Michelle Howard

Background: Family physicians provide most outpatient care in patients’ last year of life, but it remains unclear how geographic distance between patients and their physicians affects continuity of care. Patients living farther from care providers may experience reduced access and more fragmented care. This study examined whether geographic proximity to a family physician is associated with relational continuity of outpatient care in the last year of life. Methods: We conducted a population‑level retrospective cohort study of adults who died in Ontario, Canada, from 2021 to 2024, using linked health administrative data. Geographic proximity was measured using quartiles of the shortest driving distance (km) between each patient’s residence and their family physician’s practice location. Relational continuity was assessed using established measures capturing the concentration of visits, dispersion of care across physicians, and the temporal regularity of visits. Associations were estimated using proportional‑odds regression models adjusted for demographic, socioeconomic, and clinical factors. Results: Among 400,499 patients, the mean age at death was 76.0 years, 52.2% were male, the mean (SD) distance to their family physician was 25.5 (92.0) km, with the highest quartile residing more than 14.2 km away from their family physician. Greater distance was associated with modestly lower relational continuity: each one-quartile increase in distance was linked to lower odds of higher continuity with the usual family physician (cOR: 0.88, 95%CI: 0.87, 0.88). Sensitivity analyses using travel time and Euclidean distance produced consistent findings. Discussion: Geographic proximity is a significant determinant of outpatient continuity near the end of life. Patients living farther from their family physicians experience more fragmented care, underscoring the need for system-level strategies to reduce geographic barriers, including improved local access to primary care and enhanced virtual-care models.

Background: It is recognized that primary care clinicians have a critical role in ensuring access to palliative care in health systems. When primary palliative care has been examined at the system level, provision is highly variable, making the study of best practices challenging. Aim: The aim of this scoping review is to describe intervention components, implementation contexts, and evaluation approaches used in studies seeking to enhance palliative care delivery in primary care. Design: This scoping review followed the framework and stages outlined by Arskey and O’Malley, with a descriptive analysis. Data sources: Articles published from 2000 to 2025 were identified from MEDLINE, Embase, CINAHL, Cochrane Central Register of Controlled Trials, and Cochrane Database of Systematic Reviews. Studies that described interventions involving primary care clinicians and their adult patients with any medical condition and which were evaluated using controlled or uncontrolled, quantitative or qualitative methods, were included. Results: The search identified 8669 abstracts and 55 studies were included. Components of interventions included (1) specialist palliative care physician support ( n = 31), (2) multidisciplinary team-based care ( n = 36), (3) integration of new tools and resources ( n = 36), and (4) educational and training initiatives ( n = 39). Most interventions were multi-faceted. Regardless of patient population, support from palliative care specialists, multi-disciplinary roles and systematic approaches to identify patients were prevalent components. Conclusions: Supporting primary care clinicians to provide palliative care is achieved through coordinated efforts to mobilize resources within the primary care setting and intentional collaboration with external specialist teams.

