Science Spotlight

Pediatric cancer disparities in Washington State

From the Chow Group, Clinical Research Division

Why do some children with cancer survive, while others die? Yes, cancer biology and treatment options are critically important, but greater attention is being given to social determinants of health, such as race or ethnicity, socioeconomic status, health insurance coverage, and geographic location. Cancer survival disparities in children have been shown in numerous studies; however, the impact of geographic factors has received less attention. In a recent study published in Cancer Epidemiology, Biomarkers and Prevention led by Dr. Tim Ohlsen, a pediatric hematology and oncology fellow in Dr. Eric Chow’s group in the Clinical Research Division at Fred Hutch, researchers examined whether rural residence, in combination with socioeconomic factors, affected pediatric cancer survival among children living in Washington State.

The authors found that “children who were living in locations that were rural and/or classified as having high rates of socioeconomic disadvantage at the time of diagnosis experienced worse survival,” Dr. Ohlsen explained. Interestingly, “the combination of rural residence plus neighborhood socioeconomic disadvantage did not seem to confer an additive effect on mortality,” Dr. Ohlsen added. "It’s well documented that children from locations with greater socioeconomic disadvantage experience worse survival, [but] our study also demonstrates differences in survival also related to rurality,” Dr. Ohlsen noted. Their findings are especially significant for pediatric oncology, which Dr. Ohlsen describes as more centralized in large urban centers compared with medical oncology. "Even mid-size cities like Yakima, Bellingham, or the Tri-Cities have no ability to provide pediatric cancer treatment, meaning many families with a rural residence may need to travel further for pediatric care.”

To understand how non-rural/rural residence and socioeconomic factors affect pediatric cancer survival in Washington State, the authors conducted a retrospective cohort study of 4,306 pediatric cancer cases from 1993 to 2013. They gathered data from the Cancer Surveillance System (CSS) and the Washington State Cancer Registry (WSCR). The authors evaluated census block group-level socioeconomic disadvantage using the Area Deprivation Index (ADI), as well as detailed rural status. Low ADI values indicate lower neighborhood poverty whereas high ADI values indicate high neighborhood poverty. Using these data, the authors constructed a Kaplan-Meier curve which estimates the overall survival of children living with cancer in rural and non-rural areas with different classifications of neighborhood-level poverty. The authors found that children living in non-rural areas and with low neighborhood poverty had a 5-year overall survival rate of 80.9%. Children living in non-rural areas plus high neighborhood poverty had a 5-year overall survival rate of 66.4%. Children living in rural areas plus low or high neighborhood poverty had a 5-year overall survival rate of 69.4% and 66.9%, respectively. The results indicate that children living in locations that are rural and/or experiencing socioeconomic disadvantage experienced a lower survival rate than children living in non-rural areas with lower poverty.

Kaplan–Meier survival curve of children with cancer in Washington State based on geographic location (non-rural/rural) and area deprivation index (ADI).
Kaplan–Meier survival curve of children with cancer in Washington State based on geographic location (non-rural/rural) and area deprivation index (ADI). Imaged provided by Dr. Ohlsen

This study is unique in the sense that “we measured rurality and socioeconomic disadvantage at the census block group level,” Dr. Ohlsen explained. Previous studies “examine the impact of rurality on a larger scale and have done so measuring county-level rurality, which is not particularly precise and could potentially hide true relationships through data misclassification,” he continued. “There are many counties in Washington State that have a lot of heterogeneity in their rural and non-rural populations, such as Yakima, Skagit, and Stevens County”, Dr. Ohlsen added. Their study “accounted for these differences in a way that prior studies may not have”.  

Dr. Ohlsen is now interested in understanding the mechanisms behind these differences in overall survival. “Is there something about the rural environment itself, or distance to pediatric cancer centers, that is impacting time to initial diagnosis or access to quality care during emergencies? How is the medical experience different for children related to these neighborhood factors? Or could these differences just be due to unmeasured sociodemographic factors that we did not fully adjust for?” These are some of the questions he is trying to address. One way to answer these questions is to “learn more about potential differences in healthcare utilization and hospitalizations by neighborhood factors.” Additionally, Dr. Ohlsen is passionate about exploring everything from infectious complications during treatment to mental health care use in survivorship. He concluded, “I am lucky to have collaborators across Fred Hutch, Seattle Children’s, and UW who’ve made this study possible, and hope we can also apply this work to answer these outstanding questions!”

This work was supported by a grant from Alex’s Lemonade Stand Foundation for Childhood Cancer. This work was also supported by the cancer registry databases from the Cancer Surveillance System and Washington State Cancer Registry, a shared resource of the Fred Hutchinson/University of Washington Cancer Consortium.

Fred Hutch/University of Washington/Seattle Children's Cancer Consortium member Eric Chow contributed to this work.

Ohlsen TJD, Doody DR, Mueller BA, Desai AD, Chow EJ. 2023. Population-Based Impact of Rurality and Neighborhood-Level Socioeconomic Disadvantage on Pediatric Cancer Mortality in Washington State. Cancer Epidemiol Biomarkers Prev. 9;32(1):141-148.