"It was horrible for that community, but not for the way we had imagined": A qualitative study of family physicians' experiences of caring for communities experiencing marginalisation during COVID-19.

The COVID-19 response required family physicians (FPs) to adapt their practice to minimise transmission risks. Policy guidance to facilitate enacting public health measures has been generic and difficult to apply, particularly for FPs working with communities that experience marginalisation. Our objective was to explore the experiences of FPs serving communities experiencing marginalisation during COVID-19, and the impact the pandemic and pandemic response have had on physicians' ability to provide care. We conducted semi-structured qualitative interviews with FPs from four Canadian regions, October 2020 through June 2021. We employed maximum variation sampling and continued recruitment until we reached saturation. Interviews explored participants' roles/experiences during the pandemic, and the facilitators and barriers they encountered in continuing to support communities experiencing marginalisation throughout. We used a thematic approach to analyse the data. FPs working with communities experiencing marginalisation expressed the need to continue providing in-person care throughout the pandemic, often requiring them to devise innovative adaptations to their clinical settings and practice. Physicians noted the health implications for their patients, particularly where services were limited or deferred, and that pandemic response policies frequently ignored the unique needs of their patient populations. Pandemic-related precautionary measures that sought to minimise viral transmission and prevent overwhelming acute care settings may have undermined pre-existing services and superseded the ongoing harms that are disproportionately experienced by communities experiencing marginalisation. FPs are well placed to support the development of pandemic response plans that appreciate competing risks amongst their communities and must be included in pandemic planning in the future.

Introduction

From the outset of the COVID-19 pandemic, policymakers and healthcare workers understood that COVID-19 would pose increased risks for communities experiencing marginalisation1 – populations that face increased susceptibility to adverse health outcomes, including increased morbidity and mortality and decreased quality of life due to processes of social exclusion, discrimination, colonisation, and criminalisation (Flaskerud & Winslow, 1998; Hall, Stevens, & Meleis, 1994; Vasas, 2005). Such communities include people who use substances, people experiencing houselessness or underhousing, resettled refugees and newcomers to Canada, and temporary foreign workers. Within these communities, Indigenous and racialised peoples may be disproportionately represented and face intersecting experiences of marginalisation (Dhamoon & Hankivsky, 2011; Tang & Browne, 2008).

Prior to COVID-19, communities experiencing marginalisation in Canada faced a myriad of inequities that threatened their health and wellbeing – inequities that have persisted through the pandemic. These include crowded living quarters, from employer-provided housing for migrant workers to tents and shelters for people experiencing houselessness (Haley et al., 2020; Landry et al., 2021; Lima et al., 2020; Ralli, Arcangeli, & Ercoli, 2021; Migrant Workers Alliance For Change, 2020). Inadequate housing is often accompanied by limited access to sanitation facilities, making it difficult for individuals to maintain basic hygiene (Haley et al., 2020, Lima et al., 2020, Ralli, Arcangeli, & Ercoli, 2021). Social support strategies are typically insufficient and poorly coordinated (Lima et al., 2020; Ralli et al., 2021). Communities and individuals that experience marginalisation face higher rates of morbidity (e.g., high rates of HIV, hepatitis B, and hepatitis C amongst people who inject drugs (Degenhardt et al., 2017)) and premature mortality (Edmonds & Flahault, 2021; Flaskerud & Winslow, 1998; Perri, Dosani, & Hwang, 2020; Vasas, 2005; Ralli et al., 2021), yet often have inadequate access to quality health services, including poor continuity that results in fragmented care (Dunlop et al., 2020; Howells et al., 2021; Lima et al., 2020). These poor health service experiences are often tied to bureaucratic, cost-related, and language barriers; stigmatisation and discrimination by healthcare professionals; and user fears related to community experiences of criminalisation, racialisation, or citizenship status (Dunlop et al., 2020; Howells et al., 2021; Moroz, Shrestha, & Testaverde, 2020; Smith et al., 2021).

Singularly, these experiences pose a heightened risk for exposure to and poorer outcomes from COVID-19; for individuals who experience marginalisation, many of these experiences intersect and compound (Edmonds & Flahault, 2021; Perri et al., 2020; Vasas, 2005). Overcrowded and poorly ventilated housing and limited sanitation facilities make adhering to basic public health recommendations such as physical distancing and hand washing appreciably more challenging (Haley et al., 2020; MacKinnon, Socías, & Bardwell, 2020; Moroz et al., 2020; Perri et al., 2020). Other recommendations, such as stay-at-home orders, are untenable for those who rely on food banks and community food centres daily or are economically dependent on in-person work, whether formal or informal and regardless of conditions or access to appropriate personal protective equipment (PPE) (Guadagno, 2020; Melamed, Hauck, Buckley, Selby, & Mulsant, 2020). For people with substance dependence, reduced access to their regular supply, harm reduction services, and medical withdrawal management could lead to withdrawal, seizures, delirium, overdose, and death (Melamed et al., 2020; Concerned registered nurses and frontline workers of DTES, 2020). People experiencing houselessness and those who regularly use substances experience higher rates of cardiovascular disease, liver disease, chronic obstructive pulmonary disease, and cancers – all of which can place them at greater risk for acute respiratory distress syndrome and death if infected with COVID-19 (Dunlop et al., 2020; MacKinnon et al., 2020; Melamed et al., 2020; Volkow, 2020). Conversely, fentanyl and other opioids slow individuals’ respiratory rate which, when coupled with COVID-19, may increase the risk of overdose death (BC Centre for Disease Control, N.D.). Higher rates of mental health challenges amongst people experiencing houselessness, people who use substances, and resettled refugees are all likely to be exacerbated by heightened levels of collective and individual anxiety and uncertainty during a pandemic (Krawczyk et al., 2021; Lima et al., 2020; Ralli et al., 2021; Smith et al., 2021).

