Discerning conversational context in online health communities for personalized digital behavior change solutions using Pragmatics to Reveal Intent in Social Media (PRISM) framework.

BACKGROUND: Online health communities (OHCs) have emerged as prominent platforms for behavior modification, and the digitization of online peer interactions has afforded researchers with unique opportunities to model multilevel mechanisms that drive behavior change. Existing studies, however, have been limited by a lack of methods that allow the capture of conversational context and socio-behavioral dynamics at scale, as manifested in these digital platforms. OBJECTIVE: We develop, evaluate, and apply a novel methodological framework, Pragmatics to Reveal Intent in Social Media (PRISM), to facilitate granular characterization of peer interactions by combining multidimensional facets of human communication. METHODS: We developed and applied PRISM to analyze peer interactions (N = 2.23 million) in QuitNet, an OHC for tobacco cessation. First, we generated a labeled set of peer interactions (n = 2,005) through manual annotation along three dimensions: communication themes (CTs), behavior change techniques (BCTs), and speech acts (SAs). Second, we used deep learning models to apply our qualitative codes at scale. Third, we applied our validated model to perform a retrospective analysis. Finally, using social network analysis (SNA), we portrayed large-scale patterns and relationships among the aforementioned communication dimensions embedded in peer interactions in QuitNet. RESULTS: Qualitative analysis showed that the themes of social support and behavioral progress were common. The most used BCTs were feedback and monitoring and comparison of behavior, and users most commonly expressed their intentions using SAs-expressive and emotion. With additional in-domain pre-training, bidirectional encoder representations from Transformers (BERT) outperformed other deep learning models on the classification tasks. Content-specific SNA revealed that users' engagement or abstinence status is associated with the prevalence of various categories of BCTs and SAs, which also was evident from the visualization of network structures. CONCLUSIONS: Our study describes the interplay of multilevel characteristics of online communication and their association with individual health behaviors.

Digital health technologies for peripartum depression management among low-socioeconomic populations: perspectives from patients, providers, and social media channels.

BACKGROUND: Peripartum Depression (PPD) affects approximately 10-15% of perinatal women in the U.S., with those of low socioeconomic status (low-SES) more likely to develop symptoms. Multilevel treatment barriers including social stigma and not having appropriate access to mental health resources have played a major role in PPD-related disparities. Emerging advances in digital technologies and analytics provide opportunities to identify and address access barriers, knowledge gaps, and engagement issues. However, most market solutions for PPD prevention and management are produced generically without considering the specialized needs of low-SES populations. In this study, we examine and portray the information and technology needs of low-SES women by considering their unique perspectives and providers' current experiences. We supplement our understanding of women's needs by harvesting online social discourse in PPD-related forums, which we identify as valuable information resources among these populations. METHODS: We conducted (a) 2 focus groups (n = 9), (b) semi-structured interviews with care providers (n = 9) and low SES women (n = 10), and (c) secondary analysis of online messages (n = 1,424). Qualitative data were inductively analyzed using a grounded theory approach. RESULTS: A total of 134 open concepts resulted from patient interviews, 185 from provider interviews, and 106 from focus groups. These revealed six core themes for PPD management, including "Use of Technology/Features", "Access to Care", and "Pregnancy Education". Our social media analysis revealed six PPD topics of importance in online messages, including "Physical and Mental Health" (n = 725 messages), and "Social Support" (n = 674). CONCLUSION: Our data triangulation allowed us to analyze PPD information and technology needs at different levels of granularity. Differences between patients and providers included a focus from providers on needing better support from administrative staff, as well as better PPD clinical decision support. Our results can inform future research and development efforts to address PPD health disparities.

Measuring and controlling medical record abstraction (MRA) error rates in an observational study.

BACKGROUND: Studies have shown that data collection by medical record abstraction (MRA) is a significant source of error in clinical research studies relying on secondary use data. Yet, the quality of data collected using MRA is seldom assessed. We employed a novel, theory-based framework for data quality assurance and quality control of MRA. The objective of this work is to determine the potential impact of formalized MRA training and continuous quality control (QC) processes on data quality over time. METHODS: We conducted a retrospective analysis of QC data collected during a cross-sectional medical record review of mother-infant dyads with Neonatal Opioid Withdrawal Syndrome. A confidence interval approach was used to calculate crude (Wald's method) and adjusted (generalized estimating equation) error rates over time. We calculated error rates using the number of errors divided by total fields ("all-field" error rate) and populated fields ("populated-field" error rate) as the denominators, to provide both an optimistic and a conservative measurement, respectively. RESULTS: On average, the ACT NOW CE Study maintained an error rate between 1% (optimistic) and 3% (conservative). Additionally, we observed a decrease of 0.51 percentage points with each additional QC Event conducted. CONCLUSIONS: Formalized MRA training and continuous QC resulted in lower error rates than have been found in previous literature and a decrease in error rates over time. This study newly demonstrates the importance of continuous process controls for MRA within the context of a multi-site clinical research study.

