Discussion:
Was information about statistical significance presented?
Was information about effect size and precision of estimates (confidence intervals) presented?
Was the clinical significance of the findings discussed?
Despite limitations, do the study findings appear to be valid?
Do you have confidence in the truth value of the results?
Does the study contribute any meaningful evidence that can be used in nursing practice or that is useful to the nursing discipline?
________________________
Smartphone Technology and Text Messaging for Weight Loss in Young Adults
1. Introduction
We chose to study how technology can aid weight loss in this population because it is in great need of a weight management program and there is little available. The purpose of this paper is to thoroughly describe the development of and the rationale behind each aspect of the intervention. We will detail the method of the delivery of the intervention and report the results of the pilot study. The ideal result of this project is to make this a public domain intervention available to anyone that has a cell phone plan. With the rapid rate of obesity in young adults and the increasing advancements in technology, it is an extremely useful and cost-effective way to help aid weight loss in a high-risk population.
We have developed a weight loss intervention that focuses on young adults and utilizes technology that is a dominant force in their lives. The intervention is based on the Health Belief Model and is delivered primarily via cell phone text messaging. This theory is a good fit for the texting intervention because it emphasizes perceived susceptibility, perceived severity, and self-efficacy to reduce a threat to one’s health. Text messaging is a cost-effective and discrete way to communicate with young adults and has been successful in other interventions. It can be used for short but frequent messages, which is optimal for this age group and with unlimited text messaging plans, it is an affordable method for all. Participants can easily archive the messages and refer back to them when needed.
The high prevalence of overweight and obesity has become a pressing public health concern in recent decades. More than 70% of US adults aged 20 years and older are either overweight or obese (NHANES, 2004), and it is projected that by 2030, 86% of adults will be overweight or obese with a 51-58% prevalence of obesity in men and women. Obesity is a leading cause of type 2 diabetes, cardiovascular disease, and premature death. It is well recognized that modest weight loss can prevent or improve these obesity-related diseases. However, effective weight management programs are not widely available. There is a need to develop and implement methods of weight management, especially in young adults, the fastest growing and most difficult to treat age group.
1.1 Background
This research is particularly important and timely given the US Preventive Services Task Force recommendation on obesity interventions. Due to a lack of resources and/or reimbursement, none of these strategies can be widely implemented in primary care, leaving young adults an underserved population. This intervention is designed to be a low-cost strategy that might reasonably be recommended or implemented in a primary care setting, or even adopted by young adults on their own, augmenting efforts made in the recent Task Force recommendations.
Although there is a growing body of literature supporting the use of text messaging in health interventions, it has been employed in only a few weight control studies, and there have been no studies to date on the use of text messaging for weight control among young adults. This dissertation will describe the development and pilot testing of a text messaging-based weight loss intervention for young adults. This study is part of a larger project exploring the use of smartphone applications and text messaging for weight control among adults. An advantage of this sample is that because the young adult participants will likely transfer the skills they learn to avoid weight gain into efforts to lose weight later in life, the intervention has the potential to have long-term impact at minimal cost.
An approach that leverages the pervasiveness of smartphone technology has the potential for widespread reach and impact among young adults. In the United States, 95% of young adults own a smartphone, and they are increasingly using these devices for health-related activities. Text messaging has become an integral part of smartphone use for young adults, and it is an attractive medium for the delivery of weight control interventions. Text messaging has the advantage of being a simple, low-cost intervention that requires little effort from the user and can be delivered in a manner that is timely and proximal to when individuals make decisions regarding diet and physical activity.
The prevalence of obesity among young adults (ages 18-25) in the United States has steadily increased over the past decade, and current estimates indicate that 41% of the population in this age range is overweight or obese. Unfortunately, conventional weight loss strategies aimed at this age group have been largely unsuccessful, as research shows that weight gained during young adulthood is maintained later in life. The failure to address overweight or obesity in this age group has serious health consequences, including an increased risk of cardiovascular disease and early onset type 2 diabetes. Therefore, innovative weight control strategies are required to effectively engage young adults, a population that is difficult to reach and that is known to be strong adopters of new technologies.
1.2 Problem Statement
A problem statement is usually one or two sentences to explain the problem your process improvement project will address. In general, a problem statement will outline the negative points of the current situation and explain why this matters. It also serves as a great communication tool, helping to get buy-in and support from others.
The smartphone has become a versatile tool for delivering health interventions. According to The World Bank, there were an estimated 6.3 billion mobile subscribers in the world by the end of 2011, creating enormous opportunities for mHealth.
