Disparities in Cancer Stage of Diagnosis by Rurality in California, 2015 to 2019.

Abstract

Cancer rates in rural areas vary by insurance status, socioeconomic status, region, race, and ethnicity. California Cancer Registry data (2015-2019) were used to investigate the stage of diagnosis by levels of rurality for the five most common cancers. The percentage of residents in rural blocks within census tract aggregation zones was categorized into deciles up to 50%. Multivariable logistic regression was used to estimate associations with rurality, with separate models by cancer site, sex, race, and ethnicity (non-Hispanic White and Hispanic). Covariates included individual-level and zone-level factors. The percentage of late-stage cancer diagnosis was 28% for female breast, 27% for male prostate, 77% for male lung, 71% for female lung, 60% for male colorectal, 59% for female colorectal, 7.8% for male melanoma, and 5.9% for female melanoma. Increasing rurality was significantly associated with increased odds of late-stage cancer diagnosis for female breast cancer (Ptrend < 0.001), male lung cancer (Ptrend < 0.001), female lung cancer (Ptrend < 0.001), and male melanoma (Ptrend = 0.01), after adjusting for individual-level and zone-level factors. The strength of associations varied by sex and ethnicity. For males with lung cancer, odds of late-stage diagnosis in areas with >50% rural population was 1.24 (95% confidence interval, 1.06-1.45) for non-Hispanic White patients and 2.14 (95% confidence interval, 0.86-5.31) for Hispanic patients, compared with areas with 0% rural residents. Increasing rurality was associated with increased odds for late-stage diagnosis for breast cancer, lung cancer, and melanoma, with the strength of associations varying across sex and ethnicity. Our findings will inform cancer outreach to these rural subpopulations.