Pragmatics to Reveal Intent in Social Media Peer Interactions: Mixed Methods Study.

BACKGROUND: Online health communities (OHCs) have emerged as the leading venues for behavior change and health-related information seeking. The soul and success of these digital platforms lie in their ability to foster social togetherness and a sense of community by providing personalized support. However, we have a minimal understanding of how conversational posts in these settings lead to collaborative societies and ultimately result in positive health changes through social influence. OBJECTIVE: Our objective is to develop a content-specific and intent-sensitive methodological framework for analyzing peer interactions in OHCs. METHODS: We developed and applied a mixed-methods approach to understand the manifestation of expressions in peer interactions in OHCs. We applied our approach to describe online social dialogue in the context of two online communities, QuitNet (QN) and the American Diabetes Association (ADA) support community. A total of 3011 randomly selected peer interactions (n=2005 from QN, n=1006 from ADA) were analyzed. Specifically, we conducted thematic analysis to characterize communication content and linguistic expressions (speech acts) embedded within the two data sets. We also developed an empirical user persona based on their engagement levels and behavior profiles. Further, we examined the association between speech acts and communication themes across observed tiers of user engagement and self-reported behavior profiles using the chi-square test or the Fisher test. RESULTS: Although social support, the most prevalent communication theme in both communities, was expressed in several subtle manners, the prevalence of emotions was higher in the tobacco cessation community and assertions were higher in the diabetes self-management (DSM) community. Specific communication theme-speech act relationships were revealed, such as the social support theme was significantly associated (P<.05) with 9 speech acts from a total of 10 speech acts (ie, assertion, commissive, declarative, desire, directive, expressive, question, stance, and statement) within the QN community. Only four speech acts (ie, commissive, emotion, expressive, and stance) were significantly associated (P<.05) with the social support theme in the ADA community. The speech acts were also significantly associated with the users' abstinence status within the QN community and with the users' lifestyle status within the ADA community (P<.05). CONCLUSIONS: Such an overlay of communication intent implicit in online peer interactions alongside content-specific theory-linked characterizations of social media discourse can inform the development of effective digital health technologies in the field of health promotion and behavior change. Our analysis revealed a rich gradient of expressions across a standardized thematic vocabulary, with a distinct variation in emotional and informational needs, depending on the behavioral and disease management profiles within and across the communities. This signifies the need and opportunities for coupling pragmatic messaging in digital therapeutics and care management pathways for personalized support.

Diabetes-Related Topics in an Online Forum for Caregivers of Individuals Living With Alzheimer Disease and Related Dementias: Qualitative Inquiry.

BACKGROUND: Diabetes and Alzheimer disease and related dementias (ADRD) are the seventh and sixth leading causes of death in the United States, respectively, and they coexist in many older adults. Caring for a loved one with both ADRD and diabetes is challenging and burdensome. OBJECTIVE: This study aims to explore diabetes-related topics in the Alzheimer's Association ALZConnected caregiver forum by family caregivers of persons living with ADRD. METHODS: User posts on the Alzheimer's Association ALZConnected caregiver forum were extracted. A total of 528 posts related to diabetes were included in the analysis. Of the users who generated the 528 posts, approximately 96.1% (275/286) were relatives of the care recipient with ADRD (eg, child, grandchild, spouse, sibling, or unspecified relative). Two researchers analyzed the data independently using thematic analysis. Any divergence was discussed among the research team, and an agreement was reached with a senior researcher's input as deemed necessary. RESULTS: Thematic analysis revealed 7 key themes. The results showed that comorbidities of ADRD were common topics of discussions among family caregivers. Diabetes management in ADRD challenged family caregivers. Family caregivers might neglect their own health care because of the caring burden, and they reported poor health outcomes and reduced quality of life. The online forum provided a platform for family caregivers to seek support in their attempts to learn more about how to manage the ADRD of their care recipients and seek support for managing their own lives as caregivers. CONCLUSIONS: The ALZConnected forum provided a platform for caregivers to seek informational and emotional support for caring for persons living with ADRD and diabetes. The overwhelming burdens with these two health conditions were apparent for both caregivers and care recipients based on discussions from the online forum. Studies are urgently needed to provide practical guidelines and interventions for diabetes management in individuals with diabetes and ADRD. Future studies to explore delivering diabetes management interventions through online communities in caregivers and their care recipients with ADRD and diabetes are warranted.

