influence of competitive forces at play.
Prices, Competition, and Quality of Care
Standard arguments suggest that, in the absence of relevant frictions, higher prices and reimbursement rates should incentivize the delivery of higher quality services. However, the link between prices and quality in the NH market is a complex one, given that asymmetry of information weakens the signaling effect of prices and the consequences of price regulation are strictly interconnected with other regulatory measures such as limitations in production capacity and administrative standards. As a consequence of the high share of Medicaid and Medicare beneficiaries, the interplay between reimbursement rates and certificate-of-need (CON) regulation has been extensively investigated in the United States.10 Conversely, several European countries (e.g., United Kingdom, Germany, and Sweden) that previously witnessed a dominant role of bureaucratic public organizations have recorded a progressive externalization of welfare services and increased price competition (Forder & Allan, 2014).
Quality Under Medicare and Medicaid Reimbursement
In the United States, an early wave of studies exploited the cost-plus reimbursement scheme to evaluate how within-state changes in the “plus” component affect nursing quality. This literature has generally found that an increase in reimbursement rates is associated with a reduction in quality of care (Gertler, 1989, 1992; Nyman, 1985). Such apparently puzzling evidence has been rationalized on the grounds of regulatory constraints on bed supply. NHs usually treat patients covered by federal programs and private-pay patients jointly, delivering uniform quality. Consequently, increases in Medicaid reimbursement rates reduce the foregone earnings associated with this relatively less lucrative group of patients. Together with regulation-induced shortage of beds, an increase in Medicaid payments may limit incentives to attract private patients that are more responsive to high-quality services.
Later works use a different empirical strategy and nationally representative data sets to assess how different average reimbursement rates across states affect quality. The findings contrast with previous evidence pointing to a positive impact of higher reimbursement on quality (Cohen & Spector, 1996; Grabowski, 2004). In the attempt to reconcile the two approaches, Grabowski (2001) considered both sources of information—state-level and nationwide data—using facility-specific and state average reimbursement rates, respectively. The study also extended the set of quality indicators to health outcomes, in addition to input-based measures and regulatory violations, and distinguished facilities according to the share of Medicare beneficiaries, as this may induce heterogeneous shocks when Medicare reimbursement rules change.11 The resulting evidence suggests that the negative link between reimbursement rates and quality cannot be generalized when a wider set of indicators and different time spans are considered. The prevailing effect appears to be positive, although mainly small in magnitude. Moreover, it concentrates in centers treating a more balanced share of private-pay and publicly covered patients, compared to those who can be classified as “fully Medicaid” or “fully private.” Such findings are consistent also with the attenuation in the stringency of CON policies experienced since mid-1990s in the United States, which resulted in an increase in production capacity and a decline in occupancy rates of NH beds and waiting times.
The adoption of “per diem” PPS for skilled nursing facilities by Medicare has been used to provide further insights on the link between payment and NH quality. This shift in the payment scheme altered the incentives at the margin to treat Medicare patients, who typically require short-stay rehabilitation services, but it also determined a negative financial shock with possible adverse consequences also for long-term residents, largely covered by Medicaid. The concerns come from the sizeable reduction in financial transfers associated with the new reimbursement mechanism. Not being financially neutral, the new scheme was expected to curb cross-subsidization that benefits Medicaid patients. Consistently with the reduction in nurse-to-patient ratio following the introduction of PPS (Konetzka et al., 2004), the findings for health outcomes indicate that the consequences of the shift from cost reimbursement to PPS were partly transferred to long-stay non-Medicare patients. NHs suffer from a drop in health outcomes, displaying higher incidence of pressure sores and urinary tract infections (Konetzka, Norton, & Stearns, 2006).
Interestingly, some states have augmented Medicaid reimbursements to include a pass-through subsidy linked to staff expenditures. The aim of the initiative is to use financial incentives to ensure that Medicaid enrolees benefit from adequate staffing levels beyond what can be obtained using staffing requirements only. Input subsidization leads to a significant increase in average staffing level, producing a heterogeneous impact across NHs. Facilities assisting a lower share of Medicaid patients respond more intensively. As for quality, the empirical analysis reports a sizeable reduction in pressure sores caused by pass-through subsidies (Foster & Lee, 2015).
The impact of changes in Medicaid reimbursement has been recently studied also by Hackmann (2019), who developed a structural model of demand and supply of NH services based on Pennsylvania data.12 Taking advantage of the rule for setting Medicaid reimbursement, the study used variation in costs of distant facilities as an exogenous shock affecting reimbursement rates to a certain provider. The estimates show that an increase in Medicaid payment rates caused a sizeable rise in the (skilled) nurse-to-resident ratio, while no changes occurred for other inputs. Competition among NHs was based on the choice of a skilled nursing labor force as a key driver for quality and on the rate of private-pay patients. The empirical findings point to inadequate skilled nursing staff levels, whose increase would lead to a substantial rise in social welfare. Inefficiency concentrated on NHs where the share of Medicaid patients was nonnegligible, whereas there was no sign of it in NHs assisting almost only private-pay residents. Such evidence is suggestive of the fact that a low Medicaid reimbursement rate might be largely responsible for low staffing. A simulated 10% increase in Medicaid reimbursement rates was estimated to increase nurse staffing by 8.7% on average and to reduce private-pay patients by 4.5%. Such a response is expected to produce beneficial effects on quality of care and to expand access opportunities for less lucrative patients in particular.
