Patient Education and Counseling 86 (2012) 342–347

Medical decision making

Decision support systems for choosing a primary health care provider in Sweden

Agneta Ranerup a,*, Lars Noren b, Carina Sparud-Lundin c

a Department of Applied IT, University of Gothenburg, Gothenburg, Swedenb Centre for Consumer Science, School of Business, Economics and Commercial Law, University of Gothenburg, Swedenc Institute of Health and Care Sciences, The Sahlgrenska Academy, University of Gothenburg, Sweden

A R T I C L E I N F O

Article history:

Received 2 September 2010

Received in revised form 28 April 2011

Accepted 28 June 2011

Keywords:

Patient choice

Decision support

Primary care

Internet

A B S T R A C T

Objective: To evaluate how patients are supported with information and other forms of Web-based

decision support for making an informed choice of a primary health care provider.

Methods: The article is based on a comparative case study of recently developed tools provided by Web

portals in Sweden (one national, three regional). The theoretical framework for analysis is the general

calculation model, including the steps of isolating, examining, and ranking available options.

Results: The provision of information and other forms of support is reasonable when it comes to isolating

a particular alternative or presenting a general view of alternatives. As for examining and ranking, one

regional system and one national system offer support based on information from patient surveys and

waiting times, and one regional system offers support based on provider competencies.

Conclusion: One design alternative is to represent the opinions of patients about care centers, another is

to represent the characteristics of care centers, and a third is to combine the two. The general calculation

model is relevant as a framework for analysis from a practical as well as a theoretical viewpoint.

Practical implications: The study provides practical examples of decision support for patient choice in

primary health care.

� 2011 Elsevier Ireland Ltd. All rights reserved.

Contents lists available at ScienceDirect

Patient Education and Counseling

jo ur n al h o mep ag e: w ww .e lsev ier . co m / loc ate /p ated u co u

1. Introduction

1.1. Background

This article focuses on a theoretically based [1] comparativecase study of decision support for informed patient choice ofprimary health care service centers in the public health caresystem by means of Web portals. Primary health care is defined asa set of universally accessible first-level services that promotehealth, prevent disease, and provide diagnostic, curative, rehabili-tative, supportive, and palliative services [2]. The subject of patientchoice in its various forms and meanings is being increasinglydiscussed [3–9]. The concept might denote, for example, bookingappointments [10], the choice of a clinic or primary health careservice center for a longer or shorter period of time [11], the choiceof a health care plan [12], or the choice of type of treatment,especially for serious and complex illnesses [1]. An analysis ofpatient choice primarily based on experiences in a public healthcare system like that in the United Kingdom reveals that a generaltrend towards consumerism, patient-centeredness, and use ofInternet technologies are some of the most important reasons fordevoting increased attention to this topic. Furthermore, the issue

* Corresponding author. Tel.: +46 31 7862766; fax: +46 31 7724899.

E-mail address: ranerup@ituniv.se (A. Ranerup).

0738-3991/$ – see front matter � 2011 Elsevier Ireland Ltd. All rights reserved.

doi:10.1016/j.pec.2011.06.013

of providing adequate information and decision support forinformed patient choice of, for example, health care providers isa highly under-researched issue [7].

Countries like Denmark, the Netherlands, Norway, Sweden,and the United Kingdom have during the last decade introducedarrangements that allow patient choice of primary and second-ary health care within a public health care system. It isinteresting to note that information technology in general andthe Internet in particular are more and more used as a means ofinteraction between the health care system and its patients. Forexample, national, regional and local Web portals have beenintroduced to provide information and other forms of decisionsupport in both public and private health care systems tosupport a patient’s choice of physician, health care servicecenter, or treatment [11–14]. However, the preconditions forproviding information and decision support for informed [15]patient choice differ between countries depending on howprimary care is organized. Some authors such as Chauvette [2]distinguish between countries like the United States, Belgium,and the Netherlands, which use professional models focused onfamily physicians, and countries that use community modelsbased on health care service centers with responsibility for adistrict, like Finland and Sweden. The challenge confronting civilservants working within a community model is to presentinformation on primary health care in a certain geographicdistrict to citizens to enable choices between health care

A. Ranerup et al. / Patient Education and Counseling 86 (2012) 342–347 343

centers. This article describes how civil servants in a Swedishcontext are building Web portals to deal with this issue.