In Pursuit of Theoretical Ground in Behavior Change Support Systems: Analysis of Peer-to-Peer Communication in a Health-Related Online Community.

BACKGROUND: Research studies involving health-related online communities have focused on examining network structure to understand mechanisms underlying behavior change. Content analysis of the messages exchanged in these communities has been limited to the "social support" perspective. However, existing behavior change theories suggest that message content plays a prominent role reflecting several sociocognitive factors that affect an individual's efforts to make a lifestyle change. An understanding of these factors is imperative to identify and harness the mechanisms of behavior change in the Health 2.0 era. OBJECTIVE: The objective of this work is two-fold: (1) to harness digital communication data to capture essential meaning of communication and factors affecting a desired behavior change, and (2) to understand the applicability of existing behavior change theories to characterize peer-to-peer communication in online platforms. METHODS: In this paper, we describe grounded theory-based qualitative analysis of digital communication in QuitNet, an online community promoting smoking cessation. A database of 16,492 de-identified public messages from 1456 users from March 1-April 30, 2007, was used in our study. We analyzed 795 messages using grounded theory techniques to ensure thematic saturation. This analysis enabled identification of key concepts contained in the messages exchanged by QuitNet members, allowing us to understand the sociobehavioral intricacies underlying an individual's efforts to cease smoking in a group setting. We further ascertained the relevance of the identified themes to theoretical constructs in existing behavior change theories (eg, Health Belief Model) and theoretically linked techniques of behavior change taxonomy. RESULTS: We identified 43 different concepts, which were then grouped under 12 themes based on analysis of 795 messages. Examples of concepts include "sleepiness," "pledge," "patch," "spouse," and "slip." Examples of themes include "traditions," "social support," "obstacles," "relapse," and "cravings." Results indicate that themes consisting of member-generated strategies such as "virtual bonfires" and "pledges" were related to the highest number of theoretical constructs from the existing behavior change theories. In addition, results indicate that the member-generated communication content supports sociocognitive constructs from more than one behavior change model, unlike the majority of the existing theory-driven interventions. CONCLUSIONS: With the onset of mobile phones and ubiquitous Internet connectivity, online social network data reflect the intricacies of human health behavior as experienced by health consumers in real time. This study offers methodological insights for qualitative investigations that examine the various kinds of behavioral constructs prevalent in the messages exchanged among users of online communities. Theoretically, this study establishes the manifestation of existing behavior change theories in QuitNet-like online health communities. Pragmatically, it sets the stage for real-time, data-driven sociobehavioral interventions promoting healthy lifestyle modifications by allowing us to understand the emergent user needs to sustain a desired behavior change.

Content-driven analysis of an online community for smoking cessation: integration of qualitative techniques, automated text analysis, and affiliation networks.

OBJECTIVES: We identified content-specific patterns of network diffusion underlying smoking cessation in the context of online platforms, with the aim of generating targeted intervention strategies. METHODS: QuitNet is an online social network for smoking cessation. We analyzed 16 492 de-identified peer-to-peer messages from 1423 members, posted between March 1 and April 30, 2007. Our mixed-methods approach comprised qualitative coding, automated text analysis, and affiliation network analysis to identify, visualize, and analyze content-specific communication patterns underlying smoking behavior. RESULTS: Themes we identified in QuitNet messages included relapse, QuitNet-specific traditions, and cravings. QuitNet members who were exposed to other abstinent members by exchanging content related to interpersonal themes (e.g., social support, traditions, progress) tended to abstain. Themes found in other types of content did not show significant correlation with abstinence. CONCLUSIONS: Modeling health-related affiliation networks through content-driven methods can enable the identification of specific content related to higher abstinence rates, which facilitates targeted health promotion.

A pilot acceptability evaluation of MomMind: A digital health intervention for Peripartum Depression prevention and management focused on health disparities.

