Extent of Follow-Up on Abnormal Cancer Screening in Multiple California Public Hospital Systems: A Retrospective Review

Abstract

Background Inequitable follow-up of abnormal cancer screening tests may contribute to racial/ethnic disparities in colon and breast cancer outcomes. However, few multi-site studies have examined follow-up of abnormal cancer screening tests and it is unknown if racial/ethnic disparities exist. Objective This report describes patterns of performance on follow-up of abnormal colon and breast cancer screening tests and explores the extent to which racial/ethnic disparities exist in public hospital systems. Design We conducted a retrospective cohort study using data from five California public hospital systems. We used multivariable robust Poisson regression analyses to examine whether patient-level factors or site predicted receipt of follow-up test. Main Measures Using data from five public hospital systems between July 2015 and June 2017, we assessed follow-up of two screening results: (1) colonoscopy after positive fecal immunochemical tests (FIT) and (2) tissue biopsy within 21 days after a BIRADS 4/5 mammogram. Key Results Of 4132 abnormal FITs, 1736 (42%) received a follow-up colonoscopy. Older age, Medicaid insurance, lack of insurance, English language, and site were negatively associated with follow-up colonoscopy, while Hispanic ethnicity and Asian race were positively associated with follow-up colonoscopy. Of 1702 BIRADS 4/5 mammograms, 1082 (64%) received a timely biopsy; only site was associated with timely follow-up biopsy. Conclusion Despite the vulnerabilities of public-hospital-system patients, follow-up of abnormal cancer screening tests occurs at rates similar to that of patients in other healthcare settings, with colon cancer screening test follow-up occurring at lower rates than follow-up of breast cancer screening tests. Site-level factors have larger, more consistent impact on follow-up rates than patient sociodemographic traits. Resources are needed to identify health system–level factors, such as test follow-up processes or data infrastructure, that improve abnormal cancer screening test follow-up so that effective health system–level interventions can be evaluated and disseminated.

Equitable Implementation of Mailed Stool Test–Based Colorectal Cancer Screening and Patient Navigation in a Safety Net Health System

“…36,39 More recent studies have found mixed effects with respect to race/ethnicity. [40][41][42] In our study, Spanish as the patient's primary language was associated with higher rates of colonoscopy after positive FIT. Further investigation of promoting and limiting factors to colonoscopy may provide insight into these trends, though it is likely that our bilingual (English/Spanish) patient navigator role contributed to our program's success.…”

American Cancer Society’s report on the status of cancer disparities in the United States, 2023

“…Differences in stage at diagnosis comprise a major contributing factor to disparities in cancer survival and mortality. 58 screening tests [60][61][62] and receipt of surveillance testing or screening for other cancers among cancer survivors. [63][64][65] Disparities in access to care for following suspicious symptoms can also result in disparities in stage at diagnosis of cancers for which routine screening is not recommended.…”

“…In 2021, for example, the prevalence of being up to date with colorectal cancer screening ranged from 58.2% among individuals without a high school diploma to 77.5% among individuals with a college degree; by race/ethnicity, the prevalence was 73.8% among White people but ranged from 60.2% to 71.7% in other racial/ethnic groups 18 . There are also disparities by race/ethnicity, SES, and provider type in follow‐up of abnormal cancer screening tests 60–62 and receipt of surveillance testing or screening for other cancers among cancer survivors 63–65 . Disparities in access to care for following suspicious symptoms can also result in disparities in stage at diagnosis of cancers for which routine screening is not recommended 66 …”