In seeking to minimise COVID-19 exposure risks, screening protocols and restricted entry have altered access to healthcare services. Moreover, the prioritisation of public health measures during the pandemic often meant that health services have diminished capacity to address ongoing barriers to care and provide linguistically appropriate, culturally sensitive care to patients with diverse needs (Guadagno, 2020; Smith et al., 2021). The pan-Canadian shift to increasingly virtual access to care for all but “essential urgent and emergency services” at certain stages of the pandemic has also meant increasing marginalisation for those with limited access to the requisite technology (MacKinnon et al., 2020; Oetter, 2020; Smith et al., 2021; Tyndall, 2020).

Many of these challenges were recognised by public health, primary care, and community advocates early in the pandemic (Concerned registered nurses and frontline workers of DTES, 2020), spurring organisations and governments to provide supports. These targeted responses varied across Canadian jurisdictions but have included: isolation hotels to facilitate self-isolation (BC Housing, 2021; BC Ministry of Municipal Affairs and Housing, BC Housing and the Ministry of Health, 2020); quarantine requirements for temporary migrant workers, including financial compensation for employers (Agriculture and Agri-Food Canada, 2020; Employment and Social Development Canada, 2020; Migrant Workers Alliance For Change, 2020); class exemptions to the Controlled Substances Act (Health Canada, 2020); interim clinical recommendations for pandemic prescribing and withdrawal management practices to facilitate self-isolation for those who use substances (Ahamad et al., 2015; MacKinnon et al., 2020); and the extension of health insurance to all people living in Canada, regardless of their immigration status (Doyle, 2020).

Consistent access to primary care teams and services are vital to improved health outcomes (Glazier, 2007; Starfield, Shi, & Macinko, 2005), with the development of trust and provision of culturally safe care being integral to positive healthcare relationships with individuals experiencing marginalisation (Curtis et al., 2019; Ford-Gilboe et al., 2018; Schultz, Delva, & Kerr, 2012; Starfield, 2009). While existing research has highlighted the risks and harms for marginalised communities during the pandemic (e.g., Dhamoon and Hankivsky, 2011, Dunlop et al., 2020, Edmonds and Flahault, 2021, Howells et al., 2021, Landry et al., 2021, Lima et al., 2020, Ralli, Arcangeli, & Ercoli, 2021, Smith et al., 2021, Tang and Browne, 2008), there is a paucity of research highlighting the ways in which family physicians (FPs) who serve these communities have been impacted and how they could be better utilised and supported during the ongoing COVID-19 response and in future pandemics. The objective of this study is to understand the roles, responsibilities, and experiences of FPs working with communities experiencing marginalisation during the COVID-19 pandemic, and the impact that the pandemic and Canada's pandemic response have had on physicians' provision of care.

Materials and methods

Study design

The present qualitative analysis is part of a series of case studies informed by semi-structured interviews with FPs in four study regions across Canada. We sought to understand the proposed roles and actual experiences of FPs during the COVID-19 pandemic, the facilitators and barriers they encountered, and the impacts of the pandemic on both FPs and the communities they serve. The full study protocol has been documented elsewhere (Mathews et al., 2021).

Setting

Healthcare delivery in Canada is governed by individual provinces with much of primary care publicly funded through provincial health insurance and provided by physicians operating privately. While there are no out-of-pocket costs at point-of-care for those who are covered by provincial insurance, individuals with lower income and those experiencing marginalisation encounter various barriers to accessing primary care (Bloch, Rozmovits, & Giambrone, 2011). These barriers range from healthcare providers' knowledge and attitudes, inflexible clinical practice and bureaucratic requirements, structural and enacted stigma and discrimination, and individuals’ proximity to services, access to transportation, and feelings of shame (Bloch et al., 2011; Williamson et al., 2006). Given the gatekeeping and care management role that FPs hold, access to primary care frequently determines broader healthcare access, health outcomes, and equity (Glazier, 2007).

