<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
	<id>https://www.jmir.org/issue/feed</id>
	<title>Journal of Medical Internet Research</title>
			<updated>2025-01-01T11:30:03-05:00</updated>
	
		<author>
		<name>JMIR Publications</name>
				<email>editor@jmir.org</email>
			</author>
		<link rel="alternate" href="https://www.jmir.org" />
	<link rel="self" type="application/atom+xml" href="https://www.jmir.org/feed/atom" />

	<generator uri="http://pkp.sfu.ca/ojs/" version="2.2.0.0">Open Journal Systems</generator>

				    	<subtitle> The leading peer-reviewed journal for digital medicine and health and health care in the internet age.&amp;nbsp; </subtitle>



	<entry>
		<id> https://www.jmir.org/2026/1/e96844 </id>
		<title>Parents’ Perspectives on the Use and Development of Predictive AI in Neonatology: Focus Group Study</title>
		<updated>2026-11-17T16:30:24-05:00</updated>

					<author>
				<name>Lotte Schuitmaker</name>
			</author>
					<author>
				<name>Sacha Reuvers</name>
			</author>
					<author>
				<name>Sylvia A Obermann-Borst</name>
			</author>
					<author>
				<name>Manon Benders</name>
			</author>
					<author>
				<name>Karin Jongsma</name>
			</author>
					<author>
				<name>Jojanneke Drogt</name>
			</author>
				<link rel="alternate" href="https://www.jmir.org/2026/1/e96844" />
					<summary type="html" xml:base="https://www.jmir.org/2026/1/e96844">Background: Predictive AI models, such as AI-generated early warning systems for detecting critical conditions, are being praised as promising applications in neonatology. Despite technological advancements in predictive AI, the integration of AI at the bedside remains challenging, and few AI technologies are currently integrated into routine neonatal care. A better understanding of the wishes and barriers of patients and, in the case of neonatal patients, their parents, is essential to bridge this gap. Yet, parental perspectives on the use of AI remain underexplored. Objective: This study explored the perspectives of parents whose child had previously been admitted to a neonatal intensive care unit about the development and use of predictive AI in neonatal clinical settings. Methods: We conducted 3 focus groups with 18 Dutch parents of children admitted to a neonatal intensive care unit to explore their views on AI in neonatal care. Participants were purposively recruited via a neonatology parent and patient organization. Discussions focused on predictive AI in acute settings and on AI to predict neurodevelopmental outcomes in later life. A reflexive thematic analysis of the focus group data was performed. Results: Parents generally welcomed predictive AI as a supportive tool for the medical team, anticipating improvements in patient outcomes. AI for short-term predictions was especially appreciated, whereas longer-term predictions were received with some reservations. Overarching concerns regarding predictive AI systems included AI undermining parental intuition, overreliance on AI in decision-making, and undesirable ethical and psychological implications of long-term health predictions. Interestingly, in addition to predictive AI systems that support medical judgment, parents also hoped that other AI tools could be developed to better involve parental input in the decision-making process. Specifically, they envisioned applications that extend beyond clinical decision support to provide parents with information, emotional support, and tailored guidance on the right course of action. Conclusions: Parental perspectives highlight both opportunities and challenges for predictive AI integration in neonatal care. Engaging parents in the design and implementation of AI informs the development of family-centered, ethically responsible AI solutions and may identify opportunities for developing parent-centered AI tools alongside systems designed to support clinicians.</summary>
		
        
                	<content type="image/png" src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/f3a7fa33b23fe9478344b6792a5d1353" />
		
		<published>2026-11-17T16:30:24-05:00</published>
	</entry>
	<entry>
		<id> https://www.jmir.org/2026/1/e99837 </id>
		<title>Themes, Temporal Dynamics, and Public Engagement in COVID-19 Vaccination Disclosures on Twitter: Computational Content Analysis</title>
		<updated>2026-11-17T11:45:21-05:00</updated>

					<author>
				<name>Emily McKinley</name>
			</author>
					<author>
				<name>Cuihua Shen</name>
			</author>
					<author>
				<name>Kenji Sagae</name>
			</author>
					<author>
				<name>Jingwen Zhang</name>
			</author>
				<link rel="alternate" href="https://www.jmir.org/2026/1/e99837" />
					<summary type="html" xml:base="https://www.jmir.org/2026/1/e99837">Background: Social media has become a primary infrastructure for vaccine communication. Personal vaccination disclosures, or first-person announcements of vaccination decisions shared publicly online, represent an underexamined channel through which peer influence may shape vaccine attitudes and behavior. Understanding what individuals disclose when they announce vaccination, and how those disclosures evolve over time, offers actionable insights for designing peer-based communication strategies in future health emergencies. Objective: This study aimed to identify the themes present in COVID-19 vaccination disclosures on Twitter (rebranded as X in 2023), examine how those themes changed across 4 phases of the US vaccine rollout, and determine which themes were associated with greater public engagement. Methods: Guided by the disclosure decision model (DDM), we analyzed 207,647 vaccination disclosure posts on Twitter from December 14, 2020, to January 1, 2022. Posts were identified using a robustly optimized BERT pretraining approach (RoBERTa) binary classifier (-score=0.85) trained on 6348 human-coded posts. A mixed deductive-inductive coding approach, grounded in the DDM’s motivational goals, yielded 11 disclosure themes (Krippendorff =0.72–0.93). Eleven additional RoBERTa classifiers were trained to label themes across the full dataset. Network analysis examined co-occurrence of themes within each rollout phase. Negative binomial regression assessed the association between disclosure themes and an engagement metric (sum of likes, reposts, quotes, and replies), controlling for follower count, media presence, word count, sentiment, and other post-level factors. Results: Eleven distinct disclosure themes were identified, including community identity (n=31,847, 15.34%), encouraging others (n=29,619, 14.26%), and gratitude (n=29,511, 14.21%). Themes shifted meaningfully across rollout phases: gratitude dominated phase 1, community identity rose in phases 2 and 4, and encouraging vaccination became prominent in phase 3. Gratitude was associated with a 25% increase in engagement (incidence rate ratio [IRR]=1.25; 95% CI 1.23-1.28; &lt;.001), alongside relief (IRR=1.26; 95% CI 1.19-1.33; &lt;.001) and community identity (IRR=1.19; 95% CI 1.17-1.21; &lt;.001). Side effects (IRR=0.85; 95% CI 0.83-0.87; &lt;.001), political (IRR=0.90; 95% CI 0.87-0.94; &lt;.001), and social benefits disclosures (IRR=0.92; 95% CI 0.90-0.95; &lt;.001) were associated with decreased engagement. Each additional theme in a post was associated with a 7% increase in engagement (IRR=1.07; 95% CI 1.06-1.08; &lt;.001). Conclusions: Vaccination disclosures on Twitter were thematically patterned, theoretically interpretable via the DDM, and responsive to changing public health contexts. Identity-based and emotionally resonant disclosures, particularly gratitude and community identity, were among the themes most strongly associated with public engagement. These findings provide an empirically grounded framework for designing peer-based vaccine communication strategies that leverage voluntary self-disclosure in future immