Obesity has reached epidemic proportions both in the United States and other countries [1,2]. This is a major public health concern primarily because obesity increases the risk of developing chronic diseases [3], such as certain cancers [4,5], cardiovascular disease [6], and type 2 diabetes mellitus [7]. Obese adolescents are more likely to become obese adults [8], thus increasing current and future risk of chronic disease. Therefore, preventing obesity during adolescence would be of major public health significance.

Physical activity (PA) has been inversely related to adiposity [9] and decreases risk for cardiovascular disease [10], type 2 diabetes [11], and certain cancers [12]. However, research demonstrates that PA steadily decreases from childhood through adolescence [13]. Between the ages of 6 and 16, PA decreased approximately 50%, with a continued decrease into adulthood [14]. Between early and late adolescence, moderate to vigorous PA decreased substantially among both boys (6.7 hours per week to 5.1 hours per week) and girls (5.9 hours per week to 3.5 hours per week) [15], suggesting that adolescence is a particularly vulnerable time for PA. Finding ways to avoid this decline would have a significant impact on adolescent PA and obesity risk.

Walking is a popular form of PA [16]. Walking is convenient, inexpensive, and can be easily incorporated into everyday life [17,18] in multiple ways (eg, leisure time PA, active transport) [19]. Increased walking is likely a sustainable form of PA that could be maintained over time [20]. Interventions that encourage simple activities such as walking could make substantial inroads into increasing sustainable PA [18]. Brisk walking is equivalent to moderate intensity PA [21-24]. Given the potential public health significance of walking [18], research is needed to identify effective methods for promoting walking in groups at risk of low levels of PA, such as adolescents.

Pedometers monitor daily steps, thus providing a convenient way to set goals, self monitor PA, and serve as a reminder to be active [25]. Pedometer use has been associated with increased PA in adults [26] and youth [27]. Having a step goal has been associated with decreased body mass index in adults [26]. Among youth, pedometers alone may not be enough to increase PA; other strategies to augment pedometers may be needed to promote sustained PA [27]. As there is a shortage of youth pedometer based studies, there is a need to identify effective methods for incorporating pedometers into interventions promoting walking and to evaluate the long term effectiveness of these interventions [27].

Cellphone ownership and texting are high among teenagers (teens) [28]. A recent study found that 77% of 12-17 year olds had a cell phone, and that texting is the primary way 14-17 year olds communicate with others (eg, friends, family) [28]. Teens are prolific users of SMS text messaging (short message service, SMS), with a median of 100 texts sent and received each day [28]. Since texting is a familiar, convenient, and acceptable way that teens communicate with others, texting may be an effective way to increase PA in this at risk group. A number of reviews of cell phone based studies suggest this intervention mode is promising [29-32]. Some characteristics of text message based interventions that appeared to enhance effectiveness were matching text message frequency to expected frequency of behavior (eg, daily PA), tailoring to selected characteristics (eg, personal values), and interactivity (eg, ability to communicate with research team) [31]. A limitation that was cited was that few of the randomized control trials using texts were theory based [32].

Interventions that are based on psychological theories increase the likelihood of achieving behavior change [33]. The self-determination theory (SDT) [34] posits that three basic psychological needs underpin behavior: (1) competence (ie, skills, ability), (2) autonomy (ie, choice, control), and (3) relatedness (ie, connection to self and others). Need satisfaction (as related to the SDT) integrates the behavior into one’s sense of self (ie, how one defines him/herself). The higher the need satisfaction, the greater the integration with sense of self, thus increasing motivation to perform the behavior, behavioral performance, and the likelihood the behavior will be maintained over time [34]. The SDT has guided PA focused studies in both youth [35-44] and adults [45]. A systematic review of adult studies concluded the evidence supported the SDT as a framework for identifying key factors that influence and promote exercise [45]. Promising evidence is also emerging from youth studies. The SDT predicted autonomous motivation and step counts in youth [36], as well as intention to engage in leisure time PA [37]. The SDT explained changes in motivation to engage in school physical education in the transition from elementary to middle school [40]. An intervention with teens that focused on need satisfaction (ie, autonomy, competence, relatedness) enhanced cardiovascular fitness and was well received by youth [35]. Satisfaction of the basic psychological needs was also associated with autonomous motivation and PA in children [42]. Therefore, the SDT informed interventions promoting PA to youth that emphasize the basic psychological needs might be of particular utility in enhancing motivation to engage in PA, and, ultimately, PA.

Behavioral interventions should be systematically developed and tested [46] to maximize resource effectiveness and increase the likelihood of success. Ideally, interventions should be focused on behaviors that can be changed, evidence based, appropriate for the target population, and focused on realistic, achievable goals [46]. This paper describes the rationale, design, development, and methods for a systematically developed pilot intervention using pedometers, prompts to set a step goal, and SDT grounded text messages promoting walking (ie, increased daily steps) to teens.

This research included two phases: (1) formative, to develop the intervention and construct the text messages, and (2) pilot to assess the feasibility of this approach. Both of these phases are described below.

Teens for both phases of the research were recruited using the volunteer database at the Children’s Nutrition Research Center. The standard recruitment methods, (eg, newsletter and website announcements, distribution of flyers in community locations likely to be visited by parents and teens) were also used. The parents provided written informed consent for their teen to participate in the study, and teens provided written assent. The teens that participated in the formative phase of the program were not eligible to participate in the pilot study.

Eligible participants were 14-17 years old, fluent in English, with Internet access, an email address, and access to a cell phone with unlimited text messaging. Exclusionary criteria included mental (eg, learning impairments) or physical (eg, blindness, deafness, inability to be physically active, medical conditions that limited PA) conditions that impaired their ability to fully participate in the program and/or complete data collection. The families that were interested in participating in the study contacted the Children’s Nutrition Research Center’s Recruitment Coordinator. She explained the study in detail, screened interested families, and routed eligible teens to the study coordinator.