Study sample, recruitment

FPs were recruited using maximum variation sampling through faculty, practice team, and physician privileging lists and newsletters, provincial College of Physicians and Surgeons' public physician listings, social media posts, and snowball sampling. Physician privileging lists are health-authority maintained lists of family physicians with hospital or health facility admitting privileges which team members who had pre-existing access used to circulate study information. We also advertised our recruitment on various Twitter and Facebook accounts (including individual investigators’, a separate study account, and through re-posts by provincial professional organisations and collaborators) to broaden the reach of and diversify our recruitment efforts.

To be eligible, physicians must have been licensed to practice and eligible to be clinically active in 2020 in one of four regions: Vancouver Coastal health region in British Columbia, the Eastern Health region of Newfoundland and Labrador, the province of Nova Scotia, and Ontario Health West region. We included physicians along a wide variety of characteristics, including gender, primary care funding and practice models, with and without hospital or health authority affiliations, with and without academic affiliations, and from urban and rural communities. We excluded physicians with solely academic, research, or administrative roles, as well as post-graduate medical students who were permitted to practice solely during the pandemic. Recruitment continued in each region until there was sufficient data to facilitate rigorous analysis and interpretation (i.e., saturation) (Creswell, 2014; Green & Thorogood, 2018).

Data collection and analysis

Interviews were conducted (by authors SS, RB, LMe, LMo, MM) with participating FPs between October 2020 and June 2021 by either telephone or Zoom (Zoom Video Communications), depending on participant preference. Interviews were semi-structured, based upon a standardised interview guide, and conducted in private conversation with each participant. The interview guide (Supplementary File 1) was collaboratively developed by the principal investigators and pretested by our interdisciplinary study team, which includes family physicians and public health experts. Interviews ranged from 17 to 97 ​min in length and were conducted in English. With participant consent, interviews were audio recorded for accuracy. Interviews were transcribed verbatim and verified by the interviewing researchers. Additionally, interviewers took field notes during and wrote post-interview summaries highlighting emerging or significant themes.

Using a thematic analysis approach (Berg, 1995; Guest, 2012), at least two members of the research team independently read a selection of transcripts in their region and the interviewer's corresponding notes to identify key themes inductively and develop an initial regional coding framework. Up to two coders from each regional team (authors SS, RB, LMo, DR, MM) then coded one transcript from each region (four transcripts in total) using their regional coding framework before meeting to compare coding decisions and develop a unified national coding framework, resolving any coding conflicts through consensus. Through this iterative process, with codes evolving from more descriptive to more analytic, we developed broad conceptual codes by grouping similar and overlapping codes into defined thematic groups. The final unified coding framework was then used by regional research team members (authors SS, RB, LMo, DR) to code all interview transcripts using NVivo V.12 (QSR International).

Ethical considerations

Approval for this research was obtained from Research Ethics British Columbia, the Health Research Ethics Board of Newfoundland and Labrador, Nova Scotia Health Authority Research Ethics Board, and Western University Research Ethics Board. All participants were informed that their participation was voluntary and that study team members were available to answer their questions, and were interviewed after returning their signed consent form to the study team by email or fax. Responses presented in this paper have been anonymised and are attributed using participant codes.

Results

Of the total study sample (N ​= ​68), 24 participants from the four study regions expressly discussed their experiences with providing primary care services for communities experiencing marginalisation during the pandemic (participant characteristics are summarised in Table 1 ). These communities included people who use substances, people experiencing houselessness or who are underhoused, resettled refugees and newcomers to Canada, and temporary foreign workers. Key themes from these interviews are summarised in Table 2 and detailed below.

Table 1

Characteristics of participants.

	British Columbia N ​= ​9 (37.5%)	Newfoundland & Labrador N ​= ​4 (16.7%)	Nova Scotia N ​= ​6 (25%)	Ontario N ​= ​5 (20.8%)	Total N ​= ​24
Gender a
Man	2	1	4	3	10
Woman	7	3	2	2	14
Years in Practice
Range	3–20 years	5–30 years	8–32 years	5–29 years	3–32 years
Mean	13.9 years	17.8 years	19.5 years	15.8 years	16.75 years
Primary Practice Setting b
Community Health Centre	6	3	6	1	16
Physician-Owned Practice	0	1	0	2	3
Hospital-Based Clinic	2	0	0	1	3
Community Organisation c	2	0	0	1	3
Community Size d
Rural	0	0	5	3	8
Small Urban	0	4	0	0	4
Urban	9	0	1	2	12
Mix	0	0	0	0	0

- Gender was asked as an open-ended question.

- Where physicians served communities experiencing marginalisation at various practice settings, each have been included so practice settings may outnumber participants.

- Community organisations are non-profit or social enterprises for which family physicians work providing primary care services.

- Rural ≤10,000 population, Small Urban = 10,000–99,999 population, Urban ≥100,0000.

Table 2

Overview of themes.

