Abstract

Objective To assist in workforce planning by updating trends in the characteristics of near-retirement comprehensive family physicians (FPs) and their patients since the COVID-19 pandemic.

Design Population-level serial cross-sectional analysis using linked health administrative datasets.

Participants The Ontario population as of March 31, 2022 (15,023,570), and the comprehensive FPs to whom they are attached (9375). We compared these populations to pre-pandemic analyses (2008, 2013, and 2019).

Main outcome measures Temporal trends in the number, proportion, and characteristics of comprehensive FPs; comprehensive FPs nearing retirement; and patients attached to comprehensive FPs, focusing on FPs nearing retirement.

Results After 2019, growth in the overall comprehensive FP workforce stagnated (2019: 9377; 2022: 9375). For the first time during the study period, in 2022 there was a decline in the number and proportion of early-career physicians (age <35 years) and female physicians comprised the majority (51.5%) of the workforce. An increasing proportion of the workforce is age 65 and older (2008: 10.0%; 2013: 14.4%; 2019: 13.9%; 2022: 15.2%), and correspondingly, an increasing number and proportion of patients are attached to near-retirement FPs. The oldest FP cohort (age ≥70) also increased in number and proportion in 2022. Patients attached to near-retirement FPs were older and had higher levels of chronic conditions compared with patients across the overall FP workforce. Mean roster sizes remained relatively stable and female FPs consistently cared for smaller rosters than male FPs. An increasing proportion of patients had the highest level of complexity, and practices of all FP age groups comprised increasing proportions of those with the highest resource needs.

Conclusion Changes to the comprehensive FP workforce since the COVID-19 pandemic, together with increasing patient complexity, raise concerns about the workforce’s capacity to absorb patients whose FPs are poised to retire.

Virtually every nation is experiencing population aging, including Canada.1 The population of older Canadians is growing 6 times faster than that of younger Canadians, and the number of Canadians 85 years and older is expected to triple by 2046.2

In our previous work, we raised concerns about the capacity of Ontario’s comprehensive family physician (FP) supply to meet the population’s primary care needs as FPs retire and aging patients require more resources.3 Since the end of that study period (2019), the COVID-19 pandemic has occurred and surveys indicate this may be hastening FP retirements, partly due to burnout.4-6 Whereas the previous average retirement age for Ontario FPs was 70.5 years, with women retiring on average 5 years earlier than men,7 pandemic-era research suggests retirement age may now be closer to 55 years.4,6 Other literature suggests a continued trend of increasing patient complexity and clinician workload.8 All of this is occurring in the context of shifts away from comprehensive, longitudinal family medicine9 among trainees and FPs.3,10-16

In this study, we aim to assist in primary care workforce planning by updating trends in the characteristics of comprehensive FPs and the patients attached to them. We focus on the near-retirement workforce and the context of recent research indicating FP intentions to retire earlier than pre-pandemic research indicated.

METHODSStudy design and setting

Our study was designed as a population-level serial cross-sectional analysis using linked health administrative datasets.

Population

Our cohort creation methodology is published elsewhere.3 In brief, we included all Ontario residents covered by the Ontario Health Insurance Plan (OHIP) as of March 31, 2022 (N=15,023,570). With exclusions (Figure 1A3,13,17,18) our final cohort was 11,244,753 patients. At the physician level, we included all Ontario physicians who billed primary care services as of March 31, 2022 (n=16,232). With exclusions (Figure 1B3,13,17,18) our final cohort was 9375 comprehensive FPs. We compared these cohorts to our previous 2008, 2013, and 2019 analyses.

Figure 1.

Cohort creation: A) Patients and B) physicians.

Data sources

De-identified physician-level and patient-level data came from 9 databases, were linked using unique encoded identifiers, and were analyzed at ICES (formerly known as the Institute for Clinical and Evaluative Sciences). Details are in Appendix 1, available from CFPlus.*

Outcomes

The main outcomes were temporal trends in the characteristics of comprehensive FPs; the number, proportion, and characteristics of comprehensive FPs nearing retirement; and the number, proportion, and characteristics (social and medical) of patients attached to comprehensive FPs, focusing on FPs nearing retirement.

Where possible, analyses were stratified by FP age and sex. FP age was compared to years in practice and the 2 were closely correlated, rendering FP age a proxy for career stage. To examine the near-retirement FP cohorts and their patients, we used 3 near-retirement age cut-points: FPs aged 55 and older, FPs aged 65 and older, and FPs aged 70 and older.

