The Potential of Smartphone Apps in Informing Protobacco and Antitobacco Messaging Efforts Among Underserved Communities: Longitudinal Observational Study

Examining health apps and wearable use in improving physical and mental well-being across U.S., China, and Singapore

Health apps and wearables are touted to improve physical health and mental well-being. However, it is unclear from existing research the extent to which these health technologies are efficacious in improving physical and mental well-being at a population level, particularly for the underserved groups from the perspective of health equity and social determinants. Also, it is unclear if the relationship between health apps and wearables use and physical and mental well-being differs across individualistic, collectivistic, and a mix of individual-collectivistic cultures. A large-scale online survey was conducted in the U.S. (individualist culture), China (collectivist culture), and Singapore (mix of individual-collectivist culture) using quota sampling after obtaining ethical approval from the Institutional Review Board (IRB-2021-262) of Nanyang Technological University (NTU), Singapore. There was a total of 1004 respondents from the U.S., 1072 from China, and 1017 from Singapore. Data were analyzed using multiple regression and negative binomial regression. The study found that income consistently had the strongest relationship with physical and mental well-being measures in all three countries, while the use of health apps and wearables only had a moderate association with psychological well-being only in the US. Health apps and wearables were associated with the number of times people spent exercising and some mental health outcomes in China and Singapore, but they were only positively associated with psychological well-being in the US. The study emphasizes the importance of considering the social determinants, social-cultural context of the population, and the facilitating conditions for the effective use of digital health technologies. The study suggests that the combined use of both health apps and wearables is most strongly associated with better physical and mental health, though this association is less pronounced when individuals use only apps or wearables.

Examining COVID-19 Tweet Diffusion Using an Integrated Social Amplification of Risk and Issue-Attention Cycle Framework

Drawing upon the social amplification of risk (SARF) and the issue-attention cycle framework, we examined the amplification of COVID-19 risk-related tweets through (a) topics: key interests of discussion; (b) temperament: emotions of tweets; (c) topography (i.e., location); and (d) temporality (i.e., over time). We computationally analyzed 1,641,273 tweets, and conducted manual content analysis on a subset of 6,000 tweets to identify how topics, temperament, and topography of COVID-19 tweets were associated with risk amplification - retweet and favorite count - using negative binomial regression. We found 11 dominant COVID-19 topics-health impact, economic impact, reports of lockdowns, report of new cases, the need to stay home, coping with COVID-19, news about President Trump, government support, fight with COVID-19 by non-government entities, origins, and preventive measure in our corpus of tweets across the issue-attention cycle. The negative binomial regression results showed that at the pre-problem stage, topics on President Trump, speculation of origins, and initiatives to fight COVID-19 by non-government entities were most likely to be amplified, underscoring the inherent politicization of COVID-19 and erosion of trust in governments from the start of the pandemic. We also found that while tweets with negative emotions were consistently amplified throughout the issue-attention cycle, surprisingly tweets with positive emotions were amplified during the height of the pandemic - this counter-intuitive finding indicated signs of premature and misplaced optimism. Finally, our results showed that the locations of COVID-19 tweet amplification corresponded to the shifting COVID-19 hotspots across different continents across the issue-attention cycle. Theoretical and practical implications of risk amplification on social media are discussed.

Assessing the Impact of the Massachusetts Temporary Flavor Ban on Licensed Tobacco Retailers

Introduction

In 2019, the state of Massachusetts signed into law the first statewide sales restrictions of flavored ENDS/tobacco products for both physical and online shops in response to a previous executive order to curb E-Cigarette, or Vaping Product, Use Associated Lung Injury (EVALI) cases that were surging throughout the nation.

Methodology

This study obtained licensure data from the Massachusetts Department of Revenue, to observe the changes in retail licensure comparing the pre ban (October 2018-August 2019) and post ban periods (October 2020- August 2021). A series of linear regression tests were conducted on both periods using census tract data to explore potential associations with sociodemographic covariates, including median age, median household income, and population proportion by gender, age, and race/ethnicity groups.

Results

Analysis of the Massachusetts post-ban period (October 2020-August 2021) found that new tobacco retail licenses issued decreased by 52.9% (n = 968) when compared to the pre-ban period (October 2018-August 2019) of 1831. A significant positive association was discovered between change in new retailer count and proportion male population (2.48 ± 1.05, P = .018) as well as proportion Hispanic population (1.19 ± .25, P < .001) at the census tract level.

Conclusion/Discussion

Our analysis indicates that, following the temporary MA flavor sales ban, the total number of licenses decreased, though decreases were more pronounced for new licenses when compared to continuing licenses. Higher increases in new tobacco retailer density were significantly associated with concentration of male and Hispanic populations.

Toward an Extended Infodemiology Framework: Leveraging Social Media Data and Web Search Queries as Digital Pulse on Cancer Communication

This study aims to extend the infodemiology framework by postulating that effective use of digital data sources for cancer communication should consider four components: (a) content: key topics that people are concerned with, (b) congruence: how interest in cancer topics differ between public posts (i.e., tweets) and private web searches, (c) context: the influence of the information environment, and (d) information conduits. We compared tweets (n = 36, 968) and Google web searches on breast, lung, and prostate cancer between the National Cancer Prevention Month and a non-cancer awareness month in 2018. There are three key findings. First, reliance on public tweets alone may result in lost opportunities to identify potential cancer misinformation detected from private web searches. Second, lung cancer tweets were most sensitive to external information environment - tweets became substantially pessimistic after the end of cancer awareness month. Finally, the cancer communication landscape was largely democratized, with no prominent conduits dominating conversations on Twitter.

How Does Social Media Exposure and Engagement Influence College Students' Use of ENDS Products? A Cross-lagged Longitudinal Study

Electronic nicotine delivery systems (ENDS) products have been marketed heavily on social media throughout the past years, which exerts great influence on young adults' ENDS use. Despite scholars' pioneering efforts in investigating the influence of tobacco and nicotine products marketing on young adults' vaping behavior, scholarly attention has been paid primarily to passive exposure to rather than active engagement with the information on social media. In addition, the majority of existing research has been cross-sectional or focused on the unidirectional path from marketing information to behavior. To extend previous research in tobacco regulatory science on new media, we examined the bidirectional associations between self-reported exposure to and engagement with tobacco and nicotine products messaging on social media, and subsequent use of ENDS products one year later among a large, diverse sample of young adults. Results from cross-lagged panel analyses indicated that pro-tobacco/ENDS engagement and advertising exposure elevated risk whereas anti-tobacco/ENDS engagement decreased risk for the subsequent use of ENDS products one year later. On the other hand, the use of ENDS products positively predicted both pro- and anti-tobacco/ENDS engagement one year later. Findings provide empirical support for the reasoned action approach and the confirmation bias rooted in cognitive dissonance theory through rigorous longitudinal examination. Our findings not only point to the imperativeness of and offer guidance for regulating marketing information on social media, but also suggest social media as a promising platform to prevent young adults from initiating ENDS product use.

Designing Effective eHealth Interventions for Underserved Groups: Five Lessons From a Decade of eHealth Intervention Design and Deployment

Despite the proliferation of eHealth interventions, such as web portals, for health information dissemination or the use of mobile apps and wearables for health monitoring, research has shown that underserved groups do not benefit proportionately from these eHealth interventions. This is largely because of usability issues and the lack of attention to the broader structural, physical, and psychosocial barriers to technology adoption and use. The objective of this paper is to draw lessons from a decade of experience in designing different user-centered eHealth interventions (eg, web portals and health apps) to inform future work in leveraging technology to address health disparities. We draw these lessons from a series of interventions from the work we have done over 15 years in the Viswanath laboratory at the Dana-Farber Cancer Institute and Harvard TH Chan School of Public Health, focusing on three projects that used web portals and health apps targeted toward underserved groups. The projects were the following: Click to Connect, which was a community-based eHealth intervention that aimed to improve internet skills and health literacy among underserved groups by providing home access to high-speed internet, computer, and internet training classes, as well as a dedicated health web portal with ongoing technical support; PLANET MassCONECT, which was a knowledge translation project that built capacity among community-based organizations in Boston, Lawrence, and Worcester in Massachusetts to adopt evidence-based health promotion programs; and Smartphone App for Public Health, which was a mobile health research that facilitated both participatory (eg, surveys) and passive data (eg, geolocations and web-browsing behaviors) collection for the purpose of understanding tobacco message exposure in individuals’ built environment. Through our work, we distilled five key principles for researchers aiming to design eHealth interventions for underserved groups. They are as follows: develop a strategic road map to address communication inequalities (ie, a concrete action plan to identify the barriers faced by underserved groups and customize specific solutions to each of them), engage multiple stakeholders from the beginning for the long haul, design with usability—readability and navigability—in mind, build privacy safeguards into eHealth interventions and communicate privacy–utility tradeoffs in simplicity, and strive for an optimal balance between open science aspirations and protection of underserved groups.

An App-Based Mindfulness-Based Self-compassion Program to Support Caregivers of People With Dementia: Participatory Feasibility Study

BACKGROUND

The number of persons with dementia is steadily growing, as is the number of individuals supporting persons with dementia. Primary caregivers of persons with dementia are most often family members or spouses of the persons with dementia, and they are more likely to experience increased stress and other negative effects than individuals who are not primary caregivers. Although in-person support groups have been shown to help buffer the negative impacts of caregiving, some caregivers live in isolated or rural communities and are unable to make the burdensome commitment of traveling to cities. Using an interdisciplinary approach, we developed a mobile smartphone support app designed for primary caregivers of persons with dementia, with the goal of reducing caregiver burden and easing stress. The app features a 12-week intervention, largely rooted in mindfulness-based self-compassion (MBSC), because MBSC has been linked to minimizing stress, depression, and anxiety.

OBJECTIVE

The primary objectives of our program are twofold: to explore the feasibility of a 12-week mobile support program and to conduct an initial efficacy evaluation of changes in perceived caregiver burden, coping styles, and emotional well-being of caregivers before and after the program.

METHODS

Our feasibility study used a 2-phase participatory pretest and posttest design, focusing on acceptability, demand, practicality, implementation, and efficacy. At phase I, we recruited 57 primary caregivers of persons with dementia (mean age 76.3, SD 12.9 years), comprising spouses (21/57, 37%), children (21/57, 37%), and friends or relatives (15/57, 26%) of persons with dementia, of whom 29 (51%) completed all measures at both pre- and postprogram. The content of the program featured a series of MBSC podcasts. Our primary outcome measure was caregiver burden, with secondary outcome measures including coping styles and emotional well-being. Daily ecological momentary assessments enabled us to ask participants, "How are you feeling today?" Phase II of our study involved semistructured follow-up interviews with most participants (n=21) who completed phase I.

RESULTS

Our findings suggest that our app or program meets the feasibility criteria examined. Notably, participants generally accepted the program and believed it could be a useful resource. Emotional well-being increased significantly (P=.04), and emotion-based coping significantly decreased (P=.01). Participants generally considered the app or program to be a helpful resource.

CONCLUSIONS

Although there were no significant changes in caregiver burden, we were encouraged by the increased emotional well-being of our participants following the completion of our program. We also conclude that our app or program demonstrated feasibility (ie, acceptability, practicality, implementation, and efficacy) and can provide a much-needed resource for primary caregivers of persons with dementia. In the subsequent version of the program, we will respond to participant feedback by incorporating web-based weekly sessions and incorporating an outcome measure of self-compassion.

Tobacco messages encountered in real-time among low socio-economic position groups: a descriptive study

BACKGROUND

Tobacco advertising disproportionately targets low socio-economic position (SEP) groups, causing higher rates of tobacco use in this population. Anti-tobacco public health education campaigns persuade against use. This study measured real-time exposure of pro- and anti-tobacco messages from low SEP groups in two American cities.

METHODS

Individuals in low SEP groups (N = 95), aged 18-34 years old, who were smokers and non-smokers, from the Boston and Houston areas, took part in a mobile health study. They submitted images of tobacco-related messages they encountered via a mobile application for a 7-week period. Two coders analyzed the images for message characteristics. Intercoder reliability was established using Krippendorff's alpha and data were analyzed descriptively.

RESULTS

Of the submitted images (N = 131), 83 were pro-tobacco and 53 were anti-tobacco. Of the pro-tobacco messages, the majority were cigarette ads (80.7%) seen outside (36.1%) or inside (30.1%) a convenience store or gas station and used conventional themes (e.g., price promotion; 53.2%). Of the anti-tobacco messages, 56.6% were sponsored by public health campaigns or were signage prohibiting smoking in a public area (39.6%). Most focused on the health harms of smoking (28.3%).

CONCLUSION

Low SEP groups in this study encountered more pro-tobacco than anti-tobacco messages at places that were point-of-sale using price promotions to appeal to this group. Anti-tobacco messages at point-of-sale and/or advertising regulations may help combat tobacco use.