1.2. Previous research and the research question

One area of research related to the current topic concerns the useof Web-based technology to support the choice of health careprovider in a public health care system. A fairly broad analysis of e-health in primary care and a brief treatment of systems for makingappointments were conducted by Flynn et al. [16]. Green et al. [10]focused more directly on the choice of provider in U.K. public healthcare using the Choose & Book system and to what extent thissystem’s users actually felt they were offered a choice in secondarycare. It was concluded that large numbers of patients perceived noimprovement over earlier systems, possibly because of technicalproblems with online appointments or the fact that generalpractitioners (GP) can affect the information available about variouschoices. Ranerup [11,17] evaluated the decision support technolo-gies for choice of hospitals at a national level in Sweden, Norway,Denmark, and the U.K. and found the systems in Denmark and theU.K. to be the most advanced. Bastian [12] investigated a Germannational Web portal for evidence-based advice ‘‘about what citizenscould do instead of what they should do,’’ but without a particularemphasis on choice reform. Lastly, Damman [4] performed athorough search for Web portals for health care choice in all Westerncountries and found 42 instances, most of which dealt with hospitalsand of which only two (Dutch) instances included primary care.

There exists the further issue of research into more specificdetails of the design and use of information and decision supportsystems in situations of provider choice. Fasolo et al. [13] showedthat the design of comparative hospital quality scorecards affectshow people evaluate information in choice situations, whichsuggests that designers must play a critical role. Fanijang et al. [18]performed a similar, albeit theoretical, literature review on thistopic. Moser et al. [14] showed that patient evaluations such asrecommendations are still important as a basis for choice ofhospital. Ranganathan et al. [19] studied the design and use ofperformance data for the choice of doctors, emphasizing the valueof process simplicity and convenience of access.

Another relevant aspect of this topic is what people actually wantto know about primary health care as a basis for their choices. Basedon several research studies [20–23] dealing with the issue of ‘‘whatpeople think is important in primary health care,’’ these informationneeds might be summarized as follows: access-related informationthat might be translated into quality indicators (opening hours,waiting times, convenience of appointments, and so forth), aspectsof the doctor-patient relationship (continuity, friendliness, and soforth), and general judgments (such as the patient’s willingness torecommend). It appears therefore that various areas of relatedresearch exist in the literature, but there is not much that specificallydeals with the design of information and computerized decisionsupport for the choice of a primary health care provider.

In summary, there is an increasingly prevalent situation inwhich individual patients are offered a choice of primary healthcare provider in public health care that implies the initiation of asupposedly long-term relationship. Also, there is a lack of researchabout decision support for this particular choice. Therefore, ouraim is to explore in what ways does the design of Web-baseddecision support enhance the patient’s ability to make an informedchoice of primary health care provider.

1.3. Theoretical framework

This empirical study of the design of decision support systemsfor patients introduces and applies a theoretical perspective [1]that emphasizes the generality of choice behavior. It emphasizes the

importance of technology in a supporting role to enable humans toperform as informed or calculating consumers. The theoreticalframework includes the concept of calculation, which provides ageneral view of how people make well grounded and informedchoices in any situation. This is in contrast to approaches whichfocus more specifically on individuals’ psychological state anddetailed ways of reasoning, for example in hospital choice andsimilar situations [15]. In explaining their perspective on consumerchoice, Callon and Muniesa [24] describe how economists maintainthe idea of a reality of ‘‘pure’’ calculation, whereas other socialsciences try to show that real practices are infinitely more complexand leave little room for calculation practices per se. In contrast tothis perception, they look at methods for making informed choices,that is, the sources of economic calculation. They argue that materialdevices (e.g., weighing scales or supermarket shelves) as well asmore abstract tools such as those provided through the functionali-ties of Web portals are of critical importance in helping individualsto act as calculating consumers [15,24].

