|Home | About | Journals | Submit | Contact Us | Français|
The purpose of this article is to describe a pedometer-based physical activity intervention (and the research that has been conducted on this program) to provide insight into workable strategies focused on promoting lifestyle physical activity. The First Step Program (FSP) is facilitated theory-based behavior modification program, originally developed for individuals with Type 2 diabetes, which capitalizes on the unique properties of a pedometer to serve as a stimulus for walking and an instrument for individual goal-setting, self-monitoring, and feedback. Formative evaluation revealed that participants were highly enthusiastic about the program and pilot testing demonstrated an immediate and dramatic increase in walking behavior (by approximately 3,700 steps/day or 34 minutes of walking a day). A randomized and controlled evaluation produced similar results, as have head-to-head comparisons of participant outcomes produced by professional vs. peer delivery, and adaptations of the program to worksite and community-based delivery. The effectiveness of the FSP can be explained by a thoughtful consideration of pedometer characteristics, program features, and the people who participate. Sustained delivery is contingent on funding sources and administrative structures that support formalized implementation.
The recent release of the U.S. Physical Activity Guidelines highlights the multiple benefits of physical activity and provides science-based directives concerning the quantity of this behavior that is associated with diverse health outcomes.1 This effort is commendable and updates and strengthens previous guidelines.2 Despite their existence, however, less than 5% of the adult U.S. population actually achieves these objectively monitored levels of physical activity.3 To emphasize, the question remains how best to help those who can benefit most make and sustain physical activity behavioral changes. “A lack of time” remains one of the most frequently reported reasons for not engaging in healthful physical activity.4, 5 In response, there has been a growing interest in the delivery of “lifestyle-based” programs that offer individuals flexibility in how, when and where they obtain their physical activity.6 The First Step Program (FSP), originally created by Dr. Catrine Tudor-Locke as part of her doctoral training is one such lifestyle physical activity intervention. The purpose of this article is to describe the FSP and the research that has been conducted on this program to provide insight into strategies focused on promoting lifestyle physical activity. Such a discourse will be especially useful to health care practitioners interested in delivering lifestyle physical activity programs.
Inspired by the work of Yamanouchi et al.,7 the FSP uses an electronic pedometer to serve as a stimulus for walking and an instrument for individual goal-setting, self-monitoring, and feedback. Originally developed for individuals with Type 2 diabetes, the FSP grew out of a needs assessment8 that revealed that the target population (notorious for attrition from structured exercise programs) wanted a simple regimen that was flexibly scheduled and individually tailored. Further, it was clear that program deliverers were not sure about what physical activity messages were best to convey to their patients. Following a review of the literature,9 the theory underlying the FSP was fully articulated prior to its initial implementation.10 Briefly, it is a facilitated behavior modification program built primarily on the framework of Social Cognitive Theory,11 emphasizing the components of self-efficacy and social support, and also drawing from the behavioral processes of the Transtheoretical Model.12 The FSP is organized into two phases: an initial 4-week adoption phase and a subsequent 12-week adherence phase. During the adoption phase, participants use a pedometer to monitor their own walking behaviors by recording their daily values on a simple calendar, and attend weekly facilitated group meetings to review the previous week’s behaviors, discuss selected strategies for success, and set incremental steps/day goals based on perceived abilities, time constraints, and other personal issues. Each meeting also includes a short group walk integral for promoting self-efficacy through a simple mastery experience.11 Facilitators guide participants through decision balance techniques early in the behavior change process, and self-contracts are reviewed as an iterative process of augmenting individualized goals on a weekly basis. Relapse prevention and planning is practiced during the adoption phase and encouraged during the adherence phase.13 Participants are also encouraged to tap into their own unique social support networks to help sustain their behavior beyond formal group meetings. During the adherence phase, participants continue with their self-monitoring and goal-setting, practicing the strategies they have previously learned with little or no direct program contact.
Formative evaluation, or pilot testing, is a systematic appraisal of a program prototype in its developmental or draft stage, prior to full scale delivery.14 The FSP was piloted with 9 sedentary individuals with Type 2 diabetes who engaged in outcome measurements (steps/day, blood pressure, simple anthropometry) and further obliged us by permitting video tapping of weekly meetings and subsequent focus groups conducted to solicit participant experiences and opinions. The preliminary outcome evaluation demonstrated that the FSP elicited an immediate and dramatic increase in walking behavior (by approximately 3,700 steps/day or 34 minutes of walking a day15) that was sustained even two months post-intervention and after withdrawal of contact (approximately 26 minutes of walking a day16). Significant improvements in systolic blood pressure and waist girth were also noted. The formative evaluation17 indicated that participants were highly enthusiastic about the pilot project and in particular liked the group walks, personalized goal-setting, and the pedometer and the calendar used together as self-monitoring tools. Telephone follow-up was challenging for program deliverers and not always welcomed by participants, so follow-up evolved into mailed post-cards in subsequent implementations. Finally, diabetes educators were concerned about extra work demands, which ultimately raised the potential for peer delivery (covered below).
