Highlights
-
• Rural residents with Parkinson’s disease had lower virtual visit rates to neurologists, family physicians and geriatricians/psychiatrists than urban residents.
-
• Virtual care use increased early during the pandemic but later declined, suggesting sustainability challenges.
-
• Administrative data can provide real-world evidence of virtual care patterns and highlight equity gaps.
Introduction
Parkinson’s disease (PD) is considered the fastest-growing neurological disorder worldwide, beyond population aging alone. Reference Bloem, Okun and Klein1,2 Healthcare use and costs associated with PD are substantial, Reference Yang, Hamilton and Kopil3–Reference Maclagan, Marras and Sewell5 and its prevalence has increased over time. As of 2021, over 11 million individuals globally are living with PD, a 273% increase since 1990, with the burden expected to double by 2050. Reference Luo, Qiao, Li, Wen, Zhang and Li6,Reference Su, Cui and He7 In Canada, similar trends have been observed. Most people living with PD reside in urban areas. Reference Maclagan, Marras and Sewell5 National surveillance data indicate that the number of Canadians living with parkinsonism, including PD, increased from approximately 61,000 in 2004–2005 to 84,000 in 2013–2014 and is expected to continue rising with population aging. 8 In Ontario, among 18,510 prevalent cases in 2000, 86% lived in urban areas; by 2018, among 33,479 cases, approximately 89% were urban, and age-sex standardized prevalence was 14% lower in rural compared to urban areas (401 vs. 467 per 100,000). Reference Maclagan, Marras and Sewell5 Despite lower prevalence in rural compared with urban areas, rural living and pesticide exposure remain risk factors for PD. Reference Ben-Shlomo, Darweesh, Llibre-Guerra, Marras, San Luciano and Tanner9,Reference Gorell, Johnson, Rybicki, Peterson and Richardson10 Consequently, providing high-quality care to people living with PD across Canada will be increasingly challenging.
Discrepancies in care access between urban and rural regions have been reported for people living with PD in Ontario, Canada’s largest province. Reference Maclagan, Marras and Sewell5 A recent population-based study showed unequal access to overall care between the two groups, including lower rates of outpatient visits and rehabilitation admission among individuals living in rural regions compared to urban, along with higher rates of emergency department visits and long-term care admissions. Reference Maclagan, Marras and Sewell5 Another population-based study in Ontario suggested that people living with PD in urban areas had higher access to neurologists compared to rural areas. Reference Crighton, Ouédraogo, Sawada and Mestre11 PD is a multi-system disease with classical motor but also non-motor features, such as mood disturbances and autonomic dysfunction, all causing impairment in quality of life. Reference Schapira, Chaudhuri and Jenner12 Access to specialists other than neurologists, such as psychiatrists and geriatricians, is poorly described in the PD literature.
Virtual care, or telemedicine, referring to any remote interaction between patients and healthcare providers using communication technologies, including telephone and video-based care, gained increased popularity before the global pandemic. Reference Shaw, Jamieson, Agarwal, Griffin, Wong and Bhatia13 However, during the pandemic, given the need for physical distancing, expansion of virtual care was accelerated to maintain healthcare access for the population during lockdowns and public health restrictions. This rapid expansion of virtual care provided great hope with respect to ensuring care access for those living far from hospitals and healthcare providers or with significant mobility difficulties. Reference Dorsey and Topol14
Several studies indicated that virtual care was associated with a high degree of satisfaction and was beneficial to people living with PD, Reference Wilkinson, Spindler and Wood15,Reference Chen, Guan and Li16,Reference Xu, Feeney and Surface17 in part because it reduces travel burden, improves convenience and allows care to be delivered from home. However, these studies evaluated patient experiences, rather than access to care, which is a separate concept. Furthermore, prior studies had limited generalizability, as most lacked detailed demographic information about their participants who were often recruited from academic centers, foundation mailing lists or convenience samples. Reference García-Bustillo, Youn and Ahn18 As a result, it is unclear whether virtual care access is equitable across populations or as readily available to individuals in rural versus urban regions. Moreover, as the global pandemic has subsided, the sustainability of virtual care remains uncertain, and longitudinal data on the uptake and maintenance of use of virtual care in rural versus urban communities are limited.