Following this line of thought, Callon [15] emphasizes thathumans tend to use theoretical economic tools such as rationalcalculation models when they design material devices that supportcalculation. Callon and Muniesa [24] argued that designers followcertain basic principles when they construct devices. First, toenable consumer choice, objects must be detached or isolated fromtheir context and grouped into a single framework, for example acomputer screen. Second, once objects have been selected, it mustbe possible to associate them with one another. For example, theconsumer must have means available that enable comparisonsbetween objects. Third, designers must offer consumers a way toextract results to produce a new entity (a sum, an ordered list, anevaluation). An important issue, according to this perspective, isthe actual functionality or deployment of technological devices insupport of the activities of calculating consumers.

2. Methods

2.1. Context and research method

The majority of health care services in Sweden are financed bypublic, not private means. Swedish health care is regulated bynational authorities, but run by 21 county councils or regionalauthorities that are divided into municipalities. Independentpolitical bodies that also have the right of taxation govern thecounty councils. Since 2010, there has been a law making itobligatory for county councils to introduce schemes for patientchoice of primary health care provider [3,25,26], but not of anindividual doctor [20,22].

Because of the exploratory nature of this research, the selectedapproach is the comparative case study method [27] involving fourcases or contexts for choice in primary health care in Sweden andthe decision support that each provides. The focus is on the extentto which the design of these health care Web portals and the toolsthey provide enable patients to act as informed and calculatingconsumers [15,24]. The study includes only public Web portals,three operated by county councils and one national portal operatedby the Swedish Association of Local Authorities and Regions(SALAR). A purposive sampling strategy was used. The selectedfour cases include both experienced (Stockholm, Halland) andnewer county councils (Vastra Gotaland), as well as a recentnational case that most likely provides devices that will be used bythe vast majority of these and other county councils to supportchoice. Data collection was achieved by: (1) 21 semistructuredinterviews with the two to four civil servants working with Webdesign, project management, and general issues associated withchoice reform in primary health care in each of the four cases. Theinterviews lasted between 30 and 90 min and took place in the

A. Ranerup et al. / Patient Education and Counseling 86 (2012) 342–347344

office of each respondent between April 2008 and May 2010. Theywere recorded and transcribed by the researchers. The interviewfocus was on questions about the technological tools provided tosupport choice in primary health based on the three-steptheoretical model including questions about the issues of isolating,examining, ranking and choosing alternatives, as well as questionsabout the general background, sources of inspiration and futureplans for the systems. (2) The researchers’ own investigation of thetools provided as viewed on the screen. (3) Project documentationand plans related to the overall design process.

For this study, the general model of choice or calculationdeveloped by Callon and Muniesa [24] is translated into threepractical features of a decision support system: (1) isolatingalternatives—the provision of technological tools to describe theframework for choice of a primary health care provider; (2)examining alternatives—the provision of technological tools toinvestigate available choices such as waiting times and qualityindicators; and (3) ranking alternatives—the provision of techno-logical tools to compare selected alternatives that can combinedifferent qualities and create ranked choices. In the analysis theauthors have used the recommendations of Yin about ‘‘pattern-matching’’ based on theoretical frameworks [27]. One researchermade a first analysis and a second researcher checked the result toconfirm inter-rater reliability. Using this framework [15], theauthors offer a straightforward description and examination of thevarious available tools to see to what extent the tools ‘‘match’’these features. In this analysis, all sources of data were used.

3. Results

In this section, the design of Web-based decision support in thefour cases will be described (Table 1).

3.1. County Council of Halland

3.1.1. Isolating

In January 2007, the right to choose a primary health careprovider was introduced in Halland.

Basic information and an FAQ about the specifics of the choicesystem are available. The Web site includes an option to searchamong the available providers by selecting the name of one of thesix municipalities. This selection brings up a list of primary careproviders containing between 2 and 16 health care service centers.

