- Email: [email protected]
Choice of primary care provider: Results from a population survey in three Swedish counties
Health Policy 103 (2011) 31–37
Contents lists available at ScienceDirect
Health Policy journal homepage: www.elsevier.com/locate/healthpol
Choice of primary care provider: Results from a population survey in three Swedish counties Anna H. Glenngård a,b,∗ , Anders Anell c , Anders Beckman d a b c d
Institute of Economic Research, Lund University School of Economics and Management, Lund, Sweden The Swedish Institute for Health Economics (IHE), Lund, Sweden Institute of Economic Research and Department of Business Administration, Lund University School of Economics and Management, Lund, Sweden Department of Clinical Sciences and General Practice/Family Medicine, Faculty of Medicine, Lund University, Malmö, Sweden
a r t i c l e
i n f o
Keywords: Primary health care Choice Sweden Population preferences Provider competition Privatisation
a b s t r a c t Recent reforms in Swedish primary care have involved choice of provider for the population combined with freedom of establishment and privatisation of providers. This study focus to what extent individuals feel they have exercised a choice of provider, why they exercise choice and where they search for information, based on a population survey in three Swedish counties. The design of the study enabled for studying behaviour with respect to differences in time since introduction of the reform and differences in number of alternative providers and establishments of new providers in connection with the reform. About 60% of the population in the three counties felt that they had made a choice of provider in connection with or after the introduction of a reform focusing on choice and privatisation. Establishments of new providers and having enough information increased the likelihood whereas preferences for direct access to a specialist decreased the likelihood of making a choice. The data further suggests that individuals were rather passive in their search for information and tended to choose providers that they previously had been in contact with. This is in line with results from previous studies and poses challenges for county councils governance of reforms. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction During the past two decades, publicly funded health care systems in northern Europe have been subject to reforms such as purchaser–provider split, decentralisation, privatisation, the introduction of explicit measures of performance and focus on service and client orientation, often referred to as New Public Management (NPM) reforms [1,2]. Swedish health care is no exception. The responsibility of financing and provision of health care in Sweden lies with 21 county councils and reforms are therefore usually
∗ Corresponding author at: Institute of Economic Research, Lund University School of Economics and Management, PO Box 7080, 22007 Lund, Sweden. Tel.: +46 73 6348487. E-mail address: [email protected] (A.H. Glenngård). 0168-8510/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.healthpol.2011.05.014
implemented at the local level and gradually throughout the country. The first steps towards increased choice of providers for patients took place already in the early 1990s. By then, however, provider choice for patients was not combined with privatisation and freedom of establishment of providers and payment to providers did not follow the choice of patients. Consequently, there was limited alternatives for the population to choose among and limited incentives for providers to respond to expanded choice for patients [3]. More recent reforms involving expanded choice of provider have focused on primary care and combined choice of provider with freedom of establishment and privatisation of providers [4]. The new reform involving both choice and privatisation of primary care was first implemented in the county council of Halland in 2007, followed by Västmanland and Stockholm in 2008 and seven
32
A.H. Glenngård et al. / Health Policy 103 (2011) 31–37
additional counties during 2009. Since January 1st 2010, following a change in the Health Care Act decided by the Parliament, choice of primary care provider for the population and freedom of establishment for providers accredited by the local county councils is mandatory. Patient choice of provider is expected to improve the efficiency, quality and responsiveness of the health system through the threat of exit [5,6]. As patients can shift provider in case they are not satisfied with their current provider, they can cause a loss of income for the providers [7,8]. Primary health care plays a central role in most health care systems and strengthening primary care is widely seen as central in enhancing equity and efficiency in health care [9,10]. This makes choice of primary care provider especially interesting and perhaps the most important choice in health care for individuals to make. In principle, there are choice systems with and without competition of providers but for such a system to work in practice it has been argued that alternative providers for the population to choose among are essential. Of equal importance, however, is that the population is interested and adequately informed to make a choice [4–6,11]. The main argument against extended patient choice is that it might increase inequalities in access to care. People with lower income and/or education have lesser capabilities to exercise choice compared to people with higher income and/or education who are more articulate, have better skills in searching for information and are more able to travel to non-local providers [7,8,12]. Findings from previous studies in other countries suggest that people in general are interested in choice but that the actual effects of introducing choice systems are uncertain and often rather limited [7]. In practice, people tend not to shift provider even if they have the possibility [13–15]. Previous Swedish studies based on surveys and stated preferences have demonstrated that individuals in general are interested in choice and participation in primary care services [16–18]. There is however limited information about how individuals act or where they search for information in a choice-based system in practice. Recently implemented reforms in Swedish primary care enables for studying peoples actual behaviour in a choice-based system. Data from registers provides information about if individuals have shifted between providers but provide no information about to what extent individuals feel they have exercised choice of primary care provider, why they exercise choice and where they search for information. The aim of this study was to answer these questions through a population survey in nine municipalities located in three Swedish counties. Possible differences across different groups of the population were investigated. Moreover, possible barriers to making a choice and preferences regarding primary care of importance for the choice of primary care providers were investigated. The three selected counties varied in timing of introduction of the reform but all three introduced the reform before 2010. The nine selected municipalities within these counties varied in terms of number of alternative providers before the reform and establishments of new providers in connection with the reform. The overall design of reform in the three county councils was similar, however, with
Table 1 Sample of respondents. Region (date of reform) and municipality
Number of respondents
Halland (January 07) Halmstad 400 Hylte 300 Falkenberg 300 Skane (May 09) 400 Malmö Örkelljunga 300 Bromölla 300 VGR (October 09) 400 Göteborg Uddevalla 300 Skövde 300
Population
Alternative local primary hcp
90 000 10 000 40 000
Yesa No Yesa
287 000 10 000 12 000
Yesa Yesa No
500 000 51 000 51 000
Yesa Yesa Yesa
Note: Population in Sweden 9 340 682 December 2009. a Establishment of new primary hcp in connection with or after introduction of reform by December 2009.
