Skip to main content
NCBI home page
Neuro-Oncology Practice logoLink to Neuro-Oncology Practice
. 2023 Dec 14;11(2):178–187. doi: 10.1093/nop/npad076

Choreographed expansion of services results in decreased patient burden without compromise of outcomes: An assessment of the Ontario experience

Kathryn Rzadki 1, Wafa Baqri 2, Olga Yermakhanova 3, Steven Habbous 4, Sunit Das 5,6,✉
PMCID: PMC10940827  PMID: 38496909

Abstract

Background

Neuro-oncology care in Ontario, Canada has been historically centralized, at times requiring significant travel on the part of patients. Toward observing the goal of patient-centered care and reducing patient burden, 2 additional regional cancer centres (RCC) capable of neuro-oncology care delivery were introduced in 2016. This study evaluates the impact of increased regionalization of neuro-oncology services, from 11 to 13 oncology centers, on healthcare utilization and travel burden for glioblastoma (GBM) patients in Ontario.

Methods

We present a cohort of GBM patients diagnosed between 2010 and 2019. Incidence of GBM and treatment modalities were identified using provincial health administrative databases. A geographic information system and spatial analysis were used to estimate travel time from patient residences to neuro-oncology RCCs.

Results

Among the 5242 GBM patients, 79% received radiation as part of treatment. Median travel time to the closest RCC was higher for patients who did not receive radiation as part of treatment than for patients who did (P = .03). After 2016, the volume of patients receiving radiation at their local RCC increased from 62% to 69% and the median travel time to treatment RCCs decreased (P = .0072). The 2 new RCCs treated 35% and 41% of patients within their respective catchment areas. Receipt of standard of care, surgery, and chemoradiation (CRT), increased by 11%.

Conclusions

Regionalization resulted in changes in the healthcare utilization patterns in Ontario consistent with decreased patient travel burden for patients with GBM. Focused regionalization did not come at the cost of decreased quality of care, as determined by the delivery of a standard of care.

Keywords: glioblastoma, radiation, regionalization, travel burden

Glioblastoma (GBM) is a rare and aggressive primary brain tumor, affecting 4 in 100 000 Canadians.1 It most often presents in older adults, with a median age of onset of 64 years.2 The management of GBM with surgery and adjuvant chemoradiation (CRT) has been well-established.3 Under standard of care, patients receive surgery followed by a 6-week regimen of adjuvant CRT with a maintenance regimen of chemotherapy for 6 months, otherwise referred to as the “Stupp Protocol.”4 A short-course regimen has also been proven efficacious in treating the older GBM population previously deemed unfit for full-course treatment.5

Despite the benefits in survival and sustained quality of life offered by adjuvant therapy, less than half of the GBM patients in Ontario receive CRT as part of their treatment.6 While this finding is likely multifactorial, it may reflect barriers that prevent or impair access to neuro-oncology care services. Under the aegis of Canada’s single-payer universal healthcare system, patients may choose to receive care from any physician or clinic across the country, with no restriction on the number of specialists a patient can see.7 Geographical access to healthcare services, however, may limit a patient’s choice,8,9 especially in regions with disproportionate population distributions, such as Ontario. These disparities may be expected to be most severe in settings in which a patient has been diagnosed with an uncommon and complex disease like GBM, for which treatment requires significant expertise and infrastructure that makes localized efforts to deliver care difficult.

Speaking specifically to these situations, regionalization is the deliberate reorganization of cancer services based on explicit and planned processes and structures with the intent of improving the quality of care.10 In Ontario, neurosurgical care is heavily centralized to a select few hospitals across the province.11,12 This approach to resource allocation has been widely supported in the literature, where patients undergoing complex surgical treatment have been shown to have lower morbidity and mortality if care is provided at high-volume centers.12,13–20 The administration of systemic and radiation therapies for advanced and rare cancers has followed a similar pattern of allocation, with radiation and oncology specialists often concentrated within urban or academic centers.21,22 However, in contrast to surgery, the volume–outcome relationship for systemic and radiation services is less evident.10,23 Unlike single surgical events or follow-up, systemic and radiation therapies are associated with prolonged treatment, often requiring sustained daily travel.10 Traveling long distances for cancer care is a barrier to cancer treatment, and has been associated with decreased utilization of radiation therapy and worse treatment outcomes.24–27 Patients of older age, with low socioeconomic status, advanced cancer diagnosis, and those residing in rural populations are especially vulnerable to the effects of travel burden.28–34 Thus, centralization of radiation services may pose a challenge for some GBM patients seeking care in Ontario.