Theme 1	Challenges to continued provision of care
	• Ensuring continued access to care for communities • Implementing infection prevention and control (IPAC) protocols • Expanded roles for physicians
Theme 2	Innovations to ensure continued access to primary care
	• Atypical uses of virtual technologies • Modifications to clinic setup, workflow, and practice
Theme 3	Impacts on communities of care
	• Reduced access to health services • Decreased engagement • Deterioration in health status • Inadequacies of pandemic supports
Theme 4	The toll on family physicians and primary care teams
	• Distress associated with patient and clinical experiences • Moral concern regarding revised pandemic procedures

Challenges of continued provision of care

While all primary care practices and practitioners have had to alter their practice during the pandemic, FPs that work with communities experiencing marginalisation routinely faced distinct challenges to ensure that their communities would have continued access to necessary care. FPs recognised from the outset that they would have to continue providing in person care:

I would say … our threshold has been pretty low to see people in person because we just have such a clientele that we can’t trust them to be like … If they have abdominal pain, they need to come in, then we bring them in. And I know the alternative is they’re going to go to [the emergency department] and I don’t know how that makes any more sense than them seeing us in our clinic. [404]

Access challenges often varied by practice population and clinic setting. While some FPs were contending with institutional screening restrictions, others lost access to their hospital-based clinics entirely:

And so, there was a little bit of, you know, making sure I was at those [Emergency Operations Centre] meetings to say like, ‘Hey, don’t forget, there’s like a clinic that serves 1500 marginalised HIV positive patients that need to be able to roll in higher than a kite and still get care.’ And so, most of the other clinics were sort of screaming at [Name of Hospital], like, ‘don’t let everyone and their dog through the door just to walk up to the diabetes clinic and make their next appointment,’ you know? So, we were kind of having the opposite request. [404]

So … we essentially lost our clinic space in the hospital in March, which had a big impact obviously, and the clinic focuses on caring for people with HIV and Hepatitis C. … So essentially, it was myself and a social worker who was the only staff that were being on full-time, and a part-time pharmacist that were trying to get medications out to people in creative ways because we didn’t have a clinic space for them to come to and we were all spread out in different areas. So, …it definitely had a major impact on being able to provide care in that setting. [405]

For FPs operating out of practices in the community, while many of their primary care colleagues were transitioning to virtual modalities of care in spring 2020, there was immediate recognition that this would not be feasible for many of their patients: “most people don't even have phones let alone like, you know, Wi-Fi or … and people often didn't have computers” [413].

When virtual care was not a feasible alternative, the continued provision of in-person care introduced immense challenges related to PPE – particularly early in the pandemic amidst widespread shortages: “So, it is quite possible that the barriers would have been lower if we had done the [COVID test] swabbing in our own settings, but we could not because of the PPE” [303]. This could be exacerbated by some of the services offered and situations encountered in different FPs’ clinical and community care settings:

So, then, there’s all this tension about supply and then you don’t want to inappropriately use PPE just to feel better. … But I think it was affecting the whole country, of like, what does someone in community need for PPE? And then especially for overdose prevention sites, where sometimes we were told, ‘oh you can probably bag mask ventilate with surgical masks’. And then in the ER they were probably using N95s. So, I didn’t want to make this, like, hierarchy of PPE for community versus hospital. [402]

The process of ‘bag mask ventilating’ to oxygenate an individual as part of an overdose response is considered a high-risk aerosol generating procedure, yet community-based FPs and their staff were instructed to proceed with surgical masks rather than N95s despite the risk of aerosolising COVID-19.

For most FPs, the pandemic entailed reconfiguring their practices for infection, prevention, and control (IPAC). Reconfigurations ranged from the entire provision of programs (e.g., administering injectable opioid agonist treatments) to the rotation of clinicians to avoid staff rounding in multiple sites. But IPAC processes, such as implementing masks and screening protocols, were particularly challenging amongst certain populations:

And I remember the stigma amongst my patients, like the idea it was really re-triggering for previous experiences that some of my patients had had, with working with government organisations. Because this was a real government infiltration of advice and restriction and mirrored a lot of the refugee experiences that some of my patients had had, so there was a lot of fear of disclosing that you had upper respiratory tract infection symptoms. [303]

Similarly, FPs felt that pandemic policies and clinical guidelines often did not appreciate the needs in their communities. This was reflected in screening and testing guidelines where, early in the pandemic, only those who travelled internationally were eligible for COVID-19 testing and, later in the pandemic, screening questions did not adequately distinguish between COVID-19 and other realities amongst their patient population:

So, I found that had a lot of information, but the trouble was how to apply it to my patient population. Because most of the time when people have information [to support IPAC amongst their patient population], it doesn’t apply to a homeless person who’s high on crystal meth, right? [402]

And the challenge as well is that there is a lot of overlap with opioid withdrawal symptoms and flu-like symptoms, COVID symptoms, so, oftentimes people, you know, were not doing well and maybe, you know, accidentally come off their medication or missed too many doses, and their prescription was cancelled and they’re in withdrawal. So yes, they would have screened positive for COVID symptoms, but, realistically, probability-wise, you had to, I had to kind of make a judgement call about whether it was opioid withdrawal or if they might have COVID and often would still bring people in, with appropriate PPE obviously, to see people, so … [405]