Context: Arts‐based knowledge mobilisation (ABKM) is a powerful means of interpreting and communicating health services research, centreing experiential, iterative, and reflective knowledge rooted in co‐creation. We developed a research‐to‐public‐to‐research feedback loop methodology extending beyond existing ABKM practices and our previous research‐to‐public methodology, creating a novel bidirectional epistemic process wherein meaning making occurs from research to public and from public to research. This approach produces knowledge that is rich, contextually grounded, and otherwise potentially inaccessible through conventional research. Objectives: We present a case study documenting the first application of the feedback loop, with the aim of advancing and refining the methodological framework by: sharing quantitative findings with the public and eliciting public engagement via a design installation, and levering engagement to inform subsequent qualitative research. Methods: Using the design‐driven research‐to‐public‐to‐research feedback loop, we developed a participatory design installation focused on end‐of‐life care. Within an overarching mixed‐methods sequential study, we leveraged design and health services methods to: develop design principles; select installation aims and concept; prototype, refine, and iterate the design; and conduct descriptive analysis of installation outputs. Results: Borrowing characteristics of a classic diner, the installation entitled The Terminal Diner , invited individuals to complete orders for their desired end‐of‐life healthcare experience or leave reviews drawn from lived experience with a deceased loved one's end‐of‐life journey. The installation disseminated quantitative findings and engaged the general public in contemplating and co‐creating around the subject of death—outputs of which were used to inform subsequent qualitative research, while recruiting an interested audience as research participants. Conclusion: This study presents an innovative extension of our previous research‐to‐public methodology, representing a paradigmatic shift towards creative knowledge exchange and bidirectional meaning making within health services research. Patient Public Contribution: The public's role as co‐creators is foundational to the research‐to‐public‐to‐research feedback loop methodology. Those with lived experience—bereaved caregivers and patient partners—were also embedded throughout both the research and design processes, ensuring accessibility, relevance, and emotional resonance. Blending personal knowledge and public perspective, The Terminal Diner foregrounded the essential role of patient and public contributions in co‐creating meaningful, responsive, and inclusive health services research.

Patients with multiple sclerosis (MS) require increasing health care support, particularly in the last year of life. Despite the benefits of palliative care in addressing symptom burden and improving quality of life, its integration into MS care remains inconsistent. This study aimed to characterize health care utilization and palliative care access among people with MS in the last years of life and to identify key factors associated with hospital death and receipt of palliative care. A retrospective cohort study was conducted using linked health administrative databases in Ontario, Canada. The cohort consisted of all decedents aged 18 years or older who died in Ontario between April 2016 and March 2020, and we compared health care utilization in decedents with MS to the rest of the cohort. The identification of MS cases was based on a validated algorithm. Demographics, health care utilization, palliative care involvement, and medical interventions were analyzed. Predictors of palliative care receipt and hospital deaths were evaluated using multivariable logistic regression models. MS decedents (n = 1,975; mean age 68 years; 66.4% female) were younger and had fewer comorbidities than non-MS decedents (n = 500,904; mean age 77). High percentage of MS decedents had outpatient neurology visits in the last 5 years of life (49.4%) but a steep decline closer to death (2.7% in the last month). Palliative care use was similar (58.0% vs 56.9%) but occurred earlier in MS (61.9% vs 59.6% in the last 5 years). Multivariable regression showed that rural residence was associated with increased odds of hospital death (odds ratio [OR] 1.81, 95% CI 1.21-2.70) and lower odds of receiving palliative care (OR 0.74, 95% CI 0.55-0.99), while receiving palliative care within the last 5 years of life reduced the odds of hospital death (OR 0.47, 95% CI 0.37-0.60). Higher comorbidity and increased outpatient visits to urology were associated with a greater odd of receiving palliative care (OR 3.64, 95% CI 2.50-5.29 and OR 1.47, 95% CI 1.20-1.80). Although palliative care receipt was comparable between MS and non-MS decedents, rural disparities and high hospital deaths persist. Earlier palliative integration, particularly through neuropalliative care, could improve quality of life and reduce hospitalizations.

The number of older adults aged 50 years and above who are incarcerated is increasing rapidly around the world. Providing high quality healthcare for aging incarcerated populations requires tailored services and approaches that consider their unique needs in carceral environments. However, there is limited research on healthcare from the perspective of relevant interest-holders, which challenges the development of interventions to address unmet needs and improve healthcare quality. This study aimed to understand the perspectives of interest-holders on the healthcare considerations of aging prison populations. We carried out secondary reflexive thematic analysis of data from interviews and focus groups with three categories of interest-holders: people with lived experience of incarceration (including people who were incarcerated at the time of data collection), community-based advocates and researchers (including people who work in community organizations, academics, and clinicians), and correctional healthcare providers and leaders. Data were collected from 61 participants between August 2022 and November 2023. We identified four themes: need for better data tracking and comparison, healthcare needs conflict with the carceral environment, physical constraints challenge access to care, and questioning the suitability of older adults for incarceration. This study highlights factors to consider in addressing healthcare for aging incarcerated populations, as described by various individuals with a vested interest in understanding and improving correctional health and health services. Findings can be used to inform ongoing work to improve healthcare and address unmet needs for aging people in prison.