The provision of funding and allocation of cancer services across the province, including the implementation of policies such as regionalization, is overseen by a governing authority, Ontario Health (Cancer Care Ontario) (OH (CCO)).10 In an effort toward health equity and person-centered care, OH(CCO) expanded the regionalized allocation of neuro-oncology care services to include 2 additional Central Nervous System Regional Cancer Centres (CNS-RCC) in 2016. These RCCs were provided the resources to administer CRT to GBM patients under the supervision of specialized radiation and medical oncologists. In this study, we evaluate the impact of this expansion of neuro-oncology care delivery on travel burden and quality of care delivered to GBM patients in Ontario.

Methods

Study Design and Data Sources

We performed a retrospective population-based study to examine the treatment, travel, and healthcare utilization patterns of patients diagnosed with GBM between January 1, 2010, and December 31, 2019, in Ontario, Canada. Information on GBM incidence and treatment modalities were obtained from the following provincial health administrative databases: the Ontario Cancer Registry (OCR) database, which captures details regarding incident cancer cases; the Ontario Health Insurance Plan (OHIP) database, which contains all physician claims for insured services; the Ontario Drug Benefits (ODB) program database, which contains data on therapeutics funded by the provincial public drug funding system; the New Drug Funding Program (NDFP) database, which provides information on the usage of newer injectable cancer drugs administered in hospitals and cancer centers; the activity level reporting (ALR) database, which includes information on patient activity within the cancer system, specifically radiotherapy, systemic therapy, and outpatient oncology visits; the National Ambulatory Care Reporting System (NACRS), which contains data for all hospital-based and community-based ambulatory care, including day surgery, outpatient and community-based clinics, and emergency departments; and the Canadian Institute for Health Information Discharge Abstract Database (DAD), which contains administrative, clinical, and demographic information on hospital discharges, including deaths, sign-outs, and transfers.

The Postal Code Conversion File Plus (PCCF+) links postal codes to census geographic regions in Canada.35 Geospatial and road network data pertaining to Ontario’s census divisions as of the 2016 census were obtained from Statistics Canada.36 Centers providing radiation therapy were identified from a list maintained by OH(CCO).37 The point location of each cancer center was determined with latitude and longitude using Google Maps 3.35.

These studies were performed under the auspices of OH(CCO) within its mandate for quality improvement.

Study Population

Patients diagnosed with GBM between January 1, 2010 and December 31, 2019 were identified through the OCR database. Individuals with an invalid health insurance number, a non-Ontario postal code at the time of diagnosis, or who were under the age of 18 at the time of the diagnosis were excluded. Patients were also excluded if they had not accessed the healthcare system within 6 months of diagnosis, based on the absence of physician billing from OHIP. Patients were separated into two cohorts according to year of diagnosis (2010–2015 and 2016–2019) based on the OH(CCO)-mandated expansion of neuro-oncology services in 2016.

Mapping Patient and CNS-RCC Locations

Patients’ place of residence was obtained using residential postal codes at the time of diagnosis, identified through the OCR database and PCCF + (version 7a). The PCCF + links a postal code with a dissemination block face, dissemination block, dissemination area, and the longitude and latitude of a center point of those units are used as the point location of a postal code.38 Patients were excluded if their file postal code was incomplete (<6 characters), invalid, or for a non-residential address (eg, large volume receiver, PO Box, or business address). In instances where a postal code spans several dissemination areas, the point was imputed using the weighted conversion methodology based on census 2016 population counts (Statistics Canada, 2021).