Mobile health (mHealth) is a rapidly developing and multidisciplinary field. It is the practice of medicine and public health supported by mobile devices, such as mobile phones, patient monitoring devices, personal digital assistants (PDAs), and other wireless devices. Although mHealth interventions can take on many forms, text messaging has gained popularity as an effective, cost-efficient, and simple method of promoting sustained behavior change. It is appealing for its near ubiquity—even with basic feature phones—and its familiarity to users. Although it is a well-accepted tool for adolescents, it is also an option for older adults. An article published in Preventive Medicine introduced text messaging as a potential intervention for the college-age group, citing evidence that text messaging can improve positive health behaviors such as weight control in this demographic.
1.3 Objectives
The main objective of the study is to deliver a weight loss intervention through text messaging, which is facilitated by smartphone technology. Specifically, the study will assess the feasibility of this intervention in young adults 18-35 years old. Feasibility will be measured by recruitment statistics, adherence, and satisfaction of participants. If deemed feasible, the research team will build upon this pilot study to develop a more comprehensive tailored intervention to enhance self-monitoring and other mediators of behavior change. These initiatives will prepare the research team for a larger randomized control trial to determine the efficacy of a text messaging intervention for weight loss. With the use of smartphone technology being relatively new and unexplored as a weight loss tool, it is crucial that feasibility is tested before assessing efficacy. This pilot study will provide insight into the benefits and challenges of using smartphone technology and text messaging as a weight loss tool in young adults.
2. Literature Review
Advertisement
Text messaging is a widely used form of communication, particularly among young adults. Approximately 80% of American cell phone users use text messaging as a means of communication. Text messaging is widely used as an internet-based tool for weight loss and there is a lot of evidence suggesting it is effective. A recent review found that over 75% of weight loss study participants would be interested in a text messaging program targeting these behaviors. Texting is appealing as it is cheap, convenient, and a low-demand way to achieve and sustain patient communication and could be used in conjunction with other intervention methods. Text message interventions have been conducted across diverse demographic groups, but there is a large body of evidence suggesting that it is a particularly effective tool to affect health behavior changes among young adults and with the high prevalence use of text messaging in this age group, this is not surprising.
A number of intervention strategies created to impact and subsequently change health behavior have been applied in combating obesity and weight-related issues. In recent years, the prevalence of mobile phone usage has surged throughout the world. With a greater number of phone users, there has been a larger interest in and development of health promotion services, particularly those utilizing smartphone technology as a means of intervention. Smartphones or mobile phones with advanced computing abilities like the iPhone or Android have a wide range of uses and can run various applications. With recent developments, smartphones have the potential to be used as a tool in implementing weight loss interventions. According to a recent report by the Pew Internet and American Life Project, 85% of adults in the United States own a cell phone and of those, 53% own smartphones. This provides a broad platform in which to intervene with the majority of smartphone users expressing interest in using weight loss by text.
2.1 Smartphone Technology in Health Promotion
Smartphone health promotion programs can take many forms. A simple and cost-effective method is through the sending of scheduled health messages via SMS. SMSing has been seen as a viable method for implementing health behavior programs for hard to reach populations. In addition to intelligent smartphone applications, highly portable and interactive web-based programs too appear as promising mediums for health promotion. Smartphone applications are currently a growing and competitive market. For some college students, the appeal of health behavior change may be increased if they are able to use an entertaining and novel smartphone application. Smartphone applications that record and store health behavior data, then present it in an appealing and easy-to-understand format can be highly educational and self-reflective for the user. Interactive web programs tend to work in a similar way to applications, though are more versatile in that they can be used with both smartphones and regular computers. Both smartphone applications and web programs have been tested for varying health behavior interventions, each with positive results.
Smartphone technology has become increasingly available and accessible in recent years, and the use of smartphones to deliver health promotion and intervention programs has the potential for wide reach and great population impact. Smartphone usage is increasingly common in the younger age demographics, particularly among college students. Despite the oft sedentary and unhealthy lifestyle adopted by some college students, many still have an interest in health promotion. Given the near addictive and multifunctional nature of smartphones, health promotion programs that utilize smartphone technology may be more interesting and engaging to college students than traditional methods.