Price Differential and Quality in Germany
Compared to the United States, Germany is particularly interesting for assessing the link between price and quality in the NH industry. Germany is in fact characterized by a different regulatory framework and provides complementary insights about how prices affect NHs’ decision over quality. Patients are covered either by statutory or private LTC insurance, with prices resulting from negotiations between providers and sickness funds. Yet, facility-specific mark-ups over negotiated prices are covered by patients or, if they cannot afford them, by social assistance, thus making patients price-responsive. At both bargaining steps, the price range is heterogeneous across regions, with strong evidence of spatial clustering in the rates charged. Only 30% of the price difference seems to be explained by patient, structural, and other local characteristics, whereas almost 40% of it can be attributed to different negotiation practices across regions (Mennicken, Augurzky, Rothgang, & Wasem, 2014). In such a context, Herr and Hottenrott (2016) found that the price differential significantly influenced the quality of NH care. In particular, after accounting for potential endogeneity of negotiated prices and for local characteristics, they found evidence of an increase in reported quality caused by higher prices. Moreover, higher installed production capacity at the local level was in turn associated with higher quality, thus supporting new investments to increase the supply of NH beds, especially in underserved areas. While the previous analysis was based on cross-sectional, facility-level data, Reichert and Stroka (2018) exploited longitudinal patient-level information to evaluate to what extent NH prices are associated with process measures of quality, such as the use of psychotropic medications. They showed that higher NH prices reduce the use of drugs at risk of misuse and over-prescription, while no association was found with indicators of physical impairment. Such findings are consistent with the conjecture that additional resources ensured by higher per-patient compensations allow better surveillance and reduce the risk that providers rely on sedative drugs to substitute for lower staffing and inadequate patient monitoring.
On the whole, the literature on the impact of prices on NH quality consistently highlights the strict connection of the impact of price variations with the broader regulatory context. In general, an increase in the financial compensation received by NHs seems to have beneficial effects on residents. Nonetheless, to avoid unintended effects, policy initiatives in this area should be evaluated in light of the full set of constraints faced by providers, of the possibility to “pass-through” different types of patients the consequences of the interventions, and of the intensity of competition in local markets.
Information Disclosure and Quality
The problems encountered with a top-down regulation based on the enactment of standards of care and monitoring have encouraged the adoption of less intrusive, bottom-up interventions. They attempt to reduce market frictions by promoting accountability, market transparency, and competition, with the aim of improving allocative efficiency and patient welfare. Within this context, several initiatives have been designed to foster quality competition by relying on publicly released information.
In the United States in the 1990s, a pioneering initiative called the Nursing Home Compare web platform program was accessible to the general public, which initially included indicators on NH efficiency and staffing. In the early 2000s, the Nursing Home Quality Initiative (NHQI) based on report cards progressively included quality related indicators (Grabowski & Town, 2011). One of the problems encountered with the initial release of the NHQI data was the difficulty for patients to manage the available information and to choose their destination using complex report data. The “Five Star Quality Rating” program launched in 2008 tackled the issue, presenting information through a rating system based on stars grading to make the service more operational. The new system condenses information from a composite set of indicators into summary measures for the main domains of interest using a convenient, ready-to-use format (Huang & Hirth, 2016). Public reporting initiatives on NH quality are not limited to the United States. They spread out over several countries, some of which have developed comprehensive reporting schemes that collect information on a systematic basis (e.g., Germany, United Kingdom, Sweden), while others have implemented more limited pilot experiences (e.g., Austria, Finland; Rodrigues, Trigg, Schmidt, & Leichsenring, 2014).
Dissemination of quality information through report cards is deemed to improve patient sorting according to preferences, as it facilitates patient choice in a context of otherwise large information asymmetries. When quality information is more easily accessible to consumers, investments in staff, infrastructure, and organization can be rewarded by stronger attractiveness and larger market shares, thereby increasing the incentives for NHs to deliver high-quality care. Therefore, a key ingredient for the success of these initiatives is patients’ responsiveness to the new information, and estimating the elasticity of demand to quality is crucial to understand whether providers retain incentives for increasing quality.