Thirteen CNS-RCCs were identified and used in the analysis, including the 2 new centers added in 2016.37 Eight CNS-RCCs were classified as Group A and considered academic centers based on their affiliation with Ontario medical schools: Sunnybrook Health Sciences Centre (SBK), University Health Network (UHN), Hamilton Health Sciences Centre (HHSC), The Ottawa Hospital (TOH), London Health Sciences Centre (LHSC), Health Sciences North—Laurentian (HSN), Thunder Bay Regional Health Sciences Centre (TBRHSC), and Kingston General Hospital (KGH).39 The remaining 5 CNS-RCCs fell into Group B as general hospitals with over 100 beds: Grand River Hospital (GRH), Sault Ste. Marie (SSM), Windsor Regional Hospital (WRH), Royal Victoria Hospital (RVH), and Trillium Health Partners—Credit Valley Site (THP) (see Supplementary Appendix 1).39 Patients who received care at an RCC not classified as a CNS-RCC were excluded from the study. The CNS-RCC locations were mapped based on the institution address. The latitude and longitude of the institution address were manually obtained from Google Maps with the precision of 7 decimal points. We defined the localization index (LI) as the proportion of patients within a CNS-RCC service area who received treatment at a local center. We defined the market share index (MSI) as the proportion of patients treated at a CNS-RCC who are within that CNS-RCC’s service area.

Travel Time to CNS-RCCs

The exposure variable in this study was travel burden, defined as one-way travel time from a patient’s residence to the closest RCC. Travel times were calculated using the 2016 Census Road Network File36 and ArcGis version 10.6 software at OH(CCO). Patient locations were mapped to the closest entry point to the road network. Travel time calculations considered speed limits, one-way travel, avoidance of tolls, and employed a hierarchy on the type of road (eg, highways preferred over arterial roads).40 Patients were excluded if they had no access to the road network.

A total of 4484 (96%) of the patients’ locations were within 500 m of the road network and 169 (4%) more patients’ locations were found within 2 km (1.24 mi) of the road. For those that were not found, the search radius was extended to 5 km (3.11 mi). Locations for 8 patients were not identified on the road network and were excluded from further analysis. The final cohort included 4645 patients.

CNS-RCC catchment areas were estimated using the distance that can be traveled by car along roads in all directions around each center. The catchment areas were split into the following ranges: 0–15 min, 15–30 min, 30–60 min, 60–90 min, and 90–120 min. Patients with a distance of more than 120 min were not included due to a small cohort size.

Outcomes

The primary outcome was receipt of standard of care (surgery followed by CRT). The OHIP, DAD, and NACRS databases were used to identify receipt of surgery within 1 year of the diagnosis date (Supplementary Appendix 2, surgical codes; Supplementary Appendix 3, surgery date definition). Surgery dates identified in DAD and NACRS took precedence over those identified using OHIP. Receipt of radiation within 12 months of diagnosis was identified through the ALR database. Finally, receipt of systemic therapy was identified through a search in the ALR, NDFP, and ODB databases for agents with known antineoplastic activity, including chemotherapy, immunotherapy, hormonal therapy, and targeted therapy, within 12 months of diagnosis. Additional searches were conducted in DAD and NACRS for evidence of antineoplastic treatment provided in a hospital setting (see Supplementary Appendix 4 for codes). Secondary outcomes were assessed for patients who received radiation as part of treatment and included treatment at the closest CNS-RCC and receipt of radiation within 6 weeks of diagnosis.

A subgroup analysis was performed to assess the relationship between travel time to the nearest CNS-RCC and receipt of radiation. For the patients who received radiation, a second subgroup analysis assessed travel times to treatment CNS-RCC as well as utilization patterns across CNS-RCCs. In order to evaluate changes in travel patterns and receipt of radiation following the introduction of the new CNS-RCCs in 2016, patients were divided into 2 cohorts according to the year of diagnosis (2010–2015 vs. 2016–2019).

A third analysis was performed to assess the relationship between travel times to a treatment CNS-RCC and wait times for radiation. The date of the first appointment at the CNS-RCC was considered as the radiation start date. A comparison of wait times across CNS-RCCs was also conducted.

Statistical Analysis

Continuous variables were reported as median with interquartile range (IQR) and categorical variables as absolute number (n) with proportion (%). Comparison testing was performed by the Kruskal–Wallis, Wilcoxon Signed Rank, and Kolmogorov–Smirnov tests. Statistical significance was defined as P ≤ .05 and all analyses were two-sided. Statistical analyses were performed using Statistical Analysis Software (SAS). All maps were created with the geographic information system (GIS) software, ArcGis version 10.6 software.

