Financial vs. Non-Financial Rewards for Weight Loss and Weight Maintenance

Financial vs. Non-Financial Rewards for Weight Loss and Weight Maintenance: A Randomized Controlled Trial

Three of every four Veterans are overweight or obese, and weight loss is associated with reduced morbidity and mortality. The VA MOVE! program for weight loss is moderately effective but only reaches a small percentage of overweight Veterans. This proposed study will test whether a patient incentive program that gives Veterans non-financial incentives, such as Seattle Mariners baseball tickets, for losing one pound per week over 16 weeks is effective. An effective patient-incentive program could help more Veterans lose weight without requiring a substantial increase in VA staff.

Project Background: Behavioral economics suggests that the chronic inability to make the daily behavioral changes that can help us lose weight may be the result of "present bias," which is a tendency to value small, immediate rewards over large rewards in the distant future. For many of us, the immediate gratification of eating an unhealthy food is a more powerful motivator than is the elusive dissatisfaction of the long-run health consequences of an unhealthy diet. Patient rewards may overcome present bias by moving the rewards for healthy behaviors forward in time. In a patient reward program, patients are given tangible, timely rewards for achieving specific health goals, such as losing one pound per week over 16 weeks. Meta analyses of randomized trials have found that rewards for weight loss are effective during the reward period, but the weight loss was not sustained after the reward was removed. Thus, the key challenge to a reward program is not achieving weight loss, but maintaining it. The proposed study tests the hypothesis that the significant weight regain found in prior reward trials can be attributed to use of financial rewards-e.g., cash or the equivalent of cash-in those trials. Experiments in behavioral economics have found that providing participants with financial rewards for participating in a study invokes behavior defined by reciprocity-the effort the participants gave in the study was proportional to the amount of money that they were given. When participants were given non-financial rewards, they exhibited no reciprocity-the effort was consistently high and did not vary with the quantity of the non-financial reward. By using financial rewards, prior trials may have invoked money-market norms of reciprocity, such that patients' efforts toward weight loss were high when rewards were offered, and reduced when they were discontinued. The investigators hypothesize that non-financial rewards, like tickets to a Seattle Mariners baseball game, will not invoke reciprocity or the consequent weight regain. Project Objectives: The goal of this study is to test, through a randomized trial, the effectiveness of providing overweight Veterans with financial or non-financial rewards for a one pound weight loss per week over 16 weeks. The primary outcome is weight loss at 32 weeks-16 weeks after the discontinuation of the rewards. Secondary outcomes include weight loss at 16 weeks and 12 months. Project Methods: The investigators will conduct a three-armed randomized trial of patient rewards for losing one pound per week over 16 weeks. The three treatment groups will receive financial rewards, non-financial rewards, or no rewards. The investigators hypothesize that: 1) patients who receive non-financial rewards for weight loss over 16 weeks will have greater weight loss at 32 weeks than patients who do not receive rewards; 2) patients who receive non-financial rewards for weight loss over 16 weeks will experience weight loss at 16 weeks that is not inferior to the weight loss of patients who receive financial rewards; and 3) weight regain will be greater among patients who received financial rewards compared to patients who received non-financial rewards or no rewards. The investigators will also conduct post-intervention qualitative interviews and perform a cost analysis.

At enrollment, participants randomized to usual care will be given the MOVE! workbook that serves both as an information resource on diet and exercise, a resource for tools to achieve weight loss goals, and a log of participants' goals, motivations, and outcomes. Each participant will be given a chart with a personalized target weight for a loss of 1lb. per week for 16 weeks. Participants will be given a digital scale or wireless scale and counseled to weigh themselves daily. Participants will enter their weight weekly into an online portal. They will also complete the surveys in the portal. They will receive text message reminders to enter their weight.

Same as usual care, plus financial rewards that are earned in two ways: an assured and random. For the assured reward, they will receive compensation at the end of each month that they are at or below their target weight on the last day of that period. For the random portion, each week that a participant is at or below their target weight, the patient is entered in random drawing to win additional compensation. Over the first eight weeks the patient has a 1-in-8 chance of winning.Over the 17-32 week period, Veterans in this group will receive token rewards for tracking and reporting their weekly weights regardless of whether it was on target. Each week that the patient enters his/her weight into the portal, he will have a 1-in-8 chance to earn the token reward, such as a t-shirt or movie tickets. Participants will enter their weight weekly into an online portal. They will also complete the surveys in the portal. They will receive text message reminders to enter their weight.