The main empirical problem in evaluating the effect of quality disclosure arises from the need to disentangle demand- and supply-side responses. To the extent that NHs respond to the release of quality information by changing prices, this may introduce a confounding factor for the assessment of patients’ reaction to a reduction in information asymmetries. In this vein, Huang and Hirth (2016) found that, after the launch of the Five-Star Rating System, top-performing NHs raised prices for private-pay patients by nearly 5% more than NHs at the bottom of the quality ranking. Moreover, the size of the effect varied with the degree of competition providers were exposed to, and the price increases were larger for NHs operating in a more competitive environment and subject to binding capacity constraints due to public regulation. To tackle the issue, researchers often exploit the regulatory constraints of the NH market. Werner, Norton, Konetzka, and Polsky (2012) studied patients’ choices before and after information disclosure focusing on Medicare patients only, since the strict regulation of the federal program allows to control for prices and supply-side characteristics. They showed that the best-performing facilities increased attractiveness as a consequence of public reporting, even though the magnitude of the effect appeared fairly small. Moreover, the impact varied with the type of indicator considered, reductions in pain scores being the factor that influenced patients most. Consistently with expectations, patient choice was unaffected for the NHs exempted from performance reporting due to their small size.
While the online publication of report cards has also involved countries different from the United States, until now there has been only limited evidence evaluating the effect of mandatory quality disclosure in these contexts. An exception is Herr, Nguyen, and Schmitz (2016), who analyzed the relationship between public reporting and NH quality in Germany following the introduction of a reform aimed at increasing transparency with the publication of online report cards. They found that the performance of NHs, as measured in terms of a selected set of indicators, increased after policy implementation, and quality improvements implemented by the less performing providers were the main driver of the result.
The impact of information disclosure on the behavior of NHs may also vary depending on the interaction with other factors, such as ownership type and the degree of competition. While under asymmetric information NFP organizations may have relatively stronger incentives to invest in high-quality care than FP facilities, having access to greater information may allow consumers to screen facilities more accurately. Hence, in the new scenario incentives for raising quality may increase relatively more, given that stronger attraction capacity can positively affect financial performances. The expected gains (losses) from signaling high- (low-) quality levels are larger in more competitive markets, while poor-quality providers suffer from more limited adverse consequences in markets with shortage of supply and heterogeneous providers. This suggests that the impact of information disclosure should be amplified in more competitive settings.
There is evidence supporting the view that responses differ by ownership type and that the magnitude of the impact depends upon the degree of local competition. For instance, Park and Werner (2011) find a positive, albeit modest, relation between profit margins and quality, but the effect intensifies after quality information is publicly released. The result is significant for FP NHs but not for NFP NHs and becomes stronger in more competitive markets. Grabowski and Town (2011) studied the consequences of the Nursing Home Quality Initiative (NHQI) exploiting the staggered implementation of the program across U.S. states. They showed that, despite an almost negligible impact of information disclosure, NHs operating in more competitive markets react more intensively to the initiative compared to the organizations that enjoy larger market power. For them, the relative improvement in quality was sizeable for a subset of health outcomes. Exploiting more recent data, Zhao (2016) examined the interaction between competition and increased transparency and how this affects NH quality. The author exploited a change in quality reporting for U.S. NHs, captured by the improved access to quality information after the launch of the Five Star Quality Rating system. To address the endogeneity arising from simultaneity between competition and quality, the author constructed an IV for competition using exogenous variation in the geographical proximity of NHs to their potential consumers. The impact of competition on quality was identified by the differential effect of competition between the periods before and after the reform. The results show a significantly stronger impact of competition on quality after the introduction of the new rating system, suggesting the importance of accounting for the joint effect of improved information transparency and competition regulation.
Public reporting of quality also comes with potential drawbacks, such as incentives for cream skimming or for coding manipulation. Although such concerns are shared with similar programs in other settings, including hospital care, the unintended consequences of providers’ opportunistic behavior can be especially serious in the NH industry where prices are regulated and supply is subject to capacity constraints. When evaluating the costs and benefits of information disclosure, these aspects should be carefully considered. For instance, the attempt to improve quality ratings through cream skimming may substantially limit access to NH care to more complex patients, given the supply constraints that affect many local contexts.
The evidence on the unintended effects of information disclosure in the NH market is still scarce. Overall, the results suggest that the concerns about strategic patient selection based on risk have limited empirical relevance (Mukamel, Ladd, Weimer, Spector, & Zinn, 2009). Following the publication of data on NH quality, patient sorting has improved, with more (less) severe patients assisted in high-quality centers more (less) frequently than before. Yet, this does not appear to be driven by cream skimming, since the risk profiles of new patients remain fairly stable on average. Nonetheless, there is evidence that poor-quality NHs may have engaged in down-coding practices to make their patients appear in better conditions after information disclosure (Werner, Konetzka, Stuart, & Polsky, 2011).