3.1.2. Examining

Clicking on one of these providers leads the user to its own Webpage. Previously, the only information supporting more coherentcomparisons was a pdf report containing mostly internal qualityindicators from each clinic. Recently (late October 2009), a more

Table 1Technological facilities supporting the choice of primary care provider in the County C

Halland Vastra Gotaland

Isolating alternatives Information about

the rights of choice

Information about the

rights of choice

Search for clinics selecting

municipalities and number

of instances

Search for clinics selectin

municipality or using a m

Examining alternatives Comparisons of clinics based

on contact information,

competence (pediatrics,

diabetes, etc.), available

e-services

Comparisons of clinics

based on contact inform

an webpages of clinics

Ranking and choosing

alternatives

Manual ranking,

manual choice

Manual ranking,

IT-supported choice

highly developed facility was introduced. A user starts by choosinga preferred municipality and continues by selecting one to fourclinics for further comparison. The information provided includesgeneral information like opening hours, public or private owner-ship, and address, but also the percentage of patients acceptedwithin the seven-day maximum waiting-time guarantee for accessto care. There are also a number of yes–no indicators showingavailable competencies at each clinic, including the specializedcompetencies of doctors and nurses and a list of the e-servicesavailable at each clinic. This last part of the design was motivatedas follows: ‘‘[the county council of] Stockholm offers another toolbased on comparing what people think about services. We havealways thought that we must show the facts, and we don’t want tomix in what people think, but such a discussion might follow lateron’’ (Civil servant, County Council of Halland, November 26, 2009)

3.1.3. Ranking/choice

Apart from the option to click on one to four providers formaking comparisons, there are no options for sorting and ranking.Furthermore, there is no way to make the actual choice of primaryhealth care provider other than by printing out a document andfilling it in manually.

3.2. County Council of Vastra Gotaland

3.2.1. Isolating

In this county council, choice reform in primary care wasimplemented in October 2009.

This site also includes general information about the right tochoose and how to make a choice. Finding available health careproviders is made easier by a feature through which a user canchoose a preferred municipality and a type of clinic (primaryhealth care). The result is shown on a map or in a list with contactinformation and a short text that the clinic itself has provided.

3.2.2. Examining

Clicking on the name of a care center brings up a more extensivepage with a map, as well as more extensive descriptions ofcompetencies, ownership, and contact information. There is also alink to the clinic’s own Web site and a link to a function forchoosing this particular provider. The most current developmentis, according to interviewees, a project to offer patient surveys aswell as internal quality indicators to be used to manage a regionalversion of choice reform [26]. ‘‘However, while waiting for this,there are perhaps other things we can do, like showing access datafrom SALAR. [. . .] We will also engage a consulting firm to gothrough all the Web pages of health care centers looking forinformation about language competences, psychologists, male andfemale doctors, and other characteristics. We will see what is

ouncil of Halland, Vastra Gotaland, Stockholm and at national level in Sweden.

Stockholm National level

Information about the rights

of choice

Information about the rights

of choice

g

ap

Search for clinics selecting

municipalities and

number of instances

Search for clinics selecting

municipalities and type of care

ation

Comparisons of clinics based

on contact information,

telephone access, overall

impression, overall reception,

participation in

care, information

Comparisons of clinics based on

contact information, telephone

access, first visit, general reception,

experienced usefulness, information,

participation in care, confidence,

willingness to recommend,

overall perception

Manual ranking, IT-supported

choice

Manual ranking, sometimes

IT-supported choice

A. Ranerup et al. / Patient Education and Counseling 86 (2012) 342–347 345

available today while waiting for the quality indicators to becomeavailable.’’ (County Council of Vastra Gotaland, October 8, 2009)

3.2.3. Ranking/choice

There is no facility for sorting or ranking available providers.However, the actual choice of provider can be made online using aspecific password or e-identification

3.3. County Council of Stockholm

3.3.1. Isolating

The choice reform in primary care in its current form wasimplemented in January 2008.