Population-based health administrative data provide unique insights into understanding access to virtual care. Ontario’s single-payer healthcare model enables data collection on all individuals who interact with the healthcare system, regardless of their involvement with clinical research teams, which are often based in academic centers in urban areas. These data thus allow a broader representation of people living with PD and bring a complementary perspective to clinical databases. To our knowledge, no prior studies used population-based health administrative data from a single-payer healthcare system to systematically evaluate the trends in use of virtual care in PD during versus after the COVID-19 pandemic in Ontario, Canada.
In this study, we evaluated the trends in virtual care use over time among people living with PD in Ontario, Canada, by urban versus rural region of residence using population-based health administrative data.
Methods
Study design and population
This was a repeated cross-sectional study using population-based health administrative data in Ontario, Canada, conducted at approximately 6-month intervals between March 1, 2020 (index date) and August 26, 2023. Although the first recorded COVID-19 case in Ontario occurred in January 2020, March 2020 marked the onset of community transmission and was chosen as the index date. The study cohorts included community-dwelling people living with PD aged 66 years and older who were alive and eligible for health insurance coverage at the start of each week after the index date. Individuals living in nursing homes were excluded. PD was ascertained using a validated health administrative data definition: 3 physician visits for PD within 2 years with a minimum of 30-day gap between each claim. Reference Bronskill, Maclagan and Maxwell19 Individuals were censored upon death, admission to a nursing home or loss of health insurance eligibility, whichever occurred first.
Data sources
Demographic characteristics such as age, sex, death date (if applicable) and residential postal codes were obtained via the Ontario Registered Persons Database, a registry of all Ontario residents eligible for health insurance. Data on hospital admissions and emergency department visits were obtained using the Canadian Institute for Health Information Discharge Abstract Database and National Ambulatory Care Reporting System, respectively. Physician billing information was obtained via the Ontario Health Insurance Plan (OHIP) physician claims database, and dispensed medication information for the purpose of defining comorbidities was obtained via the Ontario Drug Benefit database (ODB). Data on nursing home residence were obtained via the Continuing Care Reporting System–Long-Term Care, ODB and OHIP databases. Information on primary care physician enrollment models and provider rostering was obtained via the Client Agency Program Enrolment database. Home care services were obtained via the Home Care Database. These datasets were linked using unique encoded identifiers and analyzed at ICES (formerly the Institute for Clinical Evaluative Sciences). ICES is an independent, nonprofit research institute whose legal status under Ontario’s health information privacy law allows it to collect and analyze healthcare and demographic data, without consent, for health system evaluation and improvement. The use of the data in this project is authorized under section 45 of Ontario’s Personal Health Information Protection Act and does not require review by a Research Ethics Board.
Exposure
The exposure of interest was region of residence, urban versus rural. Rurality was obtained from the Postal Code Conversion File Plus (PCCF+) version 2021, through linkage to residential postal codes. 20 Using Statistics Canada’s indicator of rural and small-town status, rural residence was defined as community size with a population size of ≤10,000 or in municipalities outside of the commuting zones of larger urban centers. 20,Reference Du Plessis, Bollman and Clemenson21 Other individuals were grouped as urban residents.
Outcomes
The primary outcomes were physician visits including total, virtual and in-person visits to family physicians, neurologists and geriatricians and/or psychiatrists to reflect the complex, multidisciplinary nature of PD management. Reference Weise, Claus and Dresel22 Rates of visits to the latter two specialties were very low and therefore combined in analyses. Physician claims associated with telephone or video consultations using OHIP fee codes were classified as virtual visits, while all other visits (e.g., office or home visits) were classified as in-person. During the pandemic, temporary “K codes” were introduced in March 2020 to facilitate care delivery, including telephone and video visits, and were expanded during the pandemic period. These temporary codes remained in use until November 2022, after which new comprehensive billing codes were implemented. 23–26 These updated codes preserved insured virtual care services for primary care physicians and specialists while introducing distinctions between comprehensive and limited virtual care.
