Do public nursing home care providers deliver higher quality than private providers? Evidence from Sweden

It is usually assumed that profit-seeking companies have stronger incentives than public or non-profit organizations to strive for cost reductions that lead to quality deterioration in the provision of public services [5, 14]. The main argument behind this assumption is that public owners have weaker incentives to implement cost reductions than private owners since they will only get a fraction, if anything, of the economic surplus, while private owners will get the whole surplus. Shleifer argues that the incentives to strive for cost reductions are stronger when (1) competition is limited (2) consumer choice is ineffective and (3) reputational mechanisms are weak [14]. The market for nursing home care is characterized by all three of these mechanisms. First, it can be argued that competition is generally weak when public service provision is contracted out. In public procurement processes, bidders are typically invited to compete for contracts, but once the contract is signed, there is essentially no further competition. Strong initial competition for contracts may reduce the providers’ profits by forcing down prices but does not restrain them from making cost reductions on non-contractible quality aspects once the contract is won. Second, for competition to have the desired effect to uphold quality, consumers must be able to make informed choices, i.e. use their exit strategy when suppliers are skimping on quality. Research has shown that older people are particularly ineffective consumers when it comes to choosing welfare services. A large percentage of the population over the age of 75 suffers from illness or cognitive impairments making it hard for them to make informed choices [15]. For the mechanism of consumer choice to have a quality-promoting effect within publicly funded welfare service systems, there also needs to be enough excess capacity in the system to make it possible for users to obtain the services they most prefer [16].

Third, a strong reputational mechanism could mitigate the risks of quality cutting in the market for nursing homes. In conventional markets, companies often have incentives to build relationships with consumers and providers by delivering high quality even for dimensions that are hard to contract or to observe, as it strengthens their reputation and sales in the long run. This sort of informal mechanism is, however, hard to rely on in cases of public contracting, at least in the relatively strict form that is employed in nursing home care in Sweden. Swedish public procurement law, drawing on EU directives, prohibits, for instance, that contracting authorities take factors such as previous experiences with a provider into account in the next procurement round, which undermines the mechanism of reputation-building. Thus, in the market for public procurement of nursing homes there seems to be no strong mechanism linking suppliers’ performance today to future demand; both the consumer and the buyer are ineffective in constraining the providers’ incentives to cut costs and thereby decrease quality. On the basis of this reasoning, it can be hypothesized that private nursing home care providers, facing stronger budget constraints and, in many cases, demands for profit, will be more likely to shirk on quality in order to reduce costs. This can be hypothesized to be true also, but to a lesser extent, in the case of private non-profit providers, as these are exposed to the same cost-cutting pressure when they compete for public contracts as for-profit firms even though their owners are less likely to expect a profit. Along the same lines, it can be presumed that the higher the expectation to generate profit on the part of private companies, the more likely they are to compromise quality in order to reduce their costs [11]. Therefore it can be expected that nursing homes owned by so-called private equity companies, that is, owners with a strong demand for short-term profit, will be more likely than the average company to reduce quality. Private equity companies are known for investing in markets with high profit potential, with the objective to improve the financial results over a short time horizon in order to resell the company [17].

Based on the reasoning above, the following hypotheses can be formulated for testing in the empirical analysis:

The expectation that ownership may exert influence on quality in nursing home care is supported by previous research in the field. A number of American studies have found that for-profit ownership is related to lower quality than non-profit ownership [9, 10, 12, 18]. In general, non-profit nursing homes have been shown to have higher staffing levels, lower staff turnover and better trained staff compared to for-profit providers. They also have better quality outcomes, such as lower prevalence of pressure ulcers, less use of physical restraints, and fewer deficiencies in governmental regulatory assessments. Furthermore, it has also been shown that for-profit providers tend to deliver lower quality care to residents that have no living relatives or are suffering from dementia, i.e. to persons who have a low "voice" [19]. This implies that profit incentives put weaker consumers at a disadvantage. In a study carried out by Harrington in 2012 it was investigated whether the largest long-term care chains in the U.S. had poorer quality after being purchased by private private equity companies. It was found in this study that there was little change in staffing levels, but that the total number of deficiencies increased after the changes in ownership [11].

Most previous studies examining the relationship between ownership and nursing home care quality have relied on cross-sectional designs, which suggests that their conclusions could be biased due to unobserved factors influencing nursing home quality [18, 20, 21]. In an attempt to isolate the causal effect of ownership change, American researchers Grabowski & Stevenson used panel-data from 1993 to 2004 covering changes in ownership from for-profit to non-profit, and vice versa. Their results, indicating that facilities changing from non-profit to for-profit ownership generally declined in quality while facilities changing ownership from for-profit to non-profit improved, seemed to confirm the relation between ownership and quality observed in previous studies [18].

