Citizen perspectives on the use of publicly reported primary care performance information: Results from citizen‐patient dialogues in three Canadian provinces

Abstract Objective Performance measurement and reporting is proliferating in all sectors of the healthcare system, including primary care, despite a dearth of evidence on how the public uses reports on primary care performance. We explored how the public might use this information, to guide the development of effective reporting systems for primary care. Methods We conducted six full‐day deliberative dialogue sessions with a purposive sample of 56 citizen‐patients across three Canadian provinces (British Columbia, Ontario and Nova Scotia). Participants identified how they would use publicly reported performance data. We conducted a thematic analysis of the data by region. Results Common uses for primary care performance information emerged across all sessions. Participants most often discussed the utility of this information for community advocacy and participation in health system decision making. Similar barriers for using performance information to choose a primary care provider were identified in each region including the perceived lack of choice of providers and the high value placed on relationships with current providers. Finally, the value of public performance reporting in enhancing trust that people would receive good care was also a common theme. Conclusions Citizen‐patient perspectives highlight that public reporting on primary care performance could promote the health system's responsiveness by enabling public engagement in decision making at the community level. The role of public reporting in promoting trust rather than empowering patient choice may reflect unique elements of the Canadian health system's context.


| INTRODUC TI ON
The increasing focus on patient experience, accountability, costeffectiveness and quality improvement in healthcare is driving investment into healthcare system performance measurement and reporting. [1][2][3][4][5][6] Public reporting has many intended purposes, including quality improvement, 1,7 public accountability, 2,[8][9][10][11] patient engagement [12][13][14] and informed decision making. 15 However, limited and mixed evidence exists on how public reporting of performance information influences the behaviours of various stakeholders. 1,14,16,17 This is especially true in primary care where there is less evidence on public reporting of performance compared to hospital and specialist sectors. [18][19][20][21] Understanding how the public uses primary care performance information may be helpful in shaping effective reporting strategies. 22,23 Countries with stronger primary care systems have healthier populations, as well as a more equitable distribution of health across the population. 24 As primary care is becoming increasingly complex, 25 helping the public understand how their primary care system performs is an important part of understanding their country's overall healthcare system. 26 Countries like the UK and Australia have conducted nationwide public reporting on primary care performance for more than a decade. They are now sharing early lessons learned on the optimal format and ways of presenting data to the public, including emphasizing easily accessible data for initially engaging the public. [27][28][29] Smaller scale studies of the utility of well-designed reports in the USA have shown that people choosing healthcare providers can engage with quality of care information which is easy to understand to select high-value choices. 30 Unique features of Canada's primary care systems, such as its public funding, expectation that healthcare is a government responsibility, and shortage of primary care providers in some communities, 9,31,32 suggest the value and type of publicly reported primary care performance information could be different in this context. As a greater number of provinces and health regions are publishing measures of primary care performance, there is a need to explore how people might use publicly reported primary care performance information to inform the design of these emerging reporting initiatives. In this study, we conducted a series of citizen-patient deliberations in three Canadian provinces focused on the uses of and optimal methods for primary care performance measurement and reporting.

| ME THODS
This project was embedded within a larger programme of research comparing primary care performance in three distinct regions in the provinces of British Columbia, Ontario and Nova Scotia. The regions were selected for their varied approach to primary care reform and performance measurement and reporting over the past decade (unpublished data) 33 and because they were identified in 2013, the first year of the study, as health regions with similar socio-demographic profiles. 34 The regions also had marked differences in ratio of general practitioners or family physicians to the general population, rates of hospitalization for ambulatory care sensitive conditions, often used as a marker of primary care performance in managing certain chronic conditions like diabetes and asthma, as well as a large spread in proportion of the population with a regular primary care provider compared with the Canadian provincial averages. 35 See Table 1 for characteristics (eg, population, health status) of the three study regions.
Six day-long deliberative dialogue sessions were held between January and May 2016, two in each of three regions. These sessions engaged people living in these different communities to explore whether common uses for primary care performance information would emerge across the regions.
Deliberative participatory methods, a well-established approach for engaging members of the public in complex topics, were chosen because they use structured facilitation to present balanced background information and enhance participants' understanding of the topic. Adequate time is given to foster learning, reflection and reasoned engagement. [36][37][38] We sought participants who had different levels of experience with the healthcare system in case experience impacted how people might use publicly reported performance information. Accordingly, one deliberation event in each region was for people with multiple medical conditions (Session 1) and the second event was for those with two or fewer medical conditions (Session 2). The study was approved by the institutional review boards of the University of British Columbia, the Nova Scotia Health Authority, the Ottawa Hospital, and Bruyère Continuing Care in Ottawa.