When tariffs vary substantially across payer source, heterogeneous patient profitability may represent an additional important driver of selection beyond individual risk. He and Konetzka (2015) considered changes in admissions rates of Medicare versus Medicaid patients.13 They showed that, due to public reporting, high-quality, capacity-constrained facilities react by increasing admissions of Medicare residents, while the share of Medicaid residents drops significantly as a consequence of selective admissions by better performing NHs. Such evidence suggests that profitability can be a more relevant source of selection than patient risk and calls for attention by policymakers to prevent patients enrolled with less generous programs from being denied access to high-quality centers following information diffusion.
Organizations may also engage in more elusive gaming strategies that exploit the institutional characteristics of information disclosure programs. Konetzka, Polsky, and Werner (2013) studied the consequences of the possibility of excluding from the quality assessment patients assisted in an NH for less than 14 days. Facilities may selectively discharge in advance patients that are at high risk of delivering poor outcomes and that might negatively affect their quality scores. Such a process likely involves complex patients who show poor initial response to treatments and for whom the facility may opt for rehospitalization. The analysis finds significant selective rehospitalizations due to the introduction of the NHQI. In this case, NHs did not ration treatments to relatively severe patients, as predicted by the standard cream-skimming hypothesis—quite the opposite: such patients were overtreated (i.e., they ended up receiving hospital care even if they could still be assisted in NHs). Such evidence suggests that organizations seem to engage in opportunistic behavior, with potentially adverse consequences on allocative efficiency and overall health expenditures, as they transfer to hospitals the burden of treating patients who could have benefitted from NH care.
Finally, very few studies examine whether information disclosure give facilities an incentive to redirect resources across different dimensions of quality. One exception is Lu (2012), who exploited an information shock after the introduction of the NHQI in 2002, which led to an increase in information about selected dimensions of NH quality. To test the impact of report cards on NH behavior, the author first focused on pilot states only and exploited the time-series variation of whether an NH was randomly inspected. Second, by using a sample of both pilot and nonpilot states, the author applied a DID approach, whereby the effects of the policy were identified by comparing changes in the outcomes of interest between pilot and nonpilot states before and after policy implementation. The first set of results suggest that quality improves along the NHQI-reported dimensions, while they become significantly worse along the NHQI-unreported ones. The second set of findings suggests that NHs do not significantly increase the total nursing inputs in response to the policy, although there was a slight increase in the hours of skilled nursing staff. Such evidence is consistent with the hypothesis that the disclosure policy incentivizes facilities to upgrade the skill mix of nursing inputs, with no decline in total nursing inputs. Finally, the results show that NHs with worse NHQI scores have a greater decline in both market shares and annual revenue, suggesting that consumer demand is sensitive to the disclosed quality measures. Taken together, the findings provide evidence for the “teaching-to-the-test” phenomenon, with the disclosure policy inducing NHs to divert resources toward reported dimensions of quality. However, overall quality improvements cannot be assessed from the existing evidence and will depend primarily on whether the reported dimensions are those that matter most to the patients.
To promote high quality of residential LTC, a strategy complementary to disclosure of performance indicators is based on NHs’ institutional accreditation. In this latter case, the certification acquired through the accreditation process should safeguard residents by providing the guarantee that preestablished structural and procedural standards have been met. The available evidence shows that accredited facilities perform significantly better than nonaccredited ones in terms of the four components of the Five Star Rating System (health inspection, quality, overall staffing, and RNs). Accredited facilities display fewer deficiencies and fines, with certification for postacute care further reinforcing performances (Williams, Morton, Braun, Longo, & Baker, 2017). However, although accreditation is a good predictor for quality even after controlling for NH size and ownership type, the direction of the causal relationship between accreditation and quality attainment is still to be established.
Even after several decades of intense debate and targeted policy initiatives, the objective of improving quality in the NH industry continues to rank high on the policy agenda. The multidimensional nature of quality and its close link with the general regulatory design of the LTC market have stimulated the investigation of its numerous facets as documented in this article. Most of the research focuses on the United States, and only recently a systematic analysis has been extended to other countries. Further enlargement of the coverage in the near future will hopefully contribute to a deeper understanding of how regulation can be improved to better address the needs of the frail elderly population in different institutional contexts.
The empirical literature studying quality-enhancing initiatives presents some relevant novelties. The advances in the design of the empirical strategy and the use of up-to-date econometric methods, aimed at identifying the causal effects of interest, represent qualifying contributions of the economics literature that have been fruitfully applied to the study of NH care. By contrast, studies from adjacent fields are usually more descriptive in nature and, in most cases, highlight correlations across the relevant indicators rather than genuine causal relationships. The increased rigor of policy evaluation has combined with the use of a broader and more accurate set of quality indicators, made possible by the great increase in publicly available information. A promising direction for future methodological improvements is the use of structural approaches jointly modeling supply and demand. Only very recently, a few such applications to the NH market have been proposed. Their main advantage is that they allow the design of proper counterfactual exercises for policy evaluation accounting for the complex set of institutional features of the market.