Background: Continuity of care (COC) measurements that reflect relational continuity have been used as quality indicators, yet their applicability near the end of life may be limited. Modified continuity indices—UPC-Team (Usual Provider of Care) and BB-Team (Bice-Boxerman)—were developed to reflect escalating care needs and capture associations with patient-centered outcomes. Objectives: To measure associations between the modified COC indices, UPC-Team, and BB-Team during the last year of life and end-of-life (EOL) health care outcomes. Methods: Retrospective cohort study of adults who died between January 1, 2018, and December 31, 2022, with advanced chronic obstructive pulmonary disease and/or heart failure prevalent ≥2 years before death, using health administrative data from Ontario, Canada. Multivariate regressions measured associations between the indices and days spent in community during the last 30 and 14 days of life, and place of death. Results: Among 175,323 included individuals (median age at death=80; 55.4% male), the median number of community days was 23 and 10 in the last 30 and 14 days of life; 56.5% died in a health care institution. Higher UPC-Team and BB-Team scores were associated with increased odds of institutional deaths and fewer community days. Conclusions: Higher continuity scores were associated with increased odds of institutional death and fewer days spent in the community, suggesting limited utility of these modified indices in predicting favorable EOL health care outcomes. Findings highlight the need for future research to incorporate all aspects of continuity (ie, relational, informational, and management) to better capture care coordination in this context.

Background: Family physicians (FP) are critical in providing outpatient end-of-life (EOL) care, yet most continuity measures do not consider temporal changes in visits. The relative variance index (RVI), which measures visit interval variation, has not been applied to the EOL context. Aim: To measure the association between outpatient FP visit regularity & acute care use in the last month of life. Design: Retrospective cohort study using linked population-level health administrative data. FP visit regularity was measured using the RVI, calculated from outpatient visits during the last two years of life (excluding the last month). Outcomes included hospitalizations, emergency department (ED) visits in the last month of life & acute care setting deaths. Setting: Adults with cardiorespiratory conditions who died in Ontario, Canada, between 2017 & 2019. Results: Patients' (N = 151 030) median age at 2-years before death was 80, 55% were male, & had a median of 9 FP outpatient visits in the last two years of life. Higher FP RVI scores were associated with increased hospitalizations (IRR [95% CI]: 1.07 [1.05,1.10]), ED visits (1.06 [1.03, 1.08]) & acute care deaths (OR [95% CI]: 1.30 [1.26, 1.35]). Sensitivity analyses identified that RVI scores varied across different look-back periods, while sensitivity model performances remained stable when adjusting for specialist visit regularity. Conclusions: Higher outpatient FP visit regularity was associated with increased EOL acute care use, despite low visit regularity during the last two years of life. However, the RVI's sensitivity to observation period lengths limits its utility as an EOL care-quality indicator.