Health disparities cause significant strain on the wellbeing of individuals and society. In this study, we focus on the health disparities present in the condition of Peripartum Depression (PPD), a significant public health issue. While PPD can be managed through therapy and medication, many women do not receive adequate PPD treatment due to issues of social stigma and limited access to healthcare resources. Digital health technologies can offer practical tools for PPD management. However, current solutions do not integrate behavior theory and are rarely responsive to the transient information needs stemming from women's unique sociodemographic, clinical and psychosocial profiles. We describe a pilot acceptability evaluation of MomMind, a health-disparities focused digital health intervention for the prevention and management of PPD. A crucial MomMind advantage is its basis on behavior change theory and patient engagement as enabled by the Digilego digital health framework. Following an internal usability evaluation, MomMind was evaluated by patients through cross-sectional acceptability surveys, pre-and-post PPD health literacy surveys, and interviews. Survey respondents included n = 30 peripartum women, of whom n = 16 (53.3%) were Hispanic and n = 17 (56.7%) of low-income. Survey results show that 96.6% of participants (n = 29) approved and welcomed MomMind, and 90% (n = 27) found MomMind to be an appealing intervention. Additionally, significant improvements (p< = 0.05) were observed in participants' PPD health literacy, specifically their ability to recognize PPD symptoms and knowledge of how to seek PPD information. Interview main themes include MomMind's straightforward design and influence of others (family members, providers) on use of technology. Results suggest that enhancement of a digital health framework with health literacy theory can support production of digital health solutions acceptable to vulnerable populations. This study incorporates existing theories from different disciplines into a unified approach for mitigating health disparities, and produced a novel solution for promotion of health in a vulnerable population.

Sequencing conversational turns in peer interactions: An integrated approach for evidence-based conversational agent for just-in-time nicotine cravings intervention.

Background: Risky health behaviors place an enormous toll on public health systems. While relapse prevention support is integrated with most behavior modification programs, the results are suboptimal. Recent advances in artificial intelligence (AI) applications provide us with unique opportunities to develop just-in-time adaptive behavior change solutions. Methods: In this study, we present an innovative framework, grounded in behavioral theory, and enhanced with social media sequencing and communications scenario builder to architect a conversational agent (CA) specialized in the prevention of relapses in the context of tobacco cessation. We modeled peer interaction data (n = 1000) using the taxonomy of behavior change techniques (BCTs) and speech act (SA) theory to uncover the socio-behavioral and linguistic context embedded within the online social discourse. Further, we uncovered the sequential patterns of BCTs and SAs from social conversations (n = 339,067). We utilized grounded theory-based techniques for extracting the scenarios that best describe individuals' needs and mapped them into the architecture of the virtual CA. Results: The frequently occurring sequential patterns for BCTs were comparison of behavior and feedback and monitoring; for SAs were directive and assertion. Five cravings-related scenarios describing users' needs as they deal with nicotine cravings were identified along with the kinds of behavior change constructs that are being elicited within those scenarios. Conclusions: AI-led virtual CAs focusing on behavior change need to employ data-driven and theory-linked approaches to address issues related to engagement, sustainability, and acceptance. The sequential patterns of theory and intent manifestations need to be considered when developing effective behavior change CAs.

Digital health technologies for high-risk pregnancy management: three case studies using Digilego framework.

Objective: High-risk pregnancy (HRP) conditions such as gestational diabetes mellitus (GDM), hypertension (HTN), and peripartum depression (PPD) affect maternal and neonatal health. Patient engagement is critical for effective HRP management (HRPM). While digital technologies and analytics hold promise, emerging research indicates limited and suboptimal support offered by the highly prevalent pregnancy digital solutions within the commercial marketplace. In this article, we describe our efforts to develop a portfolio of digital products leveraging advances in social computing, data science, and digital health. Methods: We describe three studies that leverage core methods from Digilego digital health development framework to (1) conduct large-scale social media analysis (n = 55 301 posts) to understand population-level patterns in women's needs, (2) architect a digital repository to enable women curate HRP related information, and (3) develop a digital platform to support PPD prevention. We applied a combination of qualitative coding, machine learning, theory-mapping, and programmatic implementation of theory-linked digital features. Further, we conducted preliminary testing of the resulting products for acceptance with sample of pregnant women for GDM/HTN information management (n = 10) and PPD prevention (n = 30). Results: Scalable social computing models using deep learning classifiers with reasonable accuracy have allowed us to capture and examine psychosociobehavioral drivers associated with HRPM. Our work resulted in two digital health solutions, MyPregnancyChart and MomMind are developed. Initial evaluation of both tools indicates positive acceptance from potential end users. Further evaluation with MomMind revealed statistically significant improvements (P < .05) in PPD recognition and knowledge on how to seek PPD information. Discussion: Digilego framework provides an integrative methodological lens to gain micro-macro perspective on women's needs, theory integration, engagement optimization, as well as subsequent feature and content engineering, which can be organized into core and specialized digital pathways for women engagement in disease management. Conclusion: Future works should focus on implementation and testing of digital solutions that facilitate women to capture, aggregate, preserve, and utilize, otherwise siloed, prenatal information artifacts for enhanced self-management of their high-risk conditions, ultimately leading to improved health outcomes.