The Vardguiden (‘‘Health Care Guide’’) regional Web site offers asection with basic information about the right to choose and links tofurther information. A three-step procedure has been designed tosupport the actual choice of primary health care provider; the stepsare: (1) finding information about the reform, (2) finding andcomparing providers, and (3) choosing one provider [28]. ‘‘If youwant to buy a television set, you need information about the modelsavailable, the sizes, and what is good and bad. You also want tocompare them, and later on buy one using the Internet. So we tend tolook at Pricerunner and the like. [. . .] Whether buying a TV orchoosing health care or something else, there are the samecomponents in the process of choice.’’ (Designer, Stockholm CountyCouncil, April 1, 2008).

For finding available providers, there is a feature called ‘‘findingcaregivers’’ in which the uppermost button leads to primary healthcare providers. Here, an individual makes a first choice based ongeographical location (part of the County Council of Stockholmincluding this city as well as the other 15 municipalities), afterwhich a list of available providers appears with basic contactinformation (opening hours, address, etc.). For some of these, there

Fig. 1. Device for comparing primary care

is a button leading to a Web page with maps, form of ownership,links to personal e-services, and quality data from patient surveysif available (see Section 3.3.3).

3.3.2. Examining

The health care providers can be examined more closely throughthe richer Web page with maps and other information as describedabove. Furthermore, as a part of the three-step choice procedure,there is a button labeled ‘‘Comparisons.’’ Once again, a user makes achoice based on geographical location, but as a second step, the usercan select a number of clinics. For each, contact information as wellas data from yearly patient surveys including overall impression,overall reception, participation in decisions, quality of information,and telephone access (Fig. 1, left column) appear on the same page.The other columns show the data from each clinic that has beenselected (Fig. 1, next three columns).

3.3.3. Ranking/choice

Aside from the comparison function, there is no facility forsorting or ranking facilities. However, the actual choice of providercan be made through a password-protected personal e-service (MyHealth Contacts). Here, a user can see the provider that he/she haschosen and make new choices if desired. The most recentdevelopment in the available tools involves introducing functionsfor making comparisons and choices in more areas, for exampleprenatal and postnatal clinics. For facilities offering eye care, therewill be indicators showing outcomes

3.4. National Web portal (1177.se)

3.4.1. Isolating

Since 1999, there has been a national Web portal called 1177.se,which contains information about illnesses and treatments. Since

centres in Stockholm County Council.

A. Ranerup et al. / Patient Education and Counseling 86 (2012) 342–347346

2006, a proposal has been under consideration by prominent actorsin health care to implement a full-fledged patient portal. Work hasbeen ongoing since then, and a first version of a complete portal withnational and regional information about health care, a tool forsearching all health care providers, and personal services in the formof My Health Contacts will be available at the end of 2010 [29].

Since June 2010, a tool has been available to search among alltypes of health care units in a municipality within a county council.This tool enables a user to search using terms like psychiatric care,emergency ward, child welfare center, and primary care. It forms partof the design of the national Web portal and its associateddevelopment project.

3.4.2. Examining

A national Web portal called Waiting Times in Care wasintroduced in 2000. It contains a database of waiting times forvarious types of treatments, including primary care (access by phonethe same day, first visit within seven days). The portal was created bySALAR, and its information is used in the tool for comparing primaryhealth care providers described in this section. Furthermore, sinceApril 2010, a tool has been available for further investigation ofprimary health care providers. This tool can be accessed on thenational Web portal, 1177.se, or through the Web portals of thecounty councils if they have introduced this option. The tool wascreated by SALAR and its IT development organization. It is based on apolicy document from the Ministry of Health and Social Affairs [25].

After clicking on the ‘‘Compare care’’ button on the national Webportal, 1177.se, a user must first choose a county council (if he/she isnot accessing the device from the Web portal of his/her own countycouncil). The next option is to choose among all the primary careclinics in a county council or among those within an individualmunicipality. Here one can click on one to four clinics for furthercomparison. The available information and indicators are: contactinformation, telephone access, first visit, and patient surveys aboutgeneral reception, experienced usefulness, quality of information,participation in decisions, degree of confidence, willingness torecommend, and overall perception. In support of the large amountof data from patient surveys, one interviewee said: ‘‘Researcher:Some people say that patient surveys are not facts but emotions.Interviewee: Then I get so [angry]. The patients’ experiences arealways right. [. . .] If we don’t listen to patients and compare this withwhat we measure, we will only have our own organizationalperspective’’ (Civil servant, National Board of Health and Welfare,May 24, 2010). Future versions are planned to include informationabout the specialized competencies of doctors and nurses [30].

