Identifying Patterns of Smoking Cessation App Feature Use That Predict Successful Quitting: Secondary Analysis of Experimental Data Leveraging Machine Learning.

BACKGROUND: Leveraging free smartphone apps can help expand the availability and use of evidence-based smoking cessation interventions. However, there is a need for additional research investigating how the use of different features within such apps impacts their effectiveness. OBJECTIVE: We used observational data collected from an experiment of a publicly available smoking cessation app to develop supervised machine learning (SML) algorithms intended to distinguish the app features that promote successful smoking cessation. We then assessed the extent to which patterns of app feature use accounted for variance in cessation that could not be explained by other known predictors of cessation (eg, tobacco use behaviors). METHODS: Data came from an experiment (ClinicalTrials.gov NCT04623736) testing the impacts of incentivizing ecological momentary assessments within the National Cancer Institute's quitSTART app. Participants' (N=133) app activity, including every action they took within the app and its corresponding time stamp, was recorded. Demographic and baseline tobacco use characteristics were measured at the start of the experiment, and short-term smoking cessation (7-day point prevalence abstinence) was measured at 4 weeks after baseline. Logistic regression SML modeling was used to estimate participants' probability of cessation from 28 variables reflecting participants' use of different app features, assigned experimental conditions, and phone type (iPhone [Apple Inc] or Android [Google]). The SML model was first fit in a training set (n=100) and then its accuracy was assessed in a held-aside test set (n=33). Within the test set, a likelihood ratio test (n=30) assessed whether adding individuals' SML-predicted probabilities of cessation to a logistic regression model that included demographic and tobacco use (eg, polyuse) variables explained additional variance in 4-week cessation. RESULTS: The SML model's sensitivity (0.67) and specificity (0.67) in the held-aside test set indicated that individuals' patterns of using different app features predicted cessation with reasonable accuracy. The likelihood ratio test showed that the logistic regression, which included the SML model-predicted probabilities, was statistically equivalent to the model that only included the demographic and tobacco use variables (P=.16). CONCLUSIONS: Harnessing user data through SML could help determine the features of smoking cessation apps that are most useful. This methodological approach could be applied in future research focusing on smoking cessation app features to inform the development and improvement of smoking cessation apps. TRIAL REGISTRATION: ClinicalTrials.gov NCT04623736; https://clinicaltrials.gov/study/NCT04623736.

Factors Associated With COVID-19 Behavioral Intentions: Findings From an Online Survey.

PURPOSE: The COVID-19 pandemic provides a novel context through which to evaluate salient factors for promoting behavioral change. We examined how attitudes, perceived community behaviors, and prior related behaviors predict intentions to (1) receive COVID-19 vaccination and (2) practice social distancing. DESIGN: Cross-sectional online survey administered through Amazon's Mechanical Turk in September 2020. SUBJECTS: A convenience sample of US adults (N = 1804). MEASURES: COVID-19 vaccination and social distancing intentions were measured on a 7-point Likert scale. Predictor variables included general vaccination and social distancing attitudes, perceived community mask-wearing, prior influenza vaccination, prior social distancing, and socio-demographics. ANALYSIS: Descriptive statistics and linear regressions. RESULTS: Thirty percent of respondents reported a strong willingness to receive COVID-19 vaccination, while 67% strongly intended to engage in social distancing. In regression analyses, vaccination intention was predicted by positive vaccine attitudes (b = .84; 95%CI: .78, .90; P < .001), prior influenza vaccination (b = -.47; 95%CI: -.63, -.32; P < .001), and perceived community mask-wearing (b = -.28; 95%CI: -.56, -.01; P=.049). Intention to practice social distancing was predicted by positive attitudes (b = .65; 95%CI: .61, .69; P < .001), prior social distancing (b = -.49; 95%CI: -.59, -.39; P < .001), and perceived community mask-wearing (b = -.28; 95%CI: -.46, -.09; P = .003). CONCLUSION: Findings have implications for health promotion efforts. Messages that are targeted and tailored on pre-existing attitudes may be more effective. Additionally, leveraging prior behaviors and perceived community behavior may improve communication strategies.

Discrimination, identity connectedness and tobacco use in a sample of sexual and gender minority young adults.

INTRODUCTION: Studies show that tobacco use among sexual and gender minority (SGM) populations is disproportionately higher than heterosexual or cisgender populations. However, few studies have examined tobacco use among SGM subgroups by race/ethnicity or associations between SGM-specific discrimination and connection to SGM identity and tobacco use. METHODS: This study analysed survey data from 11 313 SGM (gay, lesbian, bisexual, other sexual minority or gender minority) young adults in the USA and reported current cigarette, e-cigarette, other tobacco (cigar, smokeless tobacco, hookah) and polytobacco use. We used multinomial logistic regression to estimate associations between (a) SGM subgroup, race/ethnicity, SGM-specific discrimination and SGM identity connection and (b) each tobacco use outcome (vs never use of tobacco). We conducted postestimation testing to assess predicted probabilities of tobacco use against the sample average. RESULTS: Lesbian females (particularly black lesbian females) had higher-than-average probability of polytobacco use. White bisexual and lesbian participants had higher-than-average probability of cigarette and e-cigarette use, respectively. Higher levels of discrimination were associated with polytobacco use. Higher levels of identity connectedness were protective against certain tobacco use behaviours among gender minority participants and participants with high levels of discrimination experience. CONCLUSIONS: We found variations in tobacco use by SGM subgroups overall and by race/ethnicity. Discrimination may be a risk factor for certain tobacco use behaviours. However, SGM identity connectedness may be protective against tobacco use among gender minority individuals and individuals experiencing SGM-specific discrimination. These findings can inform targeted approaches to reach SGM subgroups at greater risk of tobacco use.

What Drives Health Professionals to Tweet About #HPVvaccine? Identifying Strategies for Effective Communication.

INTRODUCTION: We conducted this study to quantify how health professionals use Twitter to communicate about the human papillomavirus (HPV) vaccine. METHODS: We collected 193,379 tweets from August 2014 through July 2015 that contained key words related to HPV vaccine. We classified all tweets on the basis of user, audience, sentiment, content, and vaccine characteristic to examine 3 groups of tweets: 1) those sent by health professionals, 2) those intended for parents, and 3) those sent by health professionals and intended for parents. For each group, we identified the 7-day period in our sample with the most number of tweets (spikes) to report content. RESULTS: Of the 193,379 tweets, 20,451 tweets were from health professionals; 16,867 tweets were intended for parents; and 1,233 tweets overlapped both groups. The content of each spike varied per group. The largest spike in tweets from health professionals (n = 851) focused on communicating recently published scientific evidence. Most tweets were positive and were about resources and boys. The largest spike in tweets intended for parents (n = 1,043) centered on a national awareness day and were about resources, personal experiences, boys, and girls. The largest spike in tweets from health professionals to parents (n = 89) was in January and centered on an event hosted on Twitter that focused on cervical cancer awareness month. CONCLUSION: Understanding drivers of tweet spikes may help shape future communication and outreach. As more parents use social media to obtain health information, health professionals and organizations can leverage awareness events and personalize messages to maximize potential reach and parent engagement.