Race/Ethnicity and Adoption of a Population Health Management Approach to Colorectal Cancer Screening in a Community-Based Healthcare System

Abstract

Background: Screening outreach programs using population health management principles offer services uniformly to all eligible persons, but racial/ethnic colorectal cancer (CRC) screening patterns in such programs are not well known.

Objective: To examine the association between race/ethnicity and the receipt of CRC screening and timely follow-up of positive results before and after implementation of a screening program.

Design: Retrospective cohort study of screen-eligible individuals at the Kaiser Permanente Northern California community-based integrated healthcare delivery system (2004-2013).

Subjects: A total of 868,934 screen-eligible individuals 51-74 years of age at cohort entry, which included 662,872 persons in the period before program implementation (2004-2006), 654,633 during the first 3 years after implementation (2007-2009), and 665,268 in the period from 4 to 7 years (2010-2013) after program implementation.

Intervention: A comprehensive system-wide long-term effort to increase CRC that included leadership alignment, goal-setting, and quality assurance through a PHM approach, using mailed fecal immunochemical testing (FIT) along with offering screening at office visits.

Main measures: Differences over time and by race/ethnicity in up-to-date CRC screening (overall and by test type) and timely follow-up of a positive screen. Race/ethnicity categories included non-Hispanic white, non-Hispanic black, Hispanic/Latino, Asian/Pacific Islander, Native American, and multiple races.

Key results: From 2004 to 2013, age/sex-adjusted CRC screening rates increased in all groups, including 35.2 to 81.1 % among whites and 35.6 to 78.0 % among blacks. Screening rates among Hispanics (33.1 to 78.3 %) and Native Americans (29.4 to 74.5 %) remained lower than those for whites both before and after program implementation. Blacks, who had slightly higher rates before program implementation (adjusted rate ratio [RR] = 1.04, 99 % CI: 1.02-1.05), had lower rates after program implementation (RR for period from 4 to 7 years = 0.97, 99 % CI: 0.96-0.97). There were also substantial improvements in timely follow-up of positive screening results.

Conclusions: In this screening program using core PHM principles, CRC screening increased markedly in all racial/ethnic groups, but disparities persisted for some groups and developed in others, which correlated with levels of adoption of mailed FIT.

Keywords: cancer screening; colorectal cancer; health care delivery; population health; race & ethnicity.

Figure 1

Flow diagram of cohort ascertainment. ^aThe cohorts overlap in the periods before and after the program implementation.

Figure 2

Trend in the proportion of persons in each racial/ethnic group with up-to-date status on colorectal cancer screening, KPNC 2004–2013. Note: Screening receipt was defined as being up to date by the use of any (in the case of overall rate) of the US Preventive Services Task Force-recommended tests using the HEDIS criteria, defined separately for each calendar year. Patients were considered up to date on CRC screening if they had received colonoscopy within the prior 5 years, sigmoidoscopy within the prior 5 years, or gFOBT or FIT in each calendar year, without regard to the indication for the test. The vertical line on the x-axis demarcates full implementation of the outreach screening program.