respect to principles for registration, payment systems, financial responsibility for providers, and freedom of establishment for all accredited providers. The requirements from each county council primarily focus on minimum number of available clinical competencies at the primary care unit. In all three county councils, a letter with information about the reform and suggestion of primary care provider based on the last primary care visit or geographical distance (in that order) was sent to all residents prior to the introduction of the reform. If individuals did not change to an alternative provider, they were registered in accordance with this suggestion. 2. Materials and methods 2.1. Sample of respondents The study was based on information collected through a questionnaire distributed to the general population in tree municipalities in three county councils (Table 1). The county councils were selected to differ in timing of the reform, i.e. Halland in January 2007, and Skåne and Västra Götaland (VGR) in May and October 2009, respectively. The municipalities were selected to differ in terms of number of alternative providers before the reform and establishments of new providers in connection with the reform. A total of 3000 respondents were asked to participate in the survey. The sample was biased to account for an expected lower response rate in densely populated municipalities (400 in Malmö, Göteborg and Halmstad vs. 300 in the other six municipalities) and in younger age groups (1200 respondents aged 18–39 years vs. 1800 respondents aged 40–89 years). 2.2. The questionnaire The questionnaire was composed of 20 questions organised in three sections. The first section of the questionnaire contained three types of background variables, i.e. demographic and socioeconomic (age, sex, education, living conditions, occupation, country of birth), health seeking behaviour (number of health care visits in last 12 months)
A.H. Glenngård et al. / Health Policy 103 (2011) 31–37
and health status (EQ-5D and perceived health relative to others in the same age). The second section of the questionnaire related to if the respondent had made a choice of primary care provider or not and where information about the chosen provider was found. In the third section, the respondents were asked to rate different attributes and statements regarding primary care according to a Likert scale. There were 12 statements regarding preferences related to primary care (availability, friendly staff, access to different staff categories, getting information about treatment, possibility to participate in treatment decisions, help with referrals and coordination of care) and five statements regarding possible barriers to making a choice of primary care provider (interest, information and difficulties with respect to making a choice, confidence in primary care and access to alternatives to choose among). Two Likert scales from one to five (1–5) were used, i.e. 1 = “Strongly disagree” to 5 = “Strongly agree” and 1 = “Not important” to 5 = “Very important”. “Do not know”/“No opinion” was also an option. The questionnaire was tested in a pilot study including 60 respondents in November 2009. No major changes were required based on the results from the pilot study. The final questionnaires were distributed in February–April 2010. Reminders were sent out twice, once through a postcard and once with a letter together with a second copy of the questionnaire, in accordance with Dillman et al. [19]. The questionnaire is available from the authors upon request. 2.3. Statistical analysis Different analyses were performed to investigate to what extent individuals felt they had exercised choice and the reason for choosing a provider. First, descriptive statistics and non-parametric chi-square tests were used to find possible differences in answers across different groups of respondents, based on background variables. Second, binary and multilevel logistic regressions were used to analyse the relation between dependent and independent variables [20]. The dependent variable was having exercised choice or not. The independent variables were different background variables and attitudes towards statements reflecting preferences of primary care and barriers to making a choice. Variables reflecting characteristics of the municipalities (population density, number of alternative local health care providers, establishment of new local health care providers in connection with reform and distance from hospital) were also analysed. Likert scale variables were re-coded into binary variables where 1–2 = Not important/Do not agree and 3–5 = Important/Agree. The data collected was coded and analysed using SPSS 12.0 and MLwin 2.17. 3. Results 3.1. Respondents Of the 3000 questionnaires distributed, 59 were returned due to address unknown or diseased respondent and 1462 completed questionnaires were obtained (a
33
Table 2 Respondents background characteristics. Category (total 1449 respondents) Region/county Halland Skane VG Region Age (26 missing) 18–39 40–64 65–89 Sex (17 missing) Female Male Living condition (14 missing) Married/cohabitant Single Education (26 missing) Compulsory school Upper secondary school University Other Country of birth (14 missing) Sweden Other Nordic country Other European country Rest of the world Working/studying (37 missing) Yes No Self-reported health status (17 missing) compared to others in same age Worse About the same Better Number of primary care visits during last 12 months (25 missing) 0 visits 1 visit 2–5 visits More than 5 visits Level of trust/confidence in services provided at your primary care facility (49 missing) Very low or low Neither high or low High or very high
Number (%) 494 (34) 483 (33) 472 (33) 484 (34) 633 (45) 304 (21) 620 (43) 812 (57) 1085 (76) 350 (24) 422 (30) 502 (35) 429 (30) 70 (5) 1226 (85) 59 (4) 99 (7) 51 (4) 779 (55) 633 (45)
210 (15) 933 (65) 289 (20)
850 (60) 261 (18) 257 (18) 54 (4)
109 (8) 272 (20) 1017 (72)
response rate of 50%). The results were based on information from 1449 fully completed questionnaires (Table 2). As predicted, the response rate was lower in densely populated municipalities. Approximately 44% responded in Malmö, Göteborg and Halmstad compared to 47–55% in the other six municipalities. The response rate was also lower in younger respondents as 41% in the age group 18–39 responded compared to 53% in respondents 40 years and older. The response rate did not vary across counties. Almost one third of the respondents had a university education, with a higher proportion in densely populated municipalities. Among the respondents from Malmö, 43% had a university education compared to 40% according to available statistics (www.malmo.se). This suggests that the sample was slightly biased, towards higher proportion well-educated respondents compared to the actual population. Moreover, a great majority of the respondents (85%) stated that their health status was similar or better compared to other in their age.
34
A.H. Glenngård et al. / Health Policy 103 (2011) 31–37
Table 3 Proportion of respondents who answered yes to the question “Have you made a choice of primary health care unit in connection with or after the introduction of freedom of choice in your county?” by background characteristics. Category Region/municipality Halland, whereof Halmstad Hylte Falkenberg Skane, whereof Malmö Örkelljunga Bromölla VGR, whereof Göteborg Uddevalla Skövde Number of primary care visits 0 visits 1 visit 2–5 visits More than 5 visits Occupation Working/studying Other Age 18–39 40–64 65–89 Total
Proportion (%) 63 74 47 64 61 61 62 60 59 57 56 62 50 61 67 62 58 65 57 61 65 61
Only 4% of the respondents had a country of birth outside Europe. In Malmö, 6% of the respondents compared to 13% according to available statistics had a country of birth outside Europe [21]. This bias towards a lower proportion of respondents with a country of birth outside Europe was expected since the questionnaire was distributed in Swedish and means that it was not possible to analyse preferences regarding primary care for this group of the population. 3.2. Choice of provider In the total sample, 61% answered yes to the question “Have you made a choice of primary health care unit in connection with or after the introduction of freedom of choice in your county?”. The proportion that answered yes differed with respect to number of primary care visits during the last 12 months (p < 0.01), where a higher proportion among those who had made at least one such visit answered yes. The proportion that answered yes differed with respect to age (p < 0.01) and occupation (p < 0.01). A higher proportion among older respondents answered yes compared to younger respondents and a higher proportion among those who did not work or study answered yes compared to those who did work or study (see Table 3). There was no difference with respect to self-reported health status (p = 0.71), living conditions (p = 0.08), sex (p = 0.38) or education (p = 0.96). The proportion that answered yes differed with respect to municipalities (p < 0.01) but not with respect to counties (p = 0.51). The largest variation between municipalities
