Program to Improve Colorectal Cancer Screening in a Low-Income, Racially Diverse Population: A Randomized Controlled Trial

INTRODUCTION

Colorectal cancer is the second leading cause of cancer death in the United States and the third most commonly diagnosed cancer.¹ Appropriate screening and early detection can greatly reduce colorectal cancer–associated morbidity and mortality, and several national guidelines recommend regular screening for colorectal cancer among adults aged 50 years and older with high-sensitivity fecal occult blood testing (FOBT), sigmoidoscopy, or colonoscopy.^2–4 Nearly one-half of eligible adults are not up to date on colorectal cancer screening, however.⁵ Moreover, despite overall improvements in the rate of colorectal cancer screening, marked disparities persist, with lower rates of colorectal cancer screening among racial and ethnic minorities, individuals with lower income or lower educational attainment, the uninsured, and individuals who were not born in the United States.^5–7 Rates of screening are particularly low for foreign-born individuals who immigrated to the United States more recently or are less acculturated.^6–8 There is thus a clear need to identify effective strategies to improve the use of colorectal cancer screening among medically under-served populations, and particularly among immigrants and individuals with limited English proficiency.

The direct mailing of FOBT kits to patients who are due for colorectal cancer screening has been shown to be both clinically effective^9–14 and cost-effective^15,16 for increasing colorectal cancer screening rates. The vast majority of studies assessing the efficacy of this strategy have not included sizable numbers of patients with lower socioeconomic status or from racial or ethnic minority groups, however.^9–11,13,14 We therefore set out to examine the efficacy of an outreach intervention that includes the direct mailing of FOBT kits to patients who were overdue for colorectal cancer screening and were receiving care at a community health center that serves a primarily low income population with a high proportion of immigrants and refugees.

METHODS

RESULTS

DISCUSSION

In this randomized controlled trial, an outreach intervention that consisted of the mailing of FOBT kits to patients overdue for colorectal cancer screening with additional telephone outreach to initial nonresponders was efficacious for promoting colorectal cancer screening among a population with high levels of poverty, lack of health insurance, limited English proficiency, racial and ethnic diversity, and other features associated with a lower likelihood of receiving colorectal cancer screening. Although prior studies have shown that the direct-to-patient mailing of FOBT kits can lead to higher colorectal cancer screening rates in predominantly white, middle-class, or well-insured populations,^9–11,13,14 few studies have assessed the efficacy of this strategy in historically underserved populations. Those that have evaluated this intervention in more vulnerable populations have tended to focus on just 1 or 2 ethnic groups.^12,20 Our study adds to this existing literature by demonstrating that this outreach strategy can significantly improve colorectal cancer screening rates even among economically disadvantaged patients from a wide range of racial, ethnic, and cultural backgrounds.

Especially in settings with an EHR system to facilitate the ready identification of patients who are eligible and overdue for colorectal cancer screening, the direct mailing of FOBT kits to eligible patients with some additional phone outreach to initial nonresponders may be a sustainable strategy to promote colorectal cancer screening, particularly under a patient-centered medical home payment model that reimburses practices for population health management activities.²¹ Although our study did not include formal cost or cost-effectiveness analyses, very similar outreach interventions have been found to be cost-effective in other studies.^15,16

Our study demonstrated that the direct-to-patient outreach intervention was efficacious even in a health care setting that had already implemented point-of-care clinician-directed electronic clinical reminders to promote appropriate colorectal cancer screening. This finding is in keeping with prior studies that have shown the superiority of patient-directed outreach over clinician-directed reminders¹³ and the limited efficacy of clinician-directed reminders on colorectal cancer screening rates.^22–25 Although the reasons for the low and in some cases lack of efficacy of clinician-directed reminders on colorectal cancer screening are not completely clear, the evidence overall suggests that patient-directed reminders may be a crucial component of efforts to improve colorectal cancer screening rates.

The use of patient-directed outreach to promote colorectal cancer screening, particularly through FOBT, may also be an important strategy to address disparities in colorectal cancer screening. Although the US Preventive Services Task Force gives equal recommendation to colorectal cancer screening with FOBT, sigmoidoscopy, or colonoscopy,² some professional societies recommend colonoscopy as the preferred method of screening,⁴ and some physicians focus heavily on colonoscopy when discussing colorectal cancer screening with patients²⁶; however, because of the ongoing financial barriers to colonoscopy, particularly among the uninsured; the limited access to colonoscopy in some areas, even among the insured; and the preference of some patients for FOBT over other methods of colorectal cancer screening,²⁷ the use of patient-directed outreach that includes FOBT may be an effective way to promote colorectal cancer screening, especially among those who are least likely to complete endoscopic screening tests.

It is important to note that despite outreach efforts, 70% of patients assigned to the intervention arm of our study did not complete a colorectal cancer screening test during the study period. The use of simplified testing instructions,¹² culturally tailored screening promotion materials,^28,29 or single-sample fecal immunochemical tests³⁰ could be incorporated into future iterations of this intervention to address possible literacy, cultural, and logistic barriers to fecal-based colorectal cancer screening, and could potentially improve the efficacy of the intervention. Other strategies, including the use of patient navigators,^31–33 peer educators,³⁴ decision aides,^35,36 or community-based outreach efforts,^37–39 may be particularly appropriate for patients who have not completed colorectal cancer screening despite mail and telephone-based outreach efforts. Furthermore, the efficacy of FOBT screening is considered to be equivalent to endoscopic screening only if positive FOBT tests are appropriately followed up and if the adherence to repeated FOBT screening is high.⁴⁰ Further efforts will therefore be needed to ensure appropriate endoscopic follow-up of positive FOBT tests and adequate adherence, not only to 1-time, but also to repeated FOBT screening.

Our study has several limitations. First, as a small, single-center study in a historically underserved population, our results may not be generalizable; however, our findings address the potential lack of generalizability of prior studies of similar outreach strategies to vulnerable populations. Second, the use of automated queries of the EHR data warehouse to determine patient eligibility and assess outcomes, although highly standardized, is imperfect. We did not have access to colorectal cancer screening tests that were done outside of Heartland International Health Outreach, and some tests and procedures that were completed before 2006 and scanned into the EHR at the time of the 2006 EHR conversion may not have been appropriately captured. In addition, some patients classified as eligible for colorectal cancer screening by age criteria alone may have been inappropriate candidates for screening because of comorbid conditions or limited life expectancy. These potential sources of error should have affected the intervention and control groups equally given the randomized study design, however, and thus should not have affected the main study findings.

Another important limitation of our study is related to the timing of our primary outcome assessment. The prespecified 4-month follow-up period could have biased the results in favor of the outreach group; however, a post hoc analysis at 12 months of follow-up still showed a significant difference in screening completion between groups favoring the outreach group.

In conclusion, this randomized controlled study demonstrated that a cost-effective outreach intervention consisting of the direct mailing of FOBT kits to patients overdue for colorectal cancer screening and telephone outreach to initial nonresponders was an effective means to increase patient completion of colorectal cancer screening, even in a population with a high proportion of patients from racial, ethnic, socioeconomic, and English-proficiency groups that are most at risk for not receiving appropriate colorectal cancer screening. As this outreach strategy overcomes several financial barriers (eg, sliding-scale fee for preventive visit, missed wages due to time off from work) and nonfinancial barriers (eg, transportation challenges, time away from family obligations) associated with visit-initiated cancer screening, the direct mailing of FOBT kits to patients eligible for colorectal cancer screening may be a particularly desirable approach to improving adherence to colorectal cancer screening guidelines in historically underserved communities.