| Recruitment
Individuals aged 18 and older who participated in a waiting room patient experience survey at their primary care practice were also asked for their consent to be contacted for related research opportunities. Participants were primarily recruited from a convenience sample of these consenting survey participants, and we obtained their age and medical conditions from the waiting room survey data.
We had to recruit additional participants in Ontario (1) and Nova Scotia (9) through a study recruitment advertisement posted on a regional job and volunteer opportunity website to augment the number of people with no or one medical condition. We sought diversity in age, gender, types of chronic conditions, and the practice with which participants were affiliated. Participants had to speak English.
They were offered a $75 honorarium, meals during the event, and reimbursement for transportation costs. The second topic is the subject of this analysis and was informed by a rich discussion of the type of information which could be publicly reported to enable participants to contemplate uses for information they might not have considered before the dialogue sessions.

| Structure of the deliberation
The opening hour was spent reviewing the background material and orienting participants to the concepts of primary healthcare and performance measurement using interactive case discussions to both familiarize the participants with the concepts and help them identify and share their own experiences with and opinions on evaluating healthcare performance. Content experts in primary care performance measurement, patient experience, public performance reporting and deliberative dialogue methodology were present and involved in facilitating the discussion to ensure an interactive group discussion with opportunities for each participant to share their experience and learn.

| Analysis
Each deliberation session was recorded and transcribed. Observer notes were taken, and a team debrief was held after each session.
The transcripts were read in their entirety using crystallization and immersion methodology 39 by two team members experienced in qualitative data analysis (SJ, MH), one of whom attended each session and carried out data cleaning for any transcription errors (SJ). A coding template was inductively developed by the interdisciplinary analysis team (SJ, MH, JA, SW, JL), broadly informed by the study objectives and the initially emerging concepts from the in-depth transcript review. Two team members (SJ and MH) independently coded each of the transcripts. The two then compared coding of the entire transcripts to ensure agreement on all coded segments. Coded segments were then entered into NVivo qualitative data analysis software (v.10). 40 Coded segments were reviewed by the analysis team to identify shared themes (arising across all or multiple groups, supportive statements from separate participants), as well as conflicting statements, and unexpected information. Shared themes were also mapped to identify if any were specific to either the medically complex group of participants or the less medically complex groups. The team had extensive expertise in different substantive (performance measurement, patient experience, primary care), methodological (qualitative and deliberative methods) and clinical (family medicine, nursing) research areas. We carried out a thematic analysis across all deliberative dialogues as well as assessed for similarities and differences by region and for the two groups of patients (less complex and medically complex).

| RE SULTS
We involved 56 participants with a range of ages from 20 to 81 years old and no medical conditions to ten different conditions. (See Tables   2 and 3). We had a balance of men and women across the groups of each region except in Nova Scotia where all but two participants were women. Several shared themes highlighting how individuals might use publicly reported primary care performance information emerged across all three regions and six dialogue groups: to support collective health system decisions, to select providers, and to seek the best care for oneself or one's family. Additionally, some participants also noted a role for publicly reported primary care performance information in quality assurance and promoting trust.
That is, while the public may not use or engage with all performance information, it was reassuring to know that performance was being measured and monitored. No recurring themes were limited to or significantly more prominent within either the medically complex or less complex groups.

| Collective health system decisions
The most frequent theme was that publicly reported primary care performance information could be used to support collective health system decisions, strategic decisions about health services and policy.
The most common activity participants raised in relation to using performance information in each of the dialogue sessions was advocating for better primary care for their community, where they lived and expected to receive healthcare. As one participant noted, public information could be a precursor to advocacy; "I guess seeing something like this gives people the ability to start conversations when you go home…, which creates a social awareness of issues…

| Selection of providers
The use of performance information to select providers was raised in each region but with many fewer supportive statements than for collective health system decision involvement. Several participants in each region noted a significant role for primary care performance information when moving to a new area.
When the population is moving around, they're look- Participants also noted the potential of harming certain communities, such as rural areas.
… one of the things I worry a little bit about especially being part of a rural community where access usually is an issue because we have a very limited number of physicians… when we sort of almost pit one area against another is that what is the risk to the community of losing a physician who cannot perform better because perhaps they're the only one of two in an area, and can't physically take on more clients. [ON_1] The use of public performance information to compare and select a better provider was seen as potentially harming high-performing regions and poor-performing regions. Most participants finished the day stating they wanted to know that primary care performance information would be publicly reported.

| D ISCUSS I ON
The perspectives by participants in these three regions contribute to the sparse literature on how public performance reporting in primary care may influence the public's behaviour. Despite regional differences in primary care resources and rural and urban populations, similar potential uses for public performance information were raised in each region, as well as barriers or concerns over its use.

