The influence of travel time on breast cancer characteristics, receipt of primary therapy, and surveillance mammography

Abstract

Travel time has been shown to influence some aspects of cancer characteristics at diagnosis and care for women with breast cancer, but important gaps remain in our understanding of its impact. We examined the influence of travel time to the nearest radiology facility on breast cancer characteristics, treatment, and surveillance for women with early-stage invasive breast cancer. We included 1,012 women with invasive breast cancer (stages I and II) who had access to care within an integrated health care delivery system in western Washington State. The travel times to the nearest radiology facility were calculated for all the U.S. Census blocks within the study area and assigned to women based on residence at diagnosis. We collected cancer characteristics, primary and adjuvant therapies, and surveillance mammography for at least 2.5 years post diagnosis and used multivariable analyses to test the associations of travel time. The majority of women (68.6%) lived within 20 min of the nearest radiology facility, had stage I disease (72.7%), received breast conserving therapy (68.7%), and had annual surveillance mammography the first 2 years after treatment (73.7%). The travel time was not significantly associated with the stage or surveillance mammography after adjusting for covariates. Primary therapy was significantly related to travel time, with greater travel time (>30 min vs. ≤ 10 min) associated with a higher likelihood of mastectomy compared to breast conserving surgery (RR = 1.53; 95% CI, 1.16–2.01). The travel time was not associated with the stage at diagnosis or surveillance mammography receipt. The travel time does seem to influence the type of primary therapy among women with breast cancer, suggesting that women may prefer low frequency services, such as mastectomy, if geographic access to a radiology facility is limited.

Geographic disparities amongst patients with gynecologic malignancies at an urban NCI-designated cancer center

“…Additionally, a patient's socioeconomic status has been associated with their ability to access specialty centers at increasing distance [10][11][12]. Similarly, travel time to health care services has been shown to influence access, utilization and outcomes for patients with various malignancies [9,[11][12][13][14]. For example, women with breast cancer are more likely to present with advanced staged disease, less likely to receive breast conserving surgery and less likely to receive postoperative adjuvant care as the distance from their home to a health care facility increases [13,[15][16][17][18].…”

“…Similarly, travel time to health care services has been shown to influence access, utilization and outcomes for patients with various malignancies [9,[11][12][13][14]. For example, women with breast cancer are more likely to present with advanced staged disease, less likely to receive breast conserving surgery and less likely to receive postoperative adjuvant care as the distance from their home to a health care facility increases [13,[15][16][17][18]. For women with ovarian cancer, geographic proximity to a high-volume hospital has been shown to be independently associated with receipt of NCCN-adherent care in California [19].…”

Travel by public transit to mammography facilities in 6 US urban areas

“…For all census tracts, we calculated the proportion of transit time that was less than 30 minutes or 30 minutes and greater. A 30-minute travel time has been suggested as a standard for accessible travel time to health care (Onega et al, 2011; Mao & Nekorchuk 2013; Bosanac et al, 1976). For ease of reporting we have defined women with the joint characteristics of no vehicle access and transit times that were 30 minutes or greater as ‘transit marginalized’.…”

“…Geographic proximity to health care services such as mammography is strongly influenced by population density, and therefore, spatial access as measured by travel distance or service density, has been well examined for rural vs. urban areas (Doescher and Jackson, 2009; Engelman et al, 2002; Henry et al, 2013; Hyndman et al, 2000; Onega et al, 2011). In urban areas however, public transportation infrastructure and residential segregation patterns, rather than distance alone, may exert a stronger influence on access to services.…”

Computational and data sciences for health-GIS

“…zip code or census tract) to health-care locations (e.g. Onega et al 2008Onega et al , 2010aOnega et al , 2010bOnega et al , 2011Onega, Duell, Shi, Demidenko, Gottlieb, et al 2009;Celaya et al 2010). Berke and Shi (2009) found that when simply using geometric centroids of zip code polygons, which was adopted by many health-care studies, the error of the estimated average travel time can go above 10%; when more detailed population data are incorporated, the average error can go below 3%.…”