Step-2-It: Feasibility of SMS Technology to Increase Physical Activity

Overview

Rural women are more likely to be obese and have a higher risk for chronic disease than their non-rural counterparts. Inadequate physical activity (PA) at least in part contributes to this increased risk. Rural women face personal, social and environmental barriers to PA engagement. Interventions promoting walking among rural women have demonstrated success; however, few of these studies use text messaging to promote PA. Step-2-It was a pilot study to assess the feasibility, acceptability, and effectiveness of text-messaging combined with a pedometer to promote PA, specifically walking among English-speaking women, aged 40 and older, living in a rural, northwest Illinois county. There were two components to the 13-week, Step-2-It intervention: (i) participants used the pedometer to track and report their steps via text message daily; and (ii) participants received an informational or motivational text message daily. Enrolled participants completed baseline assessments, received pedometers and two types of automated text messages: motivational messages to encourage walking, and accountability messages to report pedometer steps. Participants engaged in 3, 6, 9, and 12-week follow-ups to download pedometer data, and completed post-intervention assessments at 12 weeks.

Rural women are significantly more likely to be obese than non-rural women, which is associated at least in part to inadequate physical activity (PA). It is well established that physical activity (PA) can lower the risk for heart disease, stroke, type 2 diabetes, and certain cancers and that physical inactivity is the fourth leading risk factor for mortality worldwide. Given these associations between PA engagement and mortality as well as disease risk, and that PA levels typically decrease with age, interventions to encourage PA among rural women who are middle-aged and older are essential. Rural women report personal barriers to PA engagement including the lack of time and injury, as well as barriers in their physical and social environments, including limited access to exercise facilities and lack of social support. Walking is an affordable PA option that in part addresses the barriers to PA engagement and is associated with a low injury risk. Previous interventions to encourage walking, including those implemented in rural communities, have achieved success in increasing PA, weight loss and improving risk factors for multiple chronic diseases. Mobile health (mHealth) technology and PA trackers are two technologies that can be used to promote walking in rural women. Mobile health, or mHealth, is the application of wireless devices to support medical or public health practices. Text-messaging is one mHealth strategy that has the ability to reach a large number of people at a relatively low cost.Cell phone use in the US is nearly universal, 97% of urban adults and 95% of rural adults are cell-phone users. mHealth interventions using text messaging have been used to address a variety of health concerns, including PA promotion. However, few studies have examined the use of mobile health technologies to promote PA in rural women, and to our knowledge, no such studies have focused on midlife and older, rural women. Mobile health or mHealth when used to promote walking in rural women can broaden the reach of these interventions and gives the ability to communicate with women in real-time. Physical activity trackers and wearable devices are easily available and affordable and there is evidence to show that their use is helpful in promoting increasing physical activity. In the America on the Move study, adults who were using a pedometer accumulated significantly more steps than those who were not. This finding suggests that using physical activity tacking devices such as pedometers might motivate individuals to increase their physical activity and is supported by a review that shows that pedometer-based walking programs increased participants' activity levels by 2,183 steps per day. A systematic review of 14 mobile health interventions with physical activity as an outcome showed that 7 of the 14 showed significant positive benefits on self-reported physical activity outcomes. Purpose: Step-2-It was a pilot study designed to evaluate the feasibility, acceptability and effectiveness of text-messaging combined with a pedometer to promote PA, specifically walking, among midlife and older women, residing in a rural Illinois county. This prospective, one-group, pre-post study targeted women ages 40 and older living in Stephenson County, IL. Methods: Flyers advertising the study were placed in the grocery stores and the health department, and interested participants called the lead researcher who assessed eligibility using a checklist. Women with a self-reported diagnosis of bronchitis, pneumonia, or severe asthma, or those being treated for a severe health conditions were ineligible. Eligible women were invited to one of five enrollment meetings at SCHD. At least two follow-up calls were made to interested women who did not attend their scheduled enrollment meeting to invite them to the next meeting. Women were screened and those who met the study eligibility criteria were enrolled in the study. At the enrollment meetings, participants