Abstract 12491: Impact of Augmented Treatments on Physical Activity Outcomes in a Sequential Multiple Assignment Randomized Trial

Background: Nearly half of women do not meet aerobic physical activity (PA) guidelines. Despite proven efficacy of PA interventions, response varies. Little is known about when treatment should be augmented. The Working Women Walking Program is a Sequential Multiple Assignment Randomized Trial testing two initial treatments (Fitbit vs. Fitbit+text messages) for eight weeks and two augmented treatments (personal calls vs. group meetings) added at eight weeks for non-responders to the initial treatments.

Aims: We aim to compare the two augmented treatments among non-responders to the initial treatments on PA after 34 weeks of the program.

Methods: Participants ( N = 301; women ages 18-70) were randomly assigned to an initial treatment of either Fitbit ( n = 149) or Fitbit+text messages ( n = 152). Both groups received a Fitbit and PA prescription (increase daily steps by ≥600 above baseline). Response to treatment was defined as (a) wearing the Fitbit 10 hours per day 3 of 7 days weekly in weeks 6-8 and (b) reaching a step goal ≥600 steps above baseline for 2-3 weeks in weeks 6-8. Non-responders were randomly assigned to an augmented treatment with personal calls ( n = 72; 50.7%) or group meetings ( n = 70; 49.3%). Objective PA was assessed with the waist-worn ActiGraph accelerometer (≥4 days with one weekend day and ≥10 hours daily wear time) at baseline and 34 weeks to assess daily steps and minutes of light and moderate-vigorous intensity PA (MVPA). Self-report PA was measured with the International PA Questionnaire to assess daily MVPA and walking.

Results: Across all PA measures, there were no significant differences between augmented treatments (personal calls vs. group meetings). Overall, both treatments showed non-significant increases in objective PA, including steps ( d = .15; p = .30), light PA ( d = .11; p = .39), and MVPA ( d = .22; p = .15), but showed significant increases in self-report PA, including ( d = 1.02; p < .001) and walking ( d = .73; p = .30).

Conclusion: In this sample of working women who were non-responsive to an initial PA treatment (Fitbit or Fitbit+texts), augmenting with personal calls or group meetings did not yield a difference in PA outcomes. Next, we will explore participant characteristics that may differentiate response to augmented treatments.