| The citizen role and public reporting
Our findings suggest that public performance reporting in primary care might promote the responsiveness of the health system through increased accountability and democratic participation, stimulating advocacy for high-quality community healthcare services. Participants believed that governments had the responsibility to ensure equitable access to quality primary care and could be responsive to public lobbying. This type of citizen-advocate engagement has been suggested as a mechanism for accountability in a publicly financed health system where most clinician-level costs are not borne by the individual, thus limiting the influence of market forces. 23 It is possible that the citizen-advocate role was more dominant in this study because participants were volunteers who chose to engage in a full-day session in their own region and were thus more likely than the general population to care about their community's healthcare services. There is little evidence on the public's actual or desired level of engagement in primary care advocacy for their community. However, a recent study surveying members of the public in Sweden and the UK found a stronger desire for participation in regional decision making for health services in Sweden (55%) compared to the UK (33%), suggesting that cultural and/or health system context may impact citizen health advocacy, including lower overall satisfaction with the health system in Sweden. 44 Across Canada, the publicly funded health system is seen as a valued national social service 32 and a civic entitlement. Individuals expect the public system to provide their needed care. However, Canadian seniors report being less satisfied than their peers in other countries on the overall performance of their health system. 45 Canadians also believe citizens should demand more accountability from the health system and advocate more for better health services. 32 Primary care performance reporting to the public which aims to promote citizen-advocate engagement as a mechanism for accountability could optimize public interest and use of information by addressing people's expectation for community-level civic engagement.
Current trends in performance reporting recognize the importance of incorporating local context, 46 and regional reporting approaches have been used in a number of countries such as in Australia's online primary care performance reports searchable by zipcode. 47 Reporting performance information at the community level could facilitate civic engagement by empowering members of the public to understand how their community is served and to advocate for healthcare solutions for their own neighbourhoods. Further, linking performance information with local accountable decision-makers, especially beyond government representatives, such as regional health administrators or primary care network leaders, might further support public engagement in provinces where decentralized decision making is being promoted.

| The consumer role and public reporting
Participants perceived lack of choice of primary care provider was a major barrier to people anticipating using comparative primary care performance information about providers in a consumer role to choose the best provider for their needs. This finding across the three regions was influenced by the shared perception that people had a limited choice of primary care providers despite almost all participants having their own primary care physician and significant differences in rates of family physicians per 100 000 population. This may explain why participants did not strongly support the notion that primary care performance information should be used to stimulate competition among providers. This finding contrasts with much of the efforts to promote public performance information in many countries including the UK 48 and the USA 49 which seek to empower consumers to choose the best primary care provider for themselves or their family.
Similarly, the perception of limited primary care resources may have generated concern about the potential unintended consequences of public reports stimulating "doctor shopping," and the potential for exacerbating inequities in access in regions.
Our findings also revealed an apparent tension between patient and consumer perspectives in using information on the performance of primary care. A consumer is expected to make informed choices in selecting care based on data from many patients' experiences.

| The patient experience and public reporting
Our findings also suggest that public reporting may contribute to better patient experience by promoting awareness of optimal care which people should expect, as well as by enhancing trust in clinicians and the overall health system. While continuing professional development activities have not been a major target of performance reporting, our findings suggest that this information may be empowering to patients, warranting further exploration. The finding that some participants also felt that performance information should be publicly reported to promote quality assurance suggests that some people might value public performance reporting to empower their trust rather than their decision making. This rationale for public reporting, however, relies on the data being used effectively by providers, professional organizations or quality assurance groups. This suggests that primary care performance measurement and reporting efforts should focus on building capacity to use such data for quality improvement as well as for accountability. This is a key component of "intelligent transparency" driving renewal for public reporting in the UK. 52 It would also mean that providers should be considered a target audience for public reporting alongside their patients and the public.

| CON CLUS ION
Public reporting on primary care performance to promote accountability, democratic participation and quality is receiving increasing attention as evidence continues to grow on its importance to population health, health system efficiency and individuals' experience within the healthcare system. Our findings suggest that members of the public may value and use public performance reporting to assess equity across the system in how their region is served and to advocate for their own community. Encouraging the public to utilize primary care performance reports to select optimal providers for themselves or ensure best care with their providers may require education not just on how to understand the data but also on the significance of the measures. Increasing the public's engagement in primary care as citizens, consumers and patients demands investment in effective measurement and reporting systems that enable people to effectively engage with this information, how and when they are most likely to use it. In Canada, public performance reporting to empower the public to advocate for their own communities' primary care may be a particularly meaningful accountability mechanism to promote responsive healthcare.

ACK N OWLED G EM ENTS
The authors would like to acknowledge Stephanie Blackman's invaluable help in reviewing and editing the manuscript.

CO N FLI C T O F I NTE R E S T
None declared.

DATA ACCE SS I B I LIT Y
The data that support the findings of this study are available from the corresponding author upon reasonable request.