The virtual care definition also incorporated applicable Ontario Telemedicine Network B-codes submitted with corresponding physician service codes. These codes were captured as additional billing elements associated with virtual care claims and were not counted as separate visits. The OHIP billing codes used in our study to identify virtual visits are provided in Supplementary Table 1.
Covariates
Baseline demographic characteristics were captured at the study index date including age, sex, neighborhood-level income quintile, the material resources dimension of the Ontario Marginalization Index (ON-Marg) (v2016) and comorbidities. Income quintile was captured using an area-based measure of household size-adjusted income, obtained by linking individual postal codes to census data from Statistics Canada. The material resources dimension of the ON-Marg reflects one aspect of marginalization and includes indicators such as the percentage of single-parent households, low employment and education, reliance on government financial support, low income and housing. Reference Matheson, Dunn, Smith, Moineddin and Glazier27 Comorbidities were identified among a list of 17 chronic conditions captured using validated health administrative data algorithms. Reference Mondor, Maxwell and Hogan28
Primary care enrollment models, implemented by the Ontario government in the 2000s to improve primary care delivery, reflect different physician payment and care delivery structures in Ontario. Reference Hutchison, Levesque, Strumpf and Coyle29 These models were examined to understand whether the type of primary care model influenced use of virtual care. This variable was categorized as capitation (fixed fee per individual enrolled), enhanced fee-for-service, family health teams, individuals not rostered to a primary care physician, other enrollment group and physicians not in an enrollment model as has been done in other studies. Reference McLeod, Buckley and Sweetman30 In addition, continuity of primary care was captured using the Usual Provider of Care Index, which represents relational continuity, defined as the extent to which a patient consistently visits the same provider over time. It was calculated as the ratio of the frequency of visits to a individual’s main family physician to the frequency of visits to all family physicians over a 2-year period and was categorized as low (<0.4), medium (0.4–0.8) and high (>0.8) continuity. 31 Recent registration for health insurance (<10 years prior to the index date) was used as a proxy for immigration. Health system utilization in the previous year included any acute care hospitalizations, emergency department visits, home care, family physician, neurologist and geriatrician/psychiatrist visits.
Time periods
The start of the study period matched the onset of the COVID-19 pandemic in Ontario, Canada (March 1, 2020). Given the fast pace of change in the healthcare system as the pandemic evolved, the study period was divided into subperiods of 26 weeks (approximately 6 months) until the end of the study period (August 26, 2023).
Statistical analyses
People living with PD were stratified based on residing in urban or rural regions as of each 6-month index date within each repeated cross-sectional time period. Baseline characteristics were compared between these two groups using standardized differences (Std Diff), with values above 0.1 considered meaningful differences. For each subperiod, the proportion of individuals with at least one physician visit and weekly rates of visits were calculated separately for family physician visits, neurologist visits and geriatrician/psychiatrist visits and by visit type (virtual vs. in-person). Weekly rates were defined as the number of physician visits per individual per week within each subperiod, stratified by specialty and visit type (numerator), among all individuals with PD who were alive, living in the community and OHIP eligible in each urban and rural group (denominator).
We used Poisson regression models with generalized estimating equations accounting for repeated measures to estimate rate ratios (RR) with 95% confidence intervals (CIs) comparing rates of virtual visits among people living with PD in urban versus rural regions. Models were adjusted in a stepwise fashion. Model 1 included age, sex and time period (to adjust for changes in visits over time as the pandemic progressed) as covariates. Model 2 further included income quintile, primary care enrollment model and healthcare utilization (visits to family physicians, neurologists, psychiatrists or geriatricians) in the previous year, as additional covariates. We also estimated models as described above, coding urban/rural region and time period as a single categorical variable in order to describe changes across regions over time (compared to urban regions during the first 0–6 months of the pandemic as the reference group). All analyses were conducted using SAS v9.4 (SAS Institute Inc., Cary, N.C.).