Ethics Approval

This work was undertaken at OH(CCO), under the aegis of quality improvement. Research Ethics Board (REB) approval was obtained through Clinical Trials Ontario to allow student involvement in the work (CTO Project ID: 139; Identifying barriers to completion of adjuvant therapy in patients with newly diagnosed glioblastoma multiforme: an exploratory study).

Results

GBM Incidence and Treatment Modalities (2010–2019)

A total of 5242 patients were diagnosed with GBM in Ontario from 2010 to 2019 (n = 2989 from 2010 to 2015; n = 2253 from 2016 to 2019). Standard of care (surgery with CRT) was the most common treatment regimen (57%), followed by surgery with adjuvant radiation (11%), surgery only (13%), CRT only (5%), radiation only (3%), and other treatment types and combinations (3%). A total of 249 (5%) patients had no evidence of any antineoplastic treatment (Supplementary Appendix 5). Data on treatment modality were obtained through the OCR.

Overall, 4522 (86%) of patients received surgery for treatment of GBM, most often combined with another treatment modality, such as CRT or radiation therapy, than surgery alone, and 4117 (79%) of patients received some form of radiation treatment (78% from 2010 to 2015; 79% from 2016 to 2019) (Supplementary Appendix 5). After 2016, the overall proportion of patients who received surgery combined with some form of adjuvant radiation increased slightly from 67% to 69%; significantly, the proportion of patients who received surgery with adjuvant CRT increased from 52% to 63% (Supplementary Appendix 5).

Access to CNS Radiation Services in Ontario

There were significant differences in travel times to the closest CNS-RCC from patients’ residences who received treatment with radiation, treatment without radiation, and no treatment (P = .03) (Table 1). Median travel time to the closest CNS-RCC was longer for patients who did not receive treatment than for patients who did (26 min vs 22 min), and the difference increased as travel distance increased (90th percentile, 72 min vs 84 min) (Table 1). While median travel time to the closest RCC is significant, the absolute difference makes it unlikely that travel time solely accounts for patients’ decision to reject treatment with RT. It is probable that other factors also contribute to the patient’s decisions.

Table 1.

Median Travel Time to Closest CNS-RCC by Treatment Modality and Year of Diagnosis. A Total of 5242 Patients in the Cohort were Grouped According to Whether or Not They Received Radiation Services as Part of Their Treatment. Differences in Travel Times Between 2010–2015 and 2016–2019 Were Also Evaluated. T-Tests Were Used to Determine Significance, (P<.05)

Treatment modality by year of diagnosis Number of patients, N Median travel time to closest RCC (IQR) in minutes P-value
2010–2019
Treatment with RT 4117 22 (13–41), 72 .03*
Treatment without RT 876 23 (14–44), 76
No treatment 249 26 (14–50), 84
2010–2015
Treatment with RT 2338 22 (13–42), 72 .005
Treatment without RT 494 24 (14–43), 78
No treatment 157 30 (17–58), 85
2016–2019
Treatment with RT 1779 21 (14–41), 71 .90
Treatment without RT 382 23 (13–45), 75
No treatment 92 23 (12–46), 74
Open in a new tab

RT—radiation therapy.

*Significant difference in travel time between treatment modalities using the Kolmogorov–Smirnov test.

For patients diagnosed 2010–2015, those who received no treatment had a longer travel time (median 30 min; 90th percentile 85 min) than patients receiving radiation (median 22 min; 90th percentile 72 min) or those receiving some other treatment (median 24 min; 90th percentile 78 min) (Table 1). After the expansion of radiation services in 2016, the difference in travel times to the closest CNS-RCC between the patient groups was resolved (P = .90). Overall, there was little difference in travel time to the closest CNS-RCC when comparing patients diagnosed in 2010–2015 and 2016–2019.

Patient Distribution Across CNS-RCC Service Areas

There was no significant change in the distribution of patients across CNS-RCC catchment areas after the expansion of radiation services in 2016. RCC-SBK is located in the most populated area with 24% of all GBM patients, followed by the new center RCC-THP (15%) and RCC-TOH (12%) (Supplementary Appendix 6). Only 358 (7%) patients resided in the service area of RCC-RVH. One percent of the patients resided within the remote areas around RCC-TBRHSC (n = 61) and RCC-SSM (n = 55), which serve as the only hospitals within 120 min of patients residing in these service areas. A total of 127 patients (2%) lived more than 120 min away from the nearest CNS-RCC, and the proportion of remote patients did not change with the introduction of additional radiation services in 2016 (Figure 1, Supplementary Appendix 6).