3.4.3. Ranking/choice

The current tools include no facilities for comparing orranking alternatives other than those already described.However, at the regional level of these tools, there is a linkcalled ‘‘Make a choice,’’ leading to each county council’s owntools. These offer information about how to make a choicemanually or to a tool which is part of My Health Contacts andwhich enables the user to make a choice of primary health careprovider. Approximately two-thirds of the 21 county councilshave adopted this tool

4. Discussion and conclusion

4.1. Discussion

The results of this research indicate that civil servants havedesigned tools in a way that resembles the overall model of thecalculation process as described in [24], encompassing a specifi-cation and a comparison of available providers as well astechnological support (with the exception of one case: Halland)

for making the actual choice. In the phase of isolating objectsamong which patients can choose, all four cases are similar in thedesign of tools they provide (Table 1). All Web portals start with ageographical division based on municipalities. The visitor mustfirst choose the municipality of interest. The national portal coversall municipalities, but the county councils’ portals cover only theirown. Once the patient has chosen a municipality, the health careservice centers there are displayed, and these are the main objectsto be isolated by designers. Nevertheless, it must be kept in mindthat the actual isolated object of choice is the health care servicecenter, not the individual doctor [19,20,22]. More specifically, incommunity models for primary care like the one studied here,patients might find information about the available doctors in aspecific unit, for example on the center’s own Web pages, as well aspossibly more informal evaluations on private dedicated Web sites(e.g., Doktorsguiden.se, or ‘‘the Doctor Guide’’). However, accord-ing to the regulations in the Swedish choice reform the patientcannot in fact make a choice among the doctors listed.

In the Examining phase, the challenge for designers is to presentthe differences between health care service centers that might berelevant for patient choice. One major question regarding thesedifferences is whether data from patient surveys should be used. Intwo of the case studies (Halland and the National case), there werefairly animated arguments about the pros and cons of these typesof data, differentiating between what is characterized as ‘‘facts’’and ‘‘opinions.’’ In spite of this, there are signs that the Nationalsystem and the Vastra Gotaland system might in future includeboth types of data. More specifically, in contrast to Halland that didnot want to use patient survey data, Stockholm relies heavily onsuch data. The idea of the designers in Halland was to present thekinds of specialties that the care center can provide, for examplenurses specialized in diabetes. Such information might differenti-ate between care centers with regard to the number and type ofservices available, which might facilitate the choices of patients.

This study introduces a theoretical framework for the analysisof the design of the provided tools in the form of the generalcalculation model [24]. The theoretical model might be questionedfor supposedly imposing an ‘‘economistic’’ model of thinking onhealth care [31]. However, the model seems to be useful because itis relevant for describing how civil servants actually work indesigning a market for primary health care that allows informedpatients choice. There is a certain value in applying theoreticalframeworks to designing and evaluating decision support forpatients [1]. In addition, it can be argued that the choice of primaryhealth care is not the same as the choice of treatments or hospitals[13] in situations of serious illness. Therefore, a case can be madefor choosing what is believed to be a more general choice modelrather than a model which uses psychological perspectives with amore marked focus on details of individual thinking and preferenceconstruction [15]. According to empirical investigations of patients[16,20–22], broadly speaking, access-oriented aspects such aswaiting times and general convenience, as well as relation-oriented aspects such as the willingness of other patients torecommend and patients’ judgments of other kinds, are importantto people in defining their relationship with primary health care.The results of this research show that these aspects are bothintuitively a part of the model (Isolating, Examining) along withthe implemented tools (Table 1). The high importance of thesedimensions also indicates that people strongly value having achoice [9], especially if they are not content with currentarrangements [6], a conclusion supported by both the modeland the tools.

