Patients' perceptions of quality in Swedish primary care – a study of differences between private and public ownership

Introduction

Many countries have healthcare systems that are open for both public and private care providers (Sheaff et al., 2019). Several countries have experimented with deregulation, privatization, customer-choice models and other ways of introducing private providers into the healthcare system. Sweden has been in the forefront of this wave of experiments, and has gradually transformed a system of primarily publicly owned providers to a mixed model with private providers working alongside public providers (Maun et al., 2013). The changes have been most radical in primary care, where all regions in Sweden have implemented a model where patients can choose a primary care center from a list of both private and public providers (Isaksson et al., 2016; Vengberg et al., 2019). It is also important to mention that, in Sweden, both public and private providers of primary care are funded through taxes. This raises interesting questions of whether quality has increased in the whole system, and whether there are differences between private and public primary care centers. The extant literature is not entirely conclusive, although a number of studies in a Swedish context show that private providers typically have higher quality perceptions (Glenngård, 2012; Maun et al., 2015). More data-driven analyses of patients' quality perceptions of private and public providers of primary care are clearly needed.

This paper addresses whether there are differences in patient's quality perceptions of private versus public providers of primary care. Previous research has consistently treated private providers as one, homogenous, group (Maun et al., 2015), thereby neglecting the variation between different categories of private owners. It has been argued that the distinction between for-profit and not-for-profit providers is important, and a few studies have provided results comparing the two types (Comondore et al., 2009; Eggleston et al., 2008; Kraska et al., 2016). In this paper we argue that further distinctions can be made, for instance between larger corporate groups and smaller, independent actors.

The study was done in the context of the nationwide reform in healthcare in Sweden. In 2010, primary care throughout Sweden was opened for private providers; to date, around 42% of the centers are privately owned (in 2017 there were in total 1,150 centers). This provided a good arena for a nation-wide comparison of perceived quality of care between different owner categories. Starting from a low level in 2010 (and some years before that in selected regions) the number of private providers has rapidly grown and is expected to reach 50% of the total number within a few years. The private providers are publicly funded (through the region where they operate) based on the number of patients listed at the single center (adjusted for a number of parameters, such as type of patients, socio-economic factors, and areas of specialization), and, in some cases, based on the actual amount of medical treatments and consultations.

The purpose of the paper is to describe and analyze differences in patients' quality perceptions of private and public primary care centers in Sweden. It further compares the quality perceptions for different categories of private centers.

Literature review

The research on patients' quality perceptions in healthcare show mixed results when comparing private providers and public ones. Some studies support the view that public providers have higher patients' quality perceptions (Jabnoun and Chaker, 2003), whereas others show the opposite result (Polsa et al., 2011; Shabir et al., 2016). There are also studies that show ambiguous results concerning ownership and patients' quality perceptions (Angelopoulou et al., 1998), It is important to acknowledge that the extant literature rarely focuses on primary care, with most studies comparing quality perceptions for different owners of hospitals (Eggleston et al., 2008; Owosu Kwateng et al., 2019; Tengilimoglu et al., 1999) and healthcare as a whole (Owusu-Frimpong et al., 2010). In the Swedish context, there is research (Glenngård, 2012; Maun et al., 2015) supporting the overall observation that private primary care centers have a higher patient quality perception than public centers, but little is known about the more detailed patterns.

In parallel with studies on quality perceptions, there are studies that focus on patient satisfaction. Although patient satisfaction is a different (although related) construct than quality perception (Bitner and Hubbert, 1994; Johnston, 1995; Taylor and Baker, 1994), it still sheds light on the effects of introducing private providers. Burström et al. (2017) report that patient satisfaction for primary care as a whole was unchanged after the reform. Their study, however, did not differentiate between different ownership categories. Some research shows lower patient satisfaction for private, for-profit – as compared to private not-for-profit – hospitals, at the same time as public hospitals yield lower satisfaction than their not-for-profit private counterparts (Kraska et al., 2016). Furthermore, no research has studied the differences within the broad group of private providers of primary care. Private primary care centers are usually treated as one group, and this does not reflect the variation within the broad group of private centers. This article provides a detailed analysis of different categories of private owners.

International studies on hospitals and healthcare in general show mixed results, and we agree with Eggleston et al. (2008) that comparisons between countries are difficult due to differences in institutional structures and different ways of conceptualizing quality in healthcare. In a meta-review of studies on quality of care in nursing homes, Comondore et al. (2009) showed a higher quality (as measured by a combination of input, process and outcome measures) for not-for-profit nursing homes. The study did not differentiate between public not-for-profit and private not-for-profit. Kumaraswamy (2012) provided empirical results on healthcare centers in India, showing more positive patients' quality perceptions for corporate compared to non-corporate (the study was not clear on whether non-corporate is equal to public) centers. Barhem et al. (2010) have added important insights on differences between for-profit and not-for-profit health care organizations, in their case, based on employee satisfaction in the two ownership categories.