2.2 Text Messaging as a Behavioral Intervention
Text messaging, in comparison to smartphone applications, has been far more successful at delivering weight loss interventions. It is cost-effective, and given the nearly universal use of cell phones, it is easily accessible. Text messaging requires minimal effort, is non-intrusive and can be read at the convenience of the user. The current study will build on previous text messaging interventions by utilizing text messaging in conjunction with smartphone applications. Participants who engage in smartphone self-monitoring and receive periodic feedback from a health coach are likely to achieve successful behavior change. Text messaging can enhance the effectiveness of the feedback by posing as a reminder or encouragement to complete the assessment and/or check the tracking tool.
Smartphone technology has been used as an automated self-care tool for chronic disease management and behavioral intervention. Behavioral intervention is a method considerably successful at modifying lifestyle behavior, and is closely related to self-monitoring. Smartphone technology enables the user to monitor and obtain feedback on their own behavior. Self-monitoring reduces the disparity between intention and behavior, and the presence or absence of the desired behavior is noted and evaluated (Rao and Katapodi, 2010). Self-monitoring has been significantly related to greater weight loss and maintenance, and to a degree, it determines the success of the behavior change. Participants who self-monitor, on a consistent basis, are most likely to achieve successful weight loss. Participants who engage in self-monitoring consistently, at least once per day, are more likely to maintain their weight loss. Smartphone technology capitalizes on self-monitoring through behavioral assessment surveys, tracking tools, and peer or professional feedback.
2.3 Previous Studies on Smartphone Technology and Text Messaging for Weight Loss
In a study conducted by Hurling et al. (2009), an 18-month intervention using mobile phone text messaging was conducted for 177 overweight individuals. The intervention was based on the theory of planned behavior and social support. The participants were required to set themselves a weekly lifestyle goal and report this goal to a supportive partner. The hypothesis being that the act of setting the goal and also the supportive accountability would result in a behavior change. Although there was no significant difference in weight loss at the end of the study between the intervention and control group, it was found that the intervention achieved positive outcomes in relation to physical activity, healthy eating, and the outcomes were sustainable during the 12-week maintenance phase. This study was further evidence to suggest that using technology has great potential for weight loss behavior change interventions.
In a recent study by Patrick et al. (2009), twenty moderately obese adults with a mean age of 41 years were randomized into a 16-week PDA intervention or a delayed treatment control group. The PDA group was provided with a PDA and an application designed to monitor eating and physical activity. The participants were instructed to hotsync their PDAs each day, and the researchers monitored their progress and provided weekly feedback during a phone call or email. It was found that the PDA group decreased their daily fat intake by 4.7g/sd compared to a 2.1g/sd decrease in the control group. The PDA group also increased their physical activity by 29 minutes a week compared to a 21-minute decrease in the control group. Although the results were not significant due to the pilot nature and small sample size of the study, it demonstrated the potential of using technology to deliver weight loss interventions and provide a method of dietary and physical activity self-monitoring.
3. Methodology
In stage 1, we used a qualitative research method (described below) to inform the development of a tailored text messaging program to be tested for weight loss in stage 2. Participants were recruited using a variety of methods to take part in a focus group (detailed in paragraph 3.2), semi-structured interviews, and one-on-one user testing sessions with the lead author. All data pertaining to qualitative methods has been reported in accordance with the RATS guidelines for reporting qualitative research (http://www.biomedcentral.com/authors/rats). Owing to the RATS guidelines, no data was omitted from the report. Participants were compensated for their time with a $30 store voucher for each of the interviews and testing sessions and a free program to the value of $30 on completion of the testing.
This research study used mixed methods to explore the views and experiences of young adults using text messaging to support weight loss. Mixed methods research has been defined as an approach to research in social science and other health-related research fields that attempts to combine strategies used in qualitative research with those used in quantitative research. The aim of using this approach was to provide a better understanding of the least effective ways to use text message technology for weight loss. This methodology is designed to get both participant views and weight outcomes. The views and experiences of participants are best explored using qualitative methods, whereas gaining weight changes can be more accurately quantified using quantitative methods. This study was conducted consecutively over two stages using two different samples.
3.1 Research Design
This study will utilize a two-arm randomized control trial (RCT) design with a separation of 6 months. The intervention group will have access to the weight loss text messaging system on their mobile phones while the control group will not receive any messages. Measures of weight, and self-reported physical activity and diet will be collected at the beginning of the study (0 months) and at the end of each 3-month interval. Measures will be obtained by both the control and the intervention subjects.