Objectives: To develop and describe the performance of two modified relational continuity of care indices for adults with organ failure nearing the end of life, addressing limitations of existing indices that penalise multispecialty physician care. Design: Population-based retrospective study. Setting: Linked routinely collected health administrative databases in Ontario, Canada. Participants: Adults aged ≥19 years who died between 1 January 2018 and 31 December 2022, with kidney failure on dialysis, advanced chronic obstructive pulmonary disease and/or heart failure. Primary and secondary outcome measures: We adapted the Usual Provider of Care (UPC) and Bice-Boxerman Continuity of Care (BB) indices to avoid penalising multispecialty physician care, resulting in the UPC-Team and BB-Team indices. Indices were calculated for the last 2 years of life (truncating the last month) using outpatient physician visits. Correlation coefficients were produced between the unmodified indices, the modified indices and outpatient healthcare utilisation. Results: The cohort included 199 035 individuals, with a median age of 79 years (P25, P75: 70, 86); 55.5% were male. The median modified continuity scores were higher than the unmodified scores (UPC=0.44 (0.32, 0.61), UPC-Team=0.79 (0.67, 0.89); BB=0.24 (0.15, 0.40), BB-Team=0.75 (0.52, 0.95)). The modified indices shared weaker correlations with outpatient healthcare use (ie, number of visits or specialists involved), compared with the unmodified indices. Conclusions: The UPC-Team and BB-Team indices do not penalise patients for receiving multispecialty physician care, compared with the unmodified UPC and BB indices. The modified indices demonstrated more consistency in measuring continuity across patients with varying degrees of medical complexity and less correlation with outpatient healthcare use. Use of these indices is suggested to capture different aspects of relational continuity in the context of multispecialty care nearing the end of life.

This multiple case study explored how Long-Term Care (LTC) teams in Ontario, Canada, manage informational continuity when older adults transition from community-based care to LTC. Five LTC homes, varying in size and rurality, participated, with 20 professionals interviewed across various roles, including nursing, medicine, rehabilitation, and administration. LTC providers emphasized the importance of comprehensive, accurate, and up-to-date biopsychosocial information to support effective care. However, information transferred from community and hospital sources was often incomplete or outdated. To address gaps, LTC staff sought additional details from electronic health records, families, care coordinators, and hospitalists. Their ability to obtain missing information was influenced by organizational capacity, physician’s practice location, power dynamics between providers, availability of family caregiver support, and access to electronic health records. A stronger primary/hospital-LTC collaboration, incentives for informational continuity, and a specific staff managing transition information and activities could optimize the LTC transition information exchange process.

Objectives: To compare primary care for depression among patients detected in the first 21 months of the SARS-CoV-2 pandemic to patients detected pre-pandemic, and examine whether depression care was associated with patient characteristics. Design: Retrospective pre–post study using de-identified data from electronic medical record data from the Canadian Primary Care Sentinel Surveillance Network (CPCSSN). Setting: Primary care clinics enrolled in CPCSSN from 1 January 2018 to 31 December 2021. Population: Patients with a valid CPCSSN case definition of depression detected between 01 January 2018 and 31 December 2021 were included in the cohort and categorised by detection date (pre-pandemic or during Canadian pandemic waves). Outcome measures: Primary care encounters, psychotropic prescriptions and selective serotonin reuptake inhibitor (SSRI) prescriptions were observed at 3 and 12 months post-detection. Multivariable regression evaluated associations between patients’ depression detection timing and depression care, adjusting for age, sex, rurality, neighbourhood deprivation quintile, province, number of observable comorbidities and pre-detection psychotropic use. Results: 91 453 patients with depression were identified, of whom 53% were detected pre-pandemic. Patients detected during the pandemic were younger and less comorbid than those detected pre-pandemic. Proportions of patients with any encounter, psychotropic prescriptions and SSRI prescriptions were higher for patients detected during every pandemic wave compared with patients detected pre-pandemic. The adjusted incidence rate ratios (aIRRs) of number of encounters (aIRR=1.15; 95% CI 1.13 to 1.17), psychotropics (aIRR=1.11, 95% CI 1.09 to 1.13) and SSRIs prescribed (aIRR=1.12; 95% CI 1.10 to 1.15) within 3 months of detection were higher among patients detected during the first pandemic wave compared with those detected pre-pandemic. Socio-demographic characteristics had weaker associations with outcomes compared with timing of detection. Results were similar within 12 months of detection. Conclusion: Overall, primary care for depression was maintained during the pandemic despite challenging circumstances. Increases in paediatric encounter rates and increased prescribing in younger adults warrant further investigation to understand the factors driving these patterns. Trial registration number: ClinicalTrials.gov NCT05813652 .