As Spilsbury et al. points out, research on nursing home quality has hitherto been undertaken primarily in the U.S. [21]. There are, however, a few studies conducted in Sweden which have compared performances between privately and publicly operated nursing homes. The results of these indicate a similar pattern as in the U.S., particularly with regards to the staffing levels and other structural quality indicators, such as staff education and full-time employment [22, 23]. For instance, Stolt et al., using data from 2007, found that the number of employees per resident was significantly lower (−9%) in nursing homes operated by private owners (including both non-profit and for-profit actors) than publicly operated facilities [22]. A similar result was found in the study conducted by the National Board of Health and Social Welfare (NBHW) in 2012 [23]. On the other hand, both Swedish studies also found that private providers scored higher than their public counterparts on quality indicators labeled as processual, such as the proportion of residents participating in the formulation of their care plan; the proportion of residents with a reasonable duration between evening meal and breakfast; and the proportion of residents offered different food alternatives. Worth noting, however, is that none of these studies compared for-profit with non-profit nursing homes. Nor has any scientific study investigated the role of private equity ownership in nursing home care in Sweden despite this being the most form of private ownership in recent decades.

In sum, the previous research on quality in relation to ownership in nursing home care provides fairly consistent evidence that for-profit ownership is associated with lower quality of care than both non-profit and public ownership, at least in an American context.

In particular, nursing homes operated by private for-profit providers appear to have lower staffing levels compared to publicly owned homes, a result found in Swedish studies as well. This tendency is interesting as a large stream of research emphasizes staffing levels as a particularly important determinant of quality of care in nursing homes, particularly when it comes to outcome-related quality indicators such as ulcers and fall injuries [11, 20, 24, 25]. These findings support our general hypothesis that care providers with profit motives are more likely to shirk on quality. Still, we do not know if the same pattern exists in Sweden where, so far, no distinction has been made in the scientific literature between non-profit and for-profit private providers and where the relation between ownership and quality has been shown to be more ambiguous.

Nursing home care in Sweden is provided by 290 local governing units, the so-called municipalities. The services are funded by the municipalities themselves through local income taxation, and are in most cases also provided directly by the municipalities. A legal change in 1992 made it possible for the municipalities to contract out nursing home care provision to private actors including for-profit companies [26]. Since then, the share of residents living in private nursing homes has increased steadily, reaching 20.5% in 2016 [27]. This implies that the private providers are still funded exclusively by the municipalities and subjected to direct regulatory control through contracts in the same way as public providers. In addition, they are bound to comply with the same national public regulation as municipal care providers, with regard to quality requirements, user safety regulations, and audit measures. In contrast to most other welfare systems outside the Nordic region, the Swedish elder care system is organized on the basis of universality. This means that nursing home care services are financed on the principle of solidarity  through income taxation but also that the distribution is organized on the basis of assessed need [26]. Virtually no private market for such services exists in Sweden. In principle, all Swedish citizens (or permanent residents) have access to publicly funded nursing home care services at heavily subsidized rates. User fees have a maximum ceiling and are tied to income. Needs assessment is carried out by social workers at the municipal social service authorities (Socialtjänsten) and it is also the municipal social services that decide on the placement of users in individual homes. This means that the private providers operating within the municipal systems of nursing home care provision cannot themselves decide which users to accept, but are obliged to take in all users placed by the municipality. Users placed at privately operated institutions pay the same user fees as in the public homes. In 2015, 16.2% of the population above 65 years of age received some form of elderly care services from the municipalities, while 4.2% lived in a nursing home [28]. As home care services for the elderly are well-developed in Sweden, and this has been the preferred form of care for this group since the 1970s, those assessed to have a need for nursing home care have relatively extensive social and medical needs.

The process through which nursing home care provision is delegated to private actors may be denoted as competitive contracting. This implies that interested parties compete for public contracts based on a combination of price and quality. The experience in Sweden so far is that prices have played a determining role when contracts are awarded, even if most municipalities have some form of quality “base line” which all tenders must pass in order to be considered. According to Swedish competition law, contracts between the municipality and private providers cannot be entered into without a transparent and non-discriminatory selection process, which means that non-profit organizations are obliged to compete for municipal contracts on the same basis as for-profit firms. Notably, what is contracted out in the Swedish case is the operation of nursing homes, which implies that the facilities are in most cases still owned by the municipalities. Swedish  law also stipulates that the staff employed at the nursing home in question must retain their employment for at least one year after a new provider takes over the operation. Contracts are relatively short, typically 3–4 years, with the possibility of a single extension. In practice, this means that most of the staff at nursing homes where operations are contracted out remain when there is a change in management, for instance from the municipality to a private company. It is important to note, that contracting out nursing homes is voluntary on part of the municipalities and was done in about one third of them in 2016. There is also large regional variation in this regard as some municipalities chose to contract out all local nursing homes, i.e. 100%, whereas others chose to contract out only a limited number or none at all.