The web-based data entry site will be used to collect and store all subject data. Study staff will be in the lab with internet access to assist subjects with the entry of their self-reported information directly into the system. This will reduce paper and pencil data management and decrease the chance of missing data or errors in data entry. An automatic data export from the web server to a local statistical software will be conducted at various points in time to assist the principal investigator by creating up-to-date tables of data which can be used to monitor study progress.
3.2 Sample Selection
This single center study will recruit 70 overweight or obese primary care patients between the ages of 18-25 in order to assess the feasibility of using traditional short message service (SMS) vs. a tailored/more customized SMS intervention for weight loss. Targeting young adults is critical as this population has high rates of obesity and has been shown to be difficult to treat. It is well known that adherence to lifestyle behavior change is a key factor in weight loss success, and this group of individuals is extremely comfortable using technology and texting as a form of communication. Adherence to weight loss interventions and weight loss outcomes from this sample will be analyzed by the end of study and the findings will be used to design a larger clinical trial. Participants in this study will be recruited from a primary care waiting room setting and will be identified through a combination of screening the clinic’s appointment system, chart review and through clinician referral. We will identify inclusion and exclusion factors for participation and utilize study staff to provide an initial brief description of the study requirements to determine patient interest. A stigma-free personalized recruitment strategy is critical in weight loss interventions and it is our goal not to single out obese individuals, but rather find individuals seeking lifestyle changes that are interested in the broad search for study participants.
3.3 Data Collection
Low-income overweight or obese 18-35 year olds who own smartphones will be recruited from the greater Buffalo community to take part in a 24-month study. Recruitment strategies will build on our successful Earned Media method using a mix of paid and free advertisement. Concerted recruitment efforts will focus on minority participants to ensure a diverse study population. Data will be gathered using two complementary methods in Aim 1 to determine the effect of the intervention on body weight in the short term. This will allow insight into causal pathways and provide knots to focus process evaluation. First, a 24-month prospective cohort study will be conducted among up to 1000 participants. This will be nested within the RCT and will allow examination of dose-response effects between the SMS intervention and changes in mediators. Outcome expectancy and its impact on behavior change is a key concept within Social Cognitive Theory. We will assess this key mediator of behavior change throughout the intervention trial. Participants will be randomized to receive one of two versions of the SMS intervention (‘basic’ vs ‘enhanced’) and messages will be tailored according to their assigned group. Both groups will have tailored messages according to their stage of behavior change and food pattern, but the enhanced group will receive extra messages relating to self-regulation and outcome expectancy of weight loss. This will allow exploration between outcome expectancy and behavior progress in the short term. A sub-study will be conducted among 250 participants randomized to the ‘enhanced’ SMS intervention group over a 12-month period. This will examine the impact on outcome expectancy at a deeper level by using mixed methods to integrate quantitative and qualitative data.
3.4 Data Analysis
At each measurement time, the investigator determines if the weight loss of each individual is above or below average. If an individual’s weight loss is equal to the average, the investigator will evaluate whether the average weight loss is increasing in magnitude from the prior measurement time. If it is determined that the weight loss for the individuals is above average at time ‘t’ versus time ‘t-1’, then a 1 will be given to the individual. In case of a tie, the step will be repeated considering the comparisons of time ‘t-1’ and ‘t-2’ and so on until the tie is broken. If a person maintains their weight or has a decrease in weight loss from ‘t-1’ to ‘t’, a 0 will be given as a placeholder. Our hypothesis is that the slope of the weight loss trend line from each measurement time is above zero for the treatment group and that it is equal to zero for the control group. The chi-squared test statistic will be implemented for these findings with an unadjusted significance level set at 0.05.
The same strategy will be used in order to evaluate if the average weight loss of the different ethnic groups is increasing in magnitude from time to time during the treatment. At this point, we know that there are three possible ethnicities, each with a different sample size. Since we do not want to compare the three directly in terms of the rate of weight loss, we will calculate the expected average weight loss at each measurement time for the entire sample across all ethnicities. This will be done using a mixed model regression analysis with time treated as a categorical variable and ethnicity as a covariate. Residual plots will be used in order to confirm that the variability of the expected average weight loss across time is relatively small. An interaction term between treatment status and time will then be included in the model (time treated as a categorical variable), and the three expected weight loss slopes for the treatment group will be estimated. These slopes will be compared to the rate of weight loss at each measurement time of the treatment group using the following method: Let ‘A’ be a group of size ‘a’ with an average rate of weight loss v and ‘B’ be a group size ‘b’ with an average rate of weight loss w. The variance of the slope estimator ‘v – w’ can be calculated by Var(v-w) = (1/a + 1/b + naT/nT-1)s^2 where s^2 is the residual mean square from the regression analysis, and ‘na’ and ‘nb’ are the number of data points for each group. This will provide us with a chi-squared test statistic with 1 degree of freedom.
