Optimization of Adaptive Text Messages for Cancer Survivors (OATS II)

This clinical trial evaluates whether an adaptive text-message intervention is useful in helping survivors of colorectal cancers (CRC) eat more whole grain foods and less refined grain foods. Most CRC survivors don't achieve the recommended intakes of whole grains or fiber, even though there is strong evidence that a high-fiber diet rich in whole grains lowers the risk of death from CRC. Dietary interventions are a promising approach for reducing death from CRC, and text message interventions specifically are a promising tool for reaching diverse populations. This trial evaluates a text-message based dietary intervention that continuously adapts message content to be specifically tailored for the participant for increasing whole grain consumption.

PRIMARY OBJECTIVE: I. Determine the intervention's feasibility and acceptability. SECONDARY OBJECTIVES: I. Estimate the effect of the intervention on the percent of grains consumed that are whole. II. Estimate the effect of the intervention on total fiber intake (grams per day [g/d]). EXPLORATORY OBJECTIVE: I. Assess convergence of the reinforcement learning (RL) algorithm. OUTLINE: Patients receive nutrition education materials and then receive the adaptive text message intervention for 12 weeks on study. Patients who are food insecure also receive resources for food banks, information about meal delivery programs, and support for applying for Supplemental Nutrition Assistance Program (SNAP) benefits.

Adaptive 12-week text message intervention using reinforcement learning to increase whole grain and reduce refined grain intake. Participants will also complete semi-structured follow-up interviews and study related questionnaires.

Various questionnaires evaluating participants dietary habits and use of study intervention will be administered

Participants' responses to text messages that ask for a response will be quantified using medians and interquartile range (IQR). The intervention will be determined feasible if the median response proportion to text messages that ask for a reply is >= 70%.

The system usability is comprised of 10 questions assessing participants perceived usability of the intervention. Each item is scored on a scale of 1 (strongly disagree) to 5 ( strongly agree). The SUS yields a single number representing a composite measure of the overall usability of the text messaging system. To calculate the SUS score, Each item's score contribution will ultimately range from 0 to 4 (For items 1,3,5,7,and 9 the score contribution is the item score minus 1 and for items 2,4,6,8 and 10, the contribution is the item score minus 5). The sum of the scores are then multiplied by 2.5 to obtain the overall value. SUS scores have a range of 0 to 100 with a score > 68 indicating above average usability.

The AIM is a four-item measure of implementation that assesses participants perceived acceptability of the intervention. Each item response falls on a scale of 1 (Completely disagree) to 5 (Completely agree). A scaled score is calculated by averaging responses. Scale values range from 1 to 5. The intervention will be determined acceptable if the overall median score is >= 4.

The IAM is a four-item measure of implementation that assesses participants perceived appropriateness of the intervention. Each item response falls on a scale of 1 (Completely disagree) to 5 (Completely agree). A scaled score is calculated by averaging responses. Scale values range from 1 to 5. The intervention will be determined acceptable if the overall median score is >= 4.

The FIM is a four-item measure of implementation that assesses participants perceived appropriateness of the intervention. Each item response falls on a scale of 1 (Completely disagree) to 5 (Completely agree). A scaled score is calculated by averaging responses. Scale values range from 1 to 5. The intervention will be determined acceptable if the overall median score is >= 4.

The FFQ will ask about diet in the past 3 months. For each item, participants will be asked to report how often, on average, they ate the specified portion size in the past 3 months. Participants can choose from frequency options ranging from never or less than once per month to >=6 times a day. To compute calorie and nutrient intakes, multiply the frequency of consumption of each food by the amount of each nutrient in the specified portion size using composition values from the U.S. Department of Agriculture and other sources. Percent of grains that are whole = servings per day of whole grains/(servings per day of refined grains + servings per day of whole grains) will be reported over time

The FFQ will ask about diet in the past 3 months. For each item, participants will be asked to report how often, on average, they ate the specified portion size in the past 3 months. Participants can choose from frequency options ranging from never or less than once per month to >=6 times a day. To compute calorie and nutrient intakes, multiply the frequency of consumption of each food by the amount of each nutrient in the specified portion size using composition values from the U.S. Department of Agriculture and other sources. The mean total daily fiber intake and 95% confidence intervals will be reported over time.

Inclusion Criteria: Age > 18 years. Subject has provided informed consent. Diagnosis of colon or rectal adenocarcinoma. Not on active treatment at the time of screening and not expected to receive active anticancer therapy (e.g., surgery, radiation, chemotherapy) during the study period. At least 6 weeks since a major surgery and fully recovered. Owns a mobile phone able to receive and send text messages. Able to speak/read English or Spanish. Based on a screening survey, eat grains and <50% of total grains are whole grains. Exclusion Criteria: 1. Does not meet any of the above inclusion criteria.