Factors Engaging Users of Diabetes Social Media Channels on Facebook, Twitter, and Instagram: Observational Study

Can the group effect dominate the influence of the child on the parent's decision to care for type 1 diabetes?

AIM

To understand how the interaction of peer parents in disease-specific social media groups influences their daily treatment decisions.

DESIGN AND METHODS

We examined the content of the largest Hungarian Facebook group for parents of children with type 1 diabetes mellitus, analyzing 28 days of activity using netnography. In addition, we undertook an online quantitative questionnaire to identify group members' (n = 267) attitudes.

RESULTS

In exploring the influence of the community on the decisions of peer-parents, we found that the respondents could be divided into two clusters. The main difference between the clusters was that committed parents sought the online community for advice and asked questions more frequently, and the peer-support community was a more important part of their self-image. Comparing the influence of online communities and children on the parents' decisions, we observed that the community had the most significant impact on attending diabetes-specific events, while children had the greatest influence on meals and leisure time and were often in conflict with parents.

CONCLUSIONS

Social media strongly support the integration of prescribed therapy into daily routine. The influence of the child with diabetes on parental decisions shows that diabetes education needs to become child-centered and that the parent-child relationship should be considered as a crucial element for therapy effectiveness.

PRACTICE IMPLICATIONS

Online parenting groups provide mental support in coping with diabetes and would serve as a primary non-medical information source; the healthcare staff must be supportive or even encouraging when parents join such groups.

Designing for caregiving networks: a case study of primary caregivers of children with medical complexity

OBJECTIVE

The study aimed to characterize the experiences of primary caregivers of children with medical complexity (CMC) in engaging with other members of the child's caregiving network, thereby informing the design of health information technology (IT) for the caregiving network. Caregiving networks include friends, family, community members, and other trusted individuals who provide resources, information, health, or childcare.

MATERIALS AND METHODS

We performed a secondary analysis of two qualitative studies. Primary studies conducted semi-structured interviews (n = 50) with family caregivers of CMC. Interviews were held in the Midwest (n = 30) and the mid-Atlantic region (n = 20). Interviews were transcribed verbatim for thematic analysis. Emergent themes were mapped to implications for the design of future health IT.

RESULTS

Thematic analysis identified 8 themes characterizing a wide range of primary caregivers' experiences in constructing, managing, and ensuring high-quality care delivery across the caregiving network.

DISCUSSION

Findings evidence a critical need to create flexible and customizable tools designed to support hiring/training processes, coordinating daily care across the caregiving network, communicating changing needs and care updates across the caregiving network, and creating contingency plans for instances where caregivers are unavailable to provide care to the CMC. Informaticists should additionally design accessible platforms that allow primary caregivers to connect with and learn from other caregivers while minimizing exposure to sensitive or emotional content as indicated by the user.

CONCLUSION

This article contributes to the design of health IT for CMC caregiving networks by uncovering previously underrecognized needs and experiences of CMC primary caregivers and drawing direct connections to design implications.

Supporting and Incentivizing Peer Leaders for an Internet-Based Private Peer Community for Youths With Type 1 Diabetes: Social Network and Directed Content Analysis

BACKGROUND

Youths with type 1 diabetes (T1D) frequently experience stigma. Internet-based peer communities can mitigate this through social support but require leaders to catalyze exchange. Whether nurturing potential leaders translates into a central role has not been well studied. Another issue understudied in such communities is lurking, the viewing of exchanges without commenting or posting.

OBJECTIVE

We aimed to assess the centrality of the peer leaders we selected, trained, and incentivized within the Canadian Virtual Peer Network (VPN)-T1D. This is a private Facebook (Meta Platforms, Inc) group that we created for persons aged 14 to 24 years with T1D. We specifically sought to (1) compare a quantitative estimate of network centrality between peer leaders and regular members, (2) assess the proportions of network exchanges that were social support oriented, and (3) assess proportions of high engagement (posts, comments, reactions, and votes) and low engagement (lurking) exchanges.

