Planning the Next Steps: Using an Implementation Intention Approach to Increase Daily Walking

The 5-week study consists of a 1-week baseline, where participants are asked to wear a Fitbit to document the number of steps taken that day to establish an objective measurement of steps before the intervention began. In the intervention condition, after the baseline week, during weeks 2 to 5, this condition is prompted with an email each evening for four weeks to review their schedules for the following day and identify time slots where they could add activity. They are given instructions for providing a detailed calendar of appointments and open slots for the next day using a simple daily planner. The planning implementation intention manipulation involves recording specific information about when, where, and how they will add steps to their day. They are provided with maps near their home and work with specific information about distances, estimated time to walk between different points, and number of steps for specific routes to help them in planning for specific routes.The control condition is matched for how much contact they have with the research staff (called and emailed the same amount of times) and also wear the Fitbit daily. The only difference from the intervention group is that they do not get the daily planning instructions or maps for the implementation intention strategy treatment. All activity data from the Fitbit was deidentified and aggregated with an online platform called Fitabase.

• Control Group (Behavioral): Participants will wear a Fitbit Zip for 5 weeks to record their daily activity data. The first week of the study will provide a baseline measurement of activity. After this week, participants in this condition will begin receiving daily emails asking them to report their step count in a questionnaire. Participants in the control group will be matched with the intervention group for how much contact they have with the researcher.
• Implementation Intention Condition (Behavioral): Participants will wear a Fitbit Zip for 5 weeks to record their daily activity data. The first week of the study will provide a baseline measurement of activity. After this week, participants will be asked to increase their steps incrementally by 2,000 steps each week for the subsequent four weeks. To help achieve their step goals, they will receive maps of different routes near their home and/or work of varying distances and step counts. They will be asked to review their schedule for the next day and identify times when they could add steps into their schedules, and to record their daily step data in the daily questionnaire.

Primary outcomes

• Steps [Week 1 & Week 5]: Weekly average of daily step counts with Fitbit (averaged across 7 days at Week 1 and averaged across 7 days at Week 5).
• Activity Intensity [Week 1 & Week 5]: Weekly average of daily time spent in moderate to vigorous intensity activity (averaged across 7 days at Week 1 and averaged across 7 days at Week 5).

Secondary outcomes

• Cognitive Composite Score [Baseline (Pre-Test) and Week 5 (Post-Test )]: Z-score composite on the Brief Test of Adult Cognition by Telephone (BTACT). Individual tests scores were first standardized to z-scores. The z-score composite was calculated by averaging the z-scores of the 5 tests: word list immediate, word list delayed, backwards counting, digits backwards, and category fluency. Post-test z-scores were standardized based on the mean and s.d. of the pretest scores. A higher z-score is indicative of better cognitive functioning.
• Exercise Control Beliefs [Baseline (Pre-Test) and Week 5 (Post-Test )]: Degree of perceived control over Exercise. Range from 1 (Strongly agree) to 5 (strongly agree). Reverse coded so that a higher number indicates more perceived control over exercise.
• Exercise Self-efficacy [Baseline (Pre-Test) and Week 5 (Post-Test )]: Amount of confidence in ability to exercise.
• Exercise Self-efficacy Beliefs - Time Composite [Baseline (Pre-Test) and Week 5 (Post-Test )]: Amount of confidence in ability to exercise when facing time constraints. Range from 1 (Very Sure) to 4 (Not sure at all). Reverse coded so that a higher number indicates more self-efficacy. 3 items, summed to form time-relevant composite scale.

Sex: All, ages: 35 Years to any, accepts healthy volunteers

Inclusion Criteria:

* Currently working full time
* Not engaged in regular exercise
* 35 years of age or older

Exclusion Criteria:

* Not healthy enough to engage in a walking intervention

Participant flow

Outcome results

• Steps (Primary), Mean, Daily Steps: Control Group 6337.77 +/- 2207.44; Implementation Intention Condition 6966.06 +/- 2818.05 (p =.006)
• Activity Intensity (Primary), Mean, Minutes: Control Group 18.71 +/- 20.31; Implementation Intention Condition 22.25 +/- 22.05 (p .034)
• Cognitive Composite Score (Secondary), Mean, Z-score: Control Group -.17 +/- .70; Implementation Intention Condition .17 +/- .48 (p .065)
• Exercise Control Beliefs (Secondary), Mean, units on a scale: Control Group 4.26 +/- .52; Implementation Intention Condition 4.26 +/- .72 (p .308)
• Exercise Self-efficacy (Secondary), Mean, units on a scale: Control Group 2.15 +/- 0.99; Implementation Intention Condition 2.19 +/- 0.81 (p .349)
• Exercise Self-efficacy Beliefs - Time Composite (Secondary), Mean, units on a scale: Control Group 1.96 +/- 1.05; Implementation Intention Condition 1.90 +/- .75 (p .022)

Adverse events

• Robinson SA, Lachman ME. Perceived Control and Aging: A Mini-Review and Directions for Future Research. Gerontology. 2017;63(5):435-442. doi: 10.1159/000468540. Epub 2017 Apr 8. PMID 28391279
• Robinson, S. A. & Lachman, M. E. (2016). Perceived Control and Behavior Change: A Personalized Approach. In F. Infurna & J. Reich (Eds.), Perceived Control: Theory, Research, and Practice in the First 50 Years (pp. 201-227). New York, New York: Oxford University Press.

• Margie Lachman, Ph.D., Principal investigator, Brandeis University