Figure 1.

Figure 1.

Open in a new tab

Distribution of GBM patients from the closest CNS-RCC from 2010 to 2019. Service areas shown represent the areas around a CNS-RCC within 120 min driving time. In places of overlap, the area was assigned to the closest CNS-RCC. The distribution of patients within each CNS-RCC service area was similar between 2010–2015 and 2016–2019.

Observed Travel Patterns to CNS-RCCs for Radiation Services

Of the patients who received radiation as part of their treatment (n = 4117), the median travel time to their treatment CNS-RCC was 31 (IQR, 16–55) min, with 10% of patients traveling over 93 min for care (Table 2). Median travel time to treatment CNS-RCCs decreased by 3 (IQR, 0–6; 90th percentile, 9) min following the expansion of services in 2016 from the travel time calculated in 2010–2015 (Table 2).

Table 2.

Travel Time to Treatment CNS-RCC for Patients Receiving Radiation Therapy (2010–2019)Median Travel Time to Treatment CNS-RCC for Patients Who Received Radiation as Part of Their Treatment was 31 Minutes. Travel Time to Treatment CNS-RCC Decreased Significantly, Following the Expansion of Services in 2016 (Wilcoxon, P= = .0052; Kolmogorov–Smirnov, P= = .0072)

Travel time (minutes) to treatment CNS-RCC Number of patients, N 25th Percentile Median Mean 75th Percentile 90th Percentile
All patients receiving RT (2010-2019) 4117 16 31 46 55 93
Patients diagnosed between 2010-2015 2338 16 31 47 58 96
Patients diagnosed between 2016 and 2019 1779 16 28 44 52 87
Open in a new tab

Of the patients who resided within 120 min of a CNS-RCC and received radiation, 62% (n = 1409) received radiation at the nearest CNS-RCC in 2010–2015, which increased to 69% (n = 1192) in 2016–2019 (Figure 2, Supplementary Appendices 8–9). The difference between the median travel times to the closest CNS-RCC and the CNS-RCC where patients actually received treatment was 9 (IQR, 3–12) min; however, the median difference between the travel times was 0 (IQR, 0–8) min, since the majority of patients chose the closest center as the center of treatment (Table 3). The median difference in travel times for the patients who were treated at a CNS-RCC other than the closest center to their homes was 17 (IQR, 8–39) min, which decreased slightly from 19 min in 2010–2015 to 16 min in 2016–2019 (Wilcoxon P = .0113, Kolmogorov–Smirnov P = .0442) (Table 3). Of the patients residing more than 2 h away from a CNS-RCC (n = 103), only 48% were treated at the closest center to their homes. The median additional travel time for the remaining patients who did not receive care at their local center was 51 (IQR, 37–87) min (Table 3).

Figure 2.

Figure 2.

Open in a new tab

Proportion of patients treated at a local CNS-RCC. Proportions of patients treated at the closest CNS-RCC who reside within 0–15 min, 15–30 min, 30–60 min, and 90–120 min from the closest center in (A) 2010–2015 and (B) 2016–2019 after the addition of 2 new CNS-RCCs. Delivery of local care increased from 62% in 2010–2015 to 69% in 2016–2019. In total, 65% of patients were treated at their local CNS-RCC.

Table 3.

Difference in Travel Time to Closest CNS-RCC Compared to Treatment CNS-RCC, for GBM Patients Who Received Radiation (2010–2019). Travel Times to a Patient’s Local CNS-RCC Were Compared to the Travel Times to the Treatment CNS-RCC, Where Radiation Services Were Provided. Patients Were Divided into Two Groups According to Whether or Not They Lived Within a 120-Minute Driving Time to the Closest CNS-RCC. Bypass Patterns for Patients Who Were Treated at a CNS-RCC Other Than the One Closest to Them Were Also Examined. The Difference in Travel Times for Patients Residing Within <120 min, Who Bypassed Their Local CNS-RCC, Were Somewhat Significant (Wilcoxon P = .0113; Ansari–Bradley P = .253; Kolmogorov–Smirnov P = .0442, P < .05)

