The Association Between Continuity of Care and Emergency Department Utilization Among Canadian Senior Adults

Introduction

Timely and appropriate access to primary care is a cornerstone of an effective healthcare system, particularly for older adults who often present with complex and chronic health conditions. Among the key attributes of high-quality primary care is continuity of care—the ongoing therapeutic relationship between a patient and their physician or physician team. Numerous studies have linked continuity of care to improved health outcomes and increased patient satisfaction among older adults.^1,2 It is associated with lower mortality rates,^3,4 reduced reliance on specialist services,^5,6 fewer medical interventions,^7,8 and overall reductions in healthcare costs.^9,10 Moreover, continuity of care has been shown to decrease hospital admissions and emergency department (ED) visits.^11–16

Between 2022 and 2023, there were approximately 15.1 million reported unscheduled ED visits in Canada, a figure that rose to nearly 15.5 million in the following year (2023–2024).¹⁷ Older adults (aged 65 and older) are the highest users of emergency services, often due to chronic conditions and functional limitations that necessitate coordinated, longitudinal care. According to a 2022 position statement by the Canadian Association of Emergency Physicians, individuals in this age group account for an estimated 20% to 40% of all ED visits in Canada. The volume of ED use among older adults with complex medical needs is expected to continue rising in the coming years.¹⁸

Beyond clinical need, many ED visits—particularly for less urgent conditions—reflect broader gaps in access to primary care. Unnecessary emergency visits impose a significant financial burden on the Canadian healthcare system. A 2024 report by the Montreal Economic Institute, using 2022–2023 data, estimated that the lack of timely primary care led to nearly one million avoidable ED visits across Canada, increasing healthcare costs by approximately C$145 million that year (all monetary figures reported throughout are Canadian dollars).¹⁹ Similarly, the Canadian Institute for Health Information reported that 15% of ED visits between April 2023 and March 2024 involved conditions that could potentially have been managed in primary care, highlighting ongoing strain on emergency services.²⁰ In Ontario, the direct cost of an ED visit was approximately C$165 per patient in 2019–2020—nearly three times higher than comparable care delivered in community settings.²¹

Given the financial and systemic pressures associated with high rates of ED utilization²²—particularly within a publicly funded healthcare system like Canada’s—enhancing continuity of care may offer a promising strategy to improve health outcomes while reducing unnecessary ED visits. For older adults, care continuity can facilitate more accurate clinical decision-making, foster stronger therapeutic relationships, and enable earlier interventions that prevent the escalation of health concerns.^23–26

Recognizing these benefits, several Canadian provinces have introduced policy reforms aimed at promoting primary care continuity. In British Columbia, the Longitudinal Family Practice (LFP) Payment Model, implemented in February 2023, serves as an alternative to the traditional fee-for-service system. The LFP model uses a blended compensation framework in which physician payment is determined by time spent providing care, number of patient interactions, and the size and complexity of the physician’s patient panel, rather than solely on volume of visits. The model also compensates for nonencounter time, such as care coordination, administrative work, and patient communication between visits. By rewarding long-term patient relationships, improved access, and comprehensive, person-centered care, the LFP model aligns incentives with continuity, preventive care, and proactive management.²⁷

Ontario has taken a complementary approach through the expansion of Ontario Health Teams (OHTs), which aim to integrate care across providers and settings to improve continuity and coordination. Launched in 2019, OHTs bring together primary care, hospitals, community health services, and other stakeholders to provide team-based, patient-centered care, particularly for those with complex needs.²⁸ Ontario has also expanded Family Health Teams (FHTs), interprofessional models that promote continuity through collaborative practice and comprehensive service delivery.²⁹

Despite these initiatives, empirical evidence on the relationship between continuity of care and ED use among older adults in Canada remains limited. This study addresses this gap by focusing specifically on seniors and examining how the duration of continuity with a primary care physician influences the usual place of care for minor, nonurgent health problems, including ED use. Unlike prior research that broadly links primary care access to ED utilization,^14,30 this analysis quantifies the effect of different lengths of continuous care, highlighting how continual relationships of two years or more shape care-seeking behavior. By emphasizing long-term continuity rather than access alone, the study provides more detailed insights into the role of continuous primary care in guiding appropriate healthcare utilization.