In parallel with the discussion on quality differences between private and public ownership, there has also been a fierce political discussion in Sweden, with a criticism of private actors being involved in healthcare. The discussion has focused on a number of issues, such as big company groups, and what is perceived as an unjust earning of profits in the healthcare sector, as well as an uneven utilization of healthcare in the population (Beckman and Anell, 2013; Isaksson et al., 2016; Burström et al., 2017). Less is known, however, about patients' perception and attitudes towards private primary care units, and to what extent perceptions and attitudes are formed based on the broader political discussion or on the actual care and service patients receive. It is important to acknowledge that the general attitude towards privatization is a different construct than quality perceptions. It is possible that a person can have a negative view on privatization in general, and still have a positive quality perception when given care from a private provider. Sweden being at the forefront of mixing private and public ownership in healthcare, we argue that the results from this study are germane for countries that are aiming at similar changes in their healthcare systems.

The paper analyzes data from a survey of patients in Swedish primary care (National Patient Survey), that measured their perceptions of the care they received. Selected patients had visited a primary care center and met a physician in the year of the survey. Patient quality perception is here defined as a process through which the patients' expectations are balanced with the perception of the care they receive (Alrubaiee and Alkaa'ida, 2011; Newsome and Wright, 1999). This definition is in line with the broader definition of service quality offered by the service management literature (see, for instance, Grönroos, 1990). It is generally acknowledged in the literature on service management/marketing that services are fundamentally different from goods, and that the production of services relies on a different logic. Patients nowadays are arguably more informed and active in seeking the best possible care, and this calls for a redefinition of their role and importance as a part of the healthcare system (Liff and Andersson, 2011). Joiner and Lusch (2016) make a case for introducing a Service-dominant (S-D) logic in healthcare, stressing the co-creation between patient and provider. In the survey, the patient is asked several questions on how they perceived different aspects of the last visit they made to a primary care center. Many authors argue that learning about the patients' perceptions of the services they receive is an important basis for improving the quality of healthcare (Senic and Marinkovic, 2012), and that a more patient-centered approach is called for when competition between providers is increasing (Pihlainen et al., 2019).

The paper contributes to a growing stream of research on perceived quality in healthcare in general (Joiner and Lusch, 2016; Senic and Marinkovic, 2012), and differences between different types of ownership in particular. Although it is methodologically hard to compare research between countries due to lack of data, and also due to differences in healthcare systems, we still argue that the analysis of different categories of private owners is relevant beyond a Swedish context.

Methodology

In the empirical part of the paper, some basic demographic data on the population of primary care centers are provided and results from the quality survey are presented and analyzed. With the introduction of private providers of primary care in the early 2000 (and fully implemented by 2010), the public ownership was replaced with a mixed model, with some degree of competing for patients, not only between private and public providers, but also within the group of public providers. An important aspect of the reform was to create neutral conditions for both public and private providers, which meant that the publicly owned primary care centers were expected to operate under the same conditions as privately owned. The introduction of private providers of healthcare was driven by several political arguments by the advocates of private care. Besides the financial argument that private providers would increase productivity, the reform was motivated in terms of increased freedom of choice for patients (Maun et al., 2013). It was argued that allowing patients to freely choose the primary care center would put pressure on the providers to give better services, and this would increase quality in the whole system.

At the beginning of 2017, there were a total of 1,150 primary care centers in Sweden, with 42% being private and the remaining 58% being public. The number of public centers used in the study is 662. The group of private care centers is diverse; for the purpose of this article, they were divided into four different groups, totaling 455 centers:

Independent units, for profit (Ifp): This category consists of small companies operating up to four primary care centers that do not have a clear not-for-profit profile. When the market for primary care was opened for private actors, starting gradually in a few regions already before 2010, this created opportunities for new actors to emerge in the healthcare sector. The system with public funding made it possible for completely new actors to start a primary care center. 155 out of 455 private primary care centers used in the study fall into this category.

Independent units, not for profit (Infp): This category consists of actors that own up to four primary care centers, and that operate with an explicit not-for-profit profile. It includes a variety of different types of ownership and profiles, such as idea-based organizations (for instance with a religious base) and cooperatives (personnel cooperatives or member-owned organizations). In this category we find 15 centers.

Company group, for profit (Gfp): This category consists of companies that own five or more centers. In this group we find several companies that already existed in the market, for instance operating elderly care, schools, social care and similar welfare tasks. These companies could, with the new legislation, expand into primary care. This category is the largest of the private groups and consists of 279 centers.

Company group, not for profit (Gnfp): This category has a similar profile as Infp, but owns five or more units. This category consists of 6 centers.

The distinction between the categories is not always entirely clear, with some actors having traits of two categories. A good example is the big group Praktikertjänst that can be described as a producer-cooperative, but at the same time operates with a clear for-profit mission and has the legal structure of a limited company. To be consistent, we have used the presence/absence of a profit mission as the distinguishing mark between the categories, in this case, classifying Praktikertjänst as a Company group, for profit.

The survey covers different dimensions of the perceived quality (overall impression, emotional support, respect and kindness, availability, participation and involvement, continuity and coordination, information and knowledge) that a sample of patients from each of the primary care centers that visited a physician during a specific month (September) fills in. The population consists of all patients that visited a physician in primary care during this month; from this, a random sample is drawn, and the survey is sent in digital form to the selected patients. The sample drawn from each primary care center is based on and reflects the total number of visits made to that center during the selected month. Four waves of survey results were available for analysis (2011, 2013, 2015 and 2017; data for 2019 were not yet available when the article was completed). For the 2017 survey, 80,000 surveys were completed. The data can be broken down to the level of the single primary care center, making it possible to single out the results for the private and public care centers.

For this article, it was decided to present the results for the measure Overall Impression as the basis for comparing the different owner categories. This measure asks the patient to summarize the perception of the primary care unit, and therefore provided a strong overall indication of how the patients evaluate the care they received. The remaining six measures pick up on different aspects of the patients' evaluation of the care center. A detailed presentation of all seven measures in the survey was judged to be beyond the scope of this article. This choice was motivated along the following lines. Although service quality arguably is built on different service determinants (Johnston, 1995), it is often argued in the service marketing literature that selecting a key ratio and/or measure can be used to monitor the degree to which a company is retaining its customers (Hu et al., 2009). The use of Overall Impression has been used in numerous studies of health care (Attree, 2001; Gray and Boshoff, 2004), and we furthermore draw on the argument that patients form a strong first impression of the service provider (Rimondini et al., 2018). Overall Impression asks the customer (patient) to make an overall evaluation of the quality, and can therefore be used as an indication of the perceived service quality. The main point of this article is not to make a detailed study of the determinants of patient perceptions of service quality; rather it is to compare the different categories of owners. Based on the arguments above, it was deemed sufficient to use the summary measure of overall impression.

For this article, it was decided to include all primary care centers that had valid answers on Overall Impression in the two latest surveys (2015 and 2017). Care centers with answers in only one of the two surveys were therefore excluded. The main argument for this was to allow for a check of how consistent the results were over time. After this procedure, 1,117 primary care centers remained and were used in the analysis. The empirical results are presented and discussed in three steps. In the first step, the overall results for the measure Overall Impression are presented for the two groups, private and public primary care centers in each of the 21 regions. In the second step, the private providers are more carefully examined to see whether there are any differences between the four categories of private owners, for each of the regions. In the third step, the four categories of private owners are compared with the public owners, and the differences between the owner categories are also aggregated for Sweden as a whole. Is the quality survey with completed answers from 1,117 out of 1,150 primary care centers a relevant (valid) and reliable source to reflect upon the quality of primary care in Sweden? With 80,000 responses, it is arguably one of the largest surveys conducted in Sweden; furthermore, it is done every second year, which makes it possible to create time-series. The survey instrument is freely available, and the results are open for researchers, which makes it easy to check how the survey is constructed and the methodology.

Empirical results

Discussion

The analysis of quality perceptions for private and public primary care centers in Sweden has shown that private providers are overrepresented above the median for the measure Overall Impression in most regions. In that respect, the results from this study confirm the results from previous studies of Swedish primary care, showing that private providers have higher patient perceptions of quality (Glenngård, 2012; Maun et al., 2015). Adding to previous studies, the findings from this study show interesting differences between the different categories of private owners, and especially between the Independent for-profit (Ifp) and Corporate groups (Gfp). Through the comparative analysis in Table 3, we showed a high quality perception for the independent units (with 70% of the Independent for-profit being above the median, compared to 53.8% for the corporate groups). Although the number of units in the not-for-profit category is small, it clearly shows a very high quality perception for the independent units compared to large groups and public ownership.