Procedures for the non-financial rewards arm is identical to procedures for the financial reward arm, except that the Veteran will earn points rather than cash. Each week a random drawing will be for 20 points, each monthly weigh in is worth 5 points. Veterans will be given non-financial rewards associated with the number of points they earn.

Veterans in the financial reward arm receive all of the services of the usual care arm, plus rewards that are earned in two ways: an assured portion, and a random portion. For the assured reward, they will receive compensation at the end of each month (defined as a four week period) that they are at or below their target weight. For the random portion, each week that a participant is at or below their target weight, the patient is entered into a drawing to win additional compensation. Over the first eight weeks the patient has a 1-in-8 chance of winning. Over the 17-32 week period, Veterans in this group will receive token incentives for tracking and reporting their weekly weights regardless of whether it was on target. Each week that the patient enters their weight into the portal, s/he will have a 1-in-8 chance to earn the token incentive, such as a t-shirt or movie ticket. The participant will receive reminder text messages.

Procedures for the non-financial rewards arm are identical to procedures for the financial reward arm, except that the Veteran will earn points rather than cash. Each weekly random drawing will be for 20 points. Veterans will choose non-financial rewards associated with the number of points they earn.

Measured by survey responses to the Weight Efficacy Lifestyle Questionnaire Short Form (WEL-SF).Scale goes from 0 (not at all confident) to 10 (very confident). Higher scores mean better outcomes. Missing values at follow-up were imputed using multiple imputation by chained equations (MICE)

Measured by survey responses to the Weight Efficacy Lifestyle Questionnaire Short Form (WEL-SF).Scale goes from 0 (not at all confident) to 10 (very confident).

Measured by survey responses to the Weight Efficacy Lifestyle Questionnaire Short Form (WEL-SF).Scale goes from 0 (not at all confident) to 10 (very confident).

Measured by patient survey responses to the Regulation of Eating Behavior Scale. Scale from 1 (does not agree at all) to 7 (agrees exactly). Higher values represent greater intrinsic motivation

Measured by patient survey responses to the Regulation of Eating Behavior Scale. Scale from 1 (does not agree at all) to 7 (agrees exactly). Higher values represent greater intrinsic motivation

Measured by patient survey responses to the Regulation of Eating Behavior Scale. Scale from 1 (does not agree at all) to 7 (agrees exactly). Higher values represent greater intrinsic motivation

Measured by patient survey responses to the Personal Health Questionnaire Depression Scale (PHQ-8). The scale ranges from 0 to 24. Higher scores mean more severe depression.

Measured by patient survey responses to the Personal Health Questionnaire Depression Scale (PHQ-8). The scale has the following options: not at all, several days, more than half the days, nearly every day. Scores range from 0 to 24. Higher scores represent more severe depression.

Measured by patient survey responses to the Personal Health Questionnaire Depression Scale (PHQ-8). Measured by patient survey responses to the Personal Health Questionnaire Depression Scale (PHQ-8). The scale ranges from 0 to 24. Higher scores mean more severe depression.

Inclusion Criteria: Veteran BMI>=30 at enrollment weigh less than 390lbs Active patient in primary care or women's clinic (1+ visit in last year) Access to a text capable phone in the Veteran's household or active email address live in Seattle area the entire year access to the internet Exclusion Criteria: behavioral flag serious mental illness or anti-psychotic medication eating disorder and/or sexual trauma MOVE! participation in past 4 months pregnant or planning to become pregnant prisoner/employee/student inability to independently stand inability to read insulin dependent impaired decision making no access to a cell phone or phone that can receive text messages inability to remove socks & shoes >5% of body weight lost in last 6 months (clinical data and self-report) unable to pass cognitive screening

A corresponding data repository has been created for this project. Participants must consent to both the main project and the data repository to be included in both. Otherwise the participant will only be included in this study.