Using nationally representative survey data from the Canadian Health Survey on Seniors (CHSS), the study tests the hypothesis that sustained relationships with a primary care provider are associated with lower ED utilization for minor conditions. In the context of Canada’s aging population³¹ and evolving primary care landscape, the findings aim to inform health-system planning, resource allocation, and policy development to better support seniors.

Materials and Methods

The study uses data from the 2019–2020 CHSS, a cross-sectional, nationally representative survey conducted by Statistics Canada. Administered as a supplement to the Canadian Community Health Survey (CCHS), the CHSS takes an average of 15 minutes to complete and collects information on health status, healthcare utilization, and social determinants of health among Canadians aged 65 and older. The 2019–2020 cycle was conducted between January 2019 and December 2020 and included seniors living in the ten provinces, excluding persons on reserves or other Aboriginal settlements, full-time members of the Canadian Forces, the institutionalized population, and residents of Nunavik and the Terres-Cries-de-la-Baie-James regions of Quebec.

Sampling for the CHSS was derived from the CCHS dwelling frame, using a stratified, multistage design to ensure population-level representativeness. Stratification was based on province, age group, and sex, and within each selected household, one eligible senior was randomly chosen to participate. Statistics Canada followed up with nonrespondents through reminder letters and telephone calls to minimize nonresponse bias. Of the 245,639 selected households in the 2019–2020 CCHS adult sample, 100,797 individuals responded (41.0% response rate). Among the 45,863 individuals eligible for the CHSS supplement, 41,635 participated, resulting in a response rate of 90.8%.³²

The CCHS and CHSS are cornerstone data sources for population health surveillance and research in Canada. They have been widely used to study healthcare access, utilization, and outcomes.^33,34 Both surveys are routinely relied upon by federal and provincial health agencies to monitor and evaluate health programs, inform policy development, and assess system performance, while nonprofit organizations and the media use the findings to raise public awareness about health issues.³⁵

The 2019–2020 CHSS provides valuable insights into the health-system experiences of older adults prior to the COVID-19 pandemic, making it an important resource for examining patterns of ED use and the role of continuity of care. This study analyzed de-identified, publicly available CHSS data. Because no direct participant contact occurred, the use of these anonymized data for secondary analysis complied with institutional guidelines for research involving human participants and did not require additional ethics review.

To assess the association between continuity of care with a family physician and ED use for nonurgent conditions, a multinomial logistic regression model was employed. The dependent variable—usual place of care for minor, immediate problems—was categorized into four groups: (1) doctor’s office; (2) hospital setting, comprising outpatient clinics, community health centers, and walk-in clinics; (3) hospital emergency room; and (4) not stated, which included respondents who skipped the question, refused to answer, or selected “option not included” in the defined categories. This category was retained to preserve data completeness and transparency, though it was not the focus of interpretation given its heterogeneity. The main analyses and conclusions emphasize clearly defined care settings such as doctor’s offices, hospital clinics, and EDs.

For this study, minor health problems or nonurgent conditions were defined as nonlife-threatening issues, including fever, vomiting, major headaches, sprained ankle, minor burns, cuts, skin irritation, unexplained rashes, and other minor injuries or conditions resulting from accidents. Respondents were asked whether they or a family member—defined as an individual living in the same dwelling for whose care the respondent is responsible—experienced difficulties accessing immediate care for such conditions.

The key independent variable was the length of continuous care with a family doctor, grouped into the following categories: (a) less than one year; (b) one to less than two years; (c) two years or more; and (d) not stated. The reference group comprised respondents without a regular family doctor. Additional covariates included demographic, socioeconomic, and health-related factors. These include sex (male, with female as the reference), marital status (married/living common-law, or widowed/divorced/separated/never married, with married/common-law as the reference), education (less than post-secondary, with post-secondary or higher as the reference), household income (≥ C$60,000, with < C$60,000 as the reference), and province (with Ontario as the reference).