Patient medical characteristics examined included the presence of a chronic condition (chronic obstructive pulmonary disease [COPD], congestive heart failure [CHF], diabetes), mental illness (further stratified by severe, non-severe, and substance use or addiction), frailty, and overall medical complexity based on resource utilization band (RUB), which assesses expected health care use as a measure of patient complexity or morbidity. Patient social and demographic characteristics examined included age, sex, newcomer to Ontario status, neighbourhood-level income quintile, households and dwellings, material resources, and racialized and newcomer status.

The early period of the COVID-19 pandemic resulted in altered health-seeking behaviour and availability or delivery of certain health care services.19-21 To overcome potential artifacts related to this, we extended the standard 2-year lookback period for calculating the RUB to 5 years, and for calculating mental illness to 3 years. Virtual care codes for primary care services were added to the algorithm defining patient attachment for 2022.

Ethics review

The use of the data in this project is authorized under section 45 of Ontario’s Personal Health Information Protection Act (PHIPA) and does not require review by a research ethics board.

RESULTS

Table 1 depicts the comprehensive FP workforce over time by FP age and sex.3,18 Between 2008 and 2019, the workforce grew from 7673 to 9377 physicians, but stagnated after 2019 (2022: n=9375). For the first time in the study period, in 2022, the number and proportion of early-career physicians (age <35 years) declined and female FPs comprised the majority of the workforce (51.5%). Older cohorts continued to comprise an increasing proportion of the workforce.

Table 1.

Comprehensive FP workforce over time by age and sex

A shift away from comprehensiveness and into other or focused scopes of practice (ie, noncomprehensive) continued in 2022, with the proportion of all FPs practising comprehensive family medicine declining further, from 70.7% in 2019 to 65.1% in 2022. These shifts were driven by declining proportions of younger age cohorts (those younger than 65) practising comprehensiveness between 2019 and 2022: a decline of 9.6% among FPs younger than 35, 6.7% among those aged 35 to 44, 4.7% among those aged 45 to 54, and 5.5% among FPs aged 55 to 64. In contrast, between 2019 and 2022, the proportion of FPs aged 65 and older practising comprehensiveness remained relatively stable, with a decline of 1.5% among those aged 65 to 69, and an increase of 1.0% among those aged 70 and older.

Table 2 shows mean roster sizes for the overall workforce remained relatively stable and peaked at the midcareer stage for both female and male physicians.3,18 Female FPs consistently cared for smaller rosters than males. Over time, the mean age of patients attached to comprehensive FPs increased (2008: 38.8 (SD 22.3); 2013: 40.1 (SD 22.4); 2019: 41.6 (SD 22.8); and 2022: 42.7 (SD 22.9)).

Table 2.

Mean total roster size by comprehensive FP age and sex over time

When examining the near-retirement FP cohorts, an increasing proportion was aged 65 and older (2008: 10.0%; 2013: 14.4%; 2019: 13.9%; and 2022: 15.2%) (Figure 2).3,18 Correspondingly, an increasing number and proportion of patients were attached to an FP aged 65 or older, with 15.5% of patients (1,742,601) attached to an FP aged 65 or older by 2022 (Table 3A).3,18 The 2022 workforce aged 55 and older represented 3535 FPs (Figure 2)3,18 and 4,737,493 patients (Table 3A).3,18 By 2022, the number and proportion of FPs in the oldest cohort (70 and older) grew to 674 physicians (7.2% of the comprehensive FP workforce) (Figure 2),3,18 serving a correspondingly increased number and proportion of patients (721,727, 7.2%) compared to previous years (Table 3A).3,18

Figure 2.

Comprehensive FPs by 3 near-retirement age cut-points, from 2008 to 2022: Interpretation: For example, 7.3% of comprehensive FPs in 2022 are aged 70. In 2022, 1.4% of comprehensive FPs are female and aged 70, and 5.9% are male and aged 70.

Table 3A also shows that an increasing number and proportion of patients aged 65 and older were attached to near-retirement FPs in all near-retirement FP age groups.3,18 Although proportions fluctuated, an increasing number of rural patients were attached to near-retirement FPs as well.

