Next Article in Journal
The Current Trend of Radiation Therapy for Patients with Localized Prostate Cancer
Previous Article in Journal
Upper-Limb Disability and the Severity of Lymphedema Reduce the Quality of Life of Patients with Breast Cancer-Related Lymphedema
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Current Oncology
Volume 30
Issue 9
10.3390/curroncol30090586
curroncol-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Addressing Inequity in Spatial Access to Lung Cancer Screening

by
Jonathan Simkin
1,*,
Edwin Khoo
2,
Maryam Darvishian
2,
Janette Sam
2,
Parveen Bhatti
3,4,
Stephen Lam
2 and
Ryan R. Woods
3,5
1
BC Cancer, Provincial Health Services Authority, Vancouver, BC V5Z 4C2, Canada
2
BC Cancer Screening, BC Cancer, Provincial Health Services Authority, Vancouver, BC V5Z 1G1, Canada
3
Cancer Control Research, BC Cancer Research Institute, Vancouver, BC V5Z 1G1, Canada
4
School of Population and Public Health, Faculty of Medicine, University of British Columbia, Vancouver, BC V6T 1Z3, Canada
5
Faculty of Health Sciences, Simon Fraser University, Burnaby, BC V5A 1S6, Canada
*
Author to whom correspondence should be addressed.
Curr. Oncol. 2023, 30(9), 8078-8091; https://doi.org/10.3390/curroncol30090586
Submission received: 4 August 2023 / Revised: 28 August 2023 / Accepted: 30 August 2023 / Published: 31 August 2023
(This article belongs to the Section Thoracic Oncology)
Browse Figures
Review Reports Versions Notes

Abstract

:
Background: The successful implementation of an equitable lung cancer screening program requires consideration of factors that influence accessibility to screening services. Methods: Using lung cancer cases in British Columbia (BC), Canada, as a proxy for a screen-eligible population, spatial access to 36 screening sites was examined using geospatial mapping and vehicle travel time from residential postal code at diagnosis to the nearest site. The impact of urbanization and Statistics Canada’s Canadian Index of Multiple Deprivation were examined. Results: Median travel time to the nearest screening site was 11.7 min (interquartile range 6.2–23.2 min). Urbanization was significantly associated with shorter drive time (p < 0.001). Ninety-nine percent of patients with ≥60 min drive times lived in rural areas. Drive times were associated with sex, ethnocultural composition, situational vulnerability, economic dependency, and residential instability. For example, the percentage of cases with drive times ≥60 min among the least deprived situational vulnerability group was 4.7% versus 44.4% in the most deprived group. Conclusions: Populations at risk in rural and remote regions may face more challenges accessing screening services due to increased travel times. Drive times increased with increasing sociodemographic and economic deprivations highlighting groups that may require support to ensure equitable access to lung cancer screening.

1. Introduction

Lung cancer is the most commonly diagnosed cancer and leading cause of cancer death in Canada [1] and worldwide [2]. In 2021, an estimated 29,600 Canadians were diagnosed with lung cancer and an estimated 21,000 died from this disease [1]. At 22%, the five-year net survival for lung cancer is among the lowest of all types of cancer [1]. Lung cancer screening with low dose computed tomography (LDCT) has been shown to significantly reduce lung cancer mortality. Three randomized clinical trials showed a 20% to 39% mortality reduction benefit of lung screening using LDCT versus chest X-ray or usual care [3,4,5]. Despite evidence supporting lung screening through LDCT and coverage by Medicare and most commercial insurers, screening uptake remains uneven and slow in the United States (US) [6].
An important issue in lung screening is addressing inequity by improving access to screening services to Indigenous peoples, socioeconomically deprived population groups, and those living in rural areas because of existing inequities [7,8,9,10,11]. For example, negative gradients in lung cancer risk for individual-level socioeconomic status have been reported in both Canada and the United States (US) [7,8,11]. Geographic variation has been reported at both national and provincial levels [1,9,11]. In Canada, First Nations people had lower survival for cancers of the lung and bronchus compared to non-First Nations people [10]. Differences in survival are related to stage of diagnosis, emphasizing the importance of access to screening to improve health outcomes. For example, in Canada, the three-year net survival for lung cancer diagnosed at stage 4 was 5% compared to 71% for lung cancer diagnosed at stage 1 [12]. In the US, studies suggest that while most of the screen-eligible population are within close proximity to a screening site, a small proportion has poor spatial access [13,14,15]. In Canada, increasing travel time and level of urbanization was associated with lower screening rates in breast mammography screening [16,17], and colorectal cancer screening [18,19,20,21].
To our knowledge, no studies have yet examined spatial access to fixed LDCT lung cancer screening sites by neighborhood measures of socioeconomic deprivation and few at the level of urbanization [15]. To address this gap in knowledge, we used a population-based cohort of lung cancer cases in British Columbia, Canada, as a proxy for the screen-eligible population and examined travel time, as an indicator of spatial access, to fixed lung cancer screening sites. Although eligibility for lung screening is driven by age and smoking history at present, targeting screening services to areas in relation to lung cancer incidence as a proxy may be an effective way to improve lung cancer outcomes. This is especially the case when accurate smoking history and other lung cancer risk factors used in a risk prediction tool, such as the PLCOm2012, are not available.
In Canada, health services are delivered at the provincial or territory level. British Columbia (BC) is Canada’s third largest province with an area approximately two times the size of the State of California in the United States. It is populated by over 5.1 million people [22]. BC health services are planned and delivered at various health administrative areas, with the largest consisting of five regional health authorities (HAs). Nested within HAs are 16 Health Service Delivery Areas (HSDA), 89 Local Health areas, and 218 Community Health Service Areas (CHSA) [23]. BC Cancer routinely reports screening indicators at HSDA regional levels. The CHSA health boundary was introduced by the Ministry of Health in 2019 to provide a standard geographic definition for community level analyses for health service planning [23].
We hypothesize that the majority of the screen-eligible population are within close proximity to a screening site. However, the minority with poor spatial access will also demonstrate greater socioeconomic deprivation.

