The EMBER Trial for Weight Management Engagement

This study will test whether EMBER, a self-help tool to increase weight loss treatment engagement, helps veterans engage in Veterans Health Administration (VHA) weight management programs. Participants will be randomly assigned to receive EMBER or a list of weight management programs (the control group). They will answer questions about health beliefs and behaviors and share information from their medical record at the start of the study, 2-months after the start of the study, and 6-months after the start of the study. Participants will receive compensation for each of these contacts. Participants will also receive a reminder call 10 days after they enroll in the study. All study contacts will be over the phone. The main hypothesis is that people who receive EMBER will be more likely to use VHA weight management programs than people in the control group.

Background Almost 40% of veterans using the Veterans Health Administration (VHA) have obesity. However, few patients use VHA's effective weight management programs. This study tests the effectiveness of EMBER, a self-directed tool with the goal of Enhancing Motivation for Better Engagement and Reach (EMBER) for weight management. It is available in paper and digital formats. EMBER is not a weight management program, instead it engages veterans in existing programs by informing and guiding choices about weight management. Specific Aims Assess whether veterans randomized to EMBER are more likely to have any weight management engagement at 2-month follow-up (per electronic medical record data supplemented with self-report) compared to those randomized to the control arm (information sheet listing available programs). (Primary Outcome) Assess whether veterans randomized to EMBER have greater weight management program retention, weight management behaviors (e.g., physical activity), weight loss, and quality of life gains at 6-month follow-up compared to those randomized to the control arm. (Secondary Outcomes) Assess factors likely to affect EMBER's implementation. Preliminary implementation outcomes will be assessed via RE-AIM (Reach, Effectiveness, Implementation) and the Proctor et al. implementation outcomes framework (Acceptability, Appropriateness, Costs, Fidelity). (Implementation Outcomes) Methodology Randomized two site Hybrid Type 1 Effectiveness-Implementation Trial among veteran primary care patients with obesity in VA. Participants (N=470) will be randomized to EMBER or a control condition consisting of a list of available weight management programs. Logistic regressions will be used to assess Aims 1 and 2. Aim 3 results are descriptive.

Participant outcomes will be assessed by an outside company with no knowledge of randomization status.

Dichotomous variable representing whether participant has 1+ VA weight management visits in the 2 months after baseline, per administrative data and/or self-report of use of VA weight management programs via single item question.

Dichotomous variable representing whether participant has 1+ VA weight management visits in the 6 months after baseline, per administrative data and/or self-report of use of VA weight management programs via single item question.

Count of new weight management behaviors since baseline visit via self-report to question based on NHANES item assessing whether participant tried to lose weight and how. Lowest possible value 0 new weight management behaviors, no upper limit. Any change is considered meaningful.

Self-report of quality of life as assessed with the VR12 physical and mental health composite scores (Boston University School of Public Health). Scores are standardized T-scores with mean = 50 and a standard deviation of 10. For individuals on the physical health score, a 6.5 unit change is considering clinically meaningful. For individuals on the mental health score, a 7.9 unit change is considering clinically meaningful. At the population level, a 1 unit change is considered clinically meaningful.

Based on difference between self-reported weight at baseline and 6-months, supplemented with medical record data when possible.

Measured with L-Cat (Kiernen et al, 2013), which is a 6-level categorical measure. An increase of one category is a clinically meaningful difference.

Inclusion Criteria: Veteran using primary care at VA Palo Alto or Houston in prior year BMI greater than or equal to 30 kg/m2 Exclusion Criteria: Age 80 or older Documentation of a suicide attempt in the past 30 days Hospitalization in the past 30 days Documentation or other evidence of cognitive impairment VA weight management program use in past 2 years Self-report from potential participant that they will not be in town for the majority of the 2 months following baseline Self-report from potential participant they plan to leave VA Palo Alto or Houston within the next 6 months Under age 18

Datasets meeting VA standards for disclosure to the public will be made available within 1 year of publication. Prior to distribution, a local privacy officer will certify that all datasets contains no PHI. Final data sets will be maintained locally until enterprise-level resources become available for long-term storage and access. Guidance on request and distribution processes will be provided by VA Office of Research & Development (ORD). Those requesting data will be asked to sign a Letter of Agreement.