Data sharing statement
The dataset from this study is held securely in coded form at ICES. While legal data sharing agreements between ICES and data providers (e.g., healthcare organizations and government) prohibit ICES from making the dataset publicly available, access may be granted to those who meet pre-specified criteria for confidential access, available at www.ices.on.ca/DAS (email: das@ices.on.ca). The full dataset creation plan and underlying analytic code are available from the authors upon request, understanding that the computer programs may rely upon coding templates or macros that are unique to ICES and are therefore either inaccessible or may require modification.
Results
Baseline characteristics
Among 24,736 community-dwelling people living with PD on March 1, 2020, 22,249 (89.9%) resided in urban regions and 2487 (10.1%) in rural regions (Table 1). People living with PD in urban regions were older and lived in neighborhoods associated with lower levels of material resources marginalization. Individuals in urban regions had a higher number of chronic conditions compared to those in rural regions (Table 1). The sex distribution, PD duration and neighborhood income quintiles were comparable between the two groups.
Characteristics of community-dwelling older adults aged 66 years and older living with Parkinson disease in Ontario, Canada, on March 1, 2020, by region of residence
Table 1. Long description
The table presents the sociodemographic and clinical characteristics of community-dwelling adults aged 66 years and older with Parkinson’s disease in Ontario, as of March 1, 2020, stratified by region of residence. The table contains 43 rows and 5 columns. The columns are: Characteristics, Total, Urban, Rural, and Standardized Differences. Characteristics include age, sex, income quintiles, Ontario Marginalization Index quintiles, chronic conditions, recent health insurance registration, healthcare utilization, primary care enrollment models, and continuity of care, along with their respective categories. Categorical variables are reported as counts and percentages, while continuous variables are reported as means. The Total, Urban, and Rural columns describe the distribution of each characteristic in the overall population and by region of residence. The Standardized Differences column reports differences in the distribution of characteristics between urban and rural residents.
* Bold figures indicate meaningful differences (SD < 0.10).
†FFS = fee for service; PEM = patient enrollment model; SD = standard deviation.
‡Material resources (increasing quintiles represent increasing marginalization on material resources: 1 = most material resources; 5 = least material resources).
§ Approximation due to small cells.
‖Individuals who had no core primary care fee codes for 2 years prior to the index date.
¶ Chronic conditions were identified among a list of 17 chronic diseases captured using validated algorithms: stroke, rheumatoid arthritis, dementia, congestive heart failure, osteoarthritis, mood disorders, other mental health disorders, asthma, renal failure, chronic obstructive pulmonary disease, osteoporosis, diabetes, cardiac arrhythmia, acute myocardial infarction, chronic coronary syndrome, hypertension and cancer.
Most individuals had seen a family physician in the year prior to the index date, but there was a higher proportion of individuals in rural regions without a family physician (Table 1). Continuity of care was similar between the groups. A higher proportion of individuals with multimorbidity (five or more chronic conditions) was observed in urban regions. Despite this observation, there was a lower proportion of individuals in urban regions visiting the emergency room in the year prior to the index date. Individuals in urban regions were more likely to see family physicians, neurologists and geriatricians or psychiatrists in the year prior to the index date.
Rates of virtual visits in urban versus rural regions
Differences between urban and rural regions were most pronounced for virtual visits across all specialties, while in-person visit rates were lower during the study period, consistent with the pandemic context (Figure S1). Compared to people living with PD in urban regions, those living in rural regions had significantly lower rates of virtual visits to neurologists during the study period (March 2020–August 2023) (RR = 0.93; 95% CI: 0.90–0.96). The relative difference was greater for family physicians (RR = 0.73; 95% CI: 0.70–0.77) and geriatricians/psychiatrists (RR = 0.67; 95% CI: 0.65–0.70) (Table 2).
