Abstract P418: Lapses in Physical Activity Before and During Behavioral Intervention: Results from the SMART Trial
Background: Interruptions or “lapses” in physical activity (PA) are common. Less is known about how patterns of lapse may be affected by a behavioral intervention or how patterns or “profiles” of lapse behavior may vary over time.
Methods: The SMART Trial is a behavioral weight loss trial in which 210 overweight adults were randomized with equal allocation to one of 3 self-monitoring (SM) methods: 1) paper record (PR); 2) personal digital assistant with a dietary and PA software (PDA); and 3) PDA with tailored PA feedback message every other day (PDA+FB). At baseline and 6, 12, 18 and 24 mos, PA frequency and lapses were measured with a PA interview. PA lapse (“stopped exercising for ≥ 2 weeks”) responses were classified as regular PA (≥3 sessions/wk) with (+) or without (-) lapse (RPA±L), occasional PA (<3 sessions/wk) with (+) or without (-) lapse (OPA±L) or no PA in the past 6 mos. In addition to considering individual responses at each time point, we created PA profiles using collapsed response categories over time, where we considered the following categories: missing data, no PA, occasional or lapse (including OPA±L and RPA+L), and RPA-L.
Results: The sample was 85% female and 78% White with mean (+SD) age of 46.8+9.0 years and mean baseline BMI of 34.0+4.5 kg/m2. Median PA level (MET-hr/wk) was 7.4 (IQR: 2.1, 16.3) at baseline, 13.4 (5.4, 24.8) at 6 months, 11.5 (5.2, 22.0) at 12 months, 9.4 (2.9, 19.7) at 18 months, and 7.5 (2.6, 15.8) at 24 months, with no differences by SM method at any time point. Reports of PA lapses were common at baseline with 39 (19%) reporting RPA+L and 37 (18%) reporting OPA+L. At subsequent evaluations, lapses were 31%, 26%, 20%, and 23% reporting RPA+L and 13%, 13%, 13%, and 14% reporting OPA+L at 6, 12, 18, and 24 mos, respectively. However, we noted that over time more reported RPA-L (14%, 37%, 35%, 30%, 24% at baseline, 6, 12, 18, and 24 months) and fewer reported no activity (17%, 1%, 5%, 10%, 9% at baseline, 6, 12, 18, and 24 months). Lapse patterns did not differ by SM method at baseline or any subsequent time point. A review of PA profiles over the 5 time points revealed that the most common profile was persistent occasional or lapse PA (including OPA±L and RPA+L) in 20% of the sample. Only 5% of the sample had persistent RPA-L. Smaller sub-groups demonstrated varying patterns of physical activity and lapses.
Conclusions: Our intervention increased regular PA without lapses and decreased sedentary behavior; however, lapses in PA remain common, even during an intensive behavioral intervention. We saw many profiles of lapse behavior over time, suggesting that interventions and feedback may need to be tailored to an individual’s response pattern.
Supported by NIH Grant R01 DK071817.
- © 2013 by American Heart Association, Inc.