within a county was observed in Halland, i.e. the county council that implemented the choice system already in 2007. The most common reasons for making a choice were positive opinion about the chosen provider and that a new alternative had become available. More than 70% agreed or strongly agreed that their choice of a provider was due to a positive opinion about the provider and 45% agreed or strongly agreed that their choice was due to a new alternative becoming available. The latter was more common in municipalities with establishments of new providers. Less than 20% of the respondents who answered that they had made a choice stated that it was due to dissatisfaction with the previous provider. Regarding statements about barriers, more than 70% of all respondents strongly agreed that “It is important to have the possibility to choose primary health care provider”. Moreover, two thirds of the respondents did not think that it was difficult to make a choice and felt that they had enough information to be able to make a choice of primary care provider. The most common source of information was the chosen provider. Of the respondents who stated that they had made a choice, 38% reported that they got their information in connection with a previous visit and 23% from the county council. The latter source of information was more common in Halland (33%) than VGR (24%) and Skåne (12%). Only 3% among those who had made a choice searched the Internet for information about providers. This source was slightly more common among respondents in the two biggest cities (Malmö and Göteborg), among younger respondents and among respondents with a higher education. In total, 11% strongly agreed that there were too few alternatives to choose among. This proportion was higher in municipalities with few alternative providers and where there had been no or few establishments of new providers in connection with the reform. However, in one large city with several alternative providers and new establishments (Malmö) 14% thought that there were too few alternatives. Also the overall confidence in services provided in primary care was lower in Malmö compared to other municipalities. 3.3. Regression analysis In order to investigate any hierarchical nesting, multilevel logistic regression was performed but showed no signs of clustering within municipalities or counties/regions. Logistic regression analysis was then used to assess the impact of a number of factors on the likelihood that a respondent would answer yes to the question “Have you made a choice of primary health care unit in connection with or after the introduction of freedom of choice in your county council?”. The independent variables reflected background characteristics of the respondent and the municipalities (Char), preferences regarding primary care (Pref) and possible barriers in choice of primary care provider (Barr). Variables reflecting background characteristics where chosen based on detected differences between groups with respect to having made a choice. Variables reflecting preferences and barriers were chosen based on two initial regression
A.H. Glenngård et al. / Health Policy 103 (2011) 31–37
35
Table 4 Variables used in the full logistic regression model predicting likelihood of answering yes to the question “Have you made a choice of primary health care unit in connection with or after the introduction of freedom of choice in your county council?”. Variable 2
Char Population density above 1000/km Char Alternative local primary health care providers available Char Establishment of new provider in connection with reform Char Distance to hospital less than 5 km Char Age-group Char Visited care visits during last 12 months Char Occupation – Working or studying Pref Primary health care provider close to home Pref Access to specialist for specific needs Pref Meeting same staff at follow-up visits Barr There are too few alternative primary care providers Barr It is difficult to make a choice of provider Barr I have enough information to make a choice Barr It is important to have the possibility to make a choice Barr Overall confidence in services provided at my primary care unit
Coded
p-Value
OR
No/Yes No/Yes No/Yes No/Yes 18–39/40–64/65–89 No/Yes No/Yes Not important/Important Not important/Important Not important/Important Do not agree/Agree Do not agree/Agree Do not agree/Agree Do not agree/Agree Low/High
0.308 0.847 0.027 0.911 0.377 0.084 0.992 0.445 0.000 0.064 0.013 0.196 0.068 0.007 0.065
0.908 0.932 1.896 0.932 1.133 1.413 1.002 0.825 0.292 1.777 0.622 0.756 1.514 2.087 1.449
OR, odds ratio.
analyses; one including the 12 statements about preferences and one including the five statements about barriers. All significant variables (p < 0.05) from these two separate regression models were included in a full regression model. In Table 4, the resulting 15 variables used in the full regression model are shown, together with their p-values and odds ratios. Of the 15 chosen variables, eight were significantly associated with an increased or decreased likelihood of answering yes to the question about choice of primary care provider (see Table 4). Preference for access to a specialist for specific needs and agreeing to the statement that there are too few alternative primary care providers to choose among decreased the likelihood of answering yes to the question about having made a choice. The other variables were associated with an increased likelihood of answering yes. Several independent variables were correlated. The variable “Barr There are too few alternative providers available” was excluded since it was correlated to a majority of the other independent variables. All variables reflecting barriers were correlated to one or more of “Char Visited primary care during last 12 months”, “Pref Access to specialist for specific needs” and “Pref Meeting same staff at follow-up visits”. Moreover, the variables reflecting barriers were correlated to each other. As a result of the correlation between independent variables, different combinations were tried with non-correlated independent variables in different regression models. The regression model with the highest proportion of correctly classified cases is presented in Table 5. In the final model, which correctly classified 66% of the cases, three independent variables were included. Agreeing
to having enough information to be able to make a choice increased the likelihood of making a choice of provider by almost three times. Preferences for having access to a specialist for specific need decreased the likelihood of making a choice by more than half. The establishment of a new provider/s in connection with the choice reform increased the likelihood of making a choice by about 50%. 4. Discussion Previous studies from other countries suggest that the effects of introducing a choice system are uncertain and that people tend not to exercise choice in the sense of shifting or choosing an alternative provider [7,15]. Our study suggests, however, that information about the number of individuals who has shifted provider does not capture peoples’ perceptions about choice in full. Individuals also regard choosing to stay with their current provider as exercising choice. According to the results from our survey, a majority of the population in three Swedish counties felt that they had made a choice of primary care provider in connection with or after the introduction of a choice reform. Data from registers in the same counties show that only a smaller fraction of the population have shifted provider since the introduction of the reform. For example, in VGR the proportion that had shifted provider varied between 11 and 27% across the three municipalities in our sample by December 2009 [22]. Moreover, a higher proportion among older respondents compared to younger answered that they had made a choice of provider in our study. According to registers, shifting provider is more common among younger people. For example, in the three municipalities in VGR, 13–27% among people
Table 5 Final logistic regression model predicting likelihood of answering yes to the question “Have you made a choice of primary health care unit in connection with or after the introduction of freedom of choice in your county council?”. Variable
B
p-Value
OR
Char Establishment of new provider in connection with reform Pref Access to specialist for my specific needs Barr I have enough information to make a choice Constant
0.409 −0.855 1.097 0.035
0.006 0.000 0.000 0.899
1.505 0.425 2.995 1.036
Cox and Snell R square: 0.069. Nagelkerke R square: 0.094. Correctly classified cases: 65.7%.