FPs working with communities experiencing marginalisation also recognised the new vulnerabilities introduced by the pandemic and its attendant public health measures, spurring many to expand their role beyond clinical care to support access to basic needs and community advocacy:

I mean, everyone lost their jobs and they didn’t have food. … So, there was just a lot of, like, scrambling for trying to help people pay rent, trying to help make sure that they had enough to eat, trying to – a lot of immigration processes just were immediately halted, so even people who might have been in process to like, get permanent residency and thus, healthcare access and things like that, everything just stopped. [415]

And the pandemic was like, ‘Well, I feel like I understand what’s going on with my patient population; I’m really worried about various things like, how am I going to reach them, how are they going to still access shelters, how are they going to access meals, how can they do that safely, how can we prevent an outbreak?’ And I felt deeply committed to maintaining Public Health policy, but trying to navigate that with my patients and being an advocate and a voice when many people are just not able to do that. [303]

Some of these expanded roles were more formalised within the pandemic response, such as writing clinic guidelines to inform clinicians’ risk mitigation prescribing practices of controlled substances. Similarly, FPs contributed to staffing isolation hotels, overdose and housing/tent encampment outreach teams, and migrant on-call teams. Frequently, physicians added these roles to their responsibilities while managing their regular practice which faced heightened complexity and demands in the pandemic.

Innovations to ensure continued access to primary care

While FPs supporting communities experiencing marginalisation recognised the impracticality of virtual care for many of their patients, they found innovative ways to minimise contact with patients using videoconference technologies. One medical director set up their clinic where physicians would video-call the clinic for appointments where onsite nurses were coordinating in-person care, allowing them to care for patients at multiple sites. Others used virtual modalities to connect with their patients from separate exam rooms in the same clinic:

Yeah, …in one of the community sites, we set up in the porch, you know, off the side of the building, like a virtual medicine station. So, you could come in, and then we had nurses in PPE who could assess the patient, we could put on PPE after we talked to them, to say, ‘Yeah, we’ll come into that porch and do our assessment there.’ [303]

Other physicians found utility in videoconference technology to connect with other healthcare professionals to coordinate patient care:

We would try to do a brief phone call or Zoom check-in with myself, the pharmacist and social worker and then I would do a couple of phone calls for urgent issues that were felt to be priorities. And the social worker and the pharmacist were arranging to meet patients out in the parkade of various places and try to drop off medications so that people didn’t go off their HIV medication. So that, this went on for probably about four to six months, I would say, until we were able to get back into our clinic space. [405]

Some physicians made a concerted effort to get phones to patients: “And we worked out a system where like, we would put a phone in a bag and wipe down the bag and hand it to the patient to be able to talk to us on the phone” [303].

Clinic modifications were common and varied from basic IPAC protocols such as enhanced cleaning, plexiglass shields, and removing seats from waiting rooms, to building new structures, processes, and workflows as detailed by this community-based physician:

You know, because we do care for thousands of marginalised patients [in the community] and our waiting room is full all the time. So, I think it was the first week of March, we implemented a screening process at our front gate of the clinic. And we identified … covered outdoor spaces that we transformed into COVID assessment areas if people screened positive. … we purchased walkie-talkies so that our front gate staff could communicate with like, the pharmacy, for example, someone was coming to get their methadone. Were they, did they have capacity to bring somebody in? And so, sort of to manage our clinic flow. [413]

While Ministries of Health, public health agencies, and physician regulatory colleges provided policies and best practice guidance, these were often reactionary and sometimes were developed first by physicians who recognised issues and devised innovative strategies and modifications to their clinical practice to support patients’ needs. For example, a physician who prescribe methadone wanted to ensure that their patients would not lose access if a clinic or pharmacy closed:

And what I did, the first – I think we had a day’s notice or something, or we knew the clinic was going to close in a couple of days – was to contact all the pharmacies. I sent a letter to all the pharmacies. This is before the guidance came out. Right. I just sent my own personal letter to each of the pharmacies saying that, you know, that I authorise two weeks of this for everybody … put the patients name on it and sent it out. So, then you know, I had to kind of – there was about 28 pharmacies around and so I just kind of gave them the permission to get through the first couple of weeks. [305]

Indeed, FPs that saw the early indications of the pandemic identified that altering their prescribing practices and deferring regularly scheduled care where possible was one of the best ways to reduce clinic traffic and support IPAC without affecting patients’ access to essential medicines.