Over time, a progressively increasing proportion of patients attached to a comprehensive FP had a chronic condition (Table 3B).3,18 Compared with other years, in 2022, a higher number and proportion of patients aged 65 and older, and a higher number and proportion of patients with COPD, CHF, diabetes, or frailty were attached to an FP aged 65 and older (Table 3B).3,18 An increasing proportion of patients had the highest complexity based on RUB (Table 3B),3,18 and practices of all FP age cohorts, including early-career FPs, progressively comprised increasing proportions of patients in the highest RUB (Table 4).3,18 Proportions of patients with mental illness attached to a near-retirement comprehensive FP fluctuated over the course of the study period, but the absolute numbers grew markedly (Table 3B).3,18

We compared patient characteristics among the near-retirement workforce with those among the overall workforce (Tables 3A and 3B).3,18 A comparatively higher proportion of patients aged 65 and older, or with COPD, CHF, diabetes, or frailty, were attached to a near-retirement FP. This was observed for each year, and the same was also observed for patients with a mental health disorder. The proportions of medically complex patients attached to near-retirement FPs represented large numbers of patients.

Comprehensive FPs aged 65 and older and 70 and older served progressively growing numbers and proportions of some of the most marginalized patients (lowest income, highest housing instability, lowest material resources) (Table 3C).3,18

Table 3.

Characteristics of patients attached to near-retirement FPs and all FPs over time: A) Demographic characteristics, B) health characteristics, and C) vulnerable populations.

Table 4.

Comprehensive FPs’ practice composition by patient complexity over time

DISCUSSION

Previously, we found that as of 2019, 1,695,126 patients were attached to an FP aged 65 or older.3 Since 2020, other research has found that progressively higher numbers and proportions of Ontario patients are unattached to a regular source of primary care, reaching 2.52 million (16.0%) as of 2023.22 In addition to population growth,23 this trend likely reflects, in part, the retirements anticipated from our previous near-retirement study’s findings.

The updated analyses presented here indicate that the problem of declining primary care attachment due to FP retirement is likely to worsen after the pandemic for several reasons. First, in the most recent year of the study period, we saw—for the first time—stagnation in growth of the comprehensive FP workforce and a decline in the number and proportion of early-career FPs (age <35). Second, the overall workforce is aging. Comprehensive FPs aged 65 and older represent a higher number and proportion of the workforce compared to all previous years. Correspondingly, the largest number (1.74 million) and proportion (15.5%) of patients are now attached to FPs aged 65 and older. These patients have aged with their FPs: progressively higher proportions of elderly and medically complex patients are now served by near-retirement FPs, exceeding the overall proportions of elderly and complex patients attached to comprehensive FPs. Social complexities are also increasingly prominent among patients attached to the near-retirement workforce. Third, FP groups of all career stages are serving increasingly complex rosters, including larger numbers of patients with mental illness, and for the first time, female physicians, who consistently care for smaller rosters than male FPs, comprise the majority of the workforce. Our 2022 data also indicate that mid-career (aged 45 to 54) FPs are serving smaller rosters compared with 2019—possibly a pandemic effect. Taken together, these findings suggest limited capacity to absorb the patients of near-retirement FPs. As the population continues to grow,23 as FPs retire, as fewer medical students choose family medicine as a specialty,24 as the complexity of delivering primary care increases,8,25 and as FPs of all career stages shift away from the practice of comprehensiveness,3,10 the potential for workforce growth without major policy changes may be limited.

The COVID-19 pandemic has exacerbated burnout among FPs,5,6,26,27 with other Canadian and international research now suggesting the 55 and older FP cohort (rather than 65 and older) may be more relevant to policy-makers planning for impending FP retirements.4,6,27 That said, the analyses here suggest aging FPs are actually remaining in practice longer than expected, demonstrated by the growth in the number and proportion of FPs aged 65 to 69 and 70 and older in 2022. Further, although mean roster sizes among these older FP groups have remained stable since 2019, they have increased markedly since 2008. There have been media reports of aging FPs experiencing difficulty finding a replacement for their retirement and feeling too guilty to leave their patients without primary care.28 Although some have suggested introducing incentives to retain retirement-age FPs,7 reliance on aging FPs to delay their retirement is neither a sustainable approach to workforce planning nor an ethical one.29 Instead, strategies that increase the capacity of the existing workforce and incentivize the long-term practice of comprehensive family medicine across all training and career stages have been proposed as more effective, enduring solutions. These may include funding for interprofessional teams, a reduction in system-level inefficiencies to alleviate FPs’ administrative workload, open entry for FPs into salaried or capitation-based compensation models, and remuneration for indirect patient care.3,9,27,30 In British Columbia, the recent implementation of some of these strategies appears to be increasing entry into comprehensive family medicine,31 and other provinces, such as Manitoba, Nova Scotia, and Prince Edward Island, are following suit.32-34 Ontario has recently announced $110 million in new funding for teams, although that funding represents only 0.1% of the total health care budget.35 A new Ontario medical school program that places incoming students on a direct path to family medicine holds promise,36 but will not address the near-term retirements anticipated by this study.