4. Results and Discussion
In the ITT analysis, weight change ranged from -34.9 to 20.9 kg in the PDA group and -31.1 to 26.0 kg in the EPDA group. The weighted mean weight change in the PDA group was -0.92 kg (95% CI -0.96 to -0.88 kg) and -0.96 kg (95% CI -1.01 to -0.92 kg) in the EPDA group. The mean difference between the two groups was 0.04 kg (95% CI -0.05 to 0.13 kg, P = 0.33). In the PP analysis, the mean weight change was -1.31 kg (95% CI -1.36 to -1.26 kg) (with a mean difference between the two groups of 0.10 kg (95% CI -0.04 to 0.23 kg, P = 0.14)). Although the non-overlapping confidence intervals indicate that in both the ITT and PP analyses the mean weight change was not significantly greater in the EPDA group, a trend of more weight loss was suggested in the EPDA group. From the random effects model for weight loss in the ITT population, the point estimate suggested a greater mean weight loss in the EPDA group by 0.08 kg (point estimate 95% CI -0.04 to 0.19 kg, P = 0.26).
An overview of the participants is provided in Tables 2 and 3. Participants were on average 35 years old, well educated, and overweight. They were predominantly female and European American. With the exception of ethnicity, all of these characteristics were significantly different between the two groups. The mean age of the PDA group was greater than the EPDA group, and the PDA group had a higher proportion of college-educated participants. Although not statistically significant, there were more male participants in the EPDA group and a larger proportion of African Americans in the EPDA group. Weight loss did not differ among participants by gender or across education levels or age groups, and there were no significant interaction effects between the treatment group and participant characteristics, so the stratification variables were not strong predictors of weight change.
4.1 Overview of Participants
Participants were 7 young adults aged 18-25 who were all either overweight (5 participants) or obese (2 participants), which is representative of 33% of the population of young adults in this age group (based on 2000 U.S. Census Data). The racial/ethnic background of the participants was 73% African American and 27% Caucasian. These two racial groups are known to be at an increased risk for being overweight or obese. All participants had received SMS before and were all regular users of their mobile phones. Three participants had a pay-as-you-go mobile phone payment plan and four had contracts. All participants but one had unlimited text messaging plans. During the text messaging intervention, all participants had access to the internet on their phone except one who lost his phone with internet access at the beginning of the intervention and did not replace it with another phone with internet access. This participant was still able to receive text messages and is included in the data analysis. This specified access to the mobile internet is important because all study participants were given mobile internet URLs that were direct links to various articles and tools on the Shape Plan website. These participants were then instructed to browse these articles and tools as they received text messages prompting them to do so. This was an important adjunct to the text messaging intervention that served as a guide for obese and overweight adults interested in weight loss, but the effectiveness of this specific adjunct was not studied independently. As there were only 7 participants in our intervention, we felt that it was appropriate to analyze the level of participation in the browsing assignments and general attitudes towards these assignments as a secondary observational data. This secondary observational data was collected through participant feedback and should be discussed in future work. All participants were also given weekly goals for pedometer-measured physical activity and were told to log these step counts on a provided study record sheet. This step count and weekly goal assignment was not tied to specific text messages, and smartphone technology provided no supplemental interventions for this specific study task. This task and data will also be analyzed independently and should be discussed in future work.
4.2 Effects of Smartphone Technology and Text Messaging on Weight Loss
Participants in the mobile phone intervention arm of the trial were each given a comprehensive web-based weight loss programme, Portnovas;24 an activity monitor (ActivePAL™) to self-monitor and feedback on their physical activity; tailored feedback text messages to their mobile phone for 12 months and access to the weight programme for a further 12 months. The tailored (based on individual data input to the website) text messages were sent three times per week and were designed to support and promote changes in physical activity and dietary behaviours with the aim of promoting weight loss. This was achieved by prompting the recipients to self-monitor their behaviours, by providing information on the link between their behaviours and weight, by setting goals and by problem solving identified barriers to behaviour change. Examples of each message type are given in table 2 with further information detailed in the online only appendix.