As for how people actually choose a primary health careprovider, what support they actually need, and how this supportshould be designed [13,15,18], these are all aspects that are notaddressed here and that therefore remain issues for further

A. Ranerup et al. / Patient Education and Counseling 86 (2012) 342–347 347

research. The paper is also limited due to its cross-sectional design.Further research should include a closer analysis of the set ofquality indicators that will be implemented in newly developedsystems. The law regulating the arrangements for choice ofprimary care in Sweden establishes the right to choose a primaryhealth care provider, but offers to the county councils options tomake different arrangements for its implementation [25,26].Therefore, one interesting issue for further research is a potentialcompetition between regional and national tools.

4.2. Conclusions

The aim was to explore in what ways does the design of Web-based decision support enhance patient’s capacities to make aninformed choice of primary healthcare provider. It can beconcluded that the general calculation model is relevant as aframework for analysis of tools that support the choice of healthcare provider from both practical and theoretical viewpoints. Thearticle provides several examples of how a choice situation mightbe framed by software designers in a situation where patientchoice is becoming more and more important. The study reveals adifference of opinion among designers, which sometimes becomesquite animated, regarding the publication of data about compe-tence and quality indicators on the one hand and from patientsurveys on the other hand. A third option is to combine these twoapproaches. Moreover, the study reveals a multitude of differencesamong the tools provided: national vs. regional tools, datacategories from patient surveys, waiting times, and competences,as well as choice with and without computerized support. Onetheoretical contribution of this study is to test the framework ofCallon and Muniesa [24] in the research area of health care ingeneral and decision support in health care in particular, thusopening up this area for further research and discourse.

4.3. Practical implications

This study has contributed to a broader empirical basis fordiscussing decision support for patients [1] by the inclusion oftools for choosing health care providers for primary care. This workis relevant for health care in general and the design of decisionsupport for patients in particular. However, the experiences assuch are also of value to politicians, health care managers, anddoctors who aim to provide patients who have special needs forprimary health care a tool to make an informed choice among theavailable providers.

Acknowledgement

Thanks are due to the Bank of Sweden Tercentenary Foundationfor funding this research.

References

[1] Durand M-E, Stiel M, Boivin J, Elwyn G. Where is the theory? Evaluating thetheoretical frameworks described in decision support technologies. PatientEduc Couns 2008;71:125–35.

[2] Chauvette M. Choices for change: the path for restructuring primary health-care services in Canada. Ottawa, Ontario: Canadian Health Services ResearchFoundation; 2003 , http://www.chsrf.ca/final_research/commissioned_re-search/policy_synthesis/pdf/choices_for_change_e.pdf.

[3] Burstrom B. Analysis. Looking to Europe. Will Swedish healthcare reform affectequity? Br Med J 2009;339:b4566.

[4] Damman, O. Public reporting about healthcare users’ experiences: The Con-sumer Quality Index. Dissertation. NIVEL, Utrecht; 2010.

[5] Deccache A, van Ballekom K. From patient compliance to empowerment andconsumer’s choice: evolution or regression? An overview of patient educationin French speaking European countries. Patient Educ Couns 2010;78:282–7.

[6] Foutaki M, Roland M, Boyd A, McDonald R, Scheaff R, Smith L. What benefitswill choice bring to patients? Literature review and assessment ofimplications. J Health Serv Res Policy 2008;13:178–84.

[7] Greener I. Are the assumptions underlying patients choice realistic? A reviewof the evidence. Br Med Bull 2007;1–10.

[8] Potter J. Consumerism and the public sector: how well does the coat fit? PublicAdmin 1988;66:149–64.

[9] Protheroe J, Bower P. Choosing, deciding, or participating: what do patientswant in primary care? Br J Gen Pract 2008;603–4.

[10] Green J, McDowell Z, Potts H. Does Choose & Book fail to deliver the expectedchoice to patients? A survey of patients’ experiences of outpatients booking.BMC Med Inform Dec Mak 2008;8. doi: 10.1186/1472-6947-8-36.

