JMIR Publications

ABSTRACT

Background:

Behavioral weight loss programs typically enroll 12-40 people into groups that then suffer from declining engagement over time. Online patient communities on the other hand, tend to have no limits on capacity and are fluid such that new members can enter any time. This model may be useful to boost engagement in behavioral weight loss interventions which could lead to better outcomes.

Objective:

We examined the feasibility and acceptability of continuously enrolling participants into a Facebook-delivered weight loss intervention for the first 8 of 16 weeks relative to the same intervention where no new participants were enrolled after randomization.

Methods:

We conducted a randomized pilot trial to compare a Facebook weight loss group that used open enrollment to a group that used closed enrollment on feasibility and acceptability in adults who were overweight or obese. Feasibility outcomes included retention, participant engagement, and adherence to diet tracking. We described the percent losing ≥5% of weight in both groups as an exploratory outcome. We also explored the relationship between total volume of activity in the group (from both counselors and participants) and weight loss. Participants provided feedback via a survey and focus groups.

Results:

Randomized participants (n=80; 85% female) were on average 40.2 (SD= 11.2) years old with a mean BMI of 34.4 (SD= 4.98) kg/m2. We enrolled 54 participants (93% female) into the Open Enrollment condition between weeks 1-8, bringing the total group size to 94. Retention was 98% and 100% in the Open and Closed conditions, respectively. Randomized participants across conditions did not differ in engagement (p=0.72), diet tracking adherence (p=.42), or acceptability (p’s >.05). Participant feedback in both conditions revealed sense of community was what they liked most about the program and not enough individualized feedback as what they liked the least. Weight losses of ≥5% were achieved by 30% of participants randomized to the Open Enrollment condition and 18% of participants in the Closed Enrollment condition. Exploratory analyses revealed that the Open condition (median= 385, IQR=228-536.5) had a greater volume of activity (i.e., total posts and comments) than the Closed condition (median=215, IQR=145.5-292, p=.007). Furthermore, an increase of 100 in total volume of activity in the Facebook Group each week was associated with additional 0.1% weekly weight loss among the randomized participants (p=.016), which was independent of time, individual participant engagement, and sociodemographic characteristics.

Conclusions:

Open enrollment was as feasible and acceptable as closed enrollment. Greater volume of activity in the Facebook group was associated with greater weight loss which suggests that larger groups that produce more engagement overall may be beneficial. Future research should examine the efficacy of the open enrollment approach on weight loss in a fully powered randomized trial. Clinical Trial: Clinicaltrials.gov NCT02656680