2. Materials and Methods

2.1. Study Population and Setting

This study was approved by the University of British Columbia–BC Cancer Research Ethics Board.
Lung screening programs are slowly being implemented across Canada. British Columbia (BC) implemented an organized lung cancer screening program, using low-dose computed tomography (CT) scans that are carried out through fixed sites, in 2022. All British Columbians that are 55 to 74 years of age, current or former smokers, who smoked for at least 20 years with a six-years lung cancer risk >1.5% using the PLCOm2012 risk prediction tool [24] are eligible to participate in the provincial lung cancer screening program.
This was a retrospective population-based descriptive study. A total of 36 lung cancer screening sites around the province with the necessary CT facilities have been identified. As a proxy for the potential screen eligible population, we identified all newly diagnosed (incident) lung cancer cases in BC between 2015 and 2019 aged 55 to 80 years (N = 12,886) using the BC Cancer Registry. Cases up to 80 years were included in the study cohort to include age ranges for both Canadian and US recommendations. For each case, we obtained information on sex, age at diagnosis, histologic subtype, tumor stage, and complete residential postal-code at the time of diagnosis. In Canada, six-digit postal codes typically represent one side of a city block in urban areas, while rural area postal codes are larger.

2.2. Geocoding

The Statistics Canada Postal Code Conversion File Plus (PCCF+) was used to assign spatial locations (i.e., longitude-latitude coordinates) to cases from the complete postal codes captured by the cancer registry. Records with non-valid or missing postal codes were excluded (N = 198). Records were linked from spatial coordinates to the BC CHSA boundary map to determine BC administrative health areas (CHSA and HSDA areas) at diagnosis and the BC CHSA Urban-Rural Classification [23]. The CHSA Urban-Rural classification includes seven categories: Metropolitan, Large Urban, Medium Urban, Small Urban, Rural Hub, Rural, and Remote. These categories are generated by the BC Ministry of Health and derived using two Statistics Canada Census metrics: Population Centre and Rural Area Classification and Index of Remoteness. CT screening sites (N = 36) were provided by the BC Lung Screening Program and latitude and longitude coordinates for these sites were obtained using a standardized BC healthcare facility geolocating reference file [25]. The Statistics Canada Canadian Index of Multiple Deprivation [26] (CIMD) was linked to the cases at the government census dissemination area level (DA), which is reflective of the neighborhood level. CIMD data are census-derived and geographically indexed variables that enable analyses on sociodemographic and economic inequalities. They are publicly available by Statistics Canada with the intent to support policy planning and evaluation, research and analysis and resource allocation [26].
The BC-CIMD dataset [26] used in this analysis contains four neighborhood-level dimensions, each composed of various census measures: residential instability, economic dependency, ethnocultural composition and situational vulnerability (Supplementary Table S1). The CIMD variables were available in factor scores and quintiles. In this study, we use quintile rankings. Quintiles were derived by ordering factor scores for neighborhoods (N = 7617) from smallest to largest and then divided into quintiles categorized from 1 through 5. A value of 1 corresponds to neighborhoods that were the least deprived for that dimension, and a value of 5 corresponds to neighborhoods that were the most deprived. For ethnocultural composition, a value of 5 corresponds to the highest level of ethnocultural composition.

2.3. Travel Time

The outcome variable of interest was the shortest travel time, in minutes, by car from the location associated with a patient’s residential postal code at diagnosis to the nearest screening site. Travel time was calculated using the Open Source Routing Machine (OSRM) (Version: 5.27.1, The Free DSB Project, Vancouver BC, Canada) [27] and the OSRM R package (Version: 4.1.1, Centre National de la Recherche Scientifique, Paris, France) [28].

2.4. Statistical Analysis

Travel time was treated as both a continuous and categorical variable. We reported the median and interquartile ranges for travel time to the nearest CT center in minutes. Drive time categories were as follows: <20 min, 20–<40 min, 40–<60 min and ≥60 min. These categories were informed by those used in previous travel time studies [29,30].
A bivariate analysis of travel time categories with each variable of interest was conducted including: age, sex, histologic type, stage at diagnosis, level of urbanization and the four CIMD variables. A chi-square test of independence was used to assess associations between travel time and each variable of interest. A p-value ≤ 0.05 was considered statistically significant.
Geographic variation in drive time was assessed at the HSDA-level with box plots to provide high-level distributions of drive time among these relatively large health areas. Community level variation was explored by mapping median travel time at the CHSA-level. Using the Monte Carlo simulation method [31] (N = 999 simulations), a global Moran’s I statistic was calculated to test whether or not median drive times were clustered.

3. Results

Overall, 51.6% of lung cancer patients were female (N = 6553), and the median age at time of diagnosis was 70 years (interquartile range (IQR) = 64 to 75 years) (Table 1). The most common histologic type was non-small cell lung cancer (80.3%, N = 10,194). The lung cancers were most commonly diagnosed at stage IV (43.7%, N = 5539). Most patients resided in Metropolitan areas (36.0%, N = 4573), followed by Medium Urban (18.4%, N = 2332) and Rural areas (14.8%, N = 1884) (Table 1). When examining CIMD factors, the patients generally resided in more deprived areas with low ethnocultural composition, greater residential instability and greater economic dependency (Table 1).
The median travel time to the nearest screening site was 11.7 min (Interquartile Range (IQR)= 6.2 to 23.2 min). Overall, 69.8% of cases (N = 8856) were less than 20 min from a screening site. In contrast, 8.0% of cases (N = 1014) were over 60 min from a screening site (Table 2).
In general, male sex, lower neighborhood urbanization, lower neighborhood ethnocultural composition, lower neighborhood residential instability, greater neighborhood situational vulnerability and greater neighborhood economic dependency were associated with increased drive times (p < 0.05) (Table 3).
Although cases that were 60 min or greater from a screening site only represented 8.0% of all cases (Table 2), 44.2% resided in neighborhoods with the highest situational vulnerability, 50.0% in the lowest ethnocultural composition, 5.3% in the highest residential instability, and 55.7% in the highest economic dependency (Table 3). In comparison, cases that were <20 min from a screening site represented 69.8% of all cases (Table 2). Among these cases, 22.0% resided in neighborhoods with the highest situational vulnerability, 14.6% in the lowest ethnocultural composition, 31.9% in the highest residential instability, and 22.1% in the highest economic dependency (Table 3).
Drive time distributions across HSDAs regions are provided in Figure 1. Generally, median travel times were less than 20 min across HSDAs. Some HSDA regions showed narrow distributions, such as Vancouver, Richmond, Fraser North and Fraser South. In contrast, East Kootenay, Kootenay Boundary, Northeast, among others, showed wide distributions. Most cases in Fraser East were within 40 min of a screening site. East Kootenay and Kootenay Boundary HSDAs showed most cases within 20 min or over 60 min from a screening site (Figure 1).
A choropleth map of median drive time by CHSA is shown in Figure 2. The Moran’s I statistic for median drive times was 0.47 (p = 0.002) indicating significant clustering. There were 118 CHSAs (54.1%) with a median drive time less than 20 min. Short drive times were generally located in the southwest mainland and island CHSAs and some CHSAs in the southern and central interior (Figure 2). CHSAs with a median drive time less than 20 min represented 70.4% (N = 8937) of lung cancer cases in BC (Figure 3). There were 47 CHSAs (21.6%) with a median drive over 60 min. These were generally located in the southeastern CHSAs, as well as central and northern CHSAs (Figure 3). CHSAs with drive times greater than 60 min represented 6.8% of lung cancer cases in BC (Figure 3). A choropleth map of the CHSA areas by level of urbanization was provided in Supplemental Figure S1.

