The voice of patients in system redesign: A case study of redesigning a centralized system for intake of referrals from primary care to rheumatologists for patients with suspected rheumatoid arthritis

Abstract Background The published literature demands examples of health‐care systems designed with the active engagement of patients to explore the application of this complex phenomenon in practice. Methods This case study explored how the voice of patients was incorporated into the process of redesigning an element of the health‐care system, a centralized system for intake of referrals from primary care to rheumatologists for patients with suspected rheumatoid arthritis (RA)—centralized intake. The phenomenon of patient engagement using “patient and community engagement researchers” (PaCERs) in research and the process of redesigning centralized intake were selected as the case. In‐depth evaluation of the case was undertaken through the triangulation of findings from the document review and participants’ reflection on the case. Results In this case, patients and PaCERs participated in multiple activities including an initial meeting of key stakeholders to develop the project vision; a patient‐to‐patient PaCERs study to gather perspectives of patients with RA on the challenges they face in accessing and navigating the health‐care system, and what they see as key elements of an effective system that would be responsive to their needs; the development of an evaluation framework for future centralized intake; and the choice of candidate centralized intake strategies to be evaluated. Conclusions The described feasible multistep approach to active patient engagement in health‐care system redesign contributes to an understanding of the application of this complex phenomenon in practice. Therefore, the manuscript serves as one more step towards a patient‐centred health‐care system that is redesigned with active patient engagement.


| INTRODUC TI ON
During the last decades, health-care organizations around the world have been actively advocating for patient-centred care as means of ensuring high-quality care. [1][2][3][4] Patient-centred care calls for a more holistic approach to care delivery that is focused on patients' needs and experiences of well-being and illness from a multi dimensional biopsychosocial perspective. [5][6][7][8] To achieve patient-centred care, fundamental changes, including a redesign of existing systems and/ or design of new ones, are required. 9 Integrating patients (ie, patient involvement 10 ) into research and system design has been recognized as important elements in achieving patient-centred care. 11 Patient involvement offers the potential to target research and system design to patients' needs, thus improving the patient experience with care and quality of care, and potentially reduce costs of care. [12][13][14] Despite decades of discussions about the importance and potential benefits of patient involvement in health planning, research and system design, to date, patient involvement in designing and redesigning health-care systems has often been limited to passive involvement. 10,15,16 Few examples where patients and other stakeholders have been actively engaged as partners to design and redesign the system are available in the literature. 12,17 Therefore, more examples of active engagement of patients and other stakeholders in the system redesign are needed to explore the application of this complex phenomenon in practice. 10,15,16,18 This manuscript reports on a case study of patient involvement in redesigning a centralized system for intake of referrals from primary care to rheumatologists for patients with suspected rheumatoid arthritis (RA), hereafter referred to as centralized intake (CI).

| ME THODS
The study followed the case study design approach as described by Yin. 19 In this study, the phenomenon of patient engagement using "patient and community engagement researchers" (PaCERs) in research and the process of redesigning CI were selected as the case. 19 PaCERs are citizens living with various health conditions who received formal research training that includes how to design research, engage other patients and conduct research projects using an established protocol of qualitative inquiry. [20][21][22] The case took place within the unique context of a 2-year-long research project, which aimed at "Optimizing Centralized Intake to Improve Arthritis Care," hereafter referred as the project (Figure 1). The case study aimed to explore how the voice of patients was incorporated into the process of redesigning an element of the healthcare system, CI, within the project. The case study research team consisted of multiple stakeholders (PaCERs, academic researchers, health-care professionals and health-care administrators), who were engaged in the project. The study was driven by two pre defined theoretical propositions 19 : (a) health-care systems should be responsive to patients' needs, 2 and (b) active patient engagement (ie, patient involvement 10 ) in system design and redesign is required to build a system that is responsive to patient needs. 11 A detailed reporting on each activity within the project (eg, objectives, participants, actions, decisions and results) was undertaken by the project manager (JP) and academic researcher leading the project (DAM) to facilitate the case study. All documents that reported on the case and the context of the case were reviewed by the project manager (JP) and two academic researchers (DAM, for Albertans"; The Arthritis Society, Models of Care Catalyst grant: "Creating an optimal model of care for the efficient delivery of appropriate and effective arthritis care"; Canadian Institute for Health Research (CIHR) Planning Grant (funded through Priority Announcement Health Services and Policy Research): "Evidence based planning of an optimal triaging strategy for arthritis care in Alberta"; Alberta Health Services, Research Grant (through the Bone and Joint Health Strategic Clinical Network): "Optimizing Centralized Intake to Improve Arthritis Care for Albertans." research and the process of redesigning centralized intake were selected as the case.
In-depth evaluation of the case was undertaken through the triangulation of findings from the document review and participants' reflection on the case.