Adjusted rate ratios for virtual physician visits associated with sociodemographic and clinical characteristics among community-dwelling older adults aged 66 years and older living with Parkinson’s disease in Ontario, Canada, by specialty (2020–2023)
Table 2. Long description
The table presents adjusted rate ratios for virtual physician visits among community-dwelling adults aged 66 years and older with Parkinson’s disease in Ontario from 2020 to 2023. It has 33 rows and 7 columns, with rows showing model parameters and columns showing results for family physicians, neurologists, and psychiatrists/geriatricians across two models. Variables include region, time period, age, sex, income quintile, primary care enrollment model, and healthcare utilization in the previous year. Model 1 adjusts for age, sex, region, and time period, while Model 2 additionally adjusts for income, primary care enrollment, and previous-year healthcare utilization. Reference categories are indicated by “ref.”, and adjusted rate ratios with corresponding confidence intervals are reported. Older age had fewer specialist visits but more family physician virtual visits; females and higher-income groups had higher rates of virtual family physician visits.
aRRs = adjusted rate ratios; CIs = confidence intervals; FFS = fee for service; PEM = patient enrollment model.
Adjusted rate ratios (aRRs) represent comparisons between each category and its corresponding reference category, indicated by “ref.” All models included rurality and the listed sociodemographic and clinical covariates.
* Model 1: Adjusted for age, sex, region of residence and time period; Model 2: Adjusted for age, sex, region of residence and time period, income quintile, primary care enrollment model, healthcare utilization in the previous year (family physician, neurologist and psychiatrist/geriatrician visits).
† p-value above 0.05.
$This variable was recategorized in the model to improve computational efficiency and enhance performance.
Rates of virtual visits in specific subpopulations
Older individuals (aged 75 years and older) had lower rates of virtual visits to neurologists and geriatricians/psychiatrists compared to those aged 66–74 years (Table 2). The opposite was true for virtual visits to family physicians. Females had higher rates of virtual visits than men. Individuals living in the highest income neighborhoods had higher rates of virtual visits to neurologists and family physicians compared to those living in the lowest income neighborhoods, but no such associations were observed for visits to geriatricians/psychiatrists (Table 2).
Rates of virtual visits over time
Figure 1 illustrates a decline in weekly rates of virtual visits to all physicians across the observation window, while total physician visits remained stable over time. Compared to the first 6 months after the onset of the COVID-19 pandemic, the use of virtual visits to neurologists remained stable from 6 to 12 months after the start of the pandemic but gradually declined until the end of the observation window (Table 2). The decline in the rates of virtual visits was most pronounced after public health measures were lifted in Ontario around March 2022 (24 months after the start of the pandemic). Between 24 and 30 months, the rate of virtual neurologist visits was 46% lower than at the start of the pandemic (RR = 0.54; 95% CI: 0.51–0.57). By 36–42 months, the rate of virtual neurologist visits had decreased by 77% (RR = 0.23; 95% CI: 0.22–0.25). The rate of virtual visits to family physicians and geriatricians/psychiatrists increased up to 18 months after the start of the pandemic and then gradually declined until the end of the observation window. Compared to the first 6 months of the pandemic, the rate of virtual visits at 36–42 months was 61% lower for family physician visits (RR = 0.39; 95% CI: 0.38–0.41) and 55% lower for geriatricians/psychiatrist virtual visits (RR = 0.45; 95% CI: 0.40–0.51).
Weekly rates of virtual and in-person physician visits among community-dwelling older adults aged 66 years and older living with Parkinson’s disease in Ontario, Canada, on March 1, 2020, by region of residence.
Figure 1. Long description
The figure contains six line graphs showing weekly rates of physician visits per 100 population among community-dwelling adults aged 66 years and older with Parkinson’s disease in Ontario from March 2020 to September 2023. Results are shown for urban (blue) and rural (orange) residents and for family physicians, neurologists, and psychiatrists/geriatricians. For each specialty, virtual visits are shown in the left panel and total visits (virtual and in-person) in the right panel. Following March 2020, virtual visit rates increased sharply, particularly for family physicians, and generally declined thereafter. Urban residents generally had higher virtual family physician visit rates than rural residents. Total visit rates were higher among urban residents for family physicians and psychiatrists/geriatricians, while differences were smaller for neurologists.