Number of patients, N (%) 25th Percentile Median Mean 75th Percentile 90th Percentile
Travel time (minutes) for all patients residing < 120 min of closest RCC
Travel time to closest RCC 4014 16 30 41 52  80
Travel time to treatment RCC 4014 13 21 40 40  66
Difference in travel time between closest vs. treatment RCC 4014 0 0 11 8  34
Patient bypass patterns
Difference in travel time for patients who bypassed local RCC, residing < 120 min of closest RCC, n = 4041 1413 (35%) 8 17 32 39  61
Difference in travel time for patients who bypassed local RCC, residing > 120 min of closest RCC, n = 103 49 (48%) 37 51 81 87 225
Open in a new tab

RCC—Regional Cancer Centre.

CNS-RCC Utilization by Service Area

The CNS-RCCs fell into 3 groups according to utilization: those who treat mostly patients from their catchment area, but service a low proportion of their catchment area patient population [n = 3; RVH (LI = 16%, MSI = 94%), THP (LI = 22%, MSI = 89%), and GRH (LI = 14%, MSI = 85%)]; those treating the majority of the patients in their catchment area, and mostly patients in their catchment area [n = 7; HSN (LI = 92%, MSI = 87%), WRH (LI = 93%, MSI = 96%), TOH (LI = 97%, MSI = 97%), SSM (LI = 71%, MSI = 100%), LHSC (LI = 95%, 74%), TBRHSC (LI = 96%, MSI = 100%)]; and centers treating the majority of the patients in their catchment area, but also accepting a lot of patients from other hospitals [n = 3; UHN (LI = 83%, MSI = 23%), HHSC (LI = 95%, MSI = 54%), and KGH (LI = 86%, MSI = 65%)] (Supplementary Appendix 7). SBK (LI = 64%, MSI = 73%) fell in between categories servingthe population in its service area, with almost a quarter of its volume coming from other service areas (Supplementary Appendix 7).

Prior to the expansion of services in 2016, the primary treatment center for patients residing in the new CNS-RCC service areas were academic centers. SBK (n = 74, 43%) and UHN (n = 64, 38%) treated the majority of patients in the RVH service area, and UHN (n = 182, 54%), HHSC (n = 86, 25%), and SBK (n = 51, 15%) were the main destinations for the THP patients (Table 4). With an introduction in 2016, the new CNS-RCCs, RVH treated 35% (n = 38) of the patients residing in its service area and THP treated 41% (n = 113) (Table 4). The remaining patients in their service areas traveled further for care at the academic center.

Table 4.

Changes in Hospital Utilization Patterns in RVH & THP Service Areas Between 2010–2015 and 2015–2019. (Left) Prior to 2016, the Majority of Patients Residing in the RVH Service Area (81%) Received Care at Academic Centers. Following 2016, RVH Provided Radiation Care to 35% of its Service Area. The Remainder Traveled Outside of the Service Area, Most Often UHN and SBK (65% Bypass). (Right) Prior to 2016, the Majority of Patients Residing in the THP Service Area (94%) Received Care at Academic Centers. Following 2016, THP Provided Radiation Care to 41% of its Service Area, With the Remainder Traveling Further for Care to HHSC, UHN, and SBK (59% Bypass).

Treatment CNS-RCC 2010–2015 2016–2019
N % N %
RVH 7 4 38 35
SBK 74 43 28 26
UHN 64 38 21 20
Other CNS-RCCs 25 15 20 19
Total 170 107
HHSC 86 25 37 13
SBK 51 15 35 13
THP 21 6 113 41
UHN 182 54 91 33
Total 340 276
Open in a new tab

Delivery of Adjuvant Therapy

Of the patients who received adjuvant radiation (n = 3653), 81% received CRT and 19% received radiation therapy (Supplementary Appendix 10). Overall, 85% of the patients who received adjuvant radiation, received it within 6 weeks of surgery (Supplementary Appendix 10). This proportion varied across CNS-RCCs. Generally, at high-volume hospitals, 85% or more of their patients started adjuvant radiation treatment within 6 weeks of surgery [eg, TOH (93%, n = 451), UHN (92% n = 756), and SBK (91%, n = 742)], whereas 70% or less of the patients started adjuvant radiation within the 6-week window at hospitals with less than 200 patients enrolled in adjuvant radiation [eg, 68% (n = 15) at GRH, 47% (n = 24) at TBRHSC, and 52% (n = 16) at SSM] (Supplementary Appendix 11–12).