Understanding the Predictors of Missing Location Data to Inform Smartphone Study Design: Observational Study

BACKGROUND

Smartphone location data can be used for observational health studies (to determine participant exposure or behavior) or to deliver a location-based health intervention. However, missing location data are more common when using smartphones compared to when using research-grade location trackers. Missing location data can affect study validity and intervention safety.

OBJECTIVE

The objective of this study was to investigate the distribution of missing location data and its predictors to inform design, analysis, and interpretation of future smartphone (observational and interventional) studies.

METHODS

We analyzed hourly smartphone location data collected from 9665 research participants on 488,400 participant days in a national smartphone study investigating the association between weather conditions and chronic pain in the United Kingdom. We used a generalized mixed-effects linear model with logistic regression to identify whether a successfully recorded geolocation was associated with the time of day, participants' time in study, operating system, time since previous survey completion, participant age, sex, and weather sensitivity.

RESULTS

For most participants, the app collected a median of 2 out of a maximum of 24 locations (1760/9665, 18.2% of participants), no location data (1664/9665, 17.2%), or complete location data (1575/9665, 16.3%). The median locations per day differed by the operating system: participants with an Android phone most often had complete data (a median of 24/24 locations) whereas iPhone users most often had a median of 2 out of 24 locations. The odds of a successfully recorded location for Android phones were 22.91 times higher than those for iPhones (95% CI 19.53-26.87). The odds of a successfully recorded location were lower during weekends (odds ratio [OR] 0.94, 95% CI 0.94-0.95) and nights (OR 0.37, 95% CI 0.37-0.38), if time in study was longer (OR 0.99 per additional day in study, 95% CI 0.99-1.00), and if a participant had not used the app recently (OR 0.96 per additional day since last survey entry, 95% CI 0.96-0.96). Participant age and sex did not predict missing location data.

CONCLUSIONS

The predictors of missing location data reported in our study could inform app settings and user instructions for future smartphone (observational and interventional) studies. These predictors have implications for analysis methods to deal with missing location data, such as imputation of missing values or case-only analysis. Health studies using smartphones for data collection should assess context-specific consequences of high missing data, especially among iPhone users, during the night and for disengaged participants.

Medical Specialty Recommendations by an Artificial Intelligence Chatbot on a Smartphone: Development and Deployment

Background

The COVID-19 pandemic has limited daily activities and even contact between patients and primary care providers. This makes it more difficult to provide adequate primary care services, which include connecting patients to an appropriate medical specialist. A smartphone-compatible artificial intelligence (AI) chatbot that classifies patients’ symptoms and recommends the appropriate medical specialty could provide a valuable solution.

Objective

In order to establish a contactless method of recommending the appropriate medical specialty, this study aimed to construct a deep learning–based natural language processing (NLP) pipeline and to develop an AI chatbot that can be used on a smartphone.

Methods

We collected 118,008 sentences containing information on symptoms with labels (medical specialty), conducted data cleansing, and finally constructed a pipeline of 51,134 sentences for this study. Several deep learning models, including 4 different long short-term memory (LSTM) models with or without attention and with or without a pretrained FastText embedding layer, as well as bidirectional encoder representations from transformers for NLP, were trained and validated using a randomly selected test data set. The performance of the models was evaluated on the basis of the precision, recall, F₁-score, and area under the receiver operating characteristic curve (AUC). An AI chatbot was also designed to make it easy for patients to use this specialty recommendation system. We used an open-source framework called “Alpha” to develop our AI chatbot. This takes the form of a web-based app with a frontend chat interface capable of conversing in text and a backend cloud-based server application to handle data collection, process the data with a deep learning model, and offer the medical specialty recommendation in a responsive web that is compatible with both desktops and smartphones.

Results

The bidirectional encoder representations from transformers model yielded the best performance, with an AUC of 0.964 and F₁-score of 0.768, followed by LSTM model with embedding vectors, with an AUC of 0.965 and F₁-score of 0.739. Considering the limitations of computing resources and the wide availability of smartphones, the LSTM model with embedding vectors trained on our data set was adopted for our AI chatbot service. We also deployed an Alpha version of the AI chatbot to be executed on both desktops and smartphones.

Conclusions

With the increasing need for telemedicine during the current COVID-19 pandemic, an AI chatbot with a deep learning–based NLP model that can recommend a medical specialty to patients through their smartphones would be exceedingly useful. This chatbot allows patients to identify the proper medical specialist in a rapid and contactless manner, based on their symptoms, thus potentially supporting both patients and primary care providers.