The proportion of total virtual visits was higher among individuals living in urban regions compared to those in rural regions across physician groups. Virtual visits to neurologists declined at a similar rate over time in both urban and rural regions (Table S3). Virtual family physician visits initially increased in urban regions and then declined toward the end of the observation period. A similar pattern was seen for geriatricians/psychiatrists. In contrast, virtual visits to family physicians continuously declined in rural regions. A similar pattern was seen for geriatricians/psychiatrists.
Total visits to physicians (virtual and in-person visits)
Considering both virtual and in-person visits, individuals in rural areas had overall lower rates of visits to their family physicians, geriatricians and psychiatrists compared to those in urban areas throughout the observation period (Supplementary Table 2 and Figure 1). In the first 6 months of the pandemic, 74.6% of individuals in the urban areas had at least one visit to their family physician versus 65.9% in the rural areas (Std Diff = 0.19) (Supplementary Table 2). Moreover, for the same period, 10.8% of individuals in urban areas had a visit to a geriatrician or psychiatrist versus 6.1 % in the rural areas (Std Diff = 0.17). Similar percentages were seen across the observation window (Supplementary Table 2). This was not the case for visits to neurologists, which were comparable between individuals in urban and rural areas, 52.4% versus 47.8%, respectively, in the first 6 months of the pandemic (Std Diff = 0.09). The frequency of visits overall remained stable throughout the observation window for all physicians (Figure 1).
Discussion
Our study highlights the differences in virtual care use among people living with PD across urban and rural regions in Ontario, as well as its declining use over time. People living with PD in rural regions had lower rates of virtual care compared to those residing in urban regions. Moreover, the decline over time in virtual care use was more pronounced for individuals in rural regions, especially after Ontario lifted public health measures related to COVID-19. In addition, we also found that individuals in rural areas had fewer interactions with family physicians, geriatricians and psychiatrists overall compared to individuals in urban areas, whether virtual or in-person. This finding stayed unchanged throughout the observation window and may highlight a gap in care access for individuals in rural regions.
Our findings are surprising, given that individuals in rural regions most likely live farther from hospitals and physician offices and may face greater geographic barriers to accessing care. Reference Maclagan, Marras and Sewell5 Thus, we might have expected a higher uptake and maintenance of virtual visits in these areas, given the greater travel distances for people living with PD and their caregivers to reach their physician offices. To our knowledge, this is the first study to use population-based data to characterize longitudinal trends in virtual visits over an extended period in repeated cohorts of people living with PD in Ontario, Canada. Most published studies examining the uptake of virtual care in people living with PD share one commonality: they involve active voluntary participation of people living with PD, whether through surveys, observational studies or randomized controlled studies, Reference Wilkinson, Spindler and Wood15,Reference Xu, Feeney and Surface17,Reference Sekimoto, Oyama and Hatano32 and are most often conducted in teaching hospitals. However, there is evidence to suggest that these individuals are different from the wider PD population Reference Beaulieu-Jones, Frau and Bozzi33,Reference Priti Gros, Wang, Chiu, Farkouh, Emdin and Bronskill34 from both a demographic and a disease progression perspective. Reference Priti Gros, Wang, Chiu, Farkouh, Emdin and Bronskill34 In a study comparing people living with PD seen within the first 3 years of disease in teaching hospitals versus other settings, differences with respect to age, income neighborhood and comorbidities were observed. Those seen in teaching hospitals were younger, lived in higher income neighborhoods and had fewer comorbidities. Individuals also had more subsequent PD-related interventions and lower incidence of complications such as dementia, nursing home admissions and death. Reference Priti Gros, Wang, Chiu, Farkouh, Emdin and Bronskill34 Population-based studies are necessary to capture all individuals. Prior work has not studied the trends in virtual care over time as we transitioned through major changes in the healthcare system (peri- and post-COVID-19 pandemic). Therefore, our study provides a unique perspective on these important questions.