4. Discussion

To our knowledge, this is the first study to examine spatial access to lung screening sites in a regional health care setting considering the impact of age, sex, level of urbanization, and multiple indices of deprivation. While the bulk of proxy, screen-eligible participants (69.8%) were less than 20 min from a screening site, our results do highlight important differences in spatial access to screening sites among rural versus urban participants and those with greater levels of sociodemographic and economic deprivation.
Cases that were 60 min or greater from a screening site (i.e., poor spatial access) represented a small proportion of participants (8.0%, N = 1014). Relative to cases with short spatial access (i.e., <20 min from a screening site), cases with poor spatial access were more likely to be male (51.0% vs. 48.6%) and reside in rural areas (90.4% vs. <0.5%). They also were more likely to reside in neighborhoods with lower ethnocultural composition (50.0% vs. 14.6%), greater situational deprivation (44.2% vs. 22.0%), economic dependence (55.7% vs. 22.1%), and lower residential instability (5.3% vs. 31.9%).
Consistent with our findings, studies in the US have reported that most of the target population are within close proximity to a lung screening site. Tailor et al. reported that over 80% of people who smoked in the US were within 15 miles of a CT facility [13]. Sahar et al. reported that most people 55–79 years of age in over 63% of US counties were within 40 miles of a screening site [14]. However, almost all people 55–79 years of age in approximately 6% of US counties were found to be farther than 40 miles from a screening site [14]. This figure increased to 16% when a 20-mile threshold was considered [14]. Importantly, a larger proportion of those with limited access to a screening site are rural [15]. Populations with low spatial access are less likely to participate in screening services. In Canada, increased travel time has consistently been associated with lower participation in existing screening services [16,17] and radiotherapy (RT) utilization [18,29,32,33,34].
Few previous studies have examined socioeconomic characteristics of areas with poor spatial access, which is an important consideration for program planning and implementation. In the US, counties with poor spatial access to a lung screening site were generally low resource areas with high lung cancer mortality rates [30]. Importantly, less affluent US counties generally experience a disproportionate burden of preventable cancers, e.g., lung cancer [35]. Similar disparities in lung cancer risk across socioeconomic groups have been reported in Canada [7,36].
As in our study, decreasing population density and rurality in the US has been associated with lower spatial access to a lung screening site [13,15]. Importantly, smoking prevalence and disease burden also vary across levels of urbanization [37]. In BC, smoking prevalence is generally higher in rural areas [37]. Lung cancer risk also varies geographically and has been shown to be elevated in rural versus urban population, such as BC’s Northern and Interior health administrative regions [38,39]. Across Canada, rural populations had higher age-standardized lung cancer incidence and mortality rates, as well as advanced stage specific rates, compared to urban populations [18]. Lung cancer incidence and mortality were also greater in rural compared to urban populations in the US [40]. Overall, spatial access should be considered along with socioeconomic and demographic factors, smoking prevalence, and lung cancer mortality to maximize the impact of lung screening programs.
Despite high overall spatial access in the US, the proportion of lung screen-eligible people that have undergone screening is relatively low in comparison to other screening programs. From 2010 to 2015, the figure was reported as 3.9% [41] and in 2017 it was reported as 14.4% [42]. As such, screening participation is likely influenced by a combination of factors at the patient-level including awareness, beliefs and attitudes about lung screening, stigmatization of smoking, socioeconomic status and sociocultural factors [43,44,45]. Factors related to primary care may also contribute, including provider gaps in knowledge about screening guidelines and tools to support patient-provider discussions about the benefits and harms of screening [46].
Targeting screening services to areas in relation to lung cancer incidence as a proxy may be an effective way to improve lung cancer outcomes when data on risk factors used in risk prediction tools are not available. Sahar et al. also recommended integrating regional measures of the lung cancer burden, such as lung cancer mortality rates, to guide lung screening interventions [15]. Screening eligibility criteria is an evolving science that will change as an increasing number of lung cancers will be arising from those who have never smoked or have a modest smoking history [47,48]. Targeting lung cancer screening services to areas where lung cancer patients are found may be more effective than smoking history in the long-term.