Strengths and limitations

Strengths of this study include the use of recent data and new virtual care billing codes to understand workforce impacts since the pandemic began, and the application of previously validated attachment and comprehensiveness algorithms to large, population-level datasets. Given broad trends related to population aging and the declining practice of comprehensive family medicine, our findings are relevant to other Canadian jurisdictions and support workforce planning for the complex needs of retiring FPs’ patients. Limitations include the inability to link patients seen at community health centres (CHCs) to FPs and the inability to capture nonphysician practitioners, although these populations account for an estimated small percentage (approximately 2%) of patients.22,37,38 The number of comprehensive FPs in rural areas may be underestimated due to rural physician practice patterns wherein billings for hospital-based services may exceed those for office-based primary care services. While we extended our lookback periods for patient measures (RUB, mental illness) that may have been impacted by the early period of the pandemic, there may still be some pandemic-related artifacts in our data. That said, given prior research indicating a return to the baseline volume of primary care services by the end of July 2020,19 any such artifacts are likely minor.

Conclusion

Changes to the comprehensive FP workforce since the COVID-19 pandemic, together with increasing patient complexity, raise concerns about the workforce’s capacity to absorb patients whose FPs are poised to retire.

Acknowledgment

This study was supported by the INSPIRE-PHC (Innovations Strengthening Primary Health Care Through Research) Research Program, which is funded through the Health Systems Research Program of the Ontario Ministry of Health (MOH) and the Ontario Ministry of Long-Term Care (MLTC). It was also supported by ICES, which is funded by an annual grant from the Ontario MOH and MLTC. Dr Kamila Premji was also supported by the PSI Graham Farquharson Early Career Knowledge Translation Fellowship and the Junior Clinical Research Chair in Family Medicine in the Department of Family Medicine at the University of Ottawa. The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. The analyses, conclusions, opinions, and statements expressed herein are solely those of the authors and do not reflect those of the funding or data sources; no endorsement is intended or should be inferred. Parts of this material are based on data and information compiled and provided by Ontario Health, the Canadian Institute for Health Information (CIHI), and the Ontario MOH. The analyses, conclusions, opinions, and statements expressed herein are solely those of the authors and do not reflect those of the data sources; no endorsement is intended or should be inferred. This study used data adapted from the Statistics Canada Postal CodeOM Conversion File, which is based on data licensed from Canada Post Corporation, and/or data adapted from the Ontario MOH Postal Code Conversion File, which contains data copied under license from ©Canada Post Corporation and Statistics Canada. We thank the Toronto Community Health Profiles Partnership for providing access to the Ontario Marginalization Index.

Footnotes

↵* Appendix 1 is available from https://www.cfp.ca. Go to the full text of the article online and click on the CFPlus tab.

Contributors

Drs Kamila Premji, Michael E. Green, Richard H. Glazier, and Bridget L. Ryan contributed to the study concept and design. All authors contributed to acquisition, analysis, or interpretation of data. Dr Premji drafted the manuscript. All authors critically revised the manuscript for important intellectual content. Dr Premji, Shahriar Khan, Dr Ryan, Dr Green, and Dr Glazier contributed to statistical analysis. Drs Green and Glazier obtained funding. Eliot Frymire and Shahriar Khan contributed to administrative, technical, or material support.

Competing interests

None declared

This article is eligible for Mainpro+ certified Self-Learning credits. To earn credits, go to https://www.cfp.ca and click on the Mainpro+ link.

This article has been peer reviewed.

La traduction en français de cet article se trouve à https://www.cfp.ca dans la table des matières du numéro de juin 2025 à la page e114.

Copyright © 2025 the College of Family Physicians of Canada