Similar findings in the general population showed higher uptake of virtual care in specific subgroups. A recent systematic review across higher-income countries including Canada, evaluated the use of virtual care in the general population in the first 2 years of the COVID-19 pandemic. Reference Hatef, Wilson and Hannum35 Findings suggested that virtual care was more frequently used by young to middle-age females, whites, those with higher socioeconomic status and those living in urban settings. Reference Hatef, Wilson and Hannum35,36 In other words, virtual care may improve healthcare access for those who already have access, 36 but may widen inequities for subpopulations that have had historically poor access to care. 36 These include older individuals, rural populations, people requiring translation services, racialized populations, individuals experiencing homelessness or those with low income. 36 The barriers to access virtual care are numerous, including poor digital literacy, low level of education, unreliable internet connections or digital devices, high cost, difficulty with building meaningful patient–physician relationships and concerns regarding data privacy. 36 Other barriers that are unique to people living with PD are motor and cognitive symptoms that may hamper the use of digital technologies. Reference Laar, Silva de Lima, Maas, Bloem and de Vries37
Ontario virtual care billing codes changed on December 1, 2022, with the new billing system encouraging physicians to use virtual care when a patient–physician relationship already exists, which includes but is not limited to having had an in-person visit within the past 24 months. 38 Although this change in physician billing may partly explain the decrease in virtual visits, it does not account for the significant decrease in virtual care that started before this change was implemented and does not explain the difference in virtual care observed between urban and rural regions. The decline in virtual care reflects two distinct phases: an initial rapid uptake of virtual care during the COVID-19 pandemic followed by a period of stabilization and a subsequent reduction coinciding with the December 2022 billing changes. Prior Ontario studies from the pre-COVID-19 era show that most visits to family physicians, neurologists, geriatricians and psychiatrists were in-person. Reference Bronskill, Maclagan and Maxwell19 Although virtual visits decreased over time, our study suggests that their use will likely remain an alternative mode of healthcare service delivery. Reference Bronskill, Maclagan and Maxwell19
Our study also demonstrates that individuals in rural areas interacted less frequently with family physicians, geriatricians and psychiatrists overall, compared to individuals in urban areas, across the observation window. This raises the question of overall care access to people living with PD in these regions.
Strengths
Our population-based datasets were prospectively collected from a single-payer healthcare system and therefore were agnostic to our study design. This decreased the likelihood of selection bias, reporting bias, recollection bias and differential information bias. PD was defined using a validated health administrative data case definition. Moreover, no studies have evaluated access to virtual visits with psychiatrists and geriatricians in this population. Reference Dobkin, Rubino, Friedman, Allen, Gara and Menza39
Limitations
Our study does not include a non-PD comparator group; therefore, we were unable to assess whether observed trends differed from those in the general population. We did not identify the root causes of differences in virtual care uptake between urban and rural regions among people living with PD, nor the decline in virtual care use over time, nor did we account for physician or individual preferences, particularly after more than a year of virtual care during the beginning of the pandemic. Differences in virtual care use between urban and rural regions may reflect underlying differences in overall access to healthcare services, which is generally higher in urban settings. As this study was descriptive in nature, we were not able to account for these differences in overall healthcare utilization. Therefore, further qualitative studies are needed to understand the lived experiences of people with PD with respect to virtual care. The validated PD case ascertainment used in our study has been shown to have a sensitivity of 72.3% and specificity of 100% and could therefore miss people living with PD, particularly in rural regions. Reference Bronskill, Maclagan and Maxwell19,Reference Maclagan, Visanji and Cheng40 Detailed clinical information such as PD severity was not available in the health administrative databases, which may be associated with difficulties using digital technologies. Other variables related to health inequities, such as ethnicity, education, health or digital literacy were not available in our datasets. Area-based markers, such as income, may not accurately represent individual-level measures, with a risk of ecological fallacy. We also did not capture distance between individuals’ homes and their care providers. Finally, individuals aged 65 and younger were excluded, and the pattern of virtual care use may be different in the younger populations.
Conclusion
Access to virtual care for people living with PD is heterogeneous across persons residing in urban and rural areas. People living with PD in urban centers accessed virtual care more frequently than those in rural regions. The utilization of virtual care decreased substantially following the start of the pandemic. Future studies should compare trends with the general population and explore barriers to virtual care access overall, especially among rural residents living with PD. Studies should also qualitatively explore individual and caregiver preferences with respect to PD care.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/cjn.2026.10691.