provided written consent, received a pedometer (Omron HJ-720ITC) and were trained on how to use it. In addition, they completed a series of baseline instruments detailed below and were provided educational materials on cardiovascular disease risk reduction through PA and healthy eating. There were two components to the 13-week, Step-2-It intervention: (i) participants used the pedometer to track and report their steps via text message daily; and (ii) participants received an informational or motivational text message daily. Pedometer/Step Reporting: Starting in week 0, participants self-reported daily step counts via text message, and continued to do so for the duration of the 13-week intervention. Participants received a daily text reminder to report their steps. Those who did not report steps for two days in a row received a reminder phone call. Every three weeks, participants returned to SCHD to have their pedometer data downloaded. Participants received a $5 incentive for each download, for a total of $20 over 13 weeks. Text messaging: Text messages were sent using mytapp, an online application that allowed for scheduling individual and recurring messages via a cloud service, Twilio. Text messages used for this study were limited to 160 characters and participants could choose their preferred time of day to receive texts. In week 0, the only message participants received were to remind them to report their steps. From week 1-12, in addition to the reminder to report steps, participants received one informational or motivational text message per day (7 messages/week). All participants received the same text messages each day. Motivational messages were based on social cognitive theory with the intention of increasing participant self-efficacy to engage in PA. These messages were adapted from a database of messages from a previous study to increase PA among African American breast cancer survivors. Adaptations were made to the messages based on two focus groups conducted with in the target population prior to this study (unpublished study). Sample motivational messages included, "Nothing is impossible. The word itself says "I'm Possible!"and "Always focus on how far you have come, rather than how far you have left to go." Informational messages included local, PA-related events and resources, such as walks and low-cost walking options. Sample information messages included, "Walk for a cause - sign up for charity walks" and "Find a friend to walk with." In addition, there were also messages that reminded women to walk. These messages included, "Check out what is going on outside, go for a walk", "Don't just think about it, actually go for a walk", and "Take a quick walking break." Data collected for this study along with the time-points at which they were collected are described below. Participant Characteristics (collected at baseline): participant demographic information including age, race, education, marital status, employment, and household income; and health status information including self-reported previous diagnosis of diabetes, hypertension, high blood cholesterol, and other heart health diagnoses, medication use for chronic conditions, and tobacco use. Physical Activity Questionnaire (collected at baseline and post-intervention): Self-reported PA was assessed using seven questions from the 2005 Behavioral Risk Factor Surveillance System (BRFSS) Physical Activity questionnaire. These data were used to determine whether participants met the recommended aerobic PA levels. Questionnaire reliability and validity have been established previously. Physical Activity Pedometer Readings (collected at weeks 3,6,9, and 12) and Self-Reported Steps (collected daily): The primary outcome measure was number of steps. The Omron HJ-720ITC pedometer, validated in previous studies,40-42 was used for an objective measure of steps. Data was downloaded from the pedometers every three weeks. Participants self-reported daily steps via text message. Body Weight (collected at baseline and postintervention): weight was measured using a calibrated scale at SCHD. Intervention Satisfaction (collected at post-intervention): was assessed using a survey that included questions on overall perceptions about Step-2-It, number of messages received, preferred message types, perceptions about the effectiveness of text messages in promoting health and PA, and plans to continue PA after participating in Step-2-It. Participants were also asked about barriers to reporting steps. Participants who completed this survey received a $20 incentive. Descriptive statistics were used to characterize the study population and assess post-intervention satisfaction. Baseline to post-intervention comparisons for body weight were conducted using the paired categorical Wilcoxon Sign rank test. Comparisons of BRFSS PA level was conducted using the Chi-square test. Self-reported step data were compared to downloaded step data and a Pearson correlation was calculated. Correlation testing for self-reported steps versus pedometer-recorded steps excluded observations where only one variable was present. For analyzing mean steps, missing pedometer data was filled in with self-reported step data. Mean steps by week of intervention was calculated and a box plot of mean daily steps by intervention week was produced.