The efficacy of tailored print materials in promoting colorectal cancer screening: Results from a randomized trial involving callers to the National Cancer Institute's Cancer Information Service

In this large randomized trial among callers to the Cancer Information Service (CIS), tailored print materials were tested for efficacy in promoting colorectal cancer (CRC) screening (fecal occult blood test [FOBT], flexible sigmoidoscopy, or colonoscopy). All participants completed baseline interviews at the end of their usual service calls to the CIS, as well as short-term (6-month) and longer-term (14-month) telephone follow-up interviews. The study sample (n = 4,014) was restricted to English-speaking CIS callers 50+ years of age, who would be eligible for CRC screening at 14 months follow-up and did not call the CIS about CRC or CRC screening. Four experimental conditions were compared: a single untailored (SU) mailout of print material (the control condition); a single tailored (ST) mailout of print material; four (multiple) tailored (MT) mailouts of print materials spanning 12 months, all of which were tailored to information obtained at baseline; and four (multiple) retailored (MRT) mailouts also spanning 12 months, with retailoring of the print materials (mailouts 2, 3, and 4) based on updated information obtained from the 6-month follow-up interviews. Consistent with the main hypothesis of this trial, a significant linear trend across the SU, ST, MT, and MRT groups was found at 14 months (42%, 44%, 51%, and 48%, respectively, p = 0.05). Only for MT was there a significant difference compared with SU (p = 0.03) for the sample as a whole, while no differences were found for MT vs. MRT at 14 months. Significant moderator effects in the predicted direction were found among females, younger participants, and among those with a history of CRC screening, all of which involved the SU vs. MT MRT comparisons. Only among younger participants (ages 50-59) was there a difference between SU vs. ST at 14 months. Given these results, we conclude from this trial the following: (1) the MRT intervention failed to show added benefit beyond the MT intervention, (2) the significant intervention effects involving the MT and MRT conditions can be explained by tailoring and/or the longitudinal nature of both interventions, and (3) the most compelling evidence in support of tailoring was found for the ST condition among younger participants, where a significant need for interventions exists at the national level. Directions for future research are discussed in light of the results summarized above.