Discussion

In this study, we report on travel and healthcare utilization patterns for patients in Ontario diagnosed with GBM and analyze changes in treatment patterns and demands associated with a tailored expansion of care. Over the course of the decade, treatment for GBM varied across single and multi-modal approaches. Although there was a consistent volume of patients receiving radiation as part of treatment, we observed a shift in favor of the delivery of standard of care (surgery with CRT) compared to surgery and radiation alone. Despite this trend, the volume of all GBM patients receiving adjuvant CRT was lower than expected (57%). Although medical factors may influence treatment eligibility, these volumes suggest that other barriers may be preventing patients from accessing these services. Prior to 2016, patients who elected to not receive treatment had longer travel times to their local RCC, implying that centralized allocation of systemic therapies introduces a travel burden as a barrier for some GBM patients seeking care in Ontario. Fortunately, the efforts made to increase accessibility of services for neuro-oncology patients was associated with several positive implications on patient travel and care delivery. As more patients received care at their local RCC, patient travel time decreased. The greatest decrease in travel time was observed for patients already requiring significant travel for care, suggesting that rural populations, already facing disparities in access to care and outcomes across Canada,41–43 benefited the most from expansion.

While the literature supports the conclusion that patients are willing to travel to high-volume centers for their surgical cancer care,44–46 our findings suggest that patients who require radiation therapy of their cancer favor local care: after 2016, the proportion of GBM patients receiving treatment at the closest CNS-RCC increased from 62% to 69%. Given the strain of daily travel and the clinical and demographic profile of GBM patients, it is not surprising that patients more often choose the nearest center for care.47,48

Interestingly, our findings identified a notable proportion of patients (35%) who bypassed their local RCC and traveled further for radiation, typically to an urban academic center. Conclusions drawn from the literature exploring hospital bypassing decisions are mixed. One study found that over 20% of prostate cancer patients bypassed their local hospital to receive radiation at an academic center,49 whereas a study on patients with ovarian cancer reported that nearly 20% of ovarian cancer patients preferred not to travel to referral cancer centers, even when aware of the survival benefits of doing so.50 Interestingly, a national cross-sectional study from the Netherlands reported that, regardless of the travel distance to the hospital, patients with rare cancers seem to deliberately search for the best available cancer care, while patients with more common cancers have a lower incentive to search for care beyond their regional hospital.51 Our results, along with the literature on patient preference for travel to systemic or radio-oncology care, suggest there are intrinsic motives behind patient treatment decisions that influence their travel patterns.49–54

As we move toward the goal of person-centered care, maintaining the integrity of choice becomes a critical element in preserving patient autonomy.55 We observed 2 types of GBM patients in our study: those who appeared to prioritize ease of access to care services, perhaps as a way to preserve quality of life or to alleviate caregiver burden, and a second type who actively traveled further to an academic center, likely under the perception of higher quality of care delivery or to participate in a clinical trial. Although our analysis did not account for all factors guiding patient treatment decisions (eg, preference for academic centers, physician referral patterns, enrollment in clinical trials, utilization of lodging services, or relocation for treatment), increased regionalization provided more patients, especially in rural Ontario, with the opportunity to receive care at a center that better aligned with their treatment goals.

Administration of systemic and radio-oncology services for rare or aggressive cancers by specialized oncologists and at academic centers have been favored for their ability to provide tailored care for unique cases or their capacity to quickly respond to adverse events that come with the administration of aggressive treatments. Interestingly, a recent survey of 420 South Korean oncologists revealed low endorsement for referring to high-volume centers and proposed that rare cancer treatments be administered in regional cancer centers rather than in centralized clinics.56 Our results demonstrate that quality and timely delivery of care can be maintained while expanding the scope of clinical access in the province. We observed an 11% increase in the number of patients who received standard of care following 2016, likely as a result of fewer patients selecting palliative radiation and opting instead for more active and aggressive therapy with the accessibility of care delivery. Although the increase in CRT delivery may largely be due to practice changes following the introduction of short-course CRT in 2017,5 the utilization levels of new RCCs indicate they were nonetheless advantageous in supporting the delivery of standard of care.

Centralization of any health service bears the risk of increasing volumes to levels that exceed institute and clinician capacities, which may have impeding consequences such as increased wait times and delays in starting systemic or radiation therapy.10,57 Likewise, there is a risk that a shift in patient volumes could have a detrimental effect on the quality of care that lower-volume institutions can offer to patients who choose or need to be treated there.10 The expansion of services at the new CNS-RCCs not only reduced these risks but also provided over 77% of patients with radiation within 6 weeks of surgery, in line with the current recommendations.58 Although there was no observed correlation between patient travel and wait times for radiation, there was significant variation in wait times across the RCCs. Higher volume CNS-RCCs with large corresponding catchment areas treated a greater proportion of patients within 6 weeks compared to lower-volume hospitals with smaller catchment areas. Our finding suggests that higher volume CNS-RCCs may be inherently better positioned for success, with greater capacity and infrastructure to treat a larger volume of patients at a given time. It would be advantageous to continue to monitor these trends and to evaluate the distribution of resources or referrals in a way that best supports the centers experiencing volume pressure.

There were several limitations to our study. We did not evaluate travel time as a barrier to treatment compliance or overall survival, nor did we assess the relationship between center volume and patient outcomes. We also did not evaluate data specific to patient’s regions and situation, such as the number of linear accelerators (LINACs) operating in each region of Ontario or region-specific population density, which may limit extrapolation of the findings. Future studies exploring these factors, as well as patient perspectives on treatment goals and decision-making processes, would provide additional insight into patient motivations and the implications of travel time and regionalization of radiation and systemic care on patient outcomes.

In conclusion, increased regionalization of neuro-oncology care delivery decreased the travel burden for GBM patients by facilitating ease of access to services while maintaining the integrity of patient choice and quality of care delivered. The results of this study validate efforts undertaken by OH(CCO) to expand regionalized cancer services in Ontario. Our findings may inform decision-makers in other jurisdictions who face analogous concerns of access and quality.

Supplementary material

Supplementary material is available online at Neuro-Oncology (https://academic.oup.com/neuro-oncology).

npad076_suppl_Supplementary_Appendixs
npad076_suppl_supplementary_appendixs.docx (3.9MB, docx)

Contributor Information

Kathryn Rzadki, Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada.

Wafa Baqri, Temerty Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada.

Olga Yermakhanova, Ontario Health (Cancer Care Ontario), Toronto, Ontario, Canada.

Steven Habbous, Ontario Health (Cancer Care Ontario), Toronto, Ontario, Canada.

Sunit Das, Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada; Division of Neurosurgery, Department of Surgery, St. Michael’s Hospital, University of Toronto, Toronto, Ontario, Canada.

Conflict of interest

S.D. serves as the Provincial Lead for CNS Oncology at Ontario Health (Cancer Care Ontario).

Funding

K.R. is supported by the Ontario Graduate Scholarship. S.D. is supported by an Early Researcher Award from the Province of Ontario.

Data Availability

Parts of this material are based on data and information compiled and provided by CIHI. However, the analyses, conclusions, opinions, and statements expressed herein are those of the author, and not necessarily those of CIHI. Parts of this publication are based on data provided by ICES. However, the views expressed in this publication are those of the researcher and do not necessarily represent those of ICES. This report was produced with the support of the Ontario Ministry of Health. However, the views expressed herein are those of the author, and not necessarily those of the Ontario Ministry of Health or the Government of Ontario.

References

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

npad076_suppl_Supplementary_Appendixs
npad076_suppl_supplementary_appendixs.docx (3.9MB, docx)

Data Availability Statement

Parts of this material are based on data and information compiled and provided by CIHI. However, the analyses, conclusions, opinions, and statements expressed herein are those of the author, and not necessarily those of CIHI. Parts of this publication are based on data provided by ICES. However, the views expressed in this publication are those of the researcher and do not necessarily represent those of ICES. This report was produced with the support of the Ontario Ministry of Health. However, the views expressed herein are those of the author, and not necessarily those of the Ontario Ministry of Health or the Government of Ontario.

Articles from Neuro-Oncology Practice are provided here courtesy of Oxford University Press

RESOURCES