Study Limitations

Firstly, while our analyses did benefit from the availability of CMID measures, these were assessed at the neighborhood level and may not reflect individual-level characteristics. Secondly, this study focused on descriptive analyses to quantify spatial access and associated factors. The results do not account for potential confounding and additional work is required to explore and quantify these relationships further while adjusting for confounding.
Thirdly, the scope of this study encompassed a universal health care context and was conducted within the framework of a province-wide organized screening program. While these findings may reasonably be extended to other Canadian provinces with similar organized screening programs, it is essential to consider the influence of varying health care systems, population demographics, and cancer screening delivery models. It is worth noting that our results align with studies conducted in the US [13,15], where evidence suggests a commendable coverage for lung screening on a broader scale, yet spatial access disparities persist among certain subpopulations. This alignment supports the notion that our findings hold relevance beyond national boundaries. Nevertheless, the extent to which these findings can be generalized to diverse regions hinges on factors such as population dispersion and concentration. In regions similar to that of British Columbia, where populations are dispersed across a large land mass with few high-density metropolitan areas, our findings could likely apply. Finally, in this study, a complete six-digit postal code was assigned to a geocoordinate using the Statistics Canada PCCF+ program. PCCF+ uses population weighting and random allocation to inform the geocoordinate of postal codes, specifically when postal codes match to the multiple potential reference points. Positional accuracy is better among urban versus rural areas [49,50] and therefore, drive times are likely more precise for urban versus rural individuals. We would expect positional accuracy to be the least precise for rural individuals that are 60+ min from a CT site. The magnitude of this positional error (distance error between PCCF+ geocoordinates and true reference points) is unknown, however, based on prior literature [49,50] it is typically small. For example, according to Khan et al., most urban postal codes link to a single blockface (i.e., one side of a street) or dissemination block (i.e., an area bounded on all sides by roads) and this represents nearly 75% of the population [49]. The other postal codes match to multiple representative points and the PCCF+ geocodes these postal codes randomly using population weights [51]. From their study, the percentages of the sample geocoded to within 500 m of their full street address were 56% for small communities of 10,000 to 99,999 population and more than 70% for communities of at least 500,000 [49]. Further, we expect the error would be amplified on a finer scale of drive time as opposed to our drive time categories.
Our study provides critical insights on spatial disparities in access to lung cancer screening services. Screening programs can expand screening sites strategically in underserved areas for equitable access to improve access. By pinpointing regions with poor spatial access, screening programs can allocate resources more efficiently to reach those who need it the most. Additionally, using lung cancer incidence as a proxy for the screen-eligible population, in lieu of comprehensive smoking prevalence data, can guide resource allocation for screening and health promotion efforts. Prioritizing high-risk areas, indicated by our method, can improve resource allocation and help early detection, reduce mortality, and community health. Lastly, as lung cancer screening programs are introduced in various Canadian provinces, our study underscores the importance of a strong foundational understanding of spatial access. Programs can adopt a similar methodology, utilizing open source and census-related data to identify communities facing access challenges.
While LDCT screening is inherently conducted through fixed sites requiring in-person attendance, it is important to acknowledge the potential role of telehealth in enhancing various aspects of the screening process and advancing patient care. For example, BC Cancer Lung Screening has virtual care options for individuals without a primary care physician to support eligible participants in the referral process for LDCT lung screening [52]. Notably, Magarinos et al. showcased the viability of single-encounter telemedicine lung cancer screening [53]. This approach involves leveraging telemedicine following an individual undergoing LDCT at a facility. Through telemedicine, discussions of results, coordination of follow-up procedures, and facilitation of smoking cessation counseling were effectively managed through telemedicine, thereby emphasizing the multifaceted advantages of telehealth within the broader framework of lung cancer screening.
In conclusion, we found that most of our study population were within 20 min from a lung screening site. However, spatial access differed geographically and was associated with deprivation across some sociodemographic and economic measures, highlighting populations that may require additional support to ensure equitable access to lung cancer screening services.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/curroncol30090586/s1, Figure S1: A choropleth map of community health service areas by level of urbanization. Lung screening sites are represented by black dots; Table S1: Description of CIMD Variables and corresponding index indicators.

Author Contributions

Conceptualization, J.S. (Jonathan Simkin), E.K., M.D., J.S. (Janette Sam) and R.R.W.; Data curation, J.S. (Jonathan Simkin) and E.K.; Formal analysis, J.S. (Jonathan Simkin) and E.K.; Investigation, J.S. (Jonathan Simkin), E.K. and M.D.; Methodology, J.S. (Jonathan Simkin), E.K. and M.D.; Project administration, J.S. (Jonathan Simkin); Resources, J.S. (Jonathan Simkin) and R.R.W.; Software, J.S. (Jonathan Simkin) and E.K.; Supervision, S.L. and R.R.W.; Validation, J.S. (Jonathan Simkin), E.K., M.D., J.S. (Janette Sam), P.B., S.L. and R.R.W.; Visualization, J.S. (Jonathan Simkin), E.K., M.D., J.S. (Janette Sam), P.B., S.L. and R.R.W.; Writing—original draft, J.S. (Jonathan Simkin); Writing—review and editing, J.S. (Jonathan Simkin), E.K., M.D., J.S. (Janette Sam), P.B., S.L. and R.R.W. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Research Ethics Board of University of British Columbia–BC Cancer (protocol code: H22-00223 and date of approval: 2 February 2022).

Informed Consent Statement

Patient consent was waived due to secondary data use of retrospectively collected data.

Data Availability Statement

The dataset used and analyzed during the current study were obtained from the BC Cancer Registry and are not publicly available due to privacy legislation and institutional data sharing agreements. Data however can be requested through a data access request to BC Cancer following their processes at http://www.bccancer.bc.ca/health-professionals/professional-resources/bc-cancer-registry/request-registry-data (accessed on 1 September 2021).

Conflicts of Interest

S.L. served as an Expert Advisor and was paid for this role by the Canadian Partnership Against Cancer. S.L. served on the Prevention Steering Committee of the American Association of Cancer Research and this role was unpaid. S.L. served as the Chair for the Early Detection and Screening Committee of the International Association for the Study of Lung Cancer and this role was unpaid.