Acknowledgments
This document used data adapted from the Statistics Canada Postal CodeOM Conversion File (version 2024), which is based on data licensed from Canada Post Corporation, and/or data adapted from the Ontario Ministry of Health Postal Code Conversion File, which contains data copied under license from ©Canada Post Corporation and Statistics Canada. We thank IQVIA Solutions Canada Inc. for use of their Drug Information File, and we thank the Toronto Community Health Profiles Partnership for providing access to the ON-Marg. Parts of this material are based on data and/or information compiled and provided by the Canadian Institute for Health Information (CIHI), Ontario Health and the Ontario Ministry of Health. 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.
Author contributions
PG: Contributed to study design, execution, interpreting results, writing, editing the manuscript and finalizing manuscript. LCM: Contributed to study design, execution, results interpretation, revising, editing, finalizing manuscript. TO: Contributed to study design, execution, finalizing manuscript. ZL: Contributed to study execution, analysis, finalizing manuscript. JG: Contributed to study determined methods, analysis, finalizing manuscript. CM: Contributed to study execution, reviewing, finalizing manuscript. AEL: Contributed to study execution, reviewing, finalizing manuscript. LVK: Contributed to study execution, finalizing manuscript. RHS: Contributed to study design, execution, finalizing manuscript. SEB: Contributed to study research question development, design, execution, reviewing, finalizing manuscript, administration, funding.
Funding statement
This study received funding from the Ontario Neurodegenerative Disease Research Initiative (ONDRI) through the Ontario Brain Institute, an independent nonprofit corporation, funded partially by the Ontario government. This study was also supported by ICES, which is funded by an annual grant from the Ontario Ministry of Health and the Ministry of Long-Term Care. This work was also supported by the Ontario Health Data Platform, a Province of Ontario initiative to support Ontario’s ongoing response to COVID-19 and its related impacts.
Competing interests
PG: reports graduate student salary support received from Parkinson Canada and the Canadian Institute of Health Research, and speaker fees were received from AbbVie.
CM: reports funding from the Michael J Fox Foundation, the International Parkinson and Movement Disorders Society, the Academic Health Sciences Alternating Funding Plan Innovation Fund, the Parkinson’s Foundation and the Mayvon Foundation.
AEL: reports honoraria from Sun Pharma, AbbVie, Paladin and Sunovion and consulting fees from AbbVie, Amylyx, Aprinoia, Biogen, BioAdvance, Biohaven, BioVie, BlueRock, BMS, Denali, EG427, Janssen, Lilly, Pharma 2B, Sun Pharma and UCB; serves on the advisory board for Sunovion; and receives royalties for Movement Disorders 4, Elsevier, Saunders; Parkinson‘s Disease: Non-Motor and Non-Dopaminergic Features, Wiley-Blackwell; Parkinson‘s Disease: A Complete Guide for Patients and Families, Johns Hopkins Press; Common Movement Disorders Pitfalls, Cambridge University Press.
LVK: holds the Wolfond-Krembil Chair in Parkinson’s Disease Research. In the past year, L.V.K. has received research support from Canadian Institutes of Health Research, Cure Parkinson’s, Krembil Foundation, Parkinson Canada and University of Toronto; served as a consultant for Biogen, Merz, Novo Nordisk and UCB; and received honoraria from Canadian Movement Disorders Society, International Parkinson and Movement Disorder Society and IOS Press/Sage Publications.
RHS: reports support from the Bastable-Potts Chair in Stroke Research, Sunnybrook Research Institute and University of Toronto. RHS reports grants from the Ontario Brain Institute/ONDRI. RHS is an Advisory Board member for Hoffman La Roche and a shareholder in FollowMD Inc.
SEB: reports grants from the Alzheimer Society of Canada and the Canadian Institutes of Health Research.
The remaining authors declare that there are no conflicts of interest.
Open access