95% CI for OR 1.123–2.016 0.265–0.684 2.151–4.169
36
A.H. Glenngård et al. / Health Policy 103 (2011) 31–37
aged 18–64 shifted provider compared to 7–25% in people aged 65–89 years. These differences suggest that data from registers provide a poor indicator about whether individuals have actually exercised their right to choose or not. The recently implemented reforms in Swedish primary care involve both the possibility to choose for patients and freedom of establishment for all accredited providers. The sample of nine municipalities in our survey varied both with respect to access to alternative providers and if there had been establishments of new provider/s in connection with the reform. Whereas the establishment of new providers in connection with the reform significantly increased the likelihood of making a choice, no significant association was found between the likelihood of making a choice and the number of alternative providers. This suggests that the dynamic competition created by establishments of new providers is indeed important for a choice system to work, at least initially. In the longer run, data from Halland, where the reform was implemented in 2007, i.e. two years before Skåne and VGR, indicates that the actual number of alternatives may increase in importance. The largest variation in the proportion answering that they had made a choice across municipalities was observed in Halland. The highest proportion was observed in Halmstad, with several alternatives and new establishments, and the lowest proportion was observed in Hylte, with no alternative local providers or new establishments. An alternative interpretation of these differences across municipalities in Halland is that new establishments of providers are even more important in the longer run. This study is limited to three Swedish counties and future studies with data from other counties are needed to provide additional knowledge about the validity of the two interpretations. The likelihood of making a choice of provider was also analysed with respect to individual preferences for primary care and barriers to making a choice. Preferences for having direct access to a specialist for specific needs decreased the likelihood of making a choice. This finding indicates that the idea that primary care has an added value as an agent that can refer patients to the most relevant specialist, if needed, is not supported by all individuals. Having enough information to be able to make a choice increased the likelihood of making a choice of provider by almost three times. A majority of the respondents also thought that they had enough information to be able to make such a choice. This raises the question about what information and what source of information is used. The most common source of information in all three investigated county councils was the chosen provider and the least common source was the Internet. Although searching the Internet was slightly more common in the two biggest cities in the sample, among younger respondents and among respondents with a high education, less than 10% used this source even in these groups. About 20% of the respondents in Skåne and VGR stated that they got their information from the county councils compared to one third in Halland. This suggest that individuals may search for information from the county council to a higher degree in the longer run, although the chosen provider
seems to remain the most common source of information. Hibbard and Peters [11, p. 414] argue that “to make informed choices and navigate within a complex health care system, consumers must have easily available, accurate, and timely information, and they must use it”. Although the search for information to support choice was limited across individuals in our survey, a majority actually felt that they had enough information to make a choice. This limited demand for more information raises important issues and a dilemma for the future governance of reforms by county councils. Improved comparative information about providers is indeed relevant from the perspective of the county council, however not wanted by most individuals. It is therefore probably not enough to provide comparative information about provider quality to the population through the Internet. Of equal importance is perhaps to facilitate for providers to inform patients and citizens in an adequate manner since most people tended to get their information from providers that they were already familiar with. Experiences elsewhere also suggest that comparative information about provider quality is rarely used by individuals when they choose their provider [15,23,24]. People in lower socioeconomic groups might have lesser capabilities to search for information and travel to more distant providers, thereby having lesser capabilities to exercise choice, compared to people in higher socioeconomic groups [25–27]. The respondents in our survey are biased towards higher proportion of people with a high education and a smaller proportion with a country of birth outside Europe than the general population. People with lower education and people with a country of origin outside Europe are, however, not more likely to search for information and make better-informed decisions or choices than the actual sample in our study. If individuals lack the ability to make informed choices in health care, the aim behind current reforms in Swedish primary care may be difficult to reach. A low level of health literacy, i.e. the capacity to obtain, process and understand basic health information and services needed to make appropriate health decisions, is a risk factor for deteriorating health for the individual and may lead to ineffective use of health care resources [28–30]. It may also undermine the possibility to improve the efficiency, quality and responsiveness of health systems through patient choice. Hibbard and Peters [11] suggest three processes to enhance individuals use of comparative information: lowering cognitive effort, helping people to understand the experience of a choice and highlighting the meaning of information. To support such processes will remain important challenges for county councils in their future governance of reforms involving increased choice of primary care provider for patients.
Acknowledgement This study was financially supported by research grants from Vinnvård.
A.H. Glenngård et al. / Health Policy 103 (2011) 31–37
References [1] Clark D. Public service reform: a comparative west European perspective. West European Politics 2000;23:25–44. [2] Martinussen PE, Magnussen J, In Magnussen J, Vrangbaek K, Saltman RB, editors. Nordic Health care systems – recent reforms and current policy challenges. European observatory on health systems and policies series. Berkshire: Open University Press; 2009. [3] Anell A. The monopolistic integrated model and health care reform: the Swedish experience. Health Policy 1996;37:19–33. [4] Anell A. Choice and privatisation in Swedish primary care. Health Economics Policy and Law 2010;12:1–21. [5] Le Grand J. Delivering public services through choice and competition—the other invisible hand. Princeton: Princeton University Press; 2007. [6] Le Grand J. Choice and competition in publicly funded health care. Health Economics Policy and Law 2009;4:479–88. [7] Fotaki M, Roland M, Boyd A, McDonald R, Scheaff R, Smith L. What benefits will choice bring to patients? Literature review and assessment of implications. Journal of Health Services Research and Policy 2008;13:178–84. [8] Dixon A, Le Grand J. Is greater patient choice consistent with equity? The case of the English NHS. Journal of Health Services Research and Policy 2006;11:162–6. [9] Scott A. Economics of general practice. In: Culyer AJ, Newhouse JP, editors. Handbook of health economics, vol. 1. Amsterdam: Elseiver Science; 2000. p. 1181–3. [10] Saltman RB, Rico A, Boerma W, editors. Primary care in the driver’s seat? Organisational reform in European Primary Care. Berkshire: Open University Press; 2006. [11] Hibbard JH, Peters E. Supporting informed consumer health care decisions: data presentation approaches that facilitate the use of information in choice. Annual Reviews of Public Health 2003;24:413–33. [12] Barr DA, Fenton L, Blane D. The claim for patient choice and equity. Journal of Medical Ethics 2008;34:271–4. [13] Mold JW, Fryer GW, Roberts M. When do older patients change primary care physicians? The Journal of the American Board of Family Practice 2004;17:453–60. [14] Brouwer W, van Exel J, Hermans B, Stoop A. Should I stay or should I go? Waiting lists and cross-border care in the Netherlands. Health Policy 2003;63(3):289–98. [15] Dixon A, Robertson R, Appleby J, Burge P, Devlin N, Magee H. Patient choice—how patients choose and how providers respond. London: The Kings Fund; 2010.
37
[16] Anell A, Rosén P, Hjortsberg C. Choice and participation in the health services: a survey of preferences among Swedish residents. Health Policy 1997;40:157–68. [17] Hjelmgren J, Anell A. Population preferences and choice of primary care models: a discrete choice experiment in Sweden. Health Policy 2007;83:314–22. [18] Rosén P, Anell A, Hjortsberg C. Patient views on choice and participation in primary health care. Health Policy 2001;55: 121–8. [19] Dillman D, Smyth JD, Christian SM. Internet mail and mixed-mode surveys: the tailored design method, 3rd ed. New Jersey: John Wiley & Sons, Inc.; 2007. [20] Afifi AA, Clark V. Computer-aided multivariate analysis, 2nd ed. New York: van Nostrand Reinhold; 1990. [21] City of Malmö website, www.malmo.se. [22] VGR (Västra Götalandsregionen), personal communication. Anders Lindgren, Regional Office, Skövde 2010-06-28. [23] Marshall MN, Shekelle PG, Leatherman S, Brook RH. The public release of performance data: what do we expect to gain? A review of the evidence. Journal of the American Medical Association 2000;283:1866–74. [24] Burgess S, Propper C, Wilson D. Will more choice improve outcomes in education and health care? The evidence from economic research. Bristol: Centre for Market and Public Organisation, University of Bristol; 2005. [25] Coulter A, le Maistre N, Henderson L. Patients’ experience of choosing where to undergo surgical treatment—evaluation of the London patient choice scheme. Oxford: Picker Institute; 2005. [26] Burge P, Devlin N, Appleby J, Rohr C, Grant J. Do patients always prefer quicker treatment? A discrete choice analysis of patients’ stated preferences in the London Patient Choice Project. Applied Health Economics and Health Policy 2004;4:183–94. [27] Barber S, Gordon-Dseagu V. Equity issues relating to elective care and patient choice. London: Council of Ethnic Minority Voluntary Sector Organisations, Choice and Social Exclusion Unit College of Health; 2003. [28] Institute of Medicine. Health literacy: a prescription to end confusion. Washington, DC: Institute of Medicine; 2004. [29] Eichler K, Weiser S, Brugger U. The costs of limited health literacy: a systematic review. International Journal of Public Health 2009;54:313–24. [30] Nutbeam D. The evolving concept of health literacy. Social Science and Medicine 2008;67:2072–8.