References

  1. Canadian Cancer Statistics Advisory Committee in collaboration with the Canadian Cancer Society and Statistics Canada and the Public Health Agency of Canada. Canadian Cancer Statistics 2021; Canadian Cancer Society: Toronto, ON, Canada, 2021. [Google Scholar]
  2. Sung, H.; Ferlay, J.; Siegel, R.L.; Laversanne, M.; Soerjomataram, I.; Jemal, A.; Bray, F. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J. Clin. 2021, 71, 209–249. [Google Scholar] [CrossRef] [PubMed]
  3. de Koning, H.J.; van der Aalst, C.M.; de Jong, P.A.; Scholten, E.T.; Nackaerts, K.; Heuvelmans, M.A.; Lammers, J.-W.J.; Weenink, C.; Yousaf-Khan, U.; Horeweg, N.; et al. Reduced Lung-Cancer Mortality with Volume CT Screening in a Randomized Trial. N. Engl. J. Med. 2020, 382, 503–513. [Google Scholar] [CrossRef] [PubMed]
  4. Pastorino, U.; Silva, M.; Sestini, S.; Sabia, F.; Boeri, M.; Cantarutti, A.; Sverzellati, N.; Sozzi, G.; Corrao, G.; Marchianò, A. Prolonged Lung Cancer Screening Reduced 10-Year Mortality in the MILD Trial: New Confirmation of Lung Cancer Screening Efficacy. Ann. Oncol. 2019, 30, 1162–1169. [Google Scholar] [CrossRef] [PubMed]
  5. Aberle, D.R.; Adams, A.M.; Berg, C.D.; Black, W.C.; Clapp, J.D.; Fagerstrom, R.M.; Gareen, I.F.; Gatsonis, C.; Marcus, P.M.; Sicks, J.D. Reduced Lung-Cancer Mortality with Low-Dose Computed Tomographic Screening. N. Engl. J. Med. 2011, 365, 395–409. [Google Scholar] [CrossRef] [PubMed]
  6. Fedewa, S.A.; Kazerooni, E.A.; Studts, J.L.; Smith, R.A.; Bandi, P.; Sauer, A.G.; Cotter, M.; Sineshaw, H.M.; Jemal, A.; Silvestri, G.A. State Variation in Low-Dose Computed Tomography Scanning for Lung Cancer Screening in the United States. JNCI J. Natl. Cancer Inst. 2021, 113, 1044–1052. [Google Scholar] [CrossRef]
  7. Mitra, D.; Shaw, A.; Tjepkema, M.; Peters, P. Social Determinants of Lung Cancer Incidence in Canada: A 13-Year Prospective Study. Health Rep. 2015, 26, 12–20. [Google Scholar]
  8. Singh, G.K.; Jemal, A. Socioeconomic and Racial/Ethnic Disparities in Cancer Mortality, Incidence, and Survival in the United States, 1950–2014: Over Six Decades of Changing Patterns and Widening Inequalities. J. Environ. Public Health 2017, 2017, e2819372. [Google Scholar] [CrossRef]
  9. Simkin, J.; Dummer, T.J.B.; Erickson, A.C.; Otterstatter, M.C.; Woods, R.R.; Ogilvie, G. Small Area Disease Mapping of Cancer Incidence in British Columbia Using Bayesian Spatial Models and the Smallareamapp R Package. Front. Oncol. 2022, 12, 833265. [Google Scholar] [CrossRef]
  10. Withrow, D.R.; Pole, J.D.; Nishri, E.D.; Tjepkema, M.; Marrett, L.D. Cancer Survival Disparities Between First Nation and Non-Aboriginal Adults in Canada: Follow-up of the 1991 Census Mortality Cohort. Cancer Epidemiol. Biomark. Prev. 2017, 26, 145–151. [Google Scholar] [CrossRef]
  11. Islami, F.; Guerra, C.E.; Minihan, A.; Yabroff, K.R.; Fedewa, S.A.; Sloan, K.; Wiedt, T.L.; Thomson, B.; Siegel, R.L.; Nargis, N.; et al. American Cancer Society’s Report on the Status of Cancer Disparities in the United States, 2021. CA Cancer J. Clin. 2022, 72, 112–143. [Google Scholar] [CrossRef]
  12. Canadian Cancer Statistics Advisory Committee. Canadian Cancer Statistics: A 2020 Special Report on Lung Cancer; Canadian Cancer Society: Toronto, ON, Canada, 2020. [Google Scholar]
  13. Tailor, T.D.; Choudhury, K.R.; Tong, B.C.; Christensen, J.D.; Sosa, J.A.; Rubin, G.D. Geographic Access to CT for Lung Cancer Screening: A Census Tract-Level Analysis of Cigarette Smoking in the United States and Driving Distance to a CT Facility. J. Am. Coll. Radiol. 2019, 16, 15–23. [Google Scholar] [CrossRef] [PubMed]
  14. Sahar, L.; Douangchai Wills, V.L.; Liu, K.K.; Kazerooni, E.A.; Dyer, D.S.; Smith, R.A. Using Geospatial Analysis to Evaluate Access to Lung Cancer Screening in the United States. Chest 2021, 159, 833–844. [Google Scholar] [CrossRef] [PubMed]
  15. Sahar, L.; Douangchai Wills, V.L.; Liu, K.K.; Fedewa, S.A.; Rosenthal, L.; Kazerooni, E.A.; Dyer, D.S.; Smith, R.A. Geographic Access to Lung Cancer Screening among Eligible Adults Living in Rural and Urban Environments in the United States. Cancer 2022, 128, 1584–1594. [Google Scholar] [CrossRef]
  16. St-Jacques, S.; Philibert, M.D.; Langlois, A.; Daigle, J.-M.; Pelletier, É.; Major, D.; Brisson, J. Geographic Access to Mammography Screening Centre and Participation of Women in the Quebec Breast Cancer Screening Programme. J. Epidemiol. Community Health 2013, 67, 861–867. [Google Scholar] [CrossRef]
  17. McDonald, J.T.; Wang, Y.; Liu, Z. Participation and Retention in the Breast Cancer Screening Program in New Brunswick Canada. Prev. Med. Rep. 2017, 6, 214–220. [Google Scholar] [CrossRef]
  18. Canadian Partnership Against Cancer. Examining Disparities in Cancer Control System Performance Special Focus Report; Canadian Partnership against Cancer: Toronto, ON, Canada, 2014. [Google Scholar]