A scoping review of digital health interventions for combating COVID-19 misinformation and disinformation.

OBJECTIVE: We provide a scoping review of Digital Health Interventions (DHIs) that mitigate COVID-19 misinformation and disinformation seeding and spread. MATERIALS AND METHODS: We applied our search protocol to PubMed, PsychINFO, and Web of Science to screen 1666 articles. The 17 articles included in this paper are experimental and interventional studies that developed and tested public consumer-facing DHIs. We examined these DHIs to understand digital features, incorporation of theory, the role of healthcare professionals, end-user experience, and implementation issues. RESULTS: The majority of studies (n = 11) used social media in DHIs, but there was a lack of platform-agnostic generalizability. Only half of the studies (n = 9) specified a theory, framework, or model to guide DHIs. Nine studies involve healthcare professionals as design or implementation contributors. Only one DHI was evaluated for user perceptions and acceptance. DISCUSSION: The translation of advances in online social computing to interventions is sparse. The limited application of behavioral theory and cognitive models of reasoning has resulted in suboptimal targeting of psychosocial variables and individual factors that may drive resistance to misinformation. This affects large-scale implementation and community outreach efforts. DHIs optimized through community-engaged participatory methods that enable understanding of unique needs of vulnerable communities are urgently needed. CONCLUSIONS: We recommend community engagement and theory-guided engineering of equitable DHIs. It is important to consider the problem of misinformation and disinformation through a multilevel lens that illuminates personal, clinical, cultural, and social pathways to mitigate the negative consequences of misinformation and disinformation on human health and wellness.

Lessons Learned From Interdisciplinary Efforts to Combat COVID-19 Misinformation: Development of Agile Integrative Methods From Behavioral Science, Data Science, and Implementation Science.

Background: Despite increasing awareness about and advances in addressing social media misinformation, the free flow of false COVID-19 information has continued, affecting individuals' preventive behaviors, including masking, testing, and vaccine uptake. Objective: In this paper, we describe our multidisciplinary efforts with a specific focus on methods to (1) gather community needs, (2) develop interventions, and (3) conduct large-scale agile and rapid community assessments to examine and combat COVID-19 misinformation. Methods: We used the Intervention Mapping framework to perform community needs assessment and develop theory-informed interventions. To supplement these rapid and responsive efforts through large-scale online social listening, we developed a novel methodological framework, comprising qualitative inquiry, computational methods, and quantitative network models to analyze publicly available social media data sets to model content-specific misinformation dynamics and guide content tailoring efforts. As part of community needs assessment, we conducted 11 semistructured interviews, 4 listening sessions, and 3 focus groups with community scientists. Further, we used our data repository with 416,927 COVID-19 social media posts to gather information diffusion patterns through digital channels. Results: Our results from community needs assessment revealed the complex intertwining of personal, cultural, and social influences of misinformation on individual behaviors and engagement. Our social media interventions resulted in limited community engagement and indicated the need for consumer advocacy and influencer recruitment. The linking of theoretical constructs underlying health behaviors to COVID-19-related social media interactions through semantic and syntactic features using our computational models has revealed frequent interaction typologies in factual and misleading COVID-19 posts and indicated significant differences in network metrics such as degree. The performance of our deep learning classifiers was reasonable, with an F-measure of 0.80 for speech acts and 0.81 for behavior constructs. Conclusions: Our study highlights the strengths of community-based field studies and emphasizes the utility of large-scale social media data sets in enabling rapid intervention tailoring to adapt grassroots community interventions to thwart misinformation seeding and spread among minority communities. Implications for consumer advocacy, data governance, and industry incentives are discussed for the sustainable role of social media solutions in public health.