Health-related variables included self-perceived health (categorized as excellent/very good or good/fair, with poor as the reference) and reported barriers to accessing care (including lack of provider availability in the area, no providers accepting new patients, or provider retirement/departure). These variables were included as covariates in the regression model to account for individual differences in health status and access to care. The final sample comprised 41,060 respondents, and all statistical analyses were conducted using Stata. All variables were derived from closed-ended questions in the CCHS, with detailed derivation and recoding procedures summarized in Appendix Table A.

Results

Table 1 presents the descriptive characteristics of the sample of seniors (aged 65 or older) stratified by their usual place of immediate care for minor problems. The last column of the table suggests that bivariate associations between all patient characteristics and the usual place of care are statistically significant (P < 0.001).

Seniors with long-term attachment to a family or general practitioner (≥ 2 years) were most likely to report a doctor’s office as their usual place of care (65.7%) and least likely to report the ED (7.5%). In contrast, those without a regular physician showed a higher reliance on the ED (8.7%). Differences were also observed by sex and socioeconomic status. For example, males were more likely than females to identify the ED as their usual place of care (9.4% vs 7.6%), and seniors with lower educational attainment or household income (< C$60,000) reported greater ED use (9.3% and 9.4%, respectively) compared to their more educated and affluent counterparts.

Regional variation was also evident. Seniors residing in Quebec had the lowest proportion reporting a doctor’s office as their usual source of care (41.2%) and the highest ED use (13.3%), whereas those in British Columbia showed the opposite pattern, with the highest use of doctor’s offices (70.7%) and the lowest ED use (4.8%). Finally, perceived barriers to access were a critical determinant: seniors reporting access problems were substantially more likely to use the ED (18.2%) compared to those without such problems (8.0%).

Table 2 presents a multinomial logistic regression model examining the association between continuity of care and the usual place of immediate care for minor health problems. In this unadjusted model, seniors who had been with a family physician for two years or more were significantly less likely to report using the ED than those without a regular physician (coefficient = −0.251, P < 0.001). Shorter durations of attachment (<1 year or 1–< 2 years) were not significantly associated with ED use. While these results suggest that a longer relationship with a family physician is associated with lower ED use for minor conditions, the relationship could be confounded by differences in access to care, socioeconomic status, and perceived health.

To account for these potential confounders, Table 3 presents results from the adjusted multinomial model, which includes controls for income, sex, province of residence, health status, and access to care. After adjustment, the association between long-term continuity of care and ED use remained in the same direction but was weaker in magnitude (coefficient = -0.086, P = 0.083). This pattern suggests that part of the reduction in ED visits observed among those with longer continuity of care may be explained by differences in socioeconomic factors, regional healthcare contexts, and individual health status. Other covariates, such as higher income, being male, and poorer health, are also independently associated with increased ED use for minor problems. Overall, continuity of care continued to show a consistent, though attenuated, association with reduced reliance on the ED for nonurgent issues.

The addition of covariates in Table 3 provides some interesting insights. For example, seniors with an annual income above C$60,000 are significantly less likely to use the ED for minor health problems, independent of other factors. While having a primary care physician for more than two years initially appears to reduce ED use, this protective effect is weakened after controlling for income and other covariates. This suggests that income partially explains the relationship between long-term care continuity with a primary care provider and lower ED utilization. Higher income may enable better access to healthcare resources, facilitate timely outpatient care, and support effective management of minor health issues outside of the ED.

As well, the analysis suggests a strong and significant association between better access to primary care and reduced use of the ED for minor health problems among seniors. Specifically, individuals reporting higher ease of access to care are markedly less likely to use the ED, regardless of the length of care continuity with a primary care physician. However, the duration of continuity with a primary care doctor still plays an important role: Seniors with continuity of care of more than two years tend to use the ED less frequently compared to those who do not have a family physician or those who recently started seeing a family doctor. This suggests that continued relationships with a primary care provider contribute to appropriate healthcare utilization, but only when combined with adequate access. Better access likely facilitates timely appointments and effective management of minor conditions in outpatient settings, reducing the need for emergency visits. Therefore, policies aimed at improving both access to care and continuity with a primary care doctor are essential to minimize unnecessary ED use among older adults.