  19. Honein-AbouHaidar, G.N.; Baxter, N.N.; Moineddin, R.; Urbach, D.R.; Rabeneck, L.; Bierman, A.S. Trends and Inequities in Colorectal Cancer Screening Participation in Ontario, Canada, 2005–2011. Cancer Epidemiol. 2013, 37, 946–956. [Google Scholar] [CrossRef]
  20. Kerner, J.; Liu, J.; Wang, K.; Fung, S.; Landry, C.; Lockwood, G.; Zitzelsberger, L.; Mai, V. Canadian Cancer Screening Disparities: A Recent Historical Perspective. Curr. Oncol. 2015, 22, 156–163. [Google Scholar] [CrossRef] [PubMed]
  21. Simkin, J.; Ogilvie, G.; Hanley, B.; Elliott, C. Differences in Colorectal Cancer Screening Rates across Income Strata by Levels of Urbanization: Results from the Canadian Community Health Survey (2013/2014). Can. J. Public Health. 2019, 110, 62–71. [Google Scholar] [CrossRef]
  22. Statistics Canada. Table 17-10-0005-01 Population Estimates on July 1st, by Age and Sex. Available online: https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1710000501 (accessed on 2 September 2021).
  23. Ministry of Health Community Health Service Areas-CHSA-Datasets-Data Catalogue. Available online: https://catalogue.data.gov.bc.ca/dataset/community-health-service-areas-chsa (accessed on 3 January 2020).
  24. Tammemagi, M.C.; Schmidt, H.; Martel, S.; McWilliams, A.; Goffin, J.R.; Johnston, M.R.; Nicholas, G.; Tremblay, A.; Bhatia, R.; Liu, G.; et al. Participant Selection for Lung Cancer Screening by Risk Modelling (the Pan-Canadian Early Detection of Lung Cancer [PanCan] Study): A Single-Arm, Prospective Study. Lancet Oncol. 2017, 18, 1523–1531. [Google Scholar] [CrossRef]
  25. Ministry of Health BC Health Care Facilities (Hospital)-Datasets-Data Catalogue. Available online: https://catalogue.data.gov.bc.ca/dataset/bc-health-care-facilities-hospital (accessed on 25 May 2021).
  26. Statistics Canada the Canadian Index of Multiple Deprivation. Available online: https://www150.statcan.gc.ca/n1/pub/45-20-0001/452000012019002-eng.htm (accessed on 9 February 2022).
  27. Project OSRM. Available online: http://project-osrm.org/ (accessed on 25 May 2021).
  28. Riatelab/Osrm: Shortest Paths and Travel Time from OpenStreetMap with R. Available online: https://github.com/riatelab/osrm (accessed on 25 May 2021).
  29. Liu, E.; Santibáñez, P.; Puterman, M.L.; Weber, L.; Ma, X.; Sauré, A.; Olivotto, I.A.; Halperin, R.; French, J.; Tyldesley, S. A Quantitative Analysis of the Relationship Between Radiation Therapy Use and Travel Time. Int. J. Radiat. Oncol. 2015, 93, 710–718. [Google Scholar] [CrossRef]
  30. Niranjan, S.J.; Opoku-Agyeman, W.; Carroll, N.W.; Dorsey, A.; Tipre, M.; Baskin, M.L.; Dransfield, M.T. Distribution and Geographic Accessibility of Lung Cancer Screening Centers in the United States. Ann. Am. Thorac. Soc. 2021, 18, 1577–1580. [Google Scholar] [CrossRef]
  31. Bivand, R.S.; Wong, D.W.S. Comparing Implementations of Global and Local Indicators of Spatial Association. TEST 2018, 27, 716–748. [Google Scholar] [CrossRef]
  32. Lavergne, M.R.; Johnston, G.M.; Gao, J.; Dummer, T.J.B.; Rheaume, D.E. Variation in the Use of Palliative Radiotherapy at End of Life: Examining Demographic, Clinical, Health Service, and Geographic Factors in a Population-Based Study. Palliat. Med. 2011, 25, 101–110. [Google Scholar] [CrossRef] [PubMed]
  33. Tyldesley, S.; McGahan, C. Utilisation of Radiotherapy in Rural and Urban Areas in British Columbia Compared with Evidence-Based Estimates of Radiotherapy Needs for Patients with Breast, Prostate and Lung Cancer. Clin. Oncol. 2010, 22, 526–532. [Google Scholar] [CrossRef]
  34. Huang, J.; Wai, E.S.; Lau, F.; Blood, P.A. Palliative Radiotherapy Utilization for Cancer Patients at End of Life in British Columbia: Retrospective Cohort Study. BMC Palliat. Care 2014, 13, 49. [Google Scholar] [CrossRef]
  35. Siegel, R.L.; Miller, K.D.; Jemal, A. Cancer Statistics, 2019. CA Cancer J. Clin. 2019, 69, 7–34. [Google Scholar] [CrossRef]
  36. Hystad, P.; Carpiano, R.M.; Demers, P.A.; Johnson, K.C.; Brauer, M. Neighbourhood Socioeconomic Status and Individual Lung Cancer Risk: Evaluating Long-Term Exposure Measures and Mediating Mechanisms. Soc. Sci. Med. 2013, 97, 95–103. [Google Scholar] [CrossRef]
  37. Drope, J.; Liber, A.C.; Cahn, Z.; Stoklosa, M.; Kennedy, R.; Douglas, C.E.; Henson, R.; Drope, J. Who’s Still Smoking? Disparities in Adult Cigarette Smoking Prevalence in the United States. CA Cancer J. Clin. 2018, 68, 106–115. [Google Scholar] [CrossRef] [PubMed]
  38. BC Cancer Agency. British Columbia 2011 Regional Cancer Report; BC Cancer Agency: Vancouver, BC, Canda, 2011.
  39. Simkin, J.; Woods, R.; Elliott, C. Cancer Mortality in Yukon 1999–2013: Elevated Mortality Rates and a Unique Cancer Profile. Int. J. Circumpolar Health 2017, 76, 1324231. [Google Scholar] [CrossRef]