METHODS

We recruited peer leaders and members with T1D from prior study cohorts and clinics. We trained 10 leaders, provided them with a monthly stipend, and encouraged them to post on the private Facebook group we launched on June 21, 2017. We extracted all communications (posts, messages, reactions, polls, votes, and views) that occurred until March 20, 2020. We calculated each member's centrality (80% of higher engagement communications comprising posts, comments, and reactions plus 20% of members with whom they connected). We divided each member's centrality by the highest centrality to compute the relative centrality, and compared the mean values between leaders and members (linear regression). We calculated the proportions of communications that were posts, comments, reactions, and views without reaction. We performed content analysis with a social support framework (informational, emotional, esteem-related, network, and tangible support), applying a maximum of 3 codes per communication.

RESULTS

VPN-T1D gained 212 regular members and 10 peer leaders over 33 months; of these 222 members, 26 (11.7%) exited. Peer leaders had 10-fold higher relative centrality than regular members (mean 0.53, SD 0.26 vs mean 0.04, SD 0.05; 0.49 difference; 95% CI 0.44-0.53). Overall, 91.4% (203/222) of the members connected at least once through posts, comments, or reactions. Among the 75,051 communications, there were 5109 (6.81%) posts, comments, and polls, 6233 (8.31%) reactions, and 63,709 (84.9%) views (lurking). Moreover, 54.9% (3430/6253) of codes applied were social support related, 66.4% (2277/3430) of which were informational (eg, insurance and travel preparation), and 20.4% (699/3430) of which were esteem related (eg, relieving blame).

CONCLUSIONS

Designating, training, and incentivizing peer leaders may stimulate content exchange and creation. Social support was a key VPN-T1D deliverable. Although lurking accounted for a high proportion of the overall activity, even those demonstrating this type of passive participation likely derived benefits, given that the network exit rate was low.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

RR2-10.2196/18714.

WeChat official accounts' posts on medication use of 251 community healthcare centers in Shanghai, China: content analysis and quality assessment

Background

The dissemination of online health information (OHI) on medication use via WeChat Official Accounts (WOAs) is an effective way to help primary care practitioners (PCPs) address drug-related problems (DRPs) in the community. Although an increasing number of primary care institutions in China have published WOA posts on medication use, their content and quality have not yet been assessed.

Objective

This study aimed to explore the general features and content of WOA posts on medication use published by community healthcare centers (CHCs) in Shanghai, China and to assess their quality of content. It also aimed to explore the factors associated with the number of post views.

Methods

From June 1 to October 31, 2022, two coauthors independently screened WOA posts on medication use published throughout 2021 by the CHCs in Shanghai. Content analysis was performed to analyze their general features (format, length, and source, etc.) and content (types of medicines and diseases). The QUEST tool was used to assess the quality of the posts. We compared the differences among posts published by CHCs in central urban areas and suburban areas, and used multiple linear regression to explore the factors associated with the number of post views.

Results

A total of 236 WOAs of interest published 37,147 posts in 2021, and 275 (0.74%) of them were included in the study. The median number of post views was 152. Thirty percent of the posts were reviewed by the CHCs' staff before publication and only 6% provided information on PCPs' consultations. The most commonly mentioned medicines and diseases in the posts were Chinese patent medicines (37.1%) and respiratory diseases (29.5%). The posts frequently provided information on indications (77%) and usage (56%) but rarely on follow-up (13%) and storage (11%). Of the posts, 94.9% had a total QUEST score < 17 (full score = 28). The median number of post views and total post quality scores did not significantly differ among the CHCs in central urban and suburban areas. In the multiple linear regression model, the number of post views was associated with scores of complementarity (B = 56.47, 95% CI 3.05, 109.89) and conflict of interest (B = -46.40, 95% CI -56.21, -36.60).

Conclusion

The quantity and quality of WOA posts on medication use published by CHCs in China need improvement. The quality of posts may partially impact the dissemination effect, but intrinsic causal associations merit further exploration.

A mixed-methods exploration of attitudes towards pregnant Facebook fitness influencers

BACKGROUND

Exercise during pregnancy is associated with various health benefits for both mother and child. Despite these benefits, most pregnant women do not meet physical activity recommendations. A known barrier to engaging in exercise during pregnancy is a lack of knowledge about appropriate and safe exercise. In our current era of social media, many pregnant women are turning to online information sources for guidance, including social media influencers. Little is known about attitudes towards pregnancy exercise information provided by influencers on social media platforms. This study aimed to explore attitudes towards exercise during pregnancy depicted by social media influencers on Facebook, and user engagement with posted content.

METHODS

A mixed-methods approach was used to analyse data from 10 Facebook video posts of social media influencers exercising during pregnancy. Quantitative descriptive analyses were used to report the number of views, shares, comments and emotive reactions. Qualitative analysis of user comments was achieved using an inductive thematic approach.

RESULTS

The 10 video posts analysed were viewed a total of 12,117,200 times, shared on 11,181 occasions, included 13,455 user comments and 128,804 emotive icon reactions, with the most frequently used icon being 'like' (81.48%). The thematic analysis identified three themes associated with attitudes including [1] exercise during pregnancy [2] influencers and [3] type of exercise. A fourth theme of community was also identified. Most user comments were associated with positive attitudes towards exercise during pregnancy and the influencer. However, attitudes towards the types of exercise the influencer performed were mixed (aerobic and body weight exercises were positive; resistance-based exercise with weights were negative). Finally, the online community perceived by users was mostly positive and recognised for offering social support and guidance.

CONCLUSIONS

User comments imply resistance-based exercise with weights as unsafe and unnecessary when pregnant, a perception that does not align with current best practice guidelines. Collectively, the findings from this study highlight the need for continued education regarding exercise during pregnancy and the potential for social media influencers to disseminate evidence-based material to pregnant women who are highly receptive to, and in need of reliable health information.

Opioids in Satirical News Shows: Exploring Topics, Sentiments, and Engagement in Last Week Tonight on YouTube

Prescription drug misuse and opioid overdose have become one of the most serious health crises in the US health system. In this study, we examined Last Week Tonight as hosted by John Oliver in order to explore topics, sentiments, and reactions by referring to online comments on YouTube. Using an unsupervised machine learning approach, we analyzed 33,741 user generated comments taken from three selected episodes published between 2016 and 2021. Our results revealed 13 topics: the politicization of the opioid crisis, side effects of opioid use, seeking help, legal issues, reactions to the John Oliver show, drug addiction, the US in comparison with other countries, case settlements, the drug business, addiction and pharmaceutical companies, profitability of companies from deaths, drugs and elections, as well as general awareness and education. In addition, we found that opioid side effects, seeking help, and the drug business had received the highest user engagements in terms of likes and replies. Our study suggests that comments with positive sentiments were more likely to cause a more widespread engagement from YouTube users than negative comments.

What are IBD Patients Talking About on Twitter? Using Natural Language Understanding to Investigate Patients' Tweets

This research aims to investigate what patients with inflammatory bowel disease (IBD) are talking about on Twitter and learn from the experimental knowledge they share online. The study presents a framework for analyzing patients' tweets and comparing their content to tweets of the general population. We started by constructing two datasets of tweets-a dataset of patients' tweets and a control dataset for comparison. Then, we thematically classified the tweets and obtained a subset of tweets related to health and nutrition. We used a Dirichlet regression to compare the thematic segmentations of the two groups. We continued by extracting keywords from the filtered tweets and applying entity sentiment analysis to determine the patients' sentiments towards the extracted keywords. Finally, we detected emotions within the tweets and used a Wilcoxon test to compare the emotions conveyed in each group. We found statistically significant differences between the patients' thematic segmentations and those of the control group and observed significant differences in the emotions each group expressed while talking about health. Not only do patients talk more about health in comparison to the general Twitter population, but they also address the subject with negative sentiments and express more negative emotions. The personal information IBD patients share on Twitter can be used to derive complementary knowledge about the disease and provide an additional foundation to existing medical research on IBD. The four stages of the study are also feasible to extend to other chronic conditions.

Introducing AI to the molecular tumor board: one direction toward the establishment of precision medicine using large-scale cancer clinical and biological information

Since U.S. President Barack Obama announced the Precision Medicine Initiative in his New Year's State of the Union address in 2015, the establishment of a precision medicine system has been emphasized worldwide, particularly in the field of oncology. With the advent of next-generation sequencers specifically, genome analysis technology has made remarkable progress, and there are active efforts to apply genome information to diagnosis and treatment. Generally, in the process of feeding back the results of next-generation sequencing analysis to patients, a molecular tumor board (MTB), consisting of experts in clinical oncology, genetic medicine, etc., is established to discuss the results. On the other hand, an MTB currently involves a large amount of work, with humans searching through vast databases and literature, selecting the best drug candidates, and manually confirming the status of available clinical trials. In addition, as personalized medicine advances, the burden on MTB members is expected to increase in the future. Under these circumstances, introducing cutting-edge artificial intelligence (AI) technology and information and communication technology to MTBs while reducing the burden on MTB members and building a platform that enables more accurate and personalized medical care would be of great benefit to patients. In this review, we introduced the latest status of elemental technologies that have potential for AI utilization in MTB, and discussed issues that may arise in the future as we progress with AI implementation.

Online Food Security Discussion Before and During the COVID-19 Pandemic in Native Hawaiian and Pacific Islander Community Groups and Organizations: Content Analysis of Facebook Posts

Background

The Native Hawaiian and Pacific Islander (NHPI) population experiences disproportionately higher rates of food insecurity, which is a risk factor for cardiometabolic diseases such as cardiovascular disease, type 2 diabetes, obesity, and hypertension, when compared to white individuals. Novel and effective approaches that address food insecurity are needed for the NHPI population, particularly in areas of the continental United States, which is a popular migration area for many NHPI families. Social media may serve as an opportune setting to reduce food insecurity and thus the risk factors for cardiometabolic diseases among NHPI people; however, it is unclear if and how food insecurity is discussed in online communities targeting NHPI individuals.

Objective

The objective of this study was to characterize the quantity, nature, and audience engagement of messages related to food insecurity posted online in community groups and organizations that target NHPI audiences.

Methods

Publicly accessible Facebook pages and groups focused on serving NHPI community members living in the states of Washington or Oregon served as the data source. Facebook posts between March and June 2019 (before the COVID-19 pandemic) and from March to June 2020 (during the COVID-19 pandemic) that were related to food security were identified using a set of 36 related keywords. Data on the post and any user engagement (ie, comments, shares, or digital reactions) were extracted for all relevant posts. A content analytical approach was used to identify and quantify the nature of the identified posts and any related comments. The codes resulting from the content analysis were described and compared by year, page type, and engagement.

Results

Of the 1314 nonduplicated posts in the 7 relevant Facebook groups and pages, 88 were related to food security (8 in 2019 and 80 in 2020). The nature of posts was broadly classified into literature-based codes, food assistance (the most common), perspectives of food insecurity, community gratitude and support, and macrolevel contexts. Among the 88 posts, 74% (n=65) had some form of engagement, and posts reflecting community gratitude and support or culture had more engagement than others (mean 19.9, 95% CI 11.2-28.5 vs mean 6.1, 95% CI 1.7-10.4; and mean 26.8, 95% CI 12.7-40.9 vs mean 5.3, 95% CI 3.0-7.7, respectively).

Conclusions

Food security–related posts in publicly accessible Facebook groups targeting NHPI individuals living in Washington and Oregon largely focused on food assistance, although cultural values of gratitude, maintaining NHPI culture, and supporting children were also reflected. Future work should capitalize on social media as a potential avenue to reach a unique cultural group in the United States experiencing inequitably high rates of food insecurity and risk of cardiometabolic diseases.

Cervical Cancer Prevention on Instagram: Content and Social Interaction Analysis of Brazilian Accounts

OBJECTIVE

The aim of the present study was to analyse the content of posts on Instagram about cervical cancer.

METHODS

It was conducted a qualitative analysis using the 50 most popular publicly available Portuguese-language Instagram posts, containing the hashtags #cervicalcancer, #papsmear, #hpv, #papillomavirus, and #hpvvac-cine, during the Brazilian national cervical cancer prevention campaign in March 2018.

RESULTS

Posts recruited using #cervicalcancer provided 60% of posts with contents related to secondary prevention; the #papsmear provided 46% of posts with irrelevant contents; the #hpv and #papillomavirus provided 50% and 64% of posts with informative content, respectively; and the #hpvvaccine provided 58% of posts with content related to primary prevention. The posts that received the highest number of likes were those from the hashtags #hpv and #papillomavirus with 151.33 and 78.00 likes/post, respectively. The majority of posts presented less than 05 comments/post, except for the #hpv, which had 64.76 comments/post. According to the users' profiles, the majority of the posts, regardless of the hashtag used, were made by health professionals.

CONCLUSION

The focus of Instagram posts about cervical cancer is on secondary prevention, which can contribute to the promotion of health behaviours not directed to aspects of primary prevention of the disease.

Navigating Diabetes Online Communities in Clinical Practice

OBJECTIVE

This purpose of this study was to conduct a rapid review and environmental scan of diabetes online communities (DOCs) to highlight the clinical impact of DOC engagement and provide guidance to health care providers for navigating and prescribing DOCs.

METHODS

This two-phase review included a rapid review of clinical outcomes and an assessment of DOC content. We conducted a literature search for studies evaluating DOC use and glucometric (glycated hemoglobin [A1C], time-in-range, hypoglycemia), behavioral, and psychosocial outcomes. The environmental scan of current DOC sites established key features, including available platforms, target population, content areas, and reach.

RESULTS

Twelve papers were included-eight reported DOC use within a context of do-it-yourself (DIY) noncommercial, opensource hybrid closed loop systems community and four were non-DYI related. In latter, all studies were cross-sectional, describing patient-reported outcomes. Two studies reported A1C lowering with DOC use, two DOC use was associated with high self-efficacy, one showed association with higher self-care levels, and one showed higher quality of life scores when compared with population norms. All DIY studies showed improvement in glucometrics after initiation of sensor augmented pump technology with DOC support. Of 54 DOC sites reviewed in the environmental scan, 29 were included. Diabetes online community sites were heterogenous in social media platform and type of diabetes targeted.

CONCLUSION

Diabetes online communities have the potential to improve clinical, behavioral, and psychosocial outcomes. Randomized control trials and/or longitudinal prospective studies evaluating outcomes are needed to further examine the potential benefits of DOC use. Diabetes online communities are diverse and span a variety of social media platforms, providing clinicians opportunities to individualize recommendations for DOC use.

The Influencing Factors of Nutrition and Diet Health Knowledge Dissemination Using the WeChat Official Account in Health Promotion

Background: The promotion of a healthy diet via health education is a component of the “Healthy China 2030” plan. However, few studies have reported whether health knowledge about nutrition and diet has gained public attention, and whether it is needed by the public.

Methods: The numbers of views, shares, and reads of articles published by the official WeChat account of a hospital in China were accessed. The influence index was obtained via the entropy analysis of these three indices. A questionnaire survey was developed based on the purpose of the study and the conclusion of the content analysis, which conducted to analyze users' requirements for health knowledge and their influencing factors. Moreover, risk factors were explored by logistic regression models.

Results: Of the 103 articles considered in this study, four articles in the Top 10 were related to nutrition and diet. The influence index of nutrition and diet knowledge was found to be the highest in the content analysis (p < 0.05). The higher degrees of humor (β = 0.224, p = 0.027), nutrition and diet articles (β = 0.776, p = 0.034), and cover articles (β = 0.312, p = 0.021) have significant influences on the influence index. In total, 581 questionnaires were obtained, and 78.1% of the respondents reported believing that the health knowledge of greatest concern was that related to nutrition and diet. Multivariate logistic regression analyses were conducted to explore the associations between the features of the articles and users reading nutrition and diet knowledge; it was found that gender (female, OR: 4.651, 95%Cl: 2.598, 8.325, and p < 0.001), age (young adult, OR: 0.358, 95%Cl: 0.266, 0.481, and p < 0.001), cancer precaution knowledge (OR: 4.333, 95%Cl: 2.262, 8.299, and p < 0.001), traditional Chinese medicine (OR: 2.121, 95%Cl: 1.064, 4.230, and p = 0.033), the knowledge acquisition approach [circle of friends (OR: 2.586, 95%Cl: 1.373, 4.868, and p = 0.003), social media (OR: 2.183, 95%Cl: 1.204, 3.960, and p = 0.010)), hospitals (OR: 3.194, 95%Cl: 1.793, 5.692, and p < 0.001), television media (OR: 4.348, 95%Cl: 2.341, 8.077, and p < 0.001)], and social media strategies [professionalism and authority (OR: 2.354, 95%Cl: 1.231, 4.505, and p = 0.006)] have statistically significant relationships with users reading nutrition and diet knowledge.

Conclusion: Nutrition and diet knowledge could contribute to WeChat user engagement of health information dissemination. Nutrition professionals should improve the scientific popularization ability and effectively use social media for health promotion.

TikTok as a Health Information Source: Assessment of the Quality of Information in Diabetes-Related Videos

Background

Diabetes has become one of the most prevalent chronic diseases, and many people living with diabetes use social media to seek health information. Recently, an emerging social media app, TikTok, has received much interest owing to its popularity among general health consumers. We notice that there are many videos about diabetes on TikTok. However, it remains unclear whether the information in these videos is of satisfactory quality.

Objective

This study aimed to assess the quality of the information in diabetes-related videos on TikTok.

Methods

We collected a sample of 199 diabetes-related videos in Chinese. The basic information presented in the videos was coded and analyzed. First, we identified the source of each video. Next, 2 independent raters assessed each video in terms of the completeness of six types of content (the definition of the disease, symptoms, risk factors, evaluation, management, and outcomes). Then, the 2 raters independently assessed the quality of information in the videos, using the DISCERN instrument.

Results

In regard to the sources of the videos, we found 6 distinct types of uploaders; these included 3 kinds of individual users (ie, health professionals, general users, and science communicators) and 3 types of organizational users (ie, news agencies, nonprofit organizations, and for-profit organizations). Regarding content, our results show that the videos were primarily about diabetes management and contained limited information on the definition of the disease, symptoms, risk factors, evaluation, and outcomes. The overall quality of the videos was acceptable, on average, although the quality of the information varied, depending on the sources. The videos created by nonprofit organizations had the highest information quality, while the videos contributed by for-profit organizations had the lowest information quality.

Conclusions

Although the overall quality of the information in the diabetes videos on TikTok is acceptable, TikTok might not fully meet the health information needs of patients with diabetes, and they should exercise caution when using TikTok as a source of diabetes-related information.

Evaluation of a Social Media Campaign in Saskatchewan to Promote Healthy Eating During the COVID-19 Pandemic: Social Media Analysis and Qualitative Interview Study

BACKGROUND

The beginning of the COVID-19 pandemic presented many sudden challenges regarding food, including grocery shopping changes (eg, reduced store hours, capacity restrictions, and empty store shelves due to food hoarding), restaurant closures, the need to cook more at home, and closures of food access programs. Eat Well Saskatchewan (EWS) implemented a 16-week social media campaign, #eatwellcovid19, led by a dietitian and nutrition student that focused on sharing stories submitted by the Saskatchewan public about how they were eating healthy during the COVID-19 pandemic.

OBJECTIVE

The goal of this study was to describe the implementation of the #eatwellcovid19 social media campaign and the results from the evaluation of the campaign, which included campaign performance using social media metrics and experiences and perspectives of campaign followers.

METHODS

Residents of Saskatchewan, Canada, were invited to submit personal stories and experiences to EWS about how they were eating healthy during the COVID-19 pandemic from April to August 2020. Each week, one to three stories were featured on EWS social media platforms-Facebook, Instagram, and Twitter-along with evidence-based nutrition information to help residents become more resilient to challenges related to food and nutrition experienced during the COVID-19 pandemic. Individuals who submitted stories were entered into a weekly draw for a Can $100 grocery gift card. Social media metrics and semistructured qualitative interviews of campaign followers were used to evaluate the #eatwellcovid19 campaign.

RESULTS

In total, 75 stories were submitted by 74 individuals on a variety of topics (eg, grocery shopping, traditional skills, and gardening), and 42 stories were featured on social media. EWS shared 194 #eatwellcovid19 posts across social media platforms (Facebook: n=100; Instagram: n=55; and Twitter: n=39). On Facebook, #eatawellcovid19 reached 100,571 followers and left 128,818 impressions, resulting in 9575 engagements. On Instagram, the campaign reached 11,310 followers, made 14,145 impressions, and received 823 likes and 15 comments. On Twitter, #eatwellcovid19 made 15,199 impressions and received 424 engagements. Featured story submission posts had the best engagement on Facebook and the most likes and comments on Instagram. The EWS social media pages reported increases in their following during the campaign (Instagram: +30%; Facebook: +14%; and Twitter: +12%). Results from the interviews revealed that there were two types of campaign followers: those who appreciated hearing the stories submitted by followers, as it helped them to feel connected to the community during social isolation, and those who appreciated the evidence-based information.

CONCLUSIONS

Numerous stories were submitted to the #eatwellcovid19 social media campaign on various topics. On Instagram and Facebook, posts that featured these stories had the highest engagement. During this campaign, EWS's social media following increased by more than 10% on each platform. The approach used for the #eatwellcovid19 campaign could be considered by others looking to develop health promotion campaigns.

Evaluation of Web-Based and In-Person Methods to Recruit Adults With Type 1 Diabetes for a Mobile Exercise Intervention: Prospective Observational Study

BACKGROUND

Our clinical trial of a mobile exercise intervention for adults 18 to 65 years old with type 1 diabetes (T1D) occurred during COVID-19 social distancing restrictions, prompting us to test web-based recruitment methods previously underexplored for this demographic.

OBJECTIVE

Our objectives for this study were to (1) evaluate the effectiveness and cost of using social media news feed advertisements, a clinic-based approach method, and web-based snowball sampling to reach inadequately active adults with T1D and (2) compare characteristics of enrollees against normative data.

METHODS

Participants were recruited between November 2019 and August 2020. In method #1, Facebook and Instagram news feed advertisements ran for five 1-to-8-day windows targeting adults (18 to 64 years old) in the greater New Haven and Hartford, Connecticut, areas with one or more diabetes-related profile interest. If interested, participants completed a webform so that the research team could contact them for eligibility screening. In method #2, patients 18 to 24 years old with T1D were approached in person at clinical visits in November and December 2019. Those who were interested immediately completed eligibility screening. Older patients could not be approached due to clinic restrictions. In method #3, snowball sampling was conducted by physically active individuals with T1D contacting their peers on Facebook and via email for 48 days, with details to contact the research staff to express interest and complete eligibility screening. Other methods referred participants to the study similarly to snowball sampling.

RESULTS

In method #1, advertisements were displayed to 11,738 unique viewers and attracted 274 clickers (2.33%); 20 participants from this group (7.3%) volunteered, of whom 8 (40%) were eligible. Costs averaged US $1.20 per click and US $95.88 per eligible volunteer. Men had lower click rates than women (1.71% vs 3.17%; P<.001), but their responsiveness and eligibility rates did not differ. In method #2, we approached 40 patients; 32 of these patients (80%) inquired about the study, of whom 20 (63%) volunteered, and 2 of these volunteers (10%) were eligible. Costs including personnel for in-person approaches averaged US $21.01 per inquirer and US $479.79 per eligible volunteer. In method #3, snowball sampling generated 13 inquirers; 12 of these inquirers (92%) volunteered, of whom 8 (67%) were eligible. Incremental costs to attract inquirers were negligible, and total costs averaged US $20.59 per eligible volunteer. Other methods yielded 7 inquirers; 5 of these inquirers (71%) volunteered, of whom 2 (40%) were eligible. Incremental costs to attract inquirers were negligible, and total costs averaged US $34.94 per eligible volunteer. Demographic overrepresentations emerged in the overall cohort (ie, optimal glycemic control, obesity, and low exercise), among those recruited by news feed advertisements (ie, obesity and older age), and among those recruited by snowball sampling (ie, optimal glycemic control and low exercise).

CONCLUSIONS

Web-based advertising and recruitment strategies are a promising means to attract adults with T1D to clinical trials and exercise interventions, with costs comparing favorably to prior trials despite targeting an uncommon condition (ie, T1D) and commitment to an intervention. These strategies should be tailored in future studies to increase access to higher-risk participants.

TRIAL REGISTRATION

ClinicalTrials.gov NCT04204733; https://clinicaltrials.gov/ct2/show/NCT04204733.