Undergraduate and graduate students (N=39; n=21 for static arm and n=18 for dynamic arm; see Figure 1) were recruited via convenience-based sampling at the University of Mississippi. Participants attended a baseline laboratory visit during which they provided consent, had their height and weight measured, completed preliminary surveys, and were given instructions to accept a researcher invitation to join a private Facebook page.

This pilot study was approved by the institutional review board at the University of Mississippi. Participants randomly allocated to join the static Facebook page (n=21) were exposed to a page that already contained 96 status updates targeting the precontemplation, contemplation, preparation, action, and maintenance phases of the transtheoretical model. The statuses progressed through general information tailored to participants’ community and university resources for healthy eating, stress and time management, physical activity benefits, and tips to implement positive lifestyle changes. Statuses were intended to offer participants awareness and specific tools necessary to engage cognitive processes of change (ie, consciousness raising, dramatic relief, self-evaluation, environmental evaluation, and social liberation) followed by behavioral processes of change (helping relationships, self-liberation, reinforcement, stimulus control, and counterconditioning) throughout the 8-week intervention period. Posts on the static Facebook page consisted of identical content as the dynamic group. The only distinction was that the static group received access to all 96 statuses on the first day of the study, while the dynamic group was exposed to 1 to 2 of these status updates per day across the 8-week intervention period.

Participants randomly allocated to join the dynamic Facebook page (n=18) were exposed to 2 status updates per weekday and 1 status update per weekend day for 7 weeks (96 status updates). During the final week, participants did not receive daily notifications to read status updates but were encouraged to complete online follow-up surveys. No incentives were provided to participants upon completion of this study.

The primary outcomes assessed were changes in physical activity behavior and anxiety, measured via self-report. After meeting inclusionary criteria, which included having an active Facebook account, daily access to a mobile phone, and a current University of Mississippi email account, eligible participants were sent a Facebook invitation to join either the static or dynamic Facebook group. Time spent engaging in vigorous, moderate, and walking physical activity habits was assessed using the International Physical Activity Questionnaire (IPAQ), which has demonstrated adequate reliability and validity [28]. We created a total physical activity variable by summing the minutes per week of walking, moderate-intensity, and vigorous-intensity physical activity.

Anxiety was measured using the Overall Anxiety Severity and Impairment Scale (OASIS), a 5-item psychometric instrument completed via self-reported response on a 5-point Likert scale. A sample question from the OASIS is “How much does anxiety or fear interfere with your social life and relationships?” Higher ratings indicate greater anxiety inference. This scale has been used as a valid and reliable assessment of degree of anxiety impairment in various clinical and subclinical populations [29,30]. Internal consistency was calculated using Cronbach alpha for both the static (pre: .86 and post: .76) and dynamic group (pre: .72 and post: .84).

Stata SE version 12 (StataCorp LLC) was used to conduct paired sample t tests to examine change scores between pre- and post-OASIS scores for each group. An independent samples t test was then used to compare the change scores between the 2 intervention arms. For all analyses, statistical significance was established as a nominal alpha of .05.

The dearth of research examining the link between modern media consumption and anxiety is equivocal [31]. The relationship between mobile phone use and anxiety has been demonstrated among college students exhibiting compulsive or addictive mobile phone behaviors [32,33]. Lepp et al [31] suggest that investigations should attempt to further examine whether this correlation manifests for those not engaging in problematic usage of digital technology. We did not explicitly measure Internet addiction in this pilot intervention. However, our results indicate that for those allocated to the dynamic Facebook page, self-reported anxiety levels diminished significantly. Further, average general Facebook logins per day were marginally decreased for both groups following the intervention (see Table 1), suggesting our investigation did not contribute to addictive social networking behaviors.

Regarding the main outcomes, we observed a statistically significant decrease in anxiety for the dynamic group only. This finding may be influenced by our adherence to the transtheoretical model of behavior change. All status updates followed a theoretically driven sequence of messages in alignment with the cognitive and behavioral processes of change. Participants in the dynamic group received daily status updates containing accurate and credible health information. A total of 96 wellness messages were posted over 7 weeks with content first focusing on cognitive processes of change (ie, consciousness raising, dramatic relief, self-evaluation, environmental evaluation, and social liberation) followed by behavioral processes of change (helping relationships, self-liberation, reinforcement, stimulus control, and counterconditioning). Participants randomized into the static, or passive, Facebook page received access to all 96 messages on the first day of the intervention with no daily updates occurring thereafter. We hypothesize the potency of the cognitive and behavioral messages was diluted in the static group, providing rationale for the observed reduction in anxiety in the dynamic group alone. We also found a statistically significant inverse correlation between mean total physical activity change and mean change in anxiety. We feel this finding underscores the importance of daily, dynamic exposure to Facebook wellness messages. Physical activity behavior changes are difficult to sustain over time, so perhaps consistent social media notifications were advantageous reminders for participants in the dynamic group to increase total ambulatory movement throughout the 8-week intervention period. Thus, our findings suggest that dynamic social networking sites featuring regularly updated content may be more advantageous than websites that retain static content over time.

Despite the plausibility for social media to be used effectively in physical and mental health research, there are numerous concerns that must be addressed when implementing social networking site interventions. Social media is user-centered, meaning individuals consume and produce content relevant to their unique goals and interests. Thus, efficacious social networking site research must be individually tailored. Lack of tailored content is a possible limitation of this pilot study. Although we observed increased changes in physical activity in the dynamic group, the majority of our recruited participants were exercise science or health promotion students already engaging in physical activity. Such extensive exposure to information specifically targeting initiation of positive health behaviors may have been less impactful in this sample. Cognitive and behavioral processes of change may be more appropriately tailored to individuals contemplating behavior change. Further, the risk of psychological disorders such as depression and anxiety is highest among young adults [34]. Importantly, this population is also unlikely to seek treatment, partially due to the stigmatization of mental health [26,27]. We were also unable to control for participant exposure to university courses, exercise programs, or access to wellness information outside of the pilot intervention, which may have affected physical activity or level of anxiety across the 8-week intervention period. A strength of this pilot study, as well as previous social networking site interventions, is the ability to directly target young adults, as social media use among millennials is unrivaled in today’s society [35]. However, social media inherently limits privacy, as it exists as a global platform designed to proliferate opportunities for social inclusion. Therefore, special consideration should be given to the social networking site outlet selected and precautionary measures taken to protect participant safety and anonymity. Our protocol addressed this issue appropriately by using a private, invitation-only Facebook page, although study participants were aware of fellow participants who had access to the page. The inability to completely blind participants may have limited participant interaction and comfortability in this study. Future research should design empirically based methodology that controls for participant privacy while retaining the social aspect integral to social media research.

The results of this pilot study suggest reduced anxiety among college students may be associated with exposure to dynamic, or active, Facebook wellness interventions. We propose mHealth interventions be used as self-monitoring aids rather than for mental health diagnoses. That is, mobile phone technologies should not replace face-to-face therapeutic interventions but may be complementary to clinical services. Awareness and management of psychological health status may be improved through social media, which has been suggested to enhance self-efficacy and decrease the functional impairment associated with anxiety, depression, stress, and psychological disturbances [36]. Social media research is an emerging field deserving special investigation in university settings. Communication hubs including Facebook, Twitter, Instagram, Snapchat, and YouTube are powerful modern research tools. These sites offer the novel potential for rapid dissemination of tailored messages, which may revolutionize the field of psychological and behavioral health promotion.