  40. Atkins, G.T.; Kim, T.; Munson, J. Residence in Rural Areas of the United States and Lung Cancer Mortality: Disease Incidence, Treatment Disparities, and Stage-Specific Survival. Ann. Am. Thorac. Soc. 2017, 14, 403–411. [Google Scholar] [CrossRef]
  41. Jemal, A.; Fedewa, S.A. Lung Cancer Screening with Low-Dose Computed Tomography in the United States—2010 to 2015. JAMA Oncol. 2017, 3, 1278. [Google Scholar] [CrossRef]
  42. Zahnd, W.E.; Eberth, J.M. Lung Cancer Screening Utilization: A Behavioral Risk Factor Surveillance System Analysis. Am. J. Prev. Med. 2019, 57, 250–255. [Google Scholar] [CrossRef]
  43. Rankin, N.M.; McWilliams, A.; Marshall, H.M. Lung Cancer Screening Implementation: Complexities and Priorities. Respirology 2020, 25, 5–23. [Google Scholar] [CrossRef]
  44. Jonnalagadda, S.; Bergamo, C.; Lin, J.J.; Lurslurchachai, L.; Diefenbach, M.; Smith, C.; Nelson, J.E.; Wisnivesky, J.P. Beliefs and Attitudes about Lung Cancer Screening among Smokers. Lung Cancer 2012, 77, 526–531. [Google Scholar] [CrossRef]
  45. Richmond, J.; Mbah, O.M.; Dard, S.Z.; Jordan, L.C.; Cools, K.S.; Samuel, C.A.; Khan, J.M.; Manning, M.A. Evaluating Potential Racial Inequities in Low-Dose Computed Tomography Screening for Lung Cancer. J. Natl. Med. Assoc. 2020, 112, 209–214. [Google Scholar] [CrossRef]
  46. Ersek, J.L.; Eberth, J.M.; McDonnell, K.K.; Strayer, S.M.; Sercy, E.; Cartmell, K.B.; Friedman, D.B. Knowledge of, Attitudes toward, and Use of Low-Dose Computed Tomography for Lung Cancer Screening among Family Physicians. Cancer 2016, 122, 2324–2331. [Google Scholar] [CrossRef]
  47. Jeon, J.; Holford, T.R.; Levy, D.T.; Feuer, E.J.; Cao, P.; Tam, J.; Clarke, L.; Clarke, J.; Kong, C.Y.; Meza, R. Smoking and Lung Cancer Mortality in the United States From 2015 to 2065. Ann. Intern. Med. 2018, 169, 684. [Google Scholar] [CrossRef]
  48. Gould, M.K.; Huang, B.Z.; Tammemagi, M.C.; Kinar, Y.; Shiff, R. Machine Learning for Early Lung Cancer Identification Using Routine Clinical and Laboratory Data. Am. J. Respir. Crit. Care Med. 2021, 204, 445–453. [Google Scholar] [CrossRef]
  49. Khan, S.; Pinault, L.; Tjepkema, M.; Wilkins, R. Positional Accuracy of Geocoding from Residential Postal Codes versus Full Street Addresses. Health Rep. 2018, 29, 3–9. [Google Scholar]
  50. Pinault, L.; Khan, S.; Tjepkema, M. Accuracy of Matching Residential Postal Codes to Census Geography. Health Rep. 2020, 31, 3–13. [Google Scholar] [CrossRef] [PubMed]
  51. Statistics Canada. Postal CodeOM Conversion File Plus (PCCF+), Reference Guide; Statistics Canada: Ottawa, ON, Canada, 2021.
  52. BC Cancer Get Screened. Available online: http://www.bccancer.bc.ca/screening/lung/get-screened (accessed on 25 August 2023).
  53. Magarinos, J.; Lutzow, L.; Dass, C.; Ma, G.X.; Erkmen, C.P. Feasibility of Single-Encounter Telemedicine Lung Cancer Screening: A Retrospective Cohort Study in an Underserved Population. Cancer Control 2023, 30, 10732748221121392. [Google Scholar] [CrossRef] [PubMed]
Curroncol 30 00586 g001
Figure 1. Drive time distribution by health service delivery area. The y-axis was restricted to a maximum of 120 min to better visualize drive times across all HSDA areas. There were 338 records of drive times greater than 120 min (2.6% of the analytic cohort).
Figure 1. Drive time distribution by health service delivery area. The y-axis was restricted to a maximum of 120 min to better visualize drive times across all HSDA areas. There were 338 records of drive times greater than 120 min (2.6% of the analytic cohort).
Curroncol 30 00586 g001
Curroncol 30 00586 g002
Figure 2. Median drive time to the closest screening site by community health service area. Lung screening sites are represented by black dots.
Figure 2. Median drive time to the closest screening site by community health service area. Lung screening sites are represented by black dots.
Curroncol 30 00586 g002
Curroncol 30 00586 g003
Figure 3. Proportion of lung cancer cases by community health service area. Lung screening sites are represented by black dots.
Figure 3. Proportion of lung cancer cases by community health service area. Lung screening sites are represented by black dots.
Curroncol 30 00586 g003
Table 1. Descriptive statistics on the study cohort and variables of interest such as age, sex, year of diagnosis, histology, stage at diagnosis, level of urbanization and measures of deprivation, British Columbia, 2015–2019.
Table 1. Descriptive statistics on the study cohort and variables of interest such as age, sex, year of diagnosis, histology, stage at diagnosis, level of urbanization and measures of deprivation, British Columbia, 2015–2019.
VariableTotal (N)
N = 12,688		Proportion (%)
Age (years)
Median [IQR]70 (64–75)
Age groups (years)
55–5912119.5
60–64209616.5
65–69272221.5
70–74316825
75–80349127.5
Sex
Females655351.6
Males613548.4
Year of diagnosis
2015247219.5
2016240018.9
2017252819.9
2018256720.2
2019272121.4
Histology type