The influence of mandated staffing standards and of ownership status on quality have been subject to intense debate both in the policy and academic arena. Exploiting policy shocks to identify causal effects has allowed us to draw more clear-cut implications compared to the previous literature. On the whole, the most recent contributions highlight a greater effect of nursing staff intensity and skill mix on patients’ outcomes once endogeneity is accounted for. High staffing standards appear to be beneficial for residents’ well-being when the increased labor force delivers specialized nursing services, while the impact comes out as generally negligible when it involves unskilled personnel. This calls for caution when designing mandated staffing initiatives. Unless adequately monitored, they are vulnerable to strategic manipulation by providers through variation in skill mix at the expenses of more qualified workers. Caution is also required with respect to possible exits from the market of marginal NHs that may incur in losses under more stringent standards, as this may negatively affect social welfare, particularly in underserved areas.
Although not unanimous, the empirical findings appear, by and large, supportive of the hypothesis that NFP organizations involved in personal care deliver better quality services. Still, the gap over FP providers seems to narrow after public reporting of quality data. Given the composite ownership structure in the NH industry, the more recent U.S.-based research has moved beyond the distinction between FP and NFP status to concentrate on the role of managers and of NH chains—a line of research to be fruitfully pursued further in the future. In European countries, the attention is captured mainly by the expanding role of private providers. In contexts characterized by the reorganization of welfare systems, retrenchment of direct public provision, and large differentials in prices and production capacity, it is crucial to assess private providers’ ability to take up (part of) the role previously exerted by public organizations.
The shift from a top-down to a bottom-up regulatory approach has contributed to the blossoming of information disclosure initiatives based on quality rating report cards. Even if, so far, there is no evidence that such measures induced major changes in patient choice and in NH quality, they bear a relevant welfare-enhancing potential, as they favor transparency and accountability of NHs. Yet, the risk of possible unintended consequences, such as cream skimming and effort crowding out on unmeasured dimensions of care, should not be overlooked. Given the limited-choice opportunities that the vulnerable elderly population faces in many circumstances, market-oriented actions should complement rather than substitute regulation based on administrative standards and on oversight of NH activity. Finally, as NHs regularly treat patients covered by different programs, the consequences of the policy changes are often heterogeneous across payer types. In order to provide a more comprehensive picture of NH quality, future research should investigate more in depth the equity and accessibility implications of regulatory policies.
Allan, S., & Forder, J. (2015). The determinants of care home closure. Health Economics, 34(Suppl. 1), 73–83.Find this resource:
Bardey, D., & Siciliani, L. (2018). Nursing homes’ competition and distributional implications when the market is two-sided. TSE Working Papers 18–931. Toulouse, France: Toulouse School of Economics.Find this resource:
Barron, D. N., & West, E. (2017). The quasi-market for adult residential care in the UK: Do for-profit, not-for-profit or public sector residential care and nursing homes provide better quality care? Social Science & Medicine, 179, 137–146.Find this resource:
Barsanti, S., Walker, K., Seghieri, C., Rosa, A., & Wodchis, W. P. (2017). Consistency of priorities for quality improvement for nursing homes in Italy and Canada: A comparison of optimization models of resident satisfaction. Health Policy, 121, 862–869.Find this resource:
Bergman, M. A., Johansson, P., Lundberg, S., & Spagnolo, G. (2016). Privatization and quality: Evidence from elderly care in Sweden. Journal of Health Economics, 49, 109–119.Find this resource:
Bowblis, J. R. (2011). Staffing ratios and quality: An analysis of minimum direct care staffing requirements for nursing homes. Health Services Research, 46(5), 1517–1533.Find this resource:
Bowblis, J. R. (2015). The cost of regulation: More stringent staff regulations and nursing home financial performance. Journal of Regulatory Economics, 47, 325–338.Find this resource:
Bowblis, J. R., & Ghattas, A. (2017). The impact of minimum quality standards regulations on nursing home staffing quality and exit decisions. Review of Industrial Organization, 50, 43–68.Find this resource:
Bowblis, J. R., & McHone, H. S. (2013). An instrumental variables approach to post-acute care nursing home quality: Is there a dime’s worth of evidence that continuing care communities provide a higher quality? Journal of Health Economics, 32, 980–996.Find this resource:
Brekke, K., Siciliani, L., & Straume, O. R. (2012). Quality competition with profit constraints. Journal of Economic Behavior & Organization, 84, 642–659.Find this resource:
Buljac-Samardzic, M., & van Woerkom M. (2018). Improving quality and safety of care in nursing homes by team support for strength use: A survey study. PLoS ONE, 13, 1–16.Find this resource:
Castle, N. G., & Ferguson, J. C. (2010). What is nursing home quality and how it is measured? The Gerontologist, 50(4), 426–442.Find this resource:
Ching, A. T., Hayashi, F., & Wang, H. (2015). Quantifying the impacts of limited supply: The case of nursing homes. International Economic Review, 56, 1291–1322.Find this resource:
Chen, M. N., & Grabowski, D. C. (2015). Intended and unintended consequences of minimum staffing standards for nursing homes. Health Economics, 24, 822–839.Find this resource:
Chou, S.-Y. (2002). Asymmetric information, ownership and quality of care: An empirical analysis of nursing homes. Journal of Health Economics, 21, 293–311.Find this resource:
Cohen, J. W., & Spector, W. D. (1996). The effect of Medicaid reimbursement on quality of care in nursing homes. Journal of Health Economics, 15, 23–48.Find this resource:
Comondore, V. R., Devereaux, P. J., Zhou, Q., Stone, S. B., Busse, J. W., Ravindran, N. C., . . . Guyatt, G. H. (2009). Quality of care in for-profit and not-for-profit nursing homes: Systematic review and meta-analysis. British Medical Journal, 339(7177), 381–384.Find this resource:
Dellefield, M. E., Castle, N. G., McGilton, K. S., & Spilsbury, K. (2015). The relationship between registered nurses and nursing home quality: An integrative review (2008–2014). Nursing Economics, 33, 95–108.Find this resource:
Donabedian, A. (1980). The definition of quality and approaches to its assessment. Ann Arbor, MI: Health Administration Press.Find this resource:
Dranove, D. (2012). Health care markets, regulators, and certifiers. In T. McGuire, M. Pauly, & P. P. Barros (Eds.), Handbook of health economics (Vol. 2, pp. 639–690). New York, NY: Elsevier.Find this resource:
Forder, J., & Allan, S. (2014). The impact of competition on quality and prices in the English care homes market. Journal of Health Economics, 34, 73–83.Find this resource:
Foster, A. D., & Lee, Y. S. (2015). Staffing subsidies and the quality of care in nursing homes. Journal of Health Economics, 41, 133–147.Find this resource:
Friedrich, B., & Hackmann, M. (2017). The returns to nursing: Evidence from a parental leave program. NBER Working Paper 23174. Washington, DC: National Bureau of Economic Research.Find this resource:
Gaynor, M., & Town, R. J. (2012). Competition in health care markets. In T. McGuire, M. Pauly, & P. P. Barros (Eds.), Handbook of health economics (Vol. 2, pp. 499–637). New York, NY: Elsevier.Find this resource:
Gertler, P. J. (1989). Subsidies, quality, and regulation in the nursing home industry. Journal of Public Economics, 38, 33–52.Find this resource:
Gertler, P. J. (1992). Medicaid and the cost of improving access to nursing home care. Review of Economics and Statistics, 74, 338–345.Find this resource:
Grabowski, D. C. (2001). Medicaid reimbursement and the quality of nursing home care. Journal of Health Economics, 20, 549–569.Find this resource:
Grabowski, D. C. (2004). A longitudinal study of Medicaid payment, private pay price and nursing home quality. International Journal of Health Care Finance and Economics, 4, 5–26.Find this resource:
Grabowski, D. C., Feng, Z., Hirth, R., Rahman, M., & Norton, E. C. (2013). Effect of nursing home ownership on the quality of post-acute care: An instrumental variables approach. Journal of Health Economics, 32, 12–21.Find this resource:
Grabowski, D. C., Gruber, J., & Angelelli, J. T. (2008). Nursing home quality as a common good. The Review of Economics and Statistics, 90, 754–764.Find this resource:
Grabowski, D. C., & Norton, E. C. (2012). Nursing home quality of care. In A. M. Jones (Ed.), The Elgar companion to health economics (2nd ed., pp. 307–317). Cheltenham, U.K.: Edward Elgar.Find this resource:
Grabowski, D. C., & Town, R. J. (2011). Does information matter? Competition, quality and the impact of nursing home report cards. Health Services Research, 46, 1698–1719.Find this resource:
Hackmann, M. B. (2019). Incentivizing better quality of care: The role of Medicaid and competition in the nursing home industry. American Economic Review, forthcoming.Find this resource:
Hansmann, H. B. (1980). The role of non-profit enterprises. Yale Law Journal, 89, 835–891.Find this resource:
Harrington, C. (2005). Quality of care in nursing home organisations: Establishing a health services research agenda. Nursing Outlook, 53, 300–304.Find this resource:
Harrington, C., Olney, B., Carrillo, H., & Kang, T. (2012). Nurse staffing and deficiencies in the largest for-profit nursing home chains and chains owned by private equity companies. Health Services Research, 47, 106–128.Find this resource:
Harrington, C., Schnelle, J. F., McGregor, M., & Simmons, S. F. (2016). The need for higher minimum staffing standards in U.S. nursing homes. Health Services Insights, 9, 13–19.Find this resource:
He, D., & Konetzka, T. (2015). Public reporting and demand rationing: Evidence from the nursing home industry. Health Economics, 24, 1437–1451.Find this resource:
Herr, A., & Hottenrott, H. (2016). Higher prices, higher quality? Evidence form German nursing homes. Health Policy, 120, 179–189.Find this resource:
Herr, A., Nguyen, T. V., & Schmitz, H. (2016). Public reporting and the quality of care of German nursing homes. Health Policy, 120(10), 1162–1170.Find this resource:
Hillmer, M., Wodchis, W., Gill, S., Anderson, G. M., & Rochon, P. A. (2005). Nursing home profit status and quality of care: Any evidence of an association? Medical Care and Research and Review, 62, 139–166.Find this resource:
Hirth, R. A. (1999). Consumer information and competition between nonprofit and for-profit nursing homes. Journal Health Economics, 18, 219–244.Find this resource:
Hirth, R. A., Grabowski, D. C., Feng, Z., Rahman, M., & Mor, V. (2014). Effect of nursing home ownership on hospitalisation of long-stay residents: An instrumental variables approach. International Journal of Health Care Finance and Economics, 14, 1–18.Find this resource:
Huang, S. S., & Bowblis, J. R. (2018). The principal-agent problem and owner-managers: An instrumental variables application to nursing home quality. Health Economics, 27(11), 1653–1669.Find this resource:
Huang, S. S., & Hirth, R. A. (2016). Quality rating and private prices: Evidence from the nursing home industry. Journal of Health Economics, 50, 59–70.Find this resource:
Jones, D. B., Propper, C., & Smith, S. (2017). Wolves in sheep’s clothing: Is non-profit status used to signal quality? Journal of Health Economics, 55, 108–120.Find this resource:
Kai, Y., Yue, L., Intrator, O., Stevenson, D., Hirth, R., Grabowski, D., & Banaszak-Holl, J. (2016). Do nursing home chain size and proprietary status affect experiences with care? Medical Care, 54, 229–234.Find this resource:
Konetzka, T. R. (2009). Do not-for-profit nursing homes provide better? British Medical Journal, 339(7717), 356–357.Find this resource:
Konetzka, T. R., Grabowski, D. C., Perraillon, M. C., & Werner, R. M. (2015). Nursing home 5-star rating system exacerbates disparities in quality, by payer source. Health Affairs, 34, 819–827.Find this resource:
Konetzka, T. R., Norton, E. C., & Stearns, S. (2006). Medicare payment changes and nursing home quality: Effects on long-stay residents. International Journal of Health Care Services and Economics, 6, 173–189.Find this resource:
Konetzka, T. R., Park, J., Ellis, R., & Abbo, E. (2013). Malpractice litigation and nursing home quality of care. Health Services Research, 48, 1920–38.Find this resource:
Konetzka, T. R., Polsky, D., & Werner, R. M. (2013). Shipping out instead of shaping up: Rehospitalisation from nursing homes as an unintended effect of public reporting. Journal of Health Economics, 32, 341–52.Find this resource:
Konetzka, T. R., Stearns, S. C., & Park, J. (2008). The staffing-outcomes relationship in nursing homes. Health Services Research, 43, 1025–42.Find this resource:
Konetzka, R. T., Yi, D., Norton, E. C., & Kilpatrick, K. E. (2004). Effects of Medicare payment changes on nursing home staffing and deficiencies. Health Services Research, 39, 463–88.Find this resource:
Lin, H. (2014). Revisiting the relationship between nurse staffing and quality of care in nursing homes: An institutional variables approach. Journal of Health Economics, 37, 13–24.Find this resource:
Lin, H. (2015). Quality choice and market structure: A dynamic analysis of nursing home oligopolies. International Economic Review, 56, 1261–1290.Find this resource:
Lu, S. F. (2012). Multitasking, information disclosure, and product quality: Evidence from nursing homes. Journal of Economics and Management Strategies, 21, 673–675.Find this resource:
Lu, S. F., Rui, H., & Seidmann, A. (2018). Does technology substitute for nurses? Staffing decisions in nursing homes. Management Science, 64, 1842–1859.Find this resource:
Lu, S. F., & Wedig, G. (2013). Clustering, agency costs and operating efficiency: Evidence from nursing home chain. Management Science, 59, 677–694.Find this resource:
Mennicken, R., Augurzky, B., Rothgang, H., & Wasem, J. (2014). Explaining differences in remuneration rates of nursing homes in Germany. European Journal of Health Economics, 15, 401–410.Find this resource:
Mullahy J. (2018). Health status measurement. In Oxford research encyclopedia of economics and finance. Oxford, U.K.: Oxford University Press.Find this resource:
Mukamel, D. B., Haeder, F., & Weimer, D. L. (2014). Top-down and bottom-up approaches to health care quality: The impacts of regulation and report cards. Annual Review of Public Health, 35, 477–497.Find this resource:
