Time for change: using implementation intentions to promote physical activity in a randomised pilot trial

Abstract

Objective: A common barrier to exercise is a perceived lack of time. The current pilot study examined the effects of an implementation intention intervention to enhance exercise self-efficacy, increase confidence to exercise when facing time constraints, and increase physical activity in middle-aged adults (n = 63, aged 35-69).

Design: Participants received a pedometer (Fitbit) to objectively measure activity and were randomly assigned to either a control or intervention condition. After a 1-week baseline, the intervention condition received instructions to plan how, where, and when they would add steps to their daily routine to meet their step goal, using personalised schedules and maps. Both groups were contacted nightly via email.

Main outcome measures: Physical activity (steps and time spent in moderate-to-vigorous activity), goal achievement, exercise self-efficacy, time-relevant exercise self-efficacy and affect.

Results: Compared to the control, the intervention condition significantly increased in steps, time spent in moderate-to-vigorous activity, and time-relevant exercise self-efficacy. Goal achievement was related to greater time-relevant exercise self-efficacy and more positive affect at the daily level.

Conclusion: Findings suggest that the personalised planning intervention increased physical activity and confidence in achieving physical activity goals under time constraints. Avenues for future directions, especially for producing more sustained effects, are discussed.

Keywords: adults; intervention; physical activity; randomised controlled trial; self-efficacy.

Figure 1.

Consolidated Standards or Reporting Trials (CONSORT) diagram.

Figure 2.

Multilevel interaction between conditions across weeks. Error bars represent the standard error of the mean.

Figure 3.

Mixed model ANOVA interaction between conditions across pre- and post-test scores for exercise self-efficacy time subscale. Error bars represent the standard error of the mean.