Contents lists available at ScienceDirect
Health Policy journal homepage: www.elsevier.com/locate/healthpol
Choice of primary care provider: Results from a population survey in three Swedish counties Anna H. Glenngård a,b,∗ , Anders Anell c , Anders Beckman d a b c d
Institute of Economic Research, Lund University School of Economics and Management, Lund, Sweden The Swedish Institute for Health Economics (IHE), Lund, Sweden Institute of Economic Research and Department of Business Administration, Lund University School of Economics and Management, Lund, Sweden Department of Clinical Sciences and General Practice/Family Medicine, Faculty of Medicine, Lund University, Malmö, Sweden
a r t i c l e
i n f o
Keywords: Primary health care Choice Sweden Population preferences Provider competition Privatisation
a b s t r a c t Recent reforms in Swedish primary care have involved choice of provider for the population combined with freedom of establishment and privatisation of providers. This study focus to what extent individuals feel they have exercised a choice of provider, why they exercise choice and where they search for information, based on a population survey in three Swedish counties. The design of the study enabled for studying behaviour with respect to differences in time since introduction of the reform and differences in number of alternative providers and establishments of new providers in connection with the reform. About 60% of the population in the three counties felt that they had made a choice of provider in connection with or after the introduction of a reform focusing on choice and privatisation. Establishments of new providers and having enough information increased the likelihood whereas preferences for direct access to a specialist decreased the likelihood of making a choice. The data further suggests that individuals were rather passive in their search for information and tended to choose providers that they previously had been in contact with. This is in line with results from previous studies and poses challenges for county councils governance of reforms. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction During the past two decades, publicly funded health care systems in northern Europe have been subject to reforms such as purchaser–provider split, decentralisation, privatisation, the introduction of explicit measures of performance and focus on service and client orientation, often referred to as New Public Management (NPM) reforms [1,2]. Swedish health care is no exception. The responsibility of financing and provision of health care in Sweden lies with 21 county councils and reforms are therefore usually
∗ Corresponding author at: Institute of Economic Research, Lund University School of Economics and Management, PO Box 7080, 22007 Lund, Sweden. Tel.: +46 73 6348487. E-mail address: [email protected] (A.H. Glenngård). 0168-8510/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.healthpol.2011.05.014
implemented at the local level and gradually throughout the country. The first steps towards increased choice of providers for patients took place already in the early 1990s. By then, however, provider choice for patients was not combined with privatisation and freedom of establishment of providers and payment to providers did not follow the choice of patients. Consequently, there was limited alternatives for the population to choose among and limited incentives for providers to respond to expanded choice for patients [3]. More recent reforms involving expanded choice of provider have focused on primary care and combined choice of provider with freedom of establishment and privatisation of providers [4]. The new reform involving both choice and privatisation of primary care was first implemented in the county council of Halland in 2007, followed by Västmanland and Stockholm in 2008 and seven
32
A.H. Glenngård et al. / Health Policy 103 (2011) 31–37
additional counties during 2009. Since January 1st 2010, following a change in the Health Care Act decided by the Parliament, choice of primary care provider for the population and freedom of establishment for providers accredited by the local county councils is mandatory. Patient choice of provider is expected to improve the efficiency, quality and responsiveness of the health system through the threat of exit [5,6]. As patients can shift provider in case they are not satisfied with their current provider, they can cause a loss of income for the providers [7,8]. Primary health care plays a central role in most health care systems and strengthening primary care is widely seen as central in enhancing equity and efficiency in health care [9,10]. This makes choice of primary care provider especially interesting and perhaps the most important choice in health care for individuals to make. In principle, there are choice systems with and without competition of providers but for such a system to work in practice it has been argued that alternative providers for the population to choose among are essential. Of equal importance, however, is that the population is interested and adequately informed to make a choice [4–6,11]. The main argument against extended patient choice is that it might increase inequalities in access to care. People with lower income and/or education have lesser capabilities to exercise choice compared to people with higher income and/or education who are more articulate, have better skills in searching for information and are more able to travel to non-local providers [7,8,12]. Findings from previous studies in other countries suggest that people in general are interested in choice but that the actual effects of introducing choice systems are uncertain and often rather limited [7]. In practice, people tend not to shift provider even if they have the possibility [13–15]. Previous Swedish studies based on surveys and stated preferences have demonstrated that individuals in general are interested in choice and participation in primary care services [16–18]. There is however limited information about how individuals act or where they search for information in a choice-based system in practice. Recently implemented reforms in Swedish primary care enables for studying peoples actual behaviour in a choice-based system. Data from registers provides information about if individuals have shifted between providers but provide no information about to what extent individuals feel they have exercised choice of primary care provider, why they exercise choice and where they search for information. The aim of this study was to answer these questions through a population survey in nine municipalities located in three Swedish counties. Possible differences across different groups of the population were investigated. Moreover, possible barriers to making a choice and preferences regarding primary care of importance for the choice of primary care providers were investigated. The three selected counties varied in timing of introduction of the reform but all three introduced the reform before 2010. The nine selected municipalities within these counties varied in terms of number of alternative providers before the reform and establishments of new providers in connection with the reform. The overall design of reform in the three county councils was similar, however, with
Table 1 Sample of respondents. Region (date of reform) and municipality
Number of respondents
Halland (January 07) Halmstad 400 Hylte 300 Falkenberg 300 Skane (May 09) 400 Malmö Örkelljunga 300 Bromölla 300 VGR (October 09) 400 Göteborg Uddevalla 300 Skövde 300
Population
Alternative local primary hcp
90 000 10 000 40 000
Yesa No Yesa
287 000 10 000 12 000
Yesa Yesa No
500 000 51 000 51 000
Yesa Yesa Yesa
Note: Population in Sweden 9 340 682 December 2009. a Establishment of new primary hcp in connection with or after introduction of reform by December 2009.
