Patient perspectives on quality of care for depression and anxiety in primary health care teams: A qualitative study

Abstract Background Widespread policy reforms in Canada, the United States and elsewhere over the last two decades strengthened team models of primary care by bringing together family physicians and nurse practitioners with a range of mental health and other interdisciplinary providers. Understanding how patients with depression and anxiety experience newer team‐based models of care delivery is essential to explore whether the intended impact of these reforms is achieved, identify gaps that remain and provide direction on strengthening the quality of mental health care. Objective The main study objective was to understand patients’ perspectives on the quality of care that they received for anxiety and depression in primary care teams. Methods This was a qualitative study, informed by constructivist grounded theory. We conducted focus groups and individual interviews with primary care patients about their experiences with mental health care. Focus groups and individual interviews were recorded and transcribed verbatim. Grounded theory guided an inductive analysis of the data. Results Forty patients participated in the study: 31 participated in one of four focus groups, and nine completed an individual interview. Participants in our study described their experiences with mental health care across four themes: accessibility, technical care, trusting relationships and meeting diverse needs. Conclusion Greater attention by policymakers is needed to strengthen integrated collaborative practices in primary care so that patients have similar access to mental health services across different primary care practices, and smoother continuity of care across sectors. The research team is comprised of individuals with lived experience of mental health who have participated in all aspects of the research process.


| INTRODUC TI ON
In most countries, primary care is the first point of contact in the health-care system for many individuals with mental health problems. 1 Increasingly, interprofessional primary care teams are optimally positioned to address specific mental health needs of patients along with other physical and/ behavioural needs. 2,3 Given the high prevalence of common mental disorders (CMDs)-such as anxiety and depression-in primary care 1 and the challenges of clinical management, there is substantial benefit from the collaboration between health and mental health professionals who can work together as a team. [4][5][6][7][8][9] The Patient Medical Home (PMH) is a team-based model of primary care that has continued to accelerate over the last two decades in Canada, the United States and elsewhere. [10][11][12][13] In PMHs, family physicians work in tandem with interprofessional teams to provide continuous and coordinated person-centred care. [10][11][12][13] The PMH is an optimal model for the integration of effective high-quality mental health care. 11 In Ontario, Canada, the implementation of a PMH model of primary care had consequences for people living with mental health problems. 12 This includes reforms beginning in the early 2000s involving a shift away from fee-for-service-based physician remuneration to a capitation-based system, and the expansion of collaborative team-based care mainly through the creation of Family Health Teams (FHTs). 12,13 FHTs bring family physicians together with nurses, nurse practitioners, social workers, mental health counsellors, pharmacists, dieticians, consulting psychiatrists and other health-care professionals. 12,13 Currently, there are 186 FHTs serving approximately three million Ontarians (22% of the provincial population). 12,13 FHTs were established with intention to improve access to comprehensive person-centred care, and improve continuity of care with other parts of the health-care system. [10][11][12][13][14][15] FHTs had also anticipated benefits for patients requiring mental health services for CMDs. 2 The integration of teams in primary care is advantageous for patients who can then access a range of physical and mental health services in one location, 2,16 with shorter wait times for mental health services than traditional care settings. 2,[17][18][19] Embedding mental health services in the same organization where patients see their family physician-someone with whom they have a long-standing trusting relationship-may help reduce stigmatization. 2 By reducing stigmatization, patients may be more willing to seek out mental health services when needed, especially when the mental health provider is someone who works in tandem with their family physician. 2,20,21 FHTs host a number of mental health practitioners, including social workers (92% of FHTs), psychologists (25%) and other mental health workers (13%). 22 Combining these providers and others, with family physicians, improves prevention and enhances identification, early intervention and treatment while improving patient experiences. 2,[22][23][24][25] Despite the increasing emphasis of primary care teams, we know little about the impact of this team structure on patient experiences with mental health care. 2 There is overwhelming evidence that knowledge about approaches to delivering effective mental health services is not consistently translated into action 26 with many people continuing to struggle with unmet mental health needs. [14][15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32] It is essential to examine the impact of reforms intended to improve mental healthcare delivery from the perspective of the service recipient to inform measures that can drive services that are most meaningful to patients. 27,33 Understanding how patients experience new models of care delivery is essential to evaluate whether the intended impact of these reforms is achieved, identify gaps that remain and determine whether the new approach resulted in any unintended consequences. 26,33 Our main study objective was to understand patients' perspectives on the quality of care that they received for CMDs from Ontario's FHTs.

| Study design
We used constructivist grounded theory to guide sampling, data collection and data analyses. 34,35 Constructivist grounded theory views knowledge as socially constructed, and emphasizes research that recognizes multiple viewpoints and interpretive understandings. [34][35][36][37][38] It also acknowledges the subjectivity of researchers whereby their assumptions are considered to be valuable for shaping data collection and analysis. 34,36 Team members involved in this study had different clinical or disciplinary backgrounds spanning: social work, psychiatry, mental health research, epidemiology and primary care health services delivery research.
Four members of the team had experience as advisors to provincial policy and decision-makers. Integral to our team were two individuals with lived experience of CMDs who contributed to the research process, interview questions and interpretation of results.
Sensitizing concepts acted as a starting point help to inform the research process, and provide a way of understanding and organizing participant experience. 37,38,39 The sensitizing concepts were derived through a review of the literature, from lived experience of team members and from our previous research. 27,[40][41][42][43] The data that support the findings of this study are not shared in a public repository. This study received Research Ethics Board Approval from the University of Toronto (REB#35131).

| Sampling and recruitment
We sought to engage a diversity of perspectives of people receiving care from FHTs within three different geographical regions of Ontario: Toronto Central, Central East and South East. There were 38 FHTs operating within these three regions, and these varied in terms of team size, provider composition and geographical characteristics. We selected these regions for three reasons: i) regional variation in terms of rural and urban; ii) varying diversity of populations in these regions; and, iii) existing relationships in our team with many FHTs in these regions from a previous study. 38 Potential participants self-identified as residing in one of these three regions, After completing the fourth focus group and analysis of the data, we ceased recruitment because theoretical saturation had been

| Data collection
We developed a semi-structured interview guide using sensitizing concepts. We then collected data using in-person focus groups. We chose to use focus groups for data collection because of the deep understanding of patients' perspectives that can emerge from the dynamic nature of focus groups. 44,45 Focus groups also help generate diverse views and experiences. 46 We offered individual interviews to those unable to attend a focus group. Two team members co-facilitated the focus groups and conducted the individual interviews between August 2018 and March 2019. We audio-recorded focus group and individual interviews, and transcribed verbatim immediately following the interview. We randomly assigned a code to each participant for anonymity.

| Data analysis
Analysis began immediately following the transcription of each focus group or interview. Data collection and data analysis occurred simultaneously, resulting in an iterative analysis approach. 45 The iterative analysis approach helped to inform the on-going refinement of our interview guide. For example, there were little data in the early interviews and first focus group related to diversity and quality of care, so we added a question and probes about diversity for later interviews and the three latter focus groups.
Grounded theory informed initial line-by-line coding, followed by focused and axial coding. 34,45 Two team members parallel-coded transcripts until they reached consensus in the coding process, after which one member was the primary, and the other was the secondary coder. A data analysis subcommittee met regularly to help interpret the data, discuss emerging findings, inform new coding and update the interview guide based on the emerging findings. A final interpretation of findings included all research team members. We conducted the data analysis inductively. We identified exemplar quotes as analysis proceeded. It was through prolonged engagement, reflexivity and peer debriefing that we established rigour and trustworthiness. 45,47,48 We used NVivo 11 to help organize the data analysis process.

| Sample
Forty patients participated in the study: 31 participated in one of four focus groups, and nine completed an individual interview ( Table 1). The participants in our sample varied by age, ethnicity, gender and geographical location (Tables 2 and 3).
With respect to the care experiences of patients with CMDs and areas of improvement, the following four themes emerged: "accessibility, technical care, trusting relationships and meeting diverse needs".

| Accessibility
There was consensus across all focus groups and individual interviews about the importance of accessible care. Access emerged early in all focus groups and individual interviews, with robust discussion in focus groups about the benefits of embedding mental health professionals in primary care. With mental health professionals integrated into primary care, participants were easily able to access a range multiple mental health services: 'I've seen the therapist…a psychiatrist, psychologist, and then obviously my family doctor' (I1).
All focus groups agreed that the team-based approach enhanced ac-

| Medication
There was consensus across all focus groups and many individual interviews, about the important role that medication had at one time or another in recovery. 'The first step she did when I recognized anxiety is a problem…she's connected me with medication' (FG4, P7).

| Trusting relationships
There was overwhelming agreement within and across all focus groups and individual interviews that trusting relationships with pro-

| D ISCUSS I ON
Understanding the experiences of patients is essential in order to identify how to improve care in a way that is meaningful to patients. 49 Quality of care is informed by structures-the organization of care-and clinical processes-how care is delivered by providers. 50,51 Structure is comprised of the physical characteristics of the organization and the staff, whereas processes are the technical care interventions appropriate to the condition and the interpersonal relational interactions that occur between patients and members of the health-care system. 50,51 Although all participants in our study spoke about their experiences with the processes of care, it was only through the focus groups that topics related to structure emerged. Participants in our study described their experiences with mental health care across four themes: accessibility, technical care, trusting relationships and meeting diverse needs.
Of utmost importance to our participants was the accessibility of mental health services in primary care for their own recovery. 52 Integrating mental health providers in the same location as their family physician made a profound difference in the convenience and ease to get care for CMDs when needed. Importantly, our study demonstrates that primary care teams facilitate access to mental health because by reducing patients' fears of being stigmatized relative to attending organizations known to deliver mental health services exclusively. Stigma limits access to health services. [53][54][55] Despite expressing how meaningful FHTs are for accessing mental health care, participants raised concerns about the availability 52 of some services for CMDs. 56 In particular, participants expressed concerns about capping the number of appointments for psychotherapy and reported that in some cases, the maximum cap was disruptive to their recovery. Additionally, participants noted that there was a lack of available psychiatrists working in collaborative models. This is consistent with reported trends demonstrating the limited availability of psychiatrists, particularly for patients who reside outside of urban areas. 57 Consistent with what we heard from our participants, primary care physicians are often the first point of contact for patients who experience mental health difficulties. 2 Overwhelmingly, participants wanted primary care providers to initiate discussions about mental health. Participants in our study went so far as to suggest that screening of CMDs should be implemented in primary care, which is consistent with literature on patient preferences. 58 Patients find relief having an answer for why something is happening. The Patient Health Questionnaire (PHQ-9) is one example of a screening tool for depression, although there is on-going debate about whether or not routine screening for CMDs should be implemented. 59,61 Undetected and untreated CMDs have measurable and actionable impact on numerous illnesses routinely treated in primary care, such as diabetes for instance. 59 Integrating mental health services in primary care teams provides a direct pathway from screening to treatment. Our study participants agree with recommendations supporting universal depression screening. 60 Following our participants' recommendations, validated tools such as the PHQ-9 can be used to evaluate and monitor progress, which aligns with measurement- Person-centredness is a core quality of care dimension for PMHs such as FHTs, which focuses on patients' experiences with whole-person care, therapeutic relationship and communication. [10][11][12][13][14]

| Limitations
We conducted this study with patients of one model of team-based primary care in Ontario, so findings in our study may not apply to all primary care settings. Additionally, our sample did not permit us to gain an in-depth understanding of the unique experiences of young adults or racialized patients. While focus groups shared some insights about the need for culturally appropriate, relevant and congruent care, we did not probe patients about their perspectives related to racism or sexism in care. Additionally, factors that are not always explicitly mentioned shape people's perception of the quality of care they received, notably experiences of trauma. We also did not probe for examples of care that is more or less trauma-informed, though care perceived to be of better quality might be more traumainformed, and therefore better meeting peoples' needs.

| CON CLUS ION
Integrating mental health services in primary care teams has enhanced quality of care for CMDs, namely by improving accessibility and technical care. Team-based primary care is an optimal location for mental health services because it is has a foundation of continuity and relationships, which is of utmost importance to patients. Greater attention by policymakers is needed, however, to strengthen integrated primary care so that patients have similar access to mental health services across different primary care practices.

| PATIENT CONTRIBUTI ON
Service users and individuals with lived experience of depression and anxiety were highly involved in our study. Two individuals with lived experience acted as advisors and participated as co-researchers in all aspects of the research process including developing interview questions, interpretation of results and preparation of manuscript, and are included as co-authors on this paper. Additionally, two research assistants responsible for data collection also self-identified as having lived experience with mental health. All members of our research team have participated in the development of this manuscript.

ACK N OWLED G EM ENTS
We wish to thank each participant who took the time to meet with us to share their experiences.

CO N FLI C T O F I NTE R E S T
No conflict of interest for any author.

DATA AVA I L A B I L I T Y S TAT E M E N T
Research data are not shared.