A four group design was employed: (1) no treatment control, (2) pedometer only, (3) pedometer + weekly texts to set a step goal (ie, prompts), and (4) pedometer + weekly texts to set a step goal + SDT focused text messages. This design is expected to provide insight into the additive effects of intervention components (pedometers, prompts, SDT focused text messages). Data were collected at baseline and upon completion of the 12 week intervention (post 1 data collection). The primary outcome was objectively measured PA. Randomization occurred after baseline data collection using a random numbers table. After obtaining written informed consent/assent, the participants began baseline data collection. Institutional review board approval was obtained from Baylor College of Medicine (H-27537). The study was registered with Clinicaltrials.gov prior to initiation (NCT01482234).

A 12 week pilot study was initiated in August 2012 and completed in August 2013. There were one hundred and sixty 14-17 year olds that were enrolled (40/group) using the volunteer database (ie, families interested in participating in research studies) at the Children’s Nutrition Research Center and also using standard recruitment procedures (posting of flyers, posting of study details on recruitment websites and/or electronic newsletters). The teens were randomized after completing a baseline data collection. The baseline characteristics indicated that 51.9% (83/160) were female, and 60.6% (97/160) were at the lower end of the age range (ie, less than 16 years old).  The sample was ethnically diverse (35.6%, 57/160 black; 31.3%, 50/160 white; 26.9%, 43/160 Hispanic; and 6.3%, 10/160 mixed/other), respectively.  The teens wore an activity monitor for 7 days to obtain an objective measure of PA; they also completed Web-based self report questionnaires. Baseline psychometric values of the measurement scales (Cronbach alpha) were within acceptable ranges [58] (Psychological Need Satisfaction in Exercise, .87; expanded Behavioral Regulation in Exercise Questionnaire, .84). Data are currently being analyzed to determine feasibility of this approach to promote PA to teens.

Although the pilot study was not powered to detect statistically significant differences, the group receiving pedometers + weekly prompts + SDT text messages is expected to have the greatest increase in moderate to vigorous PA and average daily steps. They are also expected to have the greatest increases in basic psychological needs, and the greatest movement towards higher levels of PA motivation. The group receiving pedometers + weekly prompts is expected to have the next greatest increase in moderate to vigorous PA, steps, psychological needs, and PA motivation, followed by the group receiving pedometers only. The control group is expected to exhibit no change in moderate to vigorous PA, steps, basic psychological needs, or PA motivation. It is also expected that recruitment goals will be met, attrition will be low, and that we will be able to collect complete data (Web-based self report, 7 days of accelerometry) in at least 75.0% (120/160) of study participants at post 1, regardless of group assignment. It is also anticipated that few technical issues will limit distribution of SMS text messaging; that program satisfaction will be high among participants, particularly in the group that received daily messages; and that the internal consistency of self report measures will be acceptable (>.70).

Teens are at risk of low levels of PA [15]. This is a major public health concern because low levels of PA, particularly during adolescence, have been shown to track into adulthood [14], thus increasing the risk of obesity [9] and multiple chronic diseases [10-12]. Effective interventions are needed to increase PA among this at risk group. Because walking is an activity that can be relatively easily incorporated into one’s other daily activities [17,18], interventions encouraging teens to take more steps throughout the day may be an effective way to enhance sustainable PA in this at risk group.

The intervention described here provides teens with a pedometer to self monitor their daily steps, encourages them to set a self endorsed daily step goal through once a week texts, and sends daily texts emphasizing the basic psychological needs. It extends the literature in several important ways. First, it systematically varies the intervention components (none, pedometers only, pedometers + prompts, pedometers + prompts + SDT informed texts), which will provide information on whether adding prompts or SDT informed text messages increase daily steps and PA over pedometers alone. Second, it will add to the body of literature on whether SDT informed interventions are an effective way to promote PA among teens. Text messages promoting obesity prevention behaviors have been shown to be acceptable and feasible with obese teens; however, their effect on behavior was not assessed, and they were developed as an adjunct to a weight management program rather than as a stand alone intervention [59]. The current research extends this by examining intervention effects on daily steps; specifically, it examines whether an SDT informed text message based intervention focusing on the basic psychological needs is feasible, acceptable, and influences average daily steps, motivation, and PA in teens. Third, it adds further support to the importance of conducting formative research with teens to help ensure the intervention meets their expectations [60]. Ultimately, this could enhance intervention effectiveness by designing interventions that are developmentally appropriate and relevant to the target audience, an important lesson learned for researchers conducting research with teens. Fourth, it extends the evidence regarding the relationship between pedometers and PA in teens [27]. And finally, it extends the literature on texting as a stand alone intervention modality versus as a supporting component. Although reviews provide suggestive evidence that text messages may be an effective way to promote health enhancing behaviors [29-32], few have examined it as a stand alone intervention method. A text message based intervention to increase PA in teens showed promising results; however, PA was self reported, and the intervention only lasted two weeks [61]. The current study will extend this research by using a stronger measure of PA (objectively measured), and sending text messages for 12 weeks.

Teens are heavy users of texting [62]. Therefore, a text based intervention encouraging teens to take more steps during the day may be an effective way to encourage them to be more physically active in a manner that has the potential to be sustained throughout life [20]. This is an important area of research; a text message based intervention would be easy to disseminate at a relatively low cost, using a familiar and convenient technology. Thus, this research ultimately has the potential for public health significance by increasing PA in an at risk group in a familiar, convenient, and relatively low cost manner.