Among private nursing home providers in Sweden, three ownership categories can be identified: non-profit, for-profit, and private equity owned nursing homes, which is a subset of the for-profit category. Non-profit refers to nursing homes run by private organizations without profit motives and with a social orientation, such as churches or charity organizations [29]. The non-profit sector of nursing home care in Sweden is very small, making up only 2.6% of all nursing homes and 10.7% of the privately operated homes in 2015. The majority (89%) of private nursing homes are instead operated by for-profit companies or private equity companies (data from Statistics Sweden from 2015, authors’ own analysis). The ownership category for-profit refers to nursing homes that are operated by for-profit firms, typically stock-holding companies but not owned by private equity companies. Private equity owned refers to nursing homes run by companies that are owned by so-called private equity companies, that are known for investing in markets with high profit potential.

The Swedish market for nursing home care is heavily consolidated. In 2012, four private equity companies controlled 52% of the private market for nursing homes, representing about 11% of the total market [16]. In 2012, the social care sector (care for the elderly and disabled) showed a return on total capital of 15%, which was quite high compared to other sectors in the same year. The median return on equity in the same sector was 38.5%, a high figure compared to the corresponding figure of 19% for the service sector as a whole (data from Statistics Sweden for 2013). Thus, the Swedish nursing home care sector is clearly attractive to private investors, even if the majority of care providers (nearly 80%) are still public.

The aim of this study is to investigate the relationship between ownership type and care quality in nursing home care in the case of Sweden. From a methodological perspective, Sweden can be seen as a least likely case. It may be expected that differences in quality between different types of care providers are relatively small, given the high level of public regulation, the public funding structure, and the high social and political premium placed on equitable care standards among all types of providers. The least likely case selection implies that if quality differences are found between various ownership categories in the Swedish case, they are likely to exist in other countries as well [30].

The data used in the study originate from the Swedish National Board of Health and Welfare (NBHW), an expert government agency in the welfare area, which conducts surveys covering a range of quality indicators of all nursing homes in Sweden on a yearly basis. The collection of data is organized so that every nursing home regardless of ownership, receives a questionnaire asking management to provide statistics on the listed quality indicators.

Following Donabedian’s well-known distinction, we differentiate in the analysis between structural and processual quality indicators [31]. Structural quality refers to institutional characteristics such as staffing and facilities, measured, for instance, as the number of employees per resident, prevalence of full-time employment or access to a kitchen. Processual quality indicators measure the activities performed by the staff, such as frequency of medication reviews or risk assessments for pressure ulcers, or malnutrition. By distinguishing between structural and processual quality measures we can obtain a more precise understanding of in what ways different types of providers make trade-offs between cost and quality.¹

The data used in the study was collected by NBHW in October 2010 and 2011. The reason for why we choose data from these years is that it was the only period when the quality indicator Employees per resident was used, a quality measure that has often been used in previous studies. The data from 2011 comprised 2324 publicly operated and 386 privately operated nursing homes, out of which 201 were private equity homes, 123 private for-profit homes and 62 private non-profit homes. In the analysis we used seven structural measures and seven processual measures, see Table 1. All of these, except for four structural measures, originate from the 2011 survey. These four measures, Employees per resident, Hourly employment, Full time employment and Advanced competence, were collected from the 2010 survey, since these indicators were not measured in 2011 by the NBHW.

As the purpose of this paper is to investigate whether ownership affects quality of care, we subdivided ownership into four categories with regard to mode of provision: (1) Public; (2) For-profit; (3) Non-Profit; and (4) Private Equity. As NBWH’s data does not divide private homes into all these categories, but only provides a dichotomous classification of facilities into public and private ownership, we used company websites and business registry databases to determine type of ownership. In order to test hypothesis 1, that publicly operated nursing homes have higher quality than private ones, we also kept the category “private” in the NBHW data which contains all three types of private actors: for-profit, non-profit and private equity. Thus, five ownership categories are presented in the tables.

We performed OLS regressions to compare mean quality differences between the different ownership groups. Dummy variables were constructed for each ownership group, using public nursing homes as a reference group (Table 3). The quality indicators were treated as dependent variables and the ownership categories as independent variables. To test hypothesis 2 and 3, for-profit homes and private equity homes were used as reference groups respectively.

Multiple OLS regressions were run to control for confounding factors that could obscure the relationship between quality of care and ownership. For example, it has been shown that privatization is more prevalent in larger cities and suburbs [32]. Privatization also tends to co-vary with political ideology as municipalities governed by right-wing parties tend to contract out nursing homes to a larger extent. In municipalities with left-wing parties, on the other hand, a strained financial situation seems to co-vary with increased privatization [33]. For these reasons all OLS models were controlled for municipality population density, political majorities, and socio-demographic, and financial differences (Appendix 1). We have adjusted the analysis for multiple testing.