Mukamel, D. B., Ladd, H., Weimer, D. L., Spector, W. D., & Zinn, J. S. (2009). Is there evidence of cream skimming among nursing homes following the publication of the nursing home compare report card? Gerontologist, 49, 793–802.Find this resource:
Norton, E. C. (2000). Long-term care. In A. J. Culyer & J. P. Newhouse (Eds.), Handbook of health economics (Vol. 1B, pp. 956–994). New York, NY: Elsevier.Find this resource:
Norton, E. C. (2018). Health and long-term care. In J. Piggott & A. Woodland (Eds.), Handbook of the economics of population aging (pp. 951–989). New York, NY: Elsevier.Find this resource:
Nyman, J. (1985). Prospective and “cost-plus” Medicaid reimbursement, excess Medicaid demand, and the quality of nursing home care. Journal of Health Economics, 4, 237–259.Find this resource:
Park, J., & Werner, R. M. (2011). Changes in the relationship between nursing home financial performance and quality of care under public reporting. Health Economics, 20, 783–801.Find this resource:
Reichert, A. R., & Stroka, M. A. (2018). Nursing home prices and quality of care: Evidence from administrative data. Health Economics, 27, 129–140.Find this resource:
Rose-Ackermann, S. (1996). Altruism, nonprofits, and economic theory. Journal of Economic Literature, 34, 701–728.Find this resource:
Rodrigues, R., Trigg, L., Schmidt, A. E., & Leichsenring, K. (2014). The public gets what the public wants: Experiences of public reporting in long-term care in Europe. Health Policy, 116(1), 84–94.Find this resource:
Siciliani, L. (2014). The economics of long term care. BE Journal of Economic Analysis and Policy, 14, 343–375.Find this resource:
Stolt, R., Blomqvist, P., & Winblad, U. (2011). Privatization of social services: Quality differences in Swedish elderly care. Social Science & Medicine, 72(4), 560–567.Find this resource:
Stone, P., Brooks, R., Brynjolfsson, E., Calo, R., Etzioni, O., Hager, G., . . . Teller, A. (2016). One hundred year study on artificial intelligence. Stanford, CA: Stanford UniversityFind this resource:
Stoyanova A., & Cantarero-Prieto D. (2018). Financing and policy for long-term care. In Oxford research encyclopedia of economics and finance. Oxford, U.K.: Oxford University Press.Find this resource:
Tong, P. K. (2011). The effects of California minimum nurse staffing laws on nurse labor and patient mortality in skilled nursing facilities. Health Economics, 20, 802–816.Find this resource:
Werner, R. M., Konetzka, R. T., Stuart, E. A., & Polsky, D. (2011). Changes in patient sorting to nursing homes under public reporting: Improved patient matching or provider gaming? Health Services Research, 46, 555–571.Find this resource:
Werner, R. M., Norton, E. C., Konetzka, R. T., & Polsky, D. (2012). Do consumers respond to publicly reported quality information? Evidence from nursing homes. Journal of Health Economics, 31, 50–61.Find this resource:
Williams, S. C., Morton, D. J., Braun, B. I., Longo, B. A., & Baker, D. W. (2017). Comparing public quality ratings for accredited and nonaccredited nursing homes. Journal of the American Medical Directors Association, 18, 24–29.Find this resource:
Zhang N., Lu, S. F., Xu, B., Wu, B., Rodriguez-Moguio, R., & Gurwitz, J. (2016). Health information technologies: Which nursing homes adopted them? Journal of the American Medical Directors Association, 17, 441–447.Find this resource:
Zhao, X. (2016). Competition, information, and quality: Evidence from nursing homes. Journal of Health Economics, 49, 136–152.Find this resource:
(2.) Seminal works building up a theoretical framework for NH behaviour are, among others, Nyman (1985), Cohen and Spector (1996), and Hirth (1999). For comprehensive reviews see Norton (2000, 2018) and Siciliani (2014).
(4.) To draw causal inference, the authors exploit individual change in eligibility associated to spending down.
(7.) Quality is measured as the distance between nurse-to-patient ratio and the minimum value set by state or federal standards.
(8.) For a more detailed discussion of the impact of public reporting initiatives on quality, see later discussion in this article.
(9.) One of the advantages of considering Medicare patients only is that the federal program covers expenses for the first 20 days of stay. In this way, possible selection effects due to a differential cost of admissions can be ruled out for postacute care.
(10.) CON regulation was introduced to curb the rise in LTC expenditures by limiting the expansion of the supply of NH beds.
(11.) In this respect, NHs display large heterogeneity with a substantial share of providers (around 25%) treating either more than 90% or less than 10% of Medicaid patients.
(12.) Structural modeling has been only seldom applied to the NH market but offers a convenient tool for evaluating the consequences of policy reform in a highly regulated context by means of counterfactual analysis. For other interesting applications, see Lin (2015) and Ching, Hayashi, and Wang (2015).
(13.) Tariffs from Medicare programs are substantially more generous than Medicaid ones, even if they are usually set below the price paid by private patients.