Results:
In this case, patients and PaCERs participated in multiple activities including an initial meeting of key stakeholders to develop the project vision; a patient-topatient PaCERs study to gather perspectives of patients with RA on the challenges they face in accessing and navigating the health-care system, and what they see as key elements of an effective system that would be responsive to their needs; the development of an evaluation framework for future centralized intake; and the choice of candidate centralized intake strategies to be evaluated.

Conclusions:
The described feasible multi step approach to active patient engagement in health-care system redesign contributes to an understanding of the application of this complex phenomenon in practice. Therefore, the manuscript serves as one more step towards a patient-centred health-care system that is redesigned with active patient engagement.

K E Y W O R D S
health system redesign, patient engagement in research and system redesign, patient needs, patient-centred care, patient-to-patient research EL) with expertise in qualitative and mixed-methods research to extract data on objectives of patient and PaCERs engagement in redesigning CI, roles of patients and PaCERs in the project, and outcomes of their engagement. An inductive narrative analysis of the extracted data was conducted by three academic researchers (DAM, EL, NM) with expertise in qualitative and mixed-methods research and two PaCERs (JLM, SRT) to summarize the in-depth description of the case and preliminary findings of the case study. 19,23 Next, other members of the team reflected on the preliminary findings and refined them from the perspective of their personal experience with participating in the project and observation of the case. Lastly, preliminary findings from the document review and the team's reflection were triangulated through team discussions to generate the final findings.
F I G U R E 1 Environment and context for the case of patient and PaCERs engagement in the "Optimizing Centralized Intake to Improve Arthritis Care" project. Sources: Badley,42 Goekoop-Ruiterman et al, 43 Pope et al, 44 Hazlewood et al, 28 Damani et al, 45 Lopatina et al, 46 Alberta Innovates, 47 Alberta Health Services, 48 Government of Canada, 49 Alberta Health Services 50 and Noseworthy et al 51 Potential solution -a redesigned centralized system for intake of referrals from primary care to rheumatologists for patients with suspected RA (CI): • CI is a system facilitating "a single point-of-entry through which referrals are received and service provision is arranged" [45][46]. • Centralized intake along with triage of referrals, pooled referrals and pooled service provision are the core components of singleentry models, which are an approach to reducing waiting times and providing timely access to appropriate care on a prioritized basis for a large referral population [45][46].
The "Optimizing Centralized Intake to Improve Arthritis Care" research project (the project): • The objective was to redesign, implement and evaluate an evidence-based approach to optimize CI systems for referrals of patients with suspected inflammatory arthritis, such as RA, and noninflammatory arthritis, such as osteoarthritis, from primary care to specialists province-wide, enabling early access and comprehensive multidisciplinary assessment for patients. • The project was funded through several sources, including the Partnership for Research and Innovation in the Health System (PRIHS), a unique funding initiative, which supports networks of health researchers and clinical practitioners, health-care administrators across the continuum of care aiming to improving health outcomes for patients across Alberta, Canada. PRIHS was developed as a partnership between a funding agency, Alberta Innovates [47], and a province-wide health-care delivery system, Alberta Health Services [48].

Unmet patient need:
• Rheumatoid arthritis (RA) is a type of arthritis that causes inflammation in joints, particularly in small joints of the hands and feet, and results in pain, swelling and stiffness [42]. • Early recognition and intervention for RA prevents or minimizes permanent, irreparable joint damage, which results in functional impairment [43][44]. • Ensuring timely access to appropriate and effective arthritis care is the first step in moving upstream to stave off the deleterious progressive effects of the disease. • In Alberta, Canada, a long wait time for a rheumatology assessment has been reported [28].
Local health-care system: • Canada has a national publicly funded health insurance system with the universal coverage for medically necessary health-care services [49]. The health-care system is predominantly administered on a provincial or territorial basis. • The province of Alberta has one province-wide health-care delivery system, Alberta Health Services, who deliver care to over 4.2 million Albertans [48]. • Strategic Clinical Networks™ were designed and introduced by the provincial health-care system as a tool for the implementation of evidence informed improvements in the health-care system [50,51]. • Strategic Clinical Networks™ represent networks of collaboratively working decision makers, health-care administrators, clinicians, researchers and patients with the mandate "to find new and innovative ways of delivering care that will provide better quality, better outcomes and better value" [50,51].
• Strong participation from patients and families is one of the key features of the work of Strategic Clinical Networks™ [50,51]. Strategic Clinical Networks™ support multiple parallel activities of PaCERs in the policy development arena.

Case context Case context
The research was approved by the University of Calgary Conjoint Health Research Ethics Board (Ethics ID number REB13-0822).

| RE SULTS
Patient and community engagement researchers were engaged in the project from the outset as equal partners with the rest of the project team members. This involved setting up the research agenda, helping to develop the funding application and applying for funding.
During the project execution, patients and PaCERs participated in multiple activities described in detail here below and presented in chronological order.

| First stakeholder meeting
At the beginning of the project execution, the project team en- 4. Ensure patients are triaged and referred to appropriate care providers based on "best practice" to achieve desired outcomes; 5. Ensure resources are optimally used in achieving desired outcomes; 6. Mitigate risks to avoid unintended or harmful results.
Once established, these targets shaped and guided subsequent project phases. In particular, the patient-centred nature of the majority of these principles highlighted the demand for further input from patients into the processes of redesign, implementation and evaluation of CI to ensure that the future system is responsive to patients' needs.

| PaCERs study
Next, PaCERs conducted a study to gather perspectives of patients with RA on the challenges they face in accessing and navigating the health-care system, and what they see as key elements of an effective system that would be responsive to their needs. Although the focus of this work was on CI, the PaCERs study explored patients' perspectives on the entire care pathway. This was done to account for the complexity of the health-care system and possible interactions between the system's elements, as CI is just one element of the care pathway for patients. The study was led by two PaCERs who are patients living with osteoarthritis with previous experience in research on care delivery for patients with musculoskeletal conditions (JLM, SRT) and a research assistant who is a patient living with RA.
Participants who self-identified as having RA were recruited through arthritis networks, posters in rheumatology clinics and rheumatologists' referrals. Those interested in the study contacted PaCERs, who sent potential participants the study description and a consent form. Next, completed consent forms were sent back to PaCERs.
The PaCERs study went through three phases: set, collect and reflect as outlined in the PaCERs methodology. 20,21 Given the iterative nature of the research, patients were recruited and data were collected until data saturation was reached at each phase of the study. Over the three phases, 15 patients were included (Table 1). factors as getting diagnosed, accessing a rheumatologist when medications need adjusting, communication between primary care providers and other providers, and maintaining patients' mental well-being.

| Set phase
Throughout the focus group, participants' points were documented on flip charts. Subsequently, PaCERs used focus group notes to develop a semi structured interview guide for the collect phase.

| Collect phase
In telephone interviews (n = 11), participants were asked about  24 PaCERs then compared and contrasted their findings and, using a collaborative, iterative process, identified a preliminary set of themes and subthemes. PaCERs also identified challenges that patients faced when accessing and navigating the system and key elements of a system that patients reported would be responsive to patients' needs. Next, preliminary findings were compiled and presented for discussion during the reflect focus groups.

| Reflect phase
There were three reflect focus groups that included a total of 10 participants: two for participants living in urban areas and one for participants from rural areas. The reflect focus groups took place locations neutral and convenient for all attending (eg, meeting room at a church or community centre) and lasted between 3 and 5 hours.
PaCERs presented preliminary findings of the collect interviews.
Then, PaCERs asked the focus group participants to reflect on the extent to which each of the identified challenges and key elements of the system that would be responsive to patients' needs fit with what had previously been discussed during the collect phase. This was followed by a discussion on centralized systems for intake of referrals and the potential role of CI in addressing the key elements of the system that would be responsive to patients' needs. The points discussed were documented on flip charts. Through a collaborative, iterative process, PaCERs examined the reflect focus group data to refine the preliminary findings and to develop the final findings. 24 As a result, five themes were identified: (a) initial access to rheumatology care; (b) on-going access to rheumatology care; (c) information about RA and resources for those living with RA; (d) fear of the future; and (e) collaborative and continuous care ( Table 2). The final set of challenges that patients face when accessing and navigating the system, and key elements of the system that would be responsive to patients' needs, were mapped to the corresponding themes (Table 3).

| Interpretation of findings
Findings of the PaCERs study suggest that patient-centred care for patients with RA should be viewed as a continuum. That continuum starts from the patients' first point of contact with the health-care system, carries through their initial appointment with a specialist and continues throughout their long-term and on-going follow-up visits with their rheumatology care team. To be responsive to patients' needs, the continuum of care should be easy to access and navigate. It should also provide patients easy access to information, education and community resources; as well as incorporate a communication infrastructure to promote collaboration among care providers. Participants reported experiencing multiple challenges when accessing and navigating the system. Among those challenges, initial access and on-going access to care were raised as the two challenges that were of primary concern to participants.
The main goal of CI, as one element of the care pathway for patients with RA, is to ensure timely initial access to appropriate care.
As such, participants found that CI had a potential to vastly improve patients' experiences with care, as well as their outcomes, and should be considered when redesigning the system to be respon-  Positive experience with the initial access to rheumatology care • One participant described her experience of getting diagnosed as "a positively deviant case." Participant's symptoms developed suddenly with painful swelling in most joints. She went to her family doctor's clinic, was sent for blood work and

| Application of the PaCERs study findings in the project
x-rays right away and started on medication to manage the inflammation and pain while waiting for results. The test results indicated inflammation, and her family doctor got a telephone consult with a rheumatologist. This was on a Friday. The rheumatologist said she would see her Monday morning and suggested that if she could stand the pain over the weekend, she not start on steroids, which would mask her symptoms, as she and her doctor had planned. The participant agreed to this. On Monday, she was examined by a rheumatologist and diagnosed with RA. On Tuesday, she attended a class on RA medications taught by a pharmacist, and by Wednesday, the swelling and pain were under control. By Thursday, she had a follow-up education session with the clinic nurse and started her medications. Since then, she has been followed regularly by both her family doctor and her rheumatologist and has good access to clinic staff if questions concerning treatment arise Suggestions for improving patients' experience with the initial access to rheumatology care • Participants thought that awareness of family doctors of RA could be higher • Participants believed that increased awareness of family doctors of RA would facilitate earlier referral to rheumatologists • During the reflection phase, the participant, who had a positive experience in the initial access to rheumatology care, thought this "positively deviant experience" happened for a number of reasons. First, she recognized something fairly significant was happening to her, and when symptoms persisted, she made an appointment at her family practice clinic. Second, her family doctor managed her care well-she had been his patient for a long time and knew she "didn't come there lightly." Third, on the follow-up appointment about her first blood test results her doctor phoned the rheumatologist on call and together they started to figure out what to do next. In addition, as a patient she is comfortable in navigating the health-care system, is not afraid to ask questions and is a strong believer in involving others in care decisions (Continues)

Subthemes
Examples of participants' experience when accessing and navigating the health-care system for management of RA 2. On-going access to rheumatology care Challenges in accessing rheumatologists in case of flare or problems with medications • All participants were aware of the need for on-going monitoring of their disease activity and medical management • One participant who was on her third biologic said: "all of a sudden it was a wonder drug, and as quickly as it started to work it stopped" • All participants mentioned they constantly worried that RA might flare up or medications might stop working. Patients worried that they would not be able to access the specialty clinic in case of such emergency Direct contact to rheumatology care team • Some participants reported to have direct telephone numbers or email addresses for the rheumatologists or nurses from the rheumatology team • One participant thought being able to talk to a nurse practitioner is "brilliant, wonderful" as it provided patients access to the resources they need to keep their disease in control • One participant had access to the pharmacist whose class she attended, describing this person as being "amazingly available," responding to emails within hours: "I feel I have access, I don't have to wait for my 3-month appointment to get access to the people who will answer my questions" • In one situation, no one at the rheumatologist's office answered the phone so the participant had to go to emergency several times in 3 mo where she saw a rheumatologist on call Suggestions for improving patients' experience with the on-going access to rheumatology care • Having a direct contact with professionals with RA expertise was the preferred option for the participants to ensure on-going access to care 3. Information about RA and resources for those living with RA Lack of information about RA and resources for those living with RA • Participants reported receiving little if any information about RA when they were first diagnosed • One patient described it as: "no nurse, no doctor, no physio, nobody has given me anything unless I find it myself" • One participant from a remote community said she was "really in the dark" because she received no information from her rheumatologist, and there were no resources in her community • Another participant described this as being left "floating in a big ocean all by myself with only a small bit of foam to stay afloat" Patients' education is a professional responsibility of health-care providers • One participant said that all newly diagnosed patients should receive a "whole package of education materials, and proper websites." She believed that once a patient was diagnosed "a bunch of doors should start opening for you"; that patients you should be referred to physiotherapy and should be "pushed to go to certain classes" • When some focus group participants mentioned had attended a medication class at one of the clinics they attended, the rest of participants started to wonder why their rheumatologists had not told them about such classes. As such, most participants concluded that RA specialists should be more aware of available resources Positive experience with education sessions for patients with RA • One participant said education sessions for people with RA taught her "how to manage pretty much any situation that can arise" Need for peer support and lack of peer support programs for those living with RA • Participants spoke of having few people to talk to about their RA and the need for peer support • Participants described hiding their RA from their friends, family and coworkers because they did not want to be seen as different • One participant said she told her friends that her bent finger was from a sports injury • Participants shared that they often held back on developing relationships and one participant said that as the disease progresses she "shuts the world out" • One person said that at one time she had been addicted to pain medications so now she would not talk to her husband about her pain as he worried she would get addicted again • Those who are in the workforce did not talk about their RA with their employers and colleagues, as they did not want to be passed up for promotions or potentially lose their jobs • An older patient who had a heart problem and diabetes as well as RA was followed up by a family doctor, a cardiologist, an endocrinologist and a rheumatologist. In her experience, each specialist is only willing to deal with his or her own area, rather than treat her as a complex patient with multiple comorbidities. She thought that all specialists involved in her care were good, but because they did not communicate with each other, she was "lost in the shuffle"

| KPIs
The set of KPIs (ie, quantifiable measures of quality of care) to evaluate CI was developed through a multistep process described in detail elsewhere. 25 Out of the final set of KPIs, 25 four KPIs addressed the identified during the PaCERs study theme of "initial access to rheumatology care," two-the theme of "on-going access to rheumatology care," one-the theme of "lack of information about RA and resources for those living with RA," and another was focused on patient experience with CI in general ( Table 3).
The project then used a multicriteria decision analysis (MCDA) process 26 to develop an aggregate performance measure for CI. 27 In the MCDA process, the KPIs identified that aligned with the themes identified during the PaCERs study were found to be of most importance (ie, were ranked higher) by all stakeholders (patients and PaCERs [n = 2], health-care professionals [n = 10], health-care administrators [n = 7] and academic researchers [n = 9]). 27 The KPI focused on patient experience with CI was ranked as the top KPI. 27 As part of the project, KPIs were developed to evaluate the quality of care TA B L E 3 Key elements of the health-care system that would be responsive to patients' needs aligned with corresponding themes identified through the patient and community engagement researchers (PaCERs) study TA B L E 4 Key performance indicators (KPIs) 25 and statements in the patient experience survey aligned with the corresponding themes identified through the patient and community engagement research (PaCERs) study delivered by an existing centralized intake and triage in rheumatology system in Calgary, Alberta, and to identify the system's gaps. 28 Next, KPIs will be used within the implemented province-wide redesigned CI to measure the impact of the change on quality of care.

| Patient experience survey
To address and measure the KPI about patient experience, the team developed a patient experience survey. 29 The survey development . 29 This process resulted in the selection of a set of 23 questions (Appendix 1). The patient experience survey will be administered to patients with suspected RA referred to rheumatology clinics through the current centralized intake and triage in rheumatology system in Calgary, Alberta. 28 These data will be compared to the patients' experience after the system optimization. for an optimal, CI identified challenges to accessing and navigating the system, and the key elements of the system that would be responsive to patients' needs. All discussion points were recorded by a scribe. The meeting notes were reviewed and analysed to select the final set of models to be evaluated.

| Future plans
Once the candidate strategies are selected, they will be tested using simulation models 34 delineating the operational features and describing the clinical pathway and the flow of patients. Each strategy will be tested for its ability to adjust for variation in the type of treatment needed by patients and the availability of various care providers and facilities to provide the different services to identify the strategy that would be most efficient and effective in directing patients to appropriate care providers, thus achieving improved patient and system outcomes (ie, optimal strategy). Finally, based on the feasibility and readiness of the clinics, there will be an opportunity to implement and evaluate the identified optimal strategy. According to this framework, our approach covers the higher end of the continuum of engagement within the level of organizational design and governance. 35 Next, according to the framework for patient and service user engagement in research, 16 our approach includes all components of patient and public involvement in research, such as patient and service user initiation, building reciprocal relationships, colearning process and re-assessment and feedback, throughout both preparatory, execution and translational phases of the project.

| D ISCUSS I ON
Limited feasibility of approaches that cover the higher end of the continuum of patient engagement and/or include all components of patient and public involvement in research has been discussed as a barrier to their application in practice. 16,35 The current manuscript presents an example of a successful application of such an approach.
Our findings should be considered within the context of the single-case study design and qualitative analysis. First, the discussed approach to the patient engagement in system redesign has emerged from a unique single case, which limits the generalizability of our findings. 19 This case study represents one approach to patient engagement in system redesign, which may not fit every question, system settings and clinical area. Nonetheless, the holistic approach of the single-case study design facilitated an in-depth description of the case and its context, allowing the reader to make conclusions regarding the feasibility of the approach in the local environment. 19 Second, the elements of the system that would be responsive to the needs of patients with RA, as well as the suggestions as to how CI can help to meet those needs, were developed through a qualitative PaCERs study with a relatively homogenous group of patients with RA. To reduce the potential for selection, coding and interpretation biases, 36 all steps of the PaCERs study were conducted using sound methodology by trained patient engagement researchers experienced in qualitative research, and data were collected until saturation was reached.
All coding, content analysis and interpretation were done by at least two PaCERs, reviewed by the team and finalized when the consensus was achieved. Lastly, although our findings on challenges experienced by patients with RA align with the published literature, 28,[37][38][39][40][41] the identified key elements of the system that would be responsive to needs of patients with RA, and suggestions as to how CI can help to meet those needs, might be specific to a universal system with a single public payer as in Alberta, Canada. Therefore, these key elements should be carefully considered within specific local environment.

| CON CLUS IONS
Overall, this study presents a feasible multistep approach to patient engagement in health-care system redesign. This manuscript contributes towards the understanding of the complex phenomenon of patient engagement and serves as one more step towards a patientcentred system that is redesigned with active patient engagement.

ACK N OWLED G EM ENTS
Authors would like to thank all participants of the patient and community engagement researchers (PaCERs) study and everyone who has been involved in the "Optimizing Centralized Intake to Improve Arthritis Care" project.

CO N FLI C T O F I NTE R E S T
The