[11] Ranerup A. In what ways does Web technology support the individual inchoice reforms in health care? A comparison among Norway, Denmark andSweden. Int J Health Care Inform Syst Inform 2008;3:48–68.

[12] Bastian H. Health literacy and patient information: developing the methodol-ogy for a national evidence-based website. Patient Educ Couns 2008;73:551–6.

[13] Fasolo B, Reutskaja E, Dixon A, Boyce T. Helping patients choose: how toimprove the design of comparative scorecards of hospital quality. Patient EducCouns 2010;78:344–9.

[14] Moser A, Korstjens I, van der Weijden T, Tange H. Themes affecting health-care consumers’ choice of a hospital for elective surgery when receiving web-based comparative consumer information. Patient Educ Couns2010;78:365–71.

[15] Callon M. The embeddedness of economic markets in economics. In: Callon M,editor. The Laws of the Market. Oxford: Blackwell Publishers; 1998. p. 1–57.

[16] Flynn D, Gregory P, Makki H, Gabbay M. Expectations and experiences ofeHealth in primary care: a qualitative practice-based investigation. Int J MedInf 2009;78:588–604.

[17] Ranerup A. Transforming patients to consumers: evaluating national health-care portals. Int J Health Care Manage 2010;23:331–9.

[18] Fanjiang G, von Glahn T, Chang H, Rogers W, Safran D. Providing patients’ web-based data to inform physician choice: if you build it, will they come? J GenIntern Med 2007;22:1463–6.

[19] Ranganathan M, Hibbard J, Rodday A, de Brantes F, Conroy K, Rogers WH, et al.Motivating public use of physician-level performance data: an experimenton the effects of message and mode. Med Care Res Rev 2009;66:68–91.

[20] Bornstein B, Marcus D, Cassidy W. Choosing a doctor: an exploratory study offactors influencing patients’ choice of a primary care doctor. Lincoln: Psychol-ogy Department of Faculty Publications, Department of Psychology, Universityof Nebraska; 2000.

[21] Cheregahi-Sohi S, Risa Holoe A, Mead N, Mc Donald R, Whalley D, Bower P,et al. What patients want from primary care consultations discrete choiceexperiment to identify patients’ priorities. Am Fam Med 2008;6:107–15.

[22] Faber M, Bosch M, Wollersheim H, Leatherman S, Grol R. Public reporting inhealth care: how do consumers use quality-in-care information? A systematicreview. Med Care 2009;47:1–8.

[23] Wong S, Watson D, Young E, Regan S. What do people think is important aboutprimary healthcare? Health Care Policy 2008;3:89–104.

[24] Callon M, Muniesa F. Economic markets as calculative collective devices.Organ Stud 2005;26:1229–50.

[25] Ministry of Health and Social Affairs. Uppdrag om oppna jamforelser avlandstingens primarvard samt utbetalning av medel.Assignment aboutopen comparisons of primary healthcare services and payment of econom-ic means to pursue this project Stockholm: Ministry of Health and SocialAffairs; 2009 .

[26] Ministry of Health and Social Affairs. Choice reform in primary care. Govern-ment Bill [Vardval i primarvarden. Proposition]. 2008/09:74. Stockholm:Department of Health and Social Affairs; 2008.

[27] Yin RK. Case study research: design and methods, 2nd ed., London: Sage;1994.

[28] Vardguiden. Vardval Stockholm - e-tjanstekoncept pa Vardguiden (Choicereform in primary care in Stockholm–an e-service concept in the HealthcareGuide). Stockholm: Vardguiden; 2007.

[29] Center for E-health in Sweden. Projektbeskrivning. Varden pa webben. Etapp4. Utveckling och leverans av 1177.se.Project description. Healthcare on theWeb. Round 4. Development and implementation of 1177.se. Stockholm:Centre for E-health in Sweden; 2010.

[30] Inera. Emma Lennestal, project leader, Roadmap for development, e-mail,November 26, 2009.

[31] Archer M, Tritter J. Rational choice theory. Resisting colonization. London andNew York: Routledge; 2000.