Bio, psycho, or social: supervised machine learning to classify discursive framing of depression in online health communities

Supervised machine learning on textual data has successful industrial/business applications, but it is an open question whether it can be utilized in social knowledge building outside the scope of hermeneutically more trivial cases. Combining sociology and data science raises several methodological and epistemological questions. In our study the discursive framing of depression is explored in online health communities. Three discursive frameworks are introduced: the bio-medical, psychological, and social framings of depression. ~80 000 posts were collected, and a sample of them was manually classified. Conventional bag-of-words models, Gradient Boosting Machine, word-embedding-based models and a state-of-the-art Transformer-based model with transfer learning, called DistilBERT were applied to expand this classification on the whole database. According to our experience ‘discursive framing’ proves to be a complex and hermeneutically difficult concept, which affects the degree of both inter-annotator agreement and predictive performance. Our finding confirms that the level of inter-annotator disagreement provides a good estimate for the objective difficulty of the classification. By identifying the most important terms, we also interpreted the classification algorithms, which is of great importance in social sciences. We are convinced that machine learning techniques can extend the horizon of qualitative text analysis. Our paper supports a smooth fit of the new techniques into the traditional toolbox of social sciences.

Social Media Analytics for Pharmacovigilance of Antiepileptic Drugs

Epilepsy is a common neurological disorder worldwide and antiepileptic drug (AED) therapy is the cornerstone of its treatment. It has a laudable aim of achieving seizure freedom with minimal, if any, adverse drug reactions (ADRs). Too often, AED treatment is a long-lasting journey, in which ADRs have a crucial role in its administration. Therefore, from a pharmacovigilance perspective, detecting the ADRs of AEDs is a task of utmost importance. Typically, this task is accomplished by analyzing relevant data from spontaneous reporting systems. Despite their wide adoption for pharmacovigilance activities, the passiveness and high underreporting ratio associated with spontaneous reporting systems have encouraged the consideration of other data sources such as electronic health databases and pharmaceutical databases. Social media is the most recent alternative data source with many promising potentials to overcome the shortcomings of traditional data sources. Although in the literature some attempts have investigated the validity and utility of social media for ADR detection of different groups of drugs, none of them was dedicated to the ADRs of AEDs. Hence, this paper presents a novel investigation of the validity and utility of social media as an alternative data source for the detection of AED ADRs. To this end, a dataset of consumer reviews from two online health communities has been collected. The dataset is preprocessed; the unigram, bigram, and trigram are generated; and the ADRs of each AED are extracted with the aid of consumer health vocabulary and ADR lexicon. Three widely used measures, namely, proportional reporting ratio, reporting odds ratio, and information component, are used to measure the association between each ADR and AED. The resulting list of signaled ADRs for each AED is validated against a widely used ADR database, called Side Effect Resource, in terms of the precision of ADR detection. The validation results indicate the validity of online health community data for the detection of AED ADRs. Furthermore, the lists of signaled AED ADRs are analyzed to answer questions related to the common ADRs of AEDs and the similarities between AEDs in terms of their signaled ADRs. The consistency of the drawn answers with the existing pharmaceutical knowledge suggests the utility of the data from online health communities for AED-related knowledge discovery tasks.

Examining Online Health Community Users’ Information Adoption Intention

Due to the doubt on information credibility, users often hesitate to adopt the health information posted on online health communities (OHC). This may undermine the sustainable development of OHC. The purpose of this research is to identify the determinants of OHC users’ information adoption intention. The results indicated that both information factors and social interaction affect the adoption intention. Information factors include argument quality and source credibility, whereas social interaction includes perceived similarity and familiarity. The results imply that OHC need to ensure information quality and support users’ interaction in order to facilitate their information adoption.

Development and Evaluation of the Canteen Connect Online Health Community: Using a Participatory Design Approach in Meeting the Needs of Young People Impacted by Cancer

Adolescent and young adults (AYAs) impacted by their own or familial cancer require information and peer support throughout the cancer journey to ameliorate feelings of isolation. Online Health Communities (OHC) provide social networks, support, and health-related content to people united by a shared health experience. Using a participatory design (PD) process, Canteen developed Canteen Connect (CC), an OHC for AYAs impacted by cancer. This manuscript outlines the process used to develop CC: (1) A mixed-methods implementation evaluation of Version I of CC (CCv.1); (2) Qualitative workshops utilizing strengths-based approaches of PD and appreciative inquiry to inform the development of CC Version 2 (CCv.2); quantitative implementation evaluation to assess the appropriateness, acceptability, and effectiveness of CCv.2. Through several iterations designed and tested in collaboration with AYAs, CCv.2 had improvements in the user experience, such as the ability to send a private message to other users and the site becoming mobile responsive. Results from the evaluation showed CCv.2 was appropriate for connecting with other AYAs. Most AYAs reported satisfaction with CCv.2 and a positive impact on their feelings of sadness, worry, and/or anxiety. CCv.2 fills an important service provision gap in providing an appropriate and acceptable OHC for AYAs impacted by cancer, with initial promising psychological outcomes.

