Aligning Colon Cancer Screening Decision Aids with the Health Literacy Needs of Older Adults

Background and Purpose: Nationally, 46% of colorectal cancer (CRC) deaths occur in adults >75 years. The benefit to risk ratio for colorectal cancer (CRC) screening for this population is uncertain. National guidelines recommend that adults >75 engage in shared decision making (SDM) around screening to weigh risks versus benefits. Decision aids (DAs) are designed to increase patient knowledge and engagement in SDM. However, most DAs are not designed for limited health literacy (LHL), which affects 39% of older adults. To address this gap, we pilot tested a modified CRC DA for older adults with LHL.

Methods: We recruited 55 participants, ages 75-85 with LHL, to participate in a pilot randomized control trial testing the effects of the modified CRC DA. Participants with LHL responded “somewhat” to “not at all confident” to the validated health literacy question, “How confident are you filling out medical forms by yourself?” and/or had some college or less. Patients were randomized to the intervention brochure (the CRC DA) or control brochure. A pre-test/post-test questionnaire that included the validated SURE test for decisional conflict, a CRC knowledge test, and a screening intentions scale was administered to patients before and after reading a randomized brochure in their doctor’s waiting room. Changes in decisional conflict, knowledge, and screening intentions were analyzed using paired and Welch’s two-sample t-tests.

Results: Participants were predominantly female (70%), Black (63%), with a high school degree or less (52%). The mean age of participants was 79 years. After randomization, 49% of the participants were assigned to the CRC DA brochure and 51% were assigned to the control brochure. Patients who received the CRC DA decreased their uncertainty (decisional conflict) about screening (Mpre=2.85; Mpost=3.52, p=0.024), as did patients in the control group (Mpre=2.96; Mpost=3.37, p=0.05). Changes to knowledge levels among those in the CRC DA group were significantly larger than in the control group (CRC DA=-0.222, Control=0.815, p=0.016). Patients who received the CRC DA shifted toward intending to not screen (Mpre=5.67; Mpost=8.26, p=0.039), compared to the control group, where there was no significant change in screening intentions (Mpre=4.48 ; Mpost=4.96, p=0.59). Finally, in the CRC DA group, 56% had a conversation with their provider about colon cancer screening during the visit compared to 41% in the control group.

Conclusion and Implications: Findings from this study suggest that a modified CRC DA developed for adults with LHL can increase knowledge about CRC screening, a key component of engagement in SDM about CRC screening. LHL is more common among historically marginalized communities, including racial and ethnic minorities and low-income populations. Making a CRC DA more accessible ensures that everyone, regardless of literacy level has the information needed to make informed health choices and reduce the risk of late-stage diagnoses and preventable deaths.