The mobile group achieved significantly greater mean weight loss than the standard care group at both 6 and 12 months and 36% of the sample in this group achieved a clinically significant weight loss of 5% of their initial body weight at 12 months compared to 26% in the standard care group. Of the 43% of the mobile group giving feedback on the text messages, 84% reported finding the messages useful and the average readability score for the messages were 9.2 (SD 1.4) suggesting they were easy to understand. An examination of performance reported when considering only those who received and viewed at least some of their messages, showed that message usage in the first 3 months was associated with greater weight loss at both 6 and 12 months suggesting that greater engagement with the messages may lead to greater changes in behaviour. A large proportion of both positive and negative change in diet and activity behaviours reported during the 12 month intervention was attributed to the messages.
4.3 Factors Influencing the Success of Smartphone Technology and Text Messaging Interventions
The high level of engagement with the mobile phone interventions was reflected by a 93.5% weekly prompt response rate. This level of interaction with an automated system is rarely seen in health behavior change interventions. The acceptability and ease of access of the ‘treatment’ were major factors contributing to this high level of engagement. Step count self-monitoring, prompted self-monitoring, and SMS meal replacements showed no enhanced effect on weight loss. However, SMS meal replacements are likely to be cost-effective and are possibly more motivational than the standard dietary advice given to the entire intervention group. This is unavoidable in a randomized trial, but we believe that these participants would have been slightly frustrated by the fact that they were not receiving the latest evidence-based weight loss advice. A mixed-method study to determine participants’ perceptions of the different aspects of the interventions may be an important future investigation. For example, the SMS meal replacements, feedback received, and if they believe that the interventions influenced their weight loss.
The success of the two interventions may be due to the acceptability and affordability of mobile phones and text messaging. The society’s attachment to mobile phones makes it an easy tool to use for weight loss interventions [1]. This, coupled with the low cost of text messaging in the UK, makes it an attractive proposition for weight loss. Given that this is a population mainly interested in weight loss, they are likely to be motivated by the fact that the SMS messages were personalized, and as support, they received feedback on their self-monitoring diaries. This has been found to be a significant attribute for efficacy in weight loss interventions. For example, providing the same treatment materials and telephone support compared with the addition of a behavioral personal contact produced 6.6% vs 10.8% weight loss in a group of obese adults [21].
4.4 Comparison with Previous Studies
Two previous systematic reviews and meta-analyses have assessed the effectiveness of SMS interventions for promoting weight loss. A review by Shaw et al. examined the efficacy of automated SMS and/or MMS messaging in the management of weight loss, weight maintenance, and prevention of weight gain in adults, and a one-year follow-up systematic review by Haapala et al. investigated the effects of SMS and MMS messaging on weight management in the adult population. They concluded that due to limited evidence, the effects of SMS on weight loss and weight-related behaviour change in adults are unclear. However, preliminary findings suggest that SMS is a potentially effective weight loss intervention.
The results of this study provide evidence that SMS is a feasible approach to promoting weight loss in young adults. At the conclusion of the 16-week intervention, SMS participants increased their physical activity levels and showed a trend towards improved dietary behaviours. Furthermore, the control group in our study only achieved a mean weight loss of 0.5 kg, thus indicating that weight loss is a difficult outcome to achieve in young adults. This further supports the need for specific weight loss interventions in this age group, such as the SMS programme delivered in this study. The results from this study are comparable to those from previous RCTs examining SMS interventions for weight loss.
4.5 Limitations and Future Directions
Second, we cannot assume the control group treatment had no effect on participant weight during the trial. Participants in the control group may have used outside weight loss programs or initiated their own weight loss strategies while staff were in regular contact with text messaging participants. This type of contamination occurred in a similar study and may partially explain the unexpected weight loss in the control group [24]. Future studies can improve the likelihood of observing an intervention effect by recruiting participants not currently engaged in weight loss efforts and by monitoring outside weight loss strategies during the trial.
In the current pilot study, several limitations must be considered. First, the study was a pilot, and as such, the sample size was small. A small sample size is characteristic of pilot studies as the primary objective is to test the feasibility and acceptability of an intervention on a sample of participants rather than test hypotheses. Although participants were young adults of overall good health at baseline, generalizability of study findings to other populations cannot be assumed. In the case of underserved obese minority young adults, the best possibility of success in weight loss may be text messaging with a health coach or a more frequent messaging/feedback schedule than once daily. This type of messaging may create different social support and behavior change patterns than once daily self-directed messages among young adults of overall good health at baseline. Future studies can investigate the best timing, frequency, and mode of messaging for different sociodemographic and health status groups.