respect to principles for registration, payment systems, financial responsibility for providers, and freedom of establishment for all accredited providers. The requirements from each county council primarily focus on minimum number of available clinical competencies at the primary care unit. In all three county councils, a letter with information about the reform and suggestion of primary care provider based on the last primary care visit or geographical distance (in that order) was sent to all residents prior to the introduction of the reform. If individuals did not change to an alternative provider, they were registered in accordance with this suggestion. 2. Materials and methods 2.1. Sample of respondents The study was based on information collected through a questionnaire distributed to the general population in tree municipalities in three county councils (Table 1). The county councils were selected to differ in timing of the reform, i.e. Halland in January 2007, and Skåne and Västra Götaland (VGR) in May and October 2009, respectively. The municipalities were selected to differ in terms of number of alternative providers before the reform and establishments of new providers in connection with the reform. A total of 3000 respondents were asked to participate in the survey. The sample was biased to account for an expected lower response rate in densely populated municipalities (400 in Malmö, Göteborg and Halmstad vs. 300 in the other six municipalities) and in younger age groups (1200 respondents aged 18–39 years vs. 1800 respondents aged 40–89 years). 2.2. The questionnaire The questionnaire was composed of 20 questions organised in three sections. The first section of the questionnaire contained three types of background variables, i.e. demographic and socioeconomic (age, sex, education, living conditions, occupation, country of birth), health seeking behaviour (number of health care visits in last 12 months)
A.H. Glenngård et al. / Health Policy 103 (2011) 31–37
and health status (EQ-5D and perceived health relative to others in the same age). The second section of the questionnaire related to if the respondent had made a choice of primary care provider or not and where information about the chosen provider was found. In the third section, the respondents were asked to rate different attributes and statements regarding primary care according to a Likert scale. There were 12 statements regarding preferences related to primary care (availability, friendly staff, access to different staff categories, getting information about treatment, possibility to participate in treatment decisions, help with referrals and coordination of care) and five statements regarding possible barriers to making a choice of primary care provider (interest, information and difficulties with respect to making a choice, confidence in primary care and access to alternatives to choose among). Two Likert scales from one to five (1–5) were used, i.e. 1 = “Strongly disagree” to 5 = “Strongly agree” and 1 = “Not important” to 5 = “Very important”. “Do not know”/“No opinion” was also an option. The questionnaire was tested in a pilot study including 60 respondents in November 2009. No major changes were required based on the results from the pilot study. The final questionnaires were distributed in February–April 2010. Reminders were sent out twice, once through a postcard and once with a letter together with a second copy of the questionnaire, in accordance with Dillman et al. [19]. The questionnaire is available from the authors upon request. 2.3. Statistical analysis Different analyses were performed to investigate to what extent individuals felt they had exercised choice and the reason for choosing a provider. First, descriptive statistics and non-parametric chi-square tests were used to find possible differences in answers across different groups of respondents, based on background variables. Second, binary and multilevel logistic regressions were used to analyse the relation between dependent and independent variables [20]. The dependent variable was having exercised choice or not. The independent variables were different background variables and attitudes towards statements reflecting preferences of primary care and barriers to making a choice. Variables reflecting characteristics of the municipalities (population density, number of alternative local health care providers, establishment of new local health care providers in connection with reform and distance from hospital) were also analysed. Likert scale variables were re-coded into binary variables where 1–2 = Not important/Do not agree and 3–5 = Important/Agree. The data collected was coded and analysed using SPSS 12.0 and MLwin 2.17. 3. Results 3.1. Respondents Of the 3000 questionnaires distributed, 59 were returned due to address unknown or diseased respondent and 1462 completed questionnaires were obtained (a
33
Table 2 Respondents background characteristics. Category (total 1449 respondents) Region/county Halland Skane VG Region Age (26 missing) 18–39 40–64 65–89 Sex (17 missing) Female Male Living condition (14 missing) Married/cohabitant Single Education (26 missing) Compulsory school Upper secondary school University Other Country of birth (14 missing) Sweden Other Nordic country Other European country Rest of the world Working/studying (37 missing) Yes No Self-reported health status (17 missing) compared to others in same age Worse About the same Better Number of primary care visits during last 12 months (25 missing) 0 visits 1 visit 2–5 visits More than 5 visits Level of trust/confidence in services provided at your primary care facility (49 missing) Very low or low Neither high or low High or very high
Number (%) 494 (34) 483 (33) 472 (33) 484 (34) 633 (45) 304 (21) 620 (43) 812 (57) 1085 (76) 350 (24) 422 (30) 502 (35) 429 (30) 70 (5) 1226 (85) 59 (4) 99 (7) 51 (4) 779 (55) 633 (45)
210 (15) 933 (65) 289 (20)
850 (60) 261 (18) 257 (18) 54 (4)
109 (8) 272 (20) 1017 (72)
response rate of 50%). The results were based on information from 1449 fully completed questionnaires (Table 2). As predicted, the response rate was lower in densely populated municipalities. Approximately 44% responded in Malmö, Göteborg and Halmstad compared to 47–55% in the other six municipalities. The response rate was also lower in younger respondents as 41% in the age group 18–39 responded compared to 53% in respondents 40 years and older. The response rate did not vary across counties. Almost one third of the respondents had a university education, with a higher proportion in densely populated municipalities. Among the respondents from Malmö, 43% had a university education compared to 40% according to available statistics (www.malmo.se). This suggests that the sample was slightly biased, towards higher proportion well-educated respondents compared to the actual population. Moreover, a great majority of the respondents (85%) stated that their health status was similar or better compared to other in their age.
34
A.H. Glenngård et al. / Health Policy 103 (2011) 31–37
Table 3 Proportion of respondents who answered yes to the question “Have you made a choice of primary health care unit in connection with or after the introduction of freedom of choice in your county?” by background characteristics. Category Region/municipality Halland, whereof Halmstad Hylte Falkenberg Skane, whereof Malmö Örkelljunga Bromölla VGR, whereof Göteborg Uddevalla Skövde Number of primary care visits 0 visits 1 visit 2–5 visits More than 5 visits Occupation Working/studying Other Age 18–39 40–64 65–89 Total
Proportion (%) 63 74 47 64 61 61 62 60 59 57 56 62 50 61 67 62 58 65 57 61 65 61
Only 4% of the respondents had a country of birth outside Europe. In Malmö, 6% of the respondents compared to 13% according to available statistics had a country of birth outside Europe [21]. This bias towards a lower proportion of respondents with a country of birth outside Europe was expected since the questionnaire was distributed in Swedish and means that it was not possible to analyse preferences regarding primary care for this group of the population. 3.2. Choice of provider In the total sample, 61% answered yes to the question “Have you made a choice of primary health care unit in connection with or after the introduction of freedom of choice in your county?”. The proportion that answered yes differed with respect to number of primary care visits during the last 12 months (p < 0.01), where a higher proportion among those who had made at least one such visit answered yes. The proportion that answered yes differed with respect to age (p < 0.01) and occupation (p < 0.01). A higher proportion among older respondents answered yes compared to younger respondents and a higher proportion among those who did not work or study answered yes compared to those who did work or study (see Table 3). There was no difference with respect to self-reported health status (p = 0.71), living conditions (p = 0.08), sex (p = 0.38) or education (p = 0.96). The proportion that answered yes differed with respect to municipalities (p < 0.01) but not with respect to counties (p = 0.51). The largest variation between municipalities