Non-small cell lung cancer10,19480.3
Small cell lung cancer12269.7
Sarcomas and other specified malignant neoplasms 150.1
Unspecified12539.9
Stage 1
I260220.5
II9797.7
III258720.4
IV553943.7
Unknown9617.6
Occult200.2
CHSA urban-rural classifications
Metropolitan457336
Large Urban167513.2
Medium Urban233218.4
Small Urban12239.6
Rural Hub8967.1
Rural188414.8
Remote1050.8
CIMD variables
Ethnocultural composition 1
Q1268721.2
Q2311424.5
Q3278121.9
Q4221817.5
Q5186914.7
Missing190.1
Situational vulnerability 2
Q1225017.7
Q2247919.5
Q3244419.3
Q4262920.7
Q5286722.6
Missing190.1
Residential instability 2
Q1193515.3
Q2235018.5
Q3248219.6
Q4266321
Q5323925.5
Missing190.1
Economic dependency 2
Q1185814.6
Q2220017.3
Q3237518.7
Q4250619.8
Q5373029.4
Missing190.1
1 For ethnocultural composition, a value of 5 corresponds to the greatest ethnocultural composition. 2 A value of 1 corresponds to the least deprived for that dimension, and a value of 5 corresponds to the most deprived.
Table 2. Descriptive statistics on travel time (in minutes) from residential postal code to CT sites, British Columbia, 2015–2019.
Table 2. Descriptive statistics on travel time (in minutes) from residential postal code to CT sites, British Columbia, 2015–2019.
VariableTotal (N)
N = 12,688		Proportion (%)
Travel time (minutes)
Median [IQR]11.7 [6.2–23.2]
Travel time categories (minutes)
<20885669.8
20 to 40224617.7
40 to 605724.5
60+10148.0
Table 3. Bivariate analysis (chi-square test of independence) between drive time (minutes) and variables of interest such as age, sex, stage at diagnosis, level of urbanization, and measures of deprivation, British Columbia, 2015–2019.
Table 3. Bivariate analysis (chi-square test of independence) between drive time (minutes) and variables of interest such as age, sex, stage at diagnosis, level of urbanization, and measures of deprivation, British Columbia, 2015–2019.
Drive Time Categoriesp Value
Variable<2020–4040–6060+
Age 0.37
55–59852 (9.6)208 (9.3)55 (9.6)96 (9.5)
60–641464 (16.5)361 (16.1)86 (15.0)185 (18.2)
65–691894 (21.4)457 (20.3)137 (24.0)234 (23.1)
70–742198 (24.8)585 (26.0)134 (23.4)251 (24.8)
75+2448 (27.6)635 (28.3)160 (28.0)248 (24.5)
Sex 0.018
Females4555 (51.4)1218 (54.2)283 (49.5)497 (49)
Males4301 (48.6)1028 (45.8)289 (50.5)517 (51)
Stage 0.28
I1847 (22.6)454 (21.8)95 (18.1)206 (22)
II677 (8.3)189 (9.1)45 (8.6)68 (7.3)
III1792 (22)466 (22.3)112 (21.3)217 (23.2)
IV3843 (47.1)978 (46.9)273 (52)445 (47.5)
Level of urbanization <0.001
Metropolitan4204 (47.5)368 (16.4)<5 (0.9)<5 (0.5)
Large Urban1616 (18.2)<60 (2.7)<5 (0.9)<5 (0.5)
Medium Urban1704 (19.2)600 (26.7)<30 (5.2)<5 (0.5)
Small Urban757 (8.5)367 (16.3)<100 (17.5)<5 (0.5)
Rural Hub124 (1.4)297 (13.2)171 (29.9)304 (30)
Rural445 (5)551 (24.5)276 (48.3)612 (60.4)
Remote6 (0.1)<5 (0.2)<5(0.9)<5 (0.5)
CIMD variables
Ethnocultural composition *
Q11289 (14.6)693 (30.9)200 (35.4)505 (50.0)<0.001
Q21796 (20.3)712 (31.8)242 (42.8)364 (36.0)
Q32011 (22.7)553 (24.7)100 (17.7)117 (11.6)
Q41965 (22.2)205 (9.1)23 (4.1)25 (2.5)
Q51791 (20.2)78 (3.5)0 (0)0 (0)
Situational vulnerability * <0.001
Q11601 (18.1)525 (23.4)76 (13.5)48 (4.7)
Q21755 (19.8)491 (21.9)117 (20.7)116 (11.5)
Q31714 (19.4)445 (19.9)124 (21.9)161 (15.9)
Q41836 (20.7)423 (18.9)131 (23.2)239 (23.6)
Q51946 (22.0)357 (15.9)117 (20.7)447 (44.2)
Residential instability * <0.001
Q11231 (13.9)460 (20.5)76 (13.5)168 (16.6)
Q21385 (15.6)551 (24.6)156 (27.6)258 (25.5)
Q31509 (17.0)467 (20.8)214 (37.9)292 (28.9)
Q41902 (21.5)427 (19.1)95 (16.8)239 (23.6)
Q52825 (31.9)336 (15)24 (4.2)54 (5.3)
Economic dependency * <0.001
Q11477 (16.7)225 (10)63 (11.2)93 (9.2)
Q21763 (19.9)320 (14.3)37 (6.5)80 (7.9)
Q31818 (20.5)372 (16.6)74 (13.1)111 (11)
Q41834 (20.7)417 (18.6)91 (16.1)164 (16.2)
Q51960 (22.1)907 (40.5)300 (53.1)563 (55.7)
* A value of Q1 corresponds to neighborhoods that were the least deprived for that dimension, and a value of Q5 corresponds to neighborhoods that were the most deprived. For ethnocultural composition, a value of Q5 corresponds to a neighborhood with the greatest ethnocultural composition.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Simkin, J.; Khoo, E.; Darvishian, M.; Sam, J.; Bhatti, P.; Lam, S.; Woods, R.R. Addressing Inequity in Spatial Access to Lung Cancer Screening. Curr. Oncol. 2023, 30, 8078-8091. https://doi.org/10.3390/curroncol30090586

AMA Style

Simkin J, Khoo E, Darvishian M, Sam J, Bhatti P, Lam S, Woods RR. Addressing Inequity in Spatial Access to Lung Cancer Screening. Current Oncology. 2023; 30(9):8078-8091. https://doi.org/10.3390/curroncol30090586

Chicago/Turabian Style

Simkin, Jonathan, Edwin Khoo, Maryam Darvishian, Janette Sam, Parveen Bhatti, Stephen Lam, and Ryan R. Woods. 2023. "Addressing Inequity in Spatial Access to Lung Cancer Screening" Current Oncology 30, no. 9: 8078-8091. https://doi.org/10.3390/curroncol30090586

Article Metrics

Back to TopTop