A stepped randomized trial to promote colorectal cancer screening in a nationwide sample of U.S. Veterans

Abstract

Background: Colorectal cancer (CRC) screening (CRCS) facilitates early detection and lowers CRC mortality.

Objectives: To increase CRCS in a randomized trial of stepped interventions. Step 1 compared three modes of delivery of theory-informed minimal cue interventions. Step 2 was designed to more intensively engage those not completing CRCS after Step 1.

Methods: Recruitment packets (60,332) were mailed to a random sample of individuals with a record of U.S. military service during the Vietnam-era. Respondents not up-to-date with CRCS were randomized to one of four Step 1 groups: automated telephone, telephone, letter, or survey-only control. Those not completing screening after Step 1 were randomized to one of three Step 2 groups: automated motivational interviewing (MI) call, counselor-delivered MI call, or Step 2 control. Intention-to-treat (ITT) analyses assessed CRCS on follow-up surveys mailed after each step.

Results: After Step 1 (n = 1784), CRCS was higher in the letter, telephone, and automated telephone groups (by 1%, 5%, 7%) than in survey-only controls (43%), although differences were not statistically significant. After Step 2 (n = 516), there were nonsignificant increases in CRCS in the two intervention groups compared with the controls. CRCS following any combination of stepped interventions overall was 7% higher (P = 0.024) than in survey-only controls (55.6%).

Conclusions: In a nationwide study of Veterans, CRCS after each of two stepped interventions of varying modes of delivery did not differ significantly from that in controls. However, combined overall, the sequence of stepped interventions significantly increased CRCS.

Keywords: Behavioral intervention; Colorectal cancer screening; Population-based.

Figure.

CONSORT flow diagram.