Despite the encouraging results from these preliminary studies, the conclusions were limited by the single group (i.e., no control group) study design. We therefore embarked on a randomized and controlled evaluation with individuals with 47 Type 2 diabetes recruited directly from a diabetes education center that included a longer follow-up (24 weeks) and collected additional indicators of cardiovascular health (resting heart rate and blood pressure), glycemic control (fasting glucose, insulin, HbA1c, glucose concentration 120 minutes post glucose load), and plasma lipid status (total cholesterol, HDL cholesterol, LDL cholesterol, and triglycerides).18 Group assignment was determined following baseline measures by drawing from envelopes containing group status. Relative to the control group, FSP participants significantly increased their physical activity approximately 3000 steps/day (or 30 minutes) during the intervention. Although waist and hip girths decreased (approximately 2–3 cm), they did not differ significantly between groups and no other changes were noted. Further, relapse by 24 weeks indicated other strategies such as booster sessions might be needed to maintain this lifestyle change. Despite the lack of response of measured health outcomes, the consistent results in terms of physical activity change supported the FSP as a practical intervention that was capable of eliciting an immediate and profound change in walking behavior. Two recent meta-analyses have included the FSP studies and given details concerning the overall efficacy of pedometer-based programming in terms of increasing walking behaviors.19, 20 Richardson et al.19 aggregated data from nine randomized controlled studies and reported that participants increased their activity by 1,800–4,500 steps/day and lost a modest amount of weight (approximately 0.05 kg per week) over the course of interventions lasting from 4 weeks to 1 year (median duration 16 weeks). Bravata et al.20 considered observational studies in addition to randomized controlled studies and reported similar changes.
Although we demonstrated FSP efficacy in the randomized and controlled study,18 we recognized that it was necessary to evaluate program effectiveness under real-world conditions, that is, delivery through existing diabetes education centers by professional diabetes educators.21 Since these same educators had earlier voiced concerns about extra time demands due to FSP program delivery,17 we also examined the effectiveness of FSP delivery by peer leaders, that is, by 210 individuals with Type 2 diabetes who had previously graduated from the FSP.21 Specifically, we considered outcomes from 157 participants (led by 13 different professionals) versus 63 participants (led by 5 peer leaders) attending FSP programs at diabetes educations centers across Canada. There were no significant differences observed in a direct comparison of participant outcomes by type of delivery (professional vs. peer). Overall, FSP participants showed an incremental change of 4,059±3,563 steps/day, translating into an extra 37 minutes of daily walking, and consistent with previous FSP deliveries.
The FSP was originally developed to address the lifestyle physical activity needs of individuals with Type 2 diabetes, however, we recognized early on the potential for adapting the program to other users. The program was first adapted to a workplace delivery.22 Participants (n=106) were recruited from 5 workplaces where jobs were mostly sedentary (i.e., focused on desk work). Using subjects as their own control, steps/day increased from approximately 7,000 steps/day at baseline to a plateau of approximately 10,500 steps/day. On average, participants experienced significant but modest decreases in BMI, waist girth, and resting heart rate. Reductions in waist girth and heart rate were significantly related to the increase in steps/day. However, reductions in BMI were predicted by the initial steps/day. To emphasize, the consistency of behavior change magnitude (in terms of steps/day) continued even with worksite delivery that was not focused on individuals with Type 2 diabetes, increasing confidence in the adaptability of the FSP. Subsequently, we evaluated a community-based adaptation of the FSP delivered under real-world conditions.23 Limited budgets, resources, time, and personnel experience and availability represent the reality of such conditions; yet critical evaluation of this state is necessary to inform practical aspects of program delivery. The largest sample yet (n=559) individuals participated in this single-group design and a subset (n=46) provided both baseline and follow-data. Steps/day increased from approximately 8,000 steps/day to 12,000 steps/day, again consistent with previous, and more rigorous, deliveries of the FSP. At a one-year follow up, those who completed the program reported steps/day values were still significantly higher than baseline values. Sixty-eight percent of those who completed the program, and 31% of those who did not, still reported wearing their pedometer regularly. Despite the low response rates, again typical of real-world program delivery, the constant results obtained support the effectiveness of adaptations of the FSP.
