A Web-Based Intervention to Encourage Walking (StepWise): Pilot Randomized Controlled Trial

Abstract

Background: Despite Internet-based interventions that incorporate pedometers with appropriate goal-setting processes and other theoretically-based behavior change strategies being proposed as a means of increasing walking behavior, few have incorporated all of these key features or assessed maintenance of behavior change.

Objective: The objective of our study was to investigate the effect of a 12-week pedometer step goal walking program individually tailored to baseline step counts, combined with an interactive support website for step counts, health parameters and motivation over 12 and 24 weeks.

Methods: Low active participants (mean [SD] 46.2 [11.2] years) were randomly assigned to the Stepwise (SW) intervention group (n=49) or a comparison (CP) group (n=48). SW received a pedometer, step goal walking program and access to the SW website (containing interactive self-monitoring and goal feedback tools, motivational messages and action and coping planning strategies). CP received a pedometer and locally available physical activity information. Step counts, BMI, resting heart rate, blood pressure and glucose, cholesterol and triglyceride levels, psychological well-being, perceived health, self-efficacy and self-determined motivation were measured at baseline, 12 and 24 weeks.

Results: Linear mixed model analysis found that both groups' step counts increased from baseline to week 12 (β = 11,002, CI 5739-16,266, P<.001) and 24 (β = 6810, CI 1190-12,431; P=.02). Group step counts were significantly different at week 24 with SW taking 8939 (CI 274-17604, P=.04) more steps compared to CP. Compared to baseline, both groups had improved triglyceride levels (0.14 mmol/L, CI -0.25 to -0.02, P=.02) at week 12, decreased diastolic blood pressure (4.22 mmHg, CI -6.73 to -1.72) at weeks 12 and 24 (3.17 mmHg, CI -5.55 to -0.78), improved positive (β = .21, CI 0.03-0.38, P=.02) and negative affect (β = -.15, CI -0.28 to -0.03, P=.02) at week 12, and perceived health at week 12 (β = 6.37, CI 2.10-10.65, P=.004) and 24 (β = 8.52, CI 3.98-13.06, P<.001). Total cholesterol increased at week 12 (0.26 mmol/L, CI 0.099-0.423, P=.006) and week 24 (0.38 mmol/L, CI 0.20-0.56, P<.001). Repeated measures ANOVA found motivation for walking improved from baseline with higher task self-efficacy (P<.001, η(2) = .13) and autonomous motivation (P<.001, η(2)=.14) at weeks 12 and 24 and decreased controlled motivation (P=.004, η(2)=.08) at week 24.

Conclusions: Both groups had similar improvements in step counts and physical and psychological health after 12 weeks but only the SW group successfully maintained the increased step-counts 24 weeks post-intervention. This suggests the step-goal based walking program combined with Internet-based behavior change tools were important for sustained behavior change.

Keywords: physical activity, behavior change strategies, pedometer, self-monitoring, step goal walking program.

Figure 1

Examples of success and failure messages shown once step counts were entered.

Figure 2

The graph showing step goals and number of steps achieved each week and the step goal for the following week along with a motivational message.

Figure 3

Flow of participants through the study.

Figure 4

Mean steps per week taken by the SW and CP groups at baseline, week 12 and 24 (* denotes significant difference between groups, P=.04).