Impacts on communities of care

While FPs have worked tirelessly to continue or expand their provision of care throughout the pandemic, communities experiencing marginalisation have nonetheless encountered barriers to accessing essential services and supports. At times, physicians who run or support overdose prevention sites, for example, faced greater challenges to staying open to their communities:

… this happened in one of our overdose response rooms, where it shut down for hours because paramedics came and they bagged someone and then, so, aerosolised COVID all over the place. And so, it shut the site down to let the COVID dust settle, and then do this COVID clean, and this is all in the same context of overdose deaths skyrocketing. [402]

IPAC protocols such as screening, appointment requirements, and removing extraneous objects from waiting rooms also meant that harm reduction supplies and essential services were less accessible:

… we provide harm reduction supplies. And so, those were essentially removed from just being available for people to pop into the clinic and take, now … they had to sort of come in and phone ahead and request for supplies or request a Naloxone kit. That’s something I worried about quite a bit is, is where we haven’t been giving out very much Naloxone, it’s just a super important medication to be getting out there. [405]

… trying to run a shelter during a pandemic is very challenging, to get everyone in the shelters who are trying to share their crystal meth to wear masks. … and then as well, people are worried that you’ll turn them away from services if they disclose that they have symptoms, so people will lie about having a cough, so that they can come, so that they’re not turned away from the overdose prevention site or from the centre. So, I mean, there’s so many layers of vulnerability and complications. [402]

In addition to preventive services, diagnostic and laboratory services such as viral load testing for HIV and hepatitis C, bloodwork, urine drug screens, mammograms, and liver scans were deferred. Several physicians noted that some patients were content to avoid contact with the health system, recognising the COVID-19 exposure risks these interactions inherently presented. Those physicians, however, also noted their concerns about backlogs in testing and loss of patients who may have only been marginally engaged with healthcare pre-pandemic:

… we changed our guidance so that in the pandemic they only need bloodwork every six months, but what I’ve started to notice is we’re now at the six-month mark, and there’s a whole bunch of people that still haven’t gotten their bloodwork … And now, you know, now we’re at seven or eight months, and I think by the end there will be a whole bunch of people that don’t get their bloodwork done for nine months or twelve months, because, you know what happens, when someone gives you like, an inch, sort of take a mile. [404]

This concern was echoed in reflections on the impact of pandemic prescribing practices, where more frequent prescription renewal appointments normally provide FPs an opportunity to “lay eyes” on patients. When individuals are provided multiple refills of the same prescription, other health issues can be missed:

So, we were seeing more, like people developing more serious like, cellulitis and infections, people reporting like their, poor adherence to their treatment medications for their mental health or worse, just a general worsening of their mental health and/or addiction because of COVID. [406]

The cumulative impact of reduced health services can be profound, with long term individual, population health, and socioeconomic implications that may perpetuate the marginalisation individuals experience: “… what we've seen in our community is that there's an increase in HIV regionally … There's been an increase in hepatitis C as well, there's been an increase in syphilis, chlamydia, gonorrhoea, there's been an increase in unplanned pregnancies” [413].

In some cases, it was the pandemic and public health measures that FPs suggested introduced new harms to their patient populations. For individuals with existing anxieties or traumas, uncertainty related to the virus, stay-at-home closures, and engagement with government services were problematic:

… and I haven’t even begun to tell you about what mental health looked like for many of my patients, who had come from like, former refugee camps where they were self-isolated from a structural point of view. Like, they were, because of trauma that was existing and then to put them back in a situation where they can’t speak a language, they can’t leave their house, they can’t go to school, it’s like the same system that, that was very oppressive now happening for a different reason; it’s just very re-triggering. [303]

… they’re always hesitant to go seek care because it’s super expensive and they sometimes get threatened, like they’re not sure who’s going to call immigration, even though it’s obviously not in any way the duty of the healthcare system to act as an arm of [Canadian Border Services Agency], but that’s always a fear, and it’s a well-founded fear, to be perfectly honest. So, they’re always worried; I think people were especially worried because they were worried that if they were to get COVID, what would happen, right? [415]

Though COVID-19 infections did affect communities experiencing marginalisation, harms that individuals and communities experienced pre-pandemic were frequently exacerbated by pandemic policies that prioritised COVID-19 infection prevention. As physicians observed, this prioritisation of COVID-19 ignored the living realities of people in the community:

I really can’t imagine what it’s like for these folks who don’t have any place to clean themselves now or, they can’t use the library [for shelter or to access computers], and so …. it was horrible for that community, but not for the way we had imagined. We had imagined it was going to be horrible for COVID infections, but it really was just horrible for everything else. [414]

In our community, I think more people have died of an opioid overdose than of COVID, right? [117]

Even attempts to support communities through targeted policies lacked understanding of what those communities needed during the pandemic. One example is the risk mitigation and pandemic withdrawal prescribing guidelines which informed the replacement of (il)licit substances with regulated, prescribed substances to support individuals to self-isolate and gave physicians the opportunity to provide their patients a safer alternative to an increasingly toxic street drug supply. While FPs recognised that these interim clinical guidelines were well-intentioned and that there were occasional successes, rarely was it an equivalent of what their patients needed:

… our patients who use substances need a safe supply. The prescribing we were allowed to do as part of the pandemic risk mitigation is not safe supply. My patients are addicted to fentanyl, not hydromorphone. So, I can prescribe things that are similar, but they’re not the things that they are actually addicted to. [404]