Unpacking Misinfodemic During a Global Health Crisis: A Qualitative Inquiry of Psychosocial Characteristics in Social Media Interactions.

The pervasiveness of health information in social media has led to a modern misinformation crisis, also known as a misinfodemic. Misinfodemics have upended public health activities as clearly evident during the COVID-19 pandemic. The objective of this study is to characterize social media content and information sources using theory-driven health behavior and psychology constructs to better understand the motifs of misinformation and their role in the dissemination of health (mis)information in Twitter posts. We analyzed 1,400 randomly selected tweets related to COVID-19 to ascertain four important variables, what is the tweet about (content), how is it structured (linguistic features), who is tweeting (source), and what is the reach of the tweet (dissemination). Results showed there was a significant difference between themes expressed, health beliefs manifested, and observed linguistic patterns in true and false information. Implications for informatics-driven digital health utilities, such as theory-informed knowledge models and context-aware risk communications, are discussed.

Latent Linguistic Motifs in Social Media Postings Resisting COVID-19 Misinformation.

Social media has become a predominant source of information for many health care consumers. However, false and misleading information is a pervasive problem in this context. Specifically, health-related misinformation has become a significant public health challenge, impeding the effectiveness of public health awareness campaigns and resulting in suboptimal responsiveness to the communication of legitimate risk-related information. Little is known about the mechanisms driving the seeding and spreading of such information. In this paper, we specifically examine COVID-19 tweets which attempt to correct misinformation. We employ a mixed-methods approach comprising qualitative coding, deep learning classification, and computerized text analysis to understand the manifestation of speech acts and other linguistic variables. Results indicate significant differences in linguistic variables (e.g., positive emotion, tone, authenticity) of corrective tweets and their dissemination level. Our deep learning classifier has a macro average performance of 0.82. Implications for effective and persuasive misinformation correction efforts are discussed.

Mobile Health Applications for Postpartum Depression Management: A Theory-Informed Analysis of Change-Use-Engagement (CUE) Criteria in the Digital Environment.

Postpartum Depression (PPD) is the most common childbirth complication, with approximately 15% of postpartum women experiencing depression symptoms. Mobile applications have potential to expand delivery of mental health interventions. However, our understanding of how these tools engage women with PPD and facilitate positive behavioral changes is limited. In our paper, we analyze 15 commercial PPD applications to understand their role as facilitators of change, engagement, and sustained use. Applications reviewed contained an average of four theory-based behavioral change techniques, and highest patient engagement level reached was to empower patients through patient-generated data. Heuristic violations were identified in areas including user control and freedom, aesthetic and minimalist design, and help and documentation. An inverse correlation was found between the number of theory-based behavior change features and patient engagement. Findings suggest underserved populations may suffer further limitations accessing relevant health resources in the current application market.

Improving Pharmacovigilance Signal Detection from Clinical Notes with Locality Sensitive Neural Concept Embeddings.

Although pharmaceutical products undergo clinical trials to profile efficacy and safety, some adverse drug reactions (ADRs) are only discovered after release to market. Post-market drug safety surveillance - pharmacovigilance - leverages information from various sources to proactively identify such ADRs. Clinical notes are one source of observational data that could assist this process, but their inherent complexity can obfuscate possible ADR signals. In previous research, embeddings trained on observational reports have improved detection of such signals over commonly used statistical measures. Moreover, neural embedding methods which further encode juxtapositional information have shown promise on analogical retrieval tasks, suggesting proximity-based alternatives to document-level modeling for signal detection. This work uses natural language processing and locality sensitive neural embeddings to increase ADR signal recovery from clinical notes, with AUCs of ~0.63-0.71. Constituting a ~50% increase over baselines, our method sets the state-of-the-art for these reference standards when solely leveraging clinical notes.

Parent Engagement with a Self-Tailored Cancer Prevention Digital Behavior Change Intervention: Exploratory Application of Affiliation Network Analysis.