International research on differences between private and public ownership is, with some exceptions, showing a higher patient quality perception for private owners. Our literature review shows that most studies give support to a higher quality perception for the private providers (Owusu-Frimpong et al., 2010; Tengilimoglu et al., 1999). The service marketing literature suggests that the quality of a given service is measured as the relation between the expectations on the service delivery and the perception of the actual outcome (Grönroos, 1990). A high score in the patient survey used in this article indicates that expectations have been met or, more probably, exceeded. A low score can mean that the patient had low expectations and they were met, or, more probably, that the expectations were not met, generating a low score. The survey does not specifically measure the gap between expectation and service perception, but it is highly probable that the private, independent centers meet or exceed expectations to a larger degree than their public and corporate group counterparts. The survey does not replace the need to have a variety of measures of quality in healthcare, for instance, focusing more on actual outcomes (as reported in the growing number of national quality registers in Sweden). The survey is specifically focused on the patient's perception of quality and our focus is on how this differs between the owner categories. In that sense, we do not offer a general answer on what is quality in healthcare. Future research can shed more light on whether patient quality perceptions correlate with other outcomes (such as more consistent and evidence-based practices, patient safety, accidents, and other measures of the actual care given).

Neither the Swedish literature on patient quality perceptions in primary care, nor the international healthcare literature, has made a distinction between different categories of private owners. Through analyzing the different categories of private owners, this paper offers a detailed description that shows a considerably higher perceived quality for private, independent primary care centers. The access to data from all primary care centers in Sweden has given a unique opportunity to analyze the effects of ownership on patients' quality perceptions. An international comparison with the Swedish data is, for obvious reasons, hard to make, but there is no major reason to assume that the private independent groups are less popular than private groups in other counties.

It is beyond the scope of this article to give reliable explanations for these results, but some observations can be made. Some authors have argued, based on limited empirical data, that there is a selection bias in the sense that people from socio-economically wealthy areas are more prone to choose a private center (Burström et al., 2017). Does this automatically transfer over to a positive perception of the care they are given at the center? The data for the Independent for-profit (Ifp) centers seem to confirm this. On the other hand, this does not explain why the Group for-profit (Gfp) category has a substantially lower score. Future research can possibly verify whether there is such a socio-economic bias in the data, but it is beyond the scope of this article to show such patterns. Is it possible that the private centers have an incentive to influence the sample to include patients that are positive? They could, but they have little control over the sampling process, as it is carried out by an independent research company.

The data confirm that the independence of the single center is an important factor, possibly creating a perception of being small and close to the patient. It does not necessarily mean that the single center is small in terms of number of listed patients; in fact, some of the most popular independent centers have a large number of listed patients. Another potential explanation could be that patients are affected by the broader, ideological discussion on making a profit in healthcare. It is primarily the large corporate groups that have been criticized for making a profit, and there could be a guilt-by-association effect that influences the patients' perception of them.

A further explanation for the patient quality perceptions in the independent centers could be that the ownership of the center is closer to the operative management. In many of the centers, the owner is working at the center, and this could be an incentive to try to make sure that patients are satisfied. This is, however, hard to verify with the data from the survey, but it is at least consistent with the results, showing higher quality perceptions with independent centers compared to centers belonging to a larger group. Agency theory suggests that managers of private health care centers try to maximize their own utility, and thus have an incentive to run the unit in an efficient way (Tiemann et al., 2012). It can be argued that a center that is run in an efficient way probably yields a higher patient quality perception. This, however, does not explain the difference between the independent and corporate groups in our material.

This research only focused on the measure Overall Impression, which summarizes patient satisfaction with the care received. In future research, it is interesting to crosscheck the results for Overall Impression with the other six measures in the national survey, to see whether there are further patterns when comparing the owner categories. Future research can also analyze whether there are socio-economic explanations for the popularity of the private, independent care centers. This would require cross-checking the address for each care center, with data on income levels for the area. This is not possible without a lot of manual work. Addresses for the care centers, for instance, are not easily accessible for research purposes, but need to be collected region by region.

Conclusions

This paper aimed at showing how patient quality perceptions differ between different ownership categories in Swedish primary care. The paper analyzed data from a national survey that measures patient quality perceptions in a number of dimensions. The paper singled out the Overall Impression dimension as the basis for comparing ownership categories. The primary care centers were divided into five different ownership categories: Public, Independent for-profit, Independent not-for-profit, Company group for-profit, and Company group not-for-profit. All five ownership categories are funded through taxes according to specific regulations in each region.

The paper shows that Independent for-profit, as a group, gets higher scores on Overall Impression, compared to Company group, for-profit, and public centers. The number of not-for-profit centers is small and it is hard to draw any conclusions for them. Based on the small number of not-for-profit centers in the sample, however, they get very good scores. Based on the findings reported in this paper, we have contributed to the discussion on perceived quality in health care and whether it differs between different ownership categories. The major contribution of the paper is to introduce the distinction between four categories of private ownership, and to analyze how perceived patient quality differs between them. Future research can potentially compare other countries, and also contribute with deeper explanations for the differences found in this paper.