Yeah, there’s an oversupply of hydromorphone tablets available in the illicit supply because a lot of people, like … you know, you want a chocolate bar and somebody gives you a chocolate chip – like, one … So, people are getting them and basically selling them to get money for fentanyl – that’s what we’re hearing from a lot of people – tainted, tainted fentanyl. [406]

The toll on family physicians and primary care teams

Physicians and their clinic teams have felt the impact of the pandemic and the pandemic response. Many FPs outlined the extents to which they and their staff went to support their communities, commenting on how that has affected them. Remarking on the impact of modified overdose response procedures, one physician noted that reverting to “low flow oxygen and Narcan” was “torturous for staff to watch someone while they're just waiting for the ambulance to come” – staff who intervene in approximately 2000 overdoses a year [402]. Normally, staff of overdose prevention sites could respond to an overdose with a Bag-Valve-Mask or high flow oxygen, but these are considered aerosol-generating medical procedures and their use was discouraged early in the pandemic (Toward the Heart, 2020). Another physician articulated how the uncertainty and complexity of working in their given community has been exacerbated by the pandemic, impacting the mental health and wellbeing of staff:

… we had lots of cases of, you know, patients going to [the emergency department] for one thing, getting a COVID test, leaving [against medical advice], showing up at clinic, not telling anyone, sitting in the waiting room for four hours and then we figure out after the fact that they were positive. So, there was lots, and that takes a toll on people a lot, so there was a lot of everyone kind of on edge with this feeling like, at any sort of given moment someone could walk in that’s COVID positive and - either due to mental illness or other things, you know, just our population isn’t always able to kind of, give that nice clear history, textbook history. So, I mean, that was – so just trying to support everyone through that was probably one of the, one of the most challenging things, is trying to keep everyone kind of calm and accepting of, that there’s a lot that we can’t control but trying to put in enough precautions to keep people as safe as possible with a whole bunch of unknowns and a whole bunch of random chaos … [412]

Another source of moral concern for physicians was the uncertainty related to the pandemic withdrawal and risk mitigation prescribing, recognising that they were not actually replacing their patients’ unregulated supply with a safer equivalent and the potential risks that presented:

… there was a huge potential for diversion, and we know we flooded the streets with the meds of safe supply, and I worry a little bit about … if we created more access for people that weren’t previously using those substances. [403]

… that took up a ton of time in the pandemic, quite honestly. It was a lot of, like, using my brain to decide what was safe or appropriate and the conversations we had around that and what’s going to happen post-pandemic and what are my ethical obligations, let’s say, if this prescribing ability changes. So now if I started prescribing someone like, 96mg of hydromorphone and the pandemic ends, then I’m going to be like, ‘So, guess what? … ’ [404]

On top of these concerns about proper implementation for patients, physicians also noted their distress and desire for clearer support from their regulatory bodies:

In other ways, I still feel a little, like a little bit unsupported because I don’t think the College of Physicians and Surgeons is fully in support of this prescribing. We’ve been getting mixed messaging from them and that, and so I do feel a little bit out on a limb. But, you know, I’ve just been trying to like, make decisions based on what feels like the right thing for the person in question and just trying to go with that, but … yeah. I think like, theoretically, the College signed off on that document, but I just know of a number of colleagues who are getting audited for that type of prescribing currently, so … [413]

Discussion

Our findings highlight that public health precautionary measures, policies, and guidance documents frequently left FPs uncertain about how to continue providing care for the communities with which they work. The widespread, rapid move to virtual modalities in primary care largely ignored those in society with limited or unreliable access to the requisite technology. At the same time, IPAC screening protocols and guidance for minimising COVID-19 transmission were not suited to physicians managing substance withdrawal and overdoses or those supporting individuals dependent on shelter-based housing, particularly with limited access to appropriate PPE. Physicians experimented with varied innovations to continue providing care while minimising clinic traffic and COVID-19 exposure risks, many of their own devising and volition and to varying degrees of success. Despite these efforts and FPs’ continued provision of care, as well as targeted pandemic policies for some communities, FPs witnessed consequences ranging from disengagement from care and health services to the development of terminal illnesses and death.

Varied efforts have been made to prevent COVID-19 from taking hold in communities that experience marginalisation – from isolation hotels for people experiencing houselessness (BC Housing, 2021; BC Ministry of Municipal Affairs and Housing, BC Housing and the Ministry of Health, 2020) to government-funded quarantine for temporary migrant workers (Agriculture and Agri-Food Canada, 2020; Employment and Social Development Canada, 2020) – and recognise the disproportionate risks COVID-19 poses. The priority in these initiatives, however, has been controlling the spread of COVID-19 rather than comprehensively curtailing the harms incurred by communities that experience marginalisation, including those introduced or exacerbated by the pandemic. Pandemic policies and initiatives that truly seek to protect communities must consider the multitude of intersecting risks faced by communities that experience marginalisation, including but not limited to an emergent virus. Effective strategies for reducing morbidity and mortality and improving quality of life amongst communities experiencing marginalisation must be balanced against precautionary measures in a pandemic, and efforts must be made to ensure access to basic needs and essential services are upheld.