To answer this question we must consider characteristics of pedometers, the programs they are imbedded within (considering the FSP and other published programs), and the people who participate. Pedometers are designed to be most sensitive to detecting ambulatory, or walking, behaviors. Walking is commonly encouraged1 and most commonly reported,24 especially in the form of walking for exercise. However, since pedometers are worn through out the day in pedometer-based programs like the FSP, they detect walking undertaken for all purposes, including for transportation, occupation, chores, and leisure pursuits beyond just exercise. Pedometers also offer this data in a simple and straightforward output that is user friendly and a direct indicator of movement as a result of behavior choices. Since pedometers are also affordable and relatively small and unobtrusive to wear, they represent an accessible technology that is immediately personalized (each individual is equipped with their own instrument). Finally, the cumulative and readily available visual feedback provides a constant and changing reflection of personal behavior choices as they occur in real time, making them the perfect accessory for promoting and tracking lifestyle physical activity. As we have found out in delivering the FSP, physical activity behaviors can be reinforced if the pedometer is used in conjunction with a simple calendar to record and track behavior over time.17, 25
It is important to underscore that pedometers are simply tools, and their effectiveness depends largely on behavior modification principles applied within the context of theory-based programs like the FSP. Unfortunately, at this time only a few other pedometer-based physical activity interventions26–30 have explicitly acknowledged using theory to guide program design and delivery. However, since pedometers are personal wear items, which reflect individual behaviors, they fit very well into a program of self-monitoring, feedback, and goal-setting, as embodied in the FSP. Programs that have promoted increased physical activity in the absence of a specified pedometer-based goal have produced no significant improvements in steps/day compared to those with increases of ≥ 2,000 steps/day in programs that have promoted the use of the 10,000 steps/day goal or other goals (although few studies have evaluated alternative goals limiting conclusions specifically about the ultimate magnitude of the goal or its efficacy).20 The FSP relies on personally set incremental goals and yet elicits similar (or better) improvements in lifestyle physical activity compared to those assembled in the published meta-analyses. 19, 20 An improved understanding is needed of what program aspects, including the optimal characteristics of goal-setting, are necessary to procure the most favorable participant responses.
The majority of pedometer-based program participants have been women,19 and this also true for the FSP, although this may reflect targeted recruitment strategies at least to some extent. However, there is some evidence to suggest that pedometers may be only appealing for short-term physical activity behavior change in men.31 The meta-analysis of pedometer-based programs conducted by Bravata et al.20 indicated that sex, BMI and race/ethnicity were not significant predictors of increased physical activity. However, we have found that participants most likely to complete the FSP program were overweight or obese class I (i.e., with a BMI between 30 and 35kg/m2).32 We also found that those FSP participants with lower baseline steps/day stand to make the greatest relative incremental increases in physical activity behavior.22 Continued efforts to define what works well for whom under what conditions will help identify the most advantageous features of pedometer-based programming.
A number of limitations must be considered with regards to making conclusions about the effectiveness of pedometer-based programming. For example, pedometers are not designed to detect physical activity intensity, an important aspect of public health recommendations.1 Further, researchers have relied upon participants to self-record their pedometer data, another potential source of bias. However, in both cases, pedometer output correlates strongly with different accelerometers which do collect time and intensity information and in a blinded manner,33 this latter lending support for the accuracy of the self-recorded step values. Another concern is that the instrument of assessment (i.e., the pedometer) in these programs is typically the same as the intervention tool. However, since walking is the intended target behavior, it follows that it must be precisely assessed. We have previously shown that a self-report instrument is not sufficiently responsive to true changes in walking behavior 15. Again, an accelerometer might be considered for assessment purposes, however, this device represents another body worn technology which does not circumvent the fact that individuals are completely aware of the monitor’s purpose. Those studies that have included control groups have supported a true change in walking behavior as assessed by pedometers.19, 20
Another concern is that since pedometers are designed to be most sensitive to vertical accelerations at the hip, rotation of the instrument off this plane will result in diminished accuracy. For example, in obese individuals it is possible that a pedometer can be tilted off the vertical axis if worn on the anterior waist, resulting in lower physical activity estimates.34, 35 To solve this problem, there are pedometers with lower force sensitivity thresholds or those possessing mechanisms which are less sensitive to tilting.34, 35 The attachment of pedometers can also adjusted (e.g., placed on the mid-axillary line or in line with the posterior thigh) to reduce chance of tilting without compromising accuracy.25, 35 Even with these concerns for accuracy related to attachment, we have noted no difficulty in the FSP documenting changes in walking behavior even in obese individuals (where tilting, if it were an issue, would lead to underestimates).16, 18, 21, 22
The FSP is clearly an effective and adaptable program that produces predictable and important changes in lifestyle physical activity. The development and on-going evaluation of the FSP benefited from a broad range of support as detailed in the Acknowledgement section. A self-help book,36 written based on experiences delivering the FSP, was published and makes the program publically available, however, when the research projects ended and related funds inevitably dried up, the FSP essentially ceased to exist since there was no other commercial mechanism or administrative structures in place to sustain its delivery. Regardless, the knowledge and experiences shared through publication have doubtless informed other formally and informally delivered pedometer-based interventions. Continued evaluation of distribution models would deepen our understanding of successful dissemination of such effective lifestyle physical activity programs.
Research related to the First Step Program was supported by a grant from the Canadian Diabetes Association, the Health Canada Population Health Fund Canadian Diabetes Strategy (Project number 6791-15-2000/0390486: Evaluation of dissemination of the First Step Program); and from the Canadian Diabetes Association Applied Research Grant #1005 (The John J Stevenson Grant: Evaluation of dissemination and implementation prototypes of the First Step Program). Adaptation of the program and related research was support by the Prince Edward Island Health Research Program and the Prince Edward Island Department of Health and Social Services. The contribution of the Prince Edward Island Active Living Alliance to marketing, organization, and delivery of the First Step Program is gratefully acknowledged. Additional support for the development of First Step Program educational resources was provided by a Canadian Diabetes Association Award supported by Bayer Corporation. This paper was presented at the 2008 Building Healthy Lifestyles Conference, Arizona State University, Mesa, AZ on February 28–29, 2008.