within a county was observed in Halland, i.e. the county council that implemented the choice system already in 2007. The most common reasons for making a choice were positive opinion about the chosen provider and that a new alternative had become available. More than 70% agreed or strongly agreed that their choice of a provider was due to a positive opinion about the provider and 45% agreed or strongly agreed that their choice was due to a new alternative becoming available. The latter was more common in municipalities with establishments of new providers. Less than 20% of the respondents who answered that they had made a choice stated that it was due to dissatisfaction with the previous provider. Regarding statements about barriers, more than 70% of all respondents strongly agreed that “It is important to have the possibility to choose primary health care provider”. Moreover, two thirds of the respondents did not think that it was difficult to make a choice and felt that they had enough information to be able to make a choice of primary care provider. The most common source of information was the chosen provider. Of the respondents who stated that they had made a choice, 38% reported that they got their information in connection with a previous visit and 23% from the county council. The latter source of information was more common in Halland (33%) than VGR (24%) and Skåne (12%). Only 3% among those who had made a choice searched the Internet for information about providers. This source was slightly more common among respondents in the two biggest cities (Malmö and Göteborg), among younger respondents and among respondents with a higher education. In total, 11% strongly agreed that there were too few alternatives to choose among. This proportion was higher in municipalities with few alternative providers and where there had been no or few establishments of new providers in connection with the reform. However, in one large city with several alternative providers and new establishments (Malmö) 14% thought that there were too few alternatives. Also the overall confidence in services provided in primary care was lower in Malmö compared to other municipalities. 3.3. Regression analysis In order to investigate any hierarchical nesting, multilevel logistic regression was performed but showed no signs of clustering within municipalities or counties/regions. Logistic regression analysis was then used to assess the impact of a number of factors on the likelihood that a respondent would answer yes to the question “Have you made a choice of primary health care unit in connection with or after the introduction of freedom of choice in your county council?”. The independent variables reflected background characteristics of the respondent and the municipalities (Char), preferences regarding primary care (Pref) and possible barriers in choice of primary care provider (Barr). Variables reflecting background characteristics where chosen based on detected differences between groups with respect to having made a choice. Variables reflecting preferences and barriers were chosen based on two initial regression
A.H. Glenngård et al. / Health Policy 103 (2011) 31–37
35
Table 4 Variables used in the full logistic regression model predicting likelihood of answering yes to the question “Have you made a choice of primary health care unit in connection with or after the introduction of freedom of choice in your county council?”. Variable 2
Char Population density above 1000/km Char Alternative local primary health care providers available Char Establishment of new provider in connection with reform Char Distance to hospital less than 5 km Char Age-group Char Visited care visits during last 12 months Char Occupation – Working or studying Pref Primary health care provider close to home Pref Access to specialist for specific needs Pref Meeting same staff at follow-up visits Barr There are too few alternative primary care providers Barr It is difficult to make a choice of provider Barr I have enough information to make a choice Barr It is important to have the possibility to make a choice Barr Overall confidence in services provided at my primary care unit
Coded
p-Value
OR
No/Yes No/Yes No/Yes No/Yes 18–39/40–64/65–89 No/Yes No/Yes Not important/Important Not important/Important Not important/Important Do not agree/Agree Do not agree/Agree Do not agree/Agree Do not agree/Agree Low/High
0.308 0.847 0.027 0.911 0.377 0.084 0.992 0.445 0.000 0.064 0.013 0.196 0.068 0.007 0.065
0.908 0.932 1.896 0.932 1.133 1.413 1.002 0.825 0.292 1.777 0.622 0.756 1.514 2.087 1.449
OR, odds ratio.
analyses; one including the 12 statements about preferences and one including the five statements about barriers. All significant variables (p < 0.05) from these two separate regression models were included in a full regression model. In Table 4, the resulting 15 variables used in the full regression model are shown, together with their p-values and odds ratios. Of the 15 chosen variables, eight were significantly associated with an increased or decreased likelihood of answering yes to the question about choice of primary care provider (see Table 4). Preference for access to a specialist for specific needs and agreeing to the statement that there are too few alternative primary care providers to choose among decreased the likelihood of answering yes to the question about having made a choice. The other variables were associated with an increased likelihood of answering yes. Several independent variables were correlated. The variable “Barr There are too few alternative providers available” was excluded since it was correlated to a majority of the other independent variables. All variables reflecting barriers were correlated to one or more of “Char Visited primary care during last 12 months”, “Pref Access to specialist for specific needs” and “Pref Meeting same staff at follow-up visits”. Moreover, the variables reflecting barriers were correlated to each other. As a result of the correlation between independent variables, different combinations were tried with non-correlated independent variables in different regression models. The regression model with the highest proportion of correctly classified cases is presented in Table 5. In the final model, which correctly classified 66% of the cases, three independent variables were included. Agreeing
to having enough information to be able to make a choice increased the likelihood of making a choice of provider by almost three times. Preferences for having access to a specialist for specific need decreased the likelihood of making a choice by more than half. The establishment of a new provider/s in connection with the choice reform increased the likelihood of making a choice by about 50%. 4. Discussion Previous studies from other countries suggest that the effects of introducing a choice system are uncertain and that people tend not to exercise choice in the sense of shifting or choosing an alternative provider [7,15]. Our study suggests, however, that information about the number of individuals who has shifted provider does not capture peoples’ perceptions about choice in full. Individuals also regard choosing to stay with their current provider as exercising choice. According to the results from our survey, a majority of the population in three Swedish counties felt that they had made a choice of primary care provider in connection with or after the introduction of a choice reform. Data from registers in the same counties show that only a smaller fraction of the population have shifted provider since the introduction of the reform. For example, in VGR the proportion that had shifted provider varied between 11 and 27% across the three municipalities in our sample by December 2009 [22]. Moreover, a higher proportion among older respondents compared to younger answered that they had made a choice of provider in our study. According to registers, shifting provider is more common among younger people. For example, in the three municipalities in VGR, 13–27% among people
Table 5 Final logistic regression model predicting likelihood of answering yes to the question “Have you made a choice of primary health care unit in connection with or after the introduction of freedom of choice in your county council?”. Variable
B
p-Value
OR
Char Establishment of new provider in connection with reform Pref Access to specialist for my specific needs Barr I have enough information to make a choice Constant
0.409 −0.855 1.097 0.035
0.006 0.000 0.000 0.899
1.505 0.425 2.995 1.036
Cox and Snell R square: 0.069. Nagelkerke R square: 0.094. Correctly classified cases: 65.7%.