Notably, the concerns expressed by FPs regarding decreased access to harm reduction supplies (e.g., Naloxone) and services (e.g., overdose prevention site shutdowns following resuscitations) in the communities they serve suggests that greater efforts are needed to mitigate the variety of harms people who use substances face during a pandemic. Despite targeted efforts and policies to support people who use substances, including modifications to the federal Controlled Substances Act (Health Canada, 2020) and clinical guidelines for risk mitigation and pandemic withdrawal (Ahamad et al., 2020), overdose related deaths have reached unprecedented rates during the pandemic (Public Health Agency of Canada, 2021). While COVID-19 altered individuals’ access to, composition, and the cost of their regular supply and increased the risks associated with consuming illicit substances, decreased access to harm reduction supplies and services due to pandemic precautionary measures contributed to these increased risks (Canadian Centre on Substance Use and Addiction, 2020; Imtiaz et al., 2021; Zolopa et al., 2021).

Additionally, many routine primary care visits have been deferred and primary care services have been operating at reduced capacity throughout the pandemic. While availability of services strongly determines access, other factors, such as individuals' comfort in accessing health services during a pandemic and messaging around limiting interactions with health services, likely played a role as well (Gilbert et al., 2021; Rawaf et al., 2020). As patients return to their routine care, FPs' identification of the health impacts on communities experiencing marginalisation, including increased rates of infections and late diagnoses of terminal illnesses, suggests that how ‘essential care’ is framed, including what is deemed ‘essential’, during a pandemic can have harmful consequences. This is particularly salient for communities that experienced barriers to accessing health services pre-pandemic, and for whom deferral of services may result in a loss of engagement.

The rapid enactment of pandemic policies, while intended to protect and support the public, may have the unintended effect of further marginalising or harming communities that may experience such policies as further encroachments into their daily existence. Pandemic response policies and services must consider the trade-off of perceived risks among communities experiencing marginalisation (e.g., using a COVID testing site if it increases the chance of interactions with police or border services), balancing individuals’ needs for protection from harmful processes against the societal need for protection from an emergent virus.

The trusting relationships between FPs and their patients are integral components to effective person-centred care, particularly when working with individuals who experience racialisation, stigmatisation, criminalisation, and discrimination (Ford-Gilboe et al., 2018; Starfield, 2009). FPs' trusting relationships, coupled with their understanding of patients’ living experiences, need to be better harnessed in the development of pandemic response policies and precautionary measures to avoid exacerbating the harms faced by communities experiencing marginalisation. This requires that FPs be engaged as equal participants in government- and health system-led conversations that ultimately inform policy and public heath responses during a pandemic (Mathews et al., 2022). Given the diversity of practice populations and contexts in which FPs work, this may require targeted outreach by policymakers to FPs working with specific communities to obtain feedback on proposed policies, targeted interventions, and identify supports that communities require to enact public health measures during a pandemic. This may also include consultation with community members to appreciate their living experiences and priorities during a public health emergency.

Limitations

This study focused on the experiences of FPs in four regions of Canada which vary in their population demographics, health system and primary care structures, their pandemic experiences (including variation in COVID-19 cases, hospitalisations, and pandemic responses), the types of communities experiencing marginalisation, and the extent to which participants worked in and with these communities. Despite these variations, there was consistency in the experiences voiced by FPs across the four regions. Additionally, data collection occurred between October 2020 and June 2021 and may not fully capture how primary care and social supports have evolved over the course of the pandemic.

This paper presents select findings from a larger project exploring the roles and experiences of FPs during the pandemic, for which participant recruitment and interviews were not specifically designed to assess experiences in caring for communities experiencing marginalisation. Accordingly, the full extent of these experiences may not have been captured in our results and we are unable to report the proportion of FPs’ whose practices involve supporting communities that experience marginalisation. Future research should explore this topic specifically to capture the full breadth of experiences. Finally, this research is focused on the experiences of FPs, rather than community members, and it will be important to gather the experiences of communities experiencing marginalisation directly. This includes racialised and Indigenous peoples, who we recognise are disproportionately represented amongst communities experiencing marginalisation and affected by COVID-19, and who continue to face discrimination in and barriers to health and social services in Canada (Dryden & Nnorom, 2021; Turpel-Lafond, 2020).

Conclusions

FPs are expected to continue providing care while adhering to public health precautions and guidelines. Yet, such guidelines have rarely accounted for the diversity of primary care practice settings, contexts, and services users’ living experiences. Moreover, the prioritisation of COVID-19 over pre-existing harms has introduced new barriers to necessary primary care services and their delivery. FPs that work with communities experiencing marginalisation are well suited to inform and provide more attuned pandemic responses that account for the diverse living experiences and healthcare needs of these communities.

Funding

This work was supported by a grant from the [VR41 72756].

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.