Finally, regression results indicate that education level is an important factor influencing ED use for minor health problems among seniors. Individuals with higher educational attainment tend to use the ED less frequently, potentially reflecting greater health literacy and more effective navigation of the healthcare system. When considering the duration-of-care continuity with a primary care physician, those with a family doctor for more than two years exhibit a further reduction in ED visits, suggesting that long-term relationships with a primary care provider complement educational advantages by facilitating preventive care and appropriate management of minor issues. Conversely, seniors with lower education and shorter or no care continuity with a doctor are more likely to rely on the ED for care, which may indicate barriers in understanding or accessing other care options. These results underscore the importance of educational interventions alongside efforts to strengthen continuous primary care relationships to reduce unnecessary emergency visits.

Discussion

This study examined the association between the duration of continuity of care with a primary care physician and ED use for minor health problems among Canadian seniors. The results indicate that longer-term continuity—specifically relationships lasting two years or more—is associated with lower reliance on the ED for nonurgent issues. After adjusting for covariates including income, sex, province of residence, self-perceived health, and access to care, the association remained in the same direction, though its magnitude was weakened. This pattern suggests that part of the observed reduction in ED visits among those with longer continuity may be explained by differences in socioeconomic factors, regional healthcare contexts, and individual health status. Other covariates, such as higher income, sex, and poorer self-perceived health, were independently associated with increased ED use for minor problems. These findings align with prior research using CHSS and other Canadian datasets, which have shown a negative association between primary care and ED use.^36,37 Building on this work, this study highlights that the length of continuity—specifically relationships lasting two years or more—plays an important role in facilitating timely management of minor health concerns outside the ED.

Long-term patient–provider relationships support more appropriate care-seeking, complementing advantages such as higher health literacy and educational attainment. Higher income was also independently associated with lower ED use, reflecting the role of socioeconomic resources in facilitating access to outpatient and preventive care. The analysis further highlights the importance of access: Seniors reporting fewer barriers to primary care were less likely to rely on the ED, regardless of care continuity. These findings suggest that continuity of care is most effective at reducing ED use when paired with timely access to primary care services. Policies that strengthen both continuity and access—such as longitudinal family practice models—may therefore be particularly effective in minimizing unnecessary ED visits among older adults.

Limitations

Despite these strengths, several limitations should be acknowledged. First, the cross-sectional design of the CHSS prevents causal inference, so the direction of associations between continuity of care and ED use cannot be confirmed. Second, age was not further categorized beyond 65+ due to the structure of the dataset, limiting the ability to explore potential variation in care-seeking patterns across older subgroups. Third, the CHSS relies on self-reported data and is not linked to administrative health records, meaning that actual healthcare visits cannot be independently verified. However, self-reported data offer valuable insights into patient experiences, perceptions, and barriers to care—dimensions that are often absent from administrative data. Fourth, the survey excludes certain populations—residents on reserves or other Aboriginal settlements, full-time members of the Canadian Forces, the institutionalized population, and residents of Nunavik and the Terres-Cries-de-la-Baie-James regions of Quebec—affecting generalizability. Fifth, ethnicity data were not collected; while household income and education serve as partial proxies, they may not fully capture structural disparities. Sixth, minor health problems were self-reported, introducing potential recall or reporting bias. In this study, “minor health problems” are defined according to CHSS wording (eg, fever, sprains, minor burns, cuts, rashes, or other nonlife-threatening conditions), with more severe or emergent issues implicitly excluded. Although the models presented include controls for self-perceived health and reported access to care, these variables only partially account for overall health differences and do not capture clinical severity. Finally, the category “some other place, valid skip, don’t know, refusal, not stated” is heterogeneous and cannot be meaningfully interpreted, though it was retained to preserve data completeness and transparency.

Conclusion

Taken together, these findings reinforce existing evidence on the value of sustained patient-provider relationships and highlight the importance of strengthening longitudinal continuity in primary care, particularly for an aging population with complex health needs. Initiatives that promote ongoing relationships between patients and family physicians—such as British Columbia’s LFP payment model—have the potential to reduce unnecessary emergency department visits and enhance the appropriateness of care. Broadening the implementation of similar models across provinces, alongside targeted strategies to address provider shortages and improve access in underserved areas, could further advance both the continuity and comprehensiveness of care for older adults.