Evaluating digital behavior change intervention engagement is complex and requires multidimensional and novel approaches that are emerging. The relationship and interdependence between engagement with the technology and engagement with the psychosocial or behavior change process often presents conceptual and evaluative challenges. Large objective data sets detailing technology use are plentiful but meaningful interpretation can be challenging at granular levels. Affiliation network analysis which describes two-mode network data may provide a novel approach to evaluate engagement of digital behavior change interventions. The purpose of this paper is to use affiliation network analysis as an exploratory method to describe, assess and visualize content-specific patterns underlying psychosocial characteristics related to HPV vaccine safety concerns of parents using the HPVcancerFree intervention. Results indicate that affiliation network analysis shows promise in supplementing existing methods to assess engagement of digital interventions.

Recovery at the edge of error: debunking the myth of the infallible expert.

The notion that human error should not be tolerated is prevalent in both the public and personal perception of the performance of clinicians. However, researchers in other safety-critical domains have long since abandoned the quest for zero defects as an impractical goal, choosing to focus instead on the development of strategies to enhance the ability to recover from error. This paper presents a cognitive framework for the study of error recovery, and the results of our empirical research into error detection and recovery in the critical care domain, using both laboratory-based and naturalistic approaches. Both attending physicians and residents were prone to commit, detect and recover from errors, but the nature of these errors was different. Experts corrected the errors as soon as they detected them and were better able to detect errors requiring integration of multiple elements in the case. Residents were more cautious in making decisions showing a slower error recovery pattern, and the detected errors were more procedural in nature with specific patient outcomes. Error detection and correction are shown to be dependent on expertise, and on the nature of the everyday tasks of the clinicians concerned. Understanding the limits and failures of human decision-making is important if we are to build robust decision-support systems to manage the boundaries of risk of error in decision-making. Detection and correction of potential error is an integral part of cognitive work in the complex, critical care workplace.

Transitions Between Electronic and Combustible Cigarettes: A Mixed Methods Analysis of Peer Interactions in an Online Community for Tobacco Cessation.

The growing popularity of e-cigarettes is a public health concern. There is an emerging need to understand the pathways between electronic and combustible modes due to the specialized nature of risks associated with each transition. Online social media has become the most dominant knowledge space for these evolving behaviors, and as such, can provide unique opportunities for modeling switching patterns. In this paper, we describe the utility of online peer interactions using qualitative inquiry and network visualizations using 500 messages to characterize (a) transition pathways and (b) psychosocial attributes as individuals contemplate and act on such transitions. Our results indicate that the E2A pathway is the most prevalent in e-cigarette-related transitions, where most of the individuals are in the "active e-cig use" stage. Perceived benefits and barriers are the most commonly held health beliefs, while counterconditioning and stimulus control behavior change processes are frequently manifested. Such insights can help in the design of personalized pathway-specific behavior change interventions.

Analysis of Online Peripartum Depression Communities: Application of Multilabel Text Classification Techniques to Inform Digitally-Mediated Prevention and Management.

Peripartum depression (PPD) is a significant public health problem, yet many women who experience PPD do not receive adequate treatment. In many cases, this is due to social stigmas surrounding PPD that prevent women from disclosing their symptoms to their providers. Examples of these are fear of being labeled a "bad mother," or having misinformed expectations regarding motherhood. Online forums dedicated to PPD can provide a practical setting where women can better manage their mental health in the peripartum period. Data from such forums can be systematically analyzed to understand the technology and information needs of women experiencing PPD. However, deeper insights are needed on how best to translate information derived from online forum data into digital health features. In this study, we aim to adapt a digital health development framework, Digilego, toward translation of our results from social media analysis to inform digital features of a mobile intervention that promotes PPD prevention and self-management. The first step in our adaption was to conduct a user need analysis through semi-automated analysis of peer interactions in two highly popular PPD online forums: What to Expect and BabyCenter. This included the development of a machine learning pipeline that allowed us to automatically classify user post content according to major communication themes that manifested in the forums. This was followed by mapping the results of our user needs analysis to existing behavior change and engagement optimization models. Our analysis has revealed major themes being discussed by users of these online forums- family and friends, medications, symptom disclosure, breastfeeding, and social support in the peripartum period. Our results indicate that Random Forest was the best performing model in automatic text classification of user posts, when compared to Support Vector Machine, and Logistic Regression models. Computerized text analysis revealed that posts had an average length of 94 words, and had a balance between positive and negative emotions. Our Digilego-powered theory mapping also indicated that digital platforms dedicated to PPD prevention and management should contain features ranging from educational content on practical aspects of the peripartum period to inclusion of collaborative care processes that support shared decision making, as well as forum moderation strategies to address issues with cyberbullying.