95% CI for OR 1.123–2.016 0.265–0.684 2.151–4.169
36
A.H. Glenngård et al. / Health Policy 103 (2011) 31–37
aged 18–64 shifted provider compared to 7–25% in people aged 65–89 years. These differences suggest that data from registers provide a poor indicator about whether individuals have actually exercised their right to choose or not. The recently implemented reforms in Swedish primary care involve both the possibility to choose for patients and freedom of establishment for all accredited providers. The sample of nine municipalities in our survey varied both with respect to access to alternative providers and if there had been establishments of new provider/s in connection with the reform. Whereas the establishment of new providers in connection with the reform significantly increased the likelihood of making a choice, no significant association was found between the likelihood of making a choice and the number of alternative providers. This suggests that the dynamic competition created by establishments of new providers is indeed important for a choice system to work, at least initially. In the longer run, data from Halland, where the reform was implemented in 2007, i.e. two years before Skåne and VGR, indicates that the actual number of alternatives may increase in importance. The largest variation in the proportion answering that they had made a choice across municipalities was observed in Halland. The highest proportion was observed in Halmstad, with several alternatives and new establishments, and the lowest proportion was observed in Hylte, with no alternative local providers or new establishments. An alternative interpretation of these differences across municipalities in Halland is that new establishments of providers are even more important in the longer run. This study is limited to three Swedish counties and future studies with data from other counties are needed to provide additional knowledge about the validity of the two interpretations. The likelihood of making a choice of provider was also analysed with respect to individual preferences for primary care and barriers to making a choice. Preferences for having direct access to a specialist for specific needs decreased the likelihood of making a choice. This finding indicates that the idea that primary care has an added value as an agent that can refer patients to the most relevant specialist, if needed, is not supported by all individuals. Having enough information to be able to make a choice increased the likelihood of making a choice of provider by almost three times. A majority of the respondents also thought that they had enough information to be able to make such a choice. This raises the question about what information and what source of information is used. The most common source of information in all three investigated county councils was the chosen provider and the least common source was the Internet. Although searching the Internet was slightly more common in the two biggest cities in the sample, among younger respondents and among respondents with a high education, less than 10% used this source even in these groups. About 20% of the respondents in Skåne and VGR stated that they got their information from the county councils compared to one third in Halland. This suggest that individuals may search for information from the county council to a higher degree in the longer run, although the chosen provider
seems to remain the most common source of information. Hibbard and Peters [11, p. 414] argue that “to make informed choices and navigate within a complex health care system, consumers must have easily available, accurate, and timely information, and they must use it”. Although the search for information to support choice was limited across individuals in our survey, a majority actually felt that they had enough information to make a choice. This limited demand for more information raises important issues and a dilemma for the future governance of reforms by county councils. Improved comparative information about providers is indeed relevant from the perspective of the county council, however not wanted by most individuals. It is therefore probably not enough to provide comparative information about provider quality to the population through the Internet. Of equal importance is perhaps to facilitate for providers to inform patients and citizens in an adequate manner since most people tended to get their information from providers that they were already familiar with. Experiences elsewhere also suggest that comparative information about provider quality is rarely used by individuals when they choose their provider [15,23,24]. People in lower socioeconomic groups might have lesser capabilities to search for information and travel to more distant providers, thereby having lesser capabilities to exercise choice, compared to people in higher socioeconomic groups [25–27]. The respondents in our survey are biased towards higher proportion of people with a high education and a smaller proportion with a country of birth outside Europe than the general population. People with lower education and people with a country of origin outside Europe are, however, not more likely to search for information and make better-informed decisions or choices than the actual sample in our study. If individuals lack the ability to make informed choices in health care, the aim behind current reforms in Swedish primary care may be difficult to reach. A low level of health literacy, i.e. the capacity to obtain, process and understand basic health information and services needed to make appropriate health decisions, is a risk factor for deteriorating health for the individual and may lead to ineffective use of health care resources [28–30]. It may also undermine the possibility to improve the efficiency, quality and responsiveness of health systems through patient choice. Hibbard and Peters [11] suggest three processes to enhance individuals use of comparative information: lowering cognitive effort, helping people to understand the experience of a choice and highlighting the meaning of information. To support such processes will remain important challenges for county councils in their future governance of reforms involving increased choice of primary care provider for patients.
Acknowledgement This study was financially supported by research grants from Vinnvård.
A.H. Glenngård et al. / Health Policy 103 (2011) 31–37
References [1] Clark D. Public service reform: a comparative west European perspective. West European Politics 2000;23:25–44. [2] Martinussen PE, Magnussen J, In Magnussen J, Vrangbaek K, Saltman RB, editors. Nordic Health care systems – recent reforms and current policy challenges. European observatory on health systems and policies series. Berkshire: Open University Press; 2009. [3] Anell A. The monopolistic integrated model and health care reform: the Swedish experience. Health Policy 1996;37:19–33. [4] Anell A. Choice and privatisation in Swedish primary care. Health Economics Policy and Law 2010;12:1–21. [5] Le Grand J. Delivering public services through choice and competition—the other invisible hand. Princeton: Princeton University Press; 2007. [6] Le Grand J. Choice and competition in publicly funded health care. Health Economics Policy and Law 2009;4:479–88. [7] Fotaki M, Roland M, Boyd A, McDonald R, Scheaff R, Smith L. What benefits will choice bring to patients? Literature review and assessment of implications. Journal of Health Services Research and Policy 2008;13:178–84. [8] Dixon A, Le Grand J. Is greater patient choice consistent with equity? The case of the English NHS. Journal of Health Services Research and Policy 2006;11:162–6. [9] Scott A. Economics of general practice. In: Culyer AJ, Newhouse JP, editors. Handbook of health economics, vol. 1. Amsterdam: Elseiver Science; 2000. p. 1181–3. [10] Saltman RB, Rico A, Boerma W, editors. Primary care in the driver’s seat? Organisational reform in European Primary Care. Berkshire: Open University Press; 2006. [11] Hibbard JH, Peters E. Supporting informed consumer health care decisions: data presentation approaches that facilitate the use of information in choice. Annual Reviews of Public Health 2003;24:413–33. [12] Barr DA, Fenton L, Blane D. The claim for patient choice and equity. Journal of Medical Ethics 2008;34:271–4. [13] Mold JW, Fryer GW, Roberts M. When do older patients change primary care physicians? The Journal of the American Board of Family Practice 2004;17:453–60. [14] Brouwer W, van Exel J, Hermans B, Stoop A. Should I stay or should I go? Waiting lists and cross-border care in the Netherlands. Health Policy 2003;63(3):289–98. [15] Dixon A, Robertson R, Appleby J, Burge P, Devlin N, Magee H. Patient choice—how patients choose and how providers respond. London: The Kings Fund; 2010.
37
[16] Anell A, Rosén P, Hjortsberg C. Choice and participation in the health services: a survey of preferences among Swedish residents. Health Policy 1997;40:157–68. [17] Hjelmgren J, Anell A. Population preferences and choice of primary care models: a discrete choice experiment in Sweden. Health Policy 2007;83:314–22. [18] Rosén P, Anell A, Hjortsberg C. Patient views on choice and participation in primary health care. Health Policy 2001;55: 121–8. [19] Dillman D, Smyth JD, Christian SM. Internet mail and mixed-mode surveys: the tailored design method, 3rd ed. New Jersey: John Wiley & Sons, Inc.; 2007. [20] Afifi AA, Clark V. Computer-aided multivariate analysis, 2nd ed. New York: van Nostrand Reinhold; 1990. [21] City of Malmö website, www.malmo.se. [22] VGR (Västra Götalandsregionen), personal communication. Anders Lindgren, Regional Office, Skövde 2010-06-28. [23] Marshall MN, Shekelle PG, Leatherman S, Brook RH. The public release of performance data: what do we expect to gain? A review of the evidence. Journal of the American Medical Association 2000;283:1866–74. [24] Burgess S, Propper C, Wilson D. Will more choice improve outcomes in education and health care? The evidence from economic research. Bristol: Centre for Market and Public Organisation, University of Bristol; 2005. [25] Coulter A, le Maistre N, Henderson L. Patients’ experience of choosing where to undergo surgical treatment—evaluation of the London patient choice scheme. Oxford: Picker Institute; 2005. [26] Burge P, Devlin N, Appleby J, Rohr C, Grant J. Do patients always prefer quicker treatment? A discrete choice analysis of patients’ stated preferences in the London Patient Choice Project. Applied Health Economics and Health Policy 2004;4:183–94. [27] Barber S, Gordon-Dseagu V. Equity issues relating to elective care and patient choice. London: Council of Ethnic Minority Voluntary Sector Organisations, Choice and Social Exclusion Unit College of Health; 2003. [28] Institute of Medicine. Health literacy: a prescription to end confusion. Washington, DC: Institute of Medicine; 2004. [29] Eichler K, Weiser S, Brugger U. The costs of limited health literacy: a systematic review. International Journal of Public Health 2009;54:313–24. [30] Nutbeam D. The evolving concept of health literacy. Social Science and Medicine 2008;67:2072–8.