Offline Reputation Matters: Physicians’ Online Reputation Bias in Online Health Communities (Preprint)

BACKGROUND Online health communities (OHCs) have gradually become an important service platform for patients to communicate with physicians and obtain treatment and related medical information. With the rise of OHCs, an increasing number of scholars have begun to study the incentive mechanism of a physician’s online contribution behavior. However, previous studies have ignored the limited effect of online reputation on a physician’s online contribution and the moderating effect of offline reputation. OBJECTIVE We hypothesized that online reputation shows an inverted U-shaped relation with a physician’s contribution and that offline reputation reduces the inverted U-shaped relation, thus weakening the positive impact of reputation on performance and strengthening the negative impact of reputation performance. METHODS Based on incentive theory and prior studies, we proposed three hypotheses. We collected the panel data of 6,648 physicians from Good Physician, one of the largest OHCs in China. An empirical model was built to test our hypotheses. Additionally, the variables representing online reputation were replaced and the data set was split for the robustness check. RESULTS First, our results support the hypothesis of an inverted U-shaped relationship between a physician’s online reputation and their online contribution behavior(β=-0.127, t=13.624, P<.001). Second, we identify the bipolar impact of online and offline reputation on a physician’s online contribution. Specifically, over a period of time, online reputation will promote the contribution of physicians(slope=1.514, P<.001), while offline reputation will inhibit the contribution. Third, our research shows that offline reputation can weaken the impact of online reputation on a physician’s online contribution. CONCLUSIONS We amended the bias of the positive linear correlation between online reputations and online contributions in OHCs. Moreover, by identifying the moderating effect of offline reputation, we ameliorated the interpreting mechanism of online reputation on a physician’s online contribution. The findings improve our understanding of physicians' online contribution behaviors and contribute to the literature on OHCs. Our research can provide guidance for an OHC to design effective personalized incentive mechanisms and encourage physicians to provide more high-quality services.

Come for Information, Stay for Support: Harnessing the Power of Online Health Communities for Social Connectedness during the COVID-19 Pandemic

The COVID-19 pandemic created a globally shared stressor that saw a rise in the emphasis on mental and emotional wellbeing. However, historically, these topics were not openly discussed, leaving those struggling without professional support. One powerful tool to bridge the gap and facilitate connectedness during times of isolation is online health communities (OHCs). This study surveyed Health Union OHC members during the pandemic to determine the degree of COVID-19 concern, social isolation, and mental health distress they are facing, as well as to assess where they are receiving information about COVID-19 and what sources of support they desire. The survey was completed in six independent waves between March 2020 and April 2021, and garnered 10,177 total responses. In the United States, OHCs were utilized significantly more during peak lockdown times, and the desire for emotional and/or mental health support increased over time. Open-ended responses demonstrated a strong desire for connection and validation, which are quintessential characteristics of OHCs. Through active moderation utilizing trained moderators, OHCs can provide a powerful, intermediate and safe space where conversations about mental and emotional wellbeing can be normalized and those in need are encouraged to seek additional assistance from healthcare professionals if warranted.

Understanding online review behaviors of patients in online health communities: an expectation-disconfirmation perspective

PurposeAlthough online health communities (OHCs) and online patient reviews can help to eliminate health information asymmetry and improve patients' health management, how patients write online reviews within OHCs is poorly understood. Thus, it is very necessary to determine the factors influencing patients' online review behavior in OHCs, including the emotional response and reviewing effort.Design/methodology/approachBased on expectation-disconfirmation theory, this study proposes a theoretical model to analyze the effects of service quality perception (i.e. outcome quality and process quality perceptions) and disconfirmation (i.e. outcome quality and process quality disconfirmations) on patients' emotional response and reviewing effort. The authors test the research model by using empirical data collected from a popular Chinese OHC and applying ordinary least squares (OLS) regression and zero-truncated negative binomial (ZTNB) regression models.FindingsBoth service quality perception and disconfirmation have a positive effect on patients' positive emotional intensity in textual reviews, and disease severity enhances these relationships of process quality. Moreover, there is an asymmetric U-shaped relationship among service quality perception, disconfirmation and reviewing effort. Patients who perceive low service quality have higher reviewing effort, while service quality disconfirmation has the opposite relationship. Specifically, patients' effort in writing textual reviews is lowest when perceived outcome quality is 3.5 (on a five-point scale), perceived process quality is 4 or outcome quality and process quality disconfirmations are −1.Originality/valueThis study is the first to examine patients' online review behavior and its motivations and contributes to the literature on online reviews and service quality. In addition, the findings of this study have important management implications for service providers and OHC managers.

What patients with eosinophilic esophagitis may not share with their providers: a qualitative assessment of online health communities

Summary Background Little is known about how patients with eosinophilic esophagitis (EoE) experience their symptoms, receive care, and cope with their disease. Patients commonly seek peer support from online communities, which provide insights on unmet needs and barriers to care. We performed a qualitative analysis of electronic health forums to characterize patient-to-patient conversations about EoE symptoms and the experience of disease. Methods We identified three publicly accessible electronic health forums hosting EoE communities. Conversation threads posted between July 2018 and June 2020 were coded using emergent and a priori codes based on the THRIVE conceptual framework of coping with chronic illness. Results Of 659 threads (4,933 posts) collected over two years, a random sample of 240 threads (30 per 3-month quarter) were selected for analysis. Thematic saturation was reached after 172 threads. Patient experience of EoE was driven by their perspectives in four key domains: (i) perception of EoE as episodic rather than chronic, (ii) treatment choices, (iii) personal definitions of success in the disease, and (iv) views of providers. Conclusion Online health communities are a valuable and unfiltered source of patient perspectives that can be used to understand patient needs and goals. EoE patients interpret their disease as sporadic events and lack reliable sources of knowledge, which may influence how patients prioritize treatment. If providers are to succeed in providing high-quality EoE care, they need to equip themselves with evidence-based knowledge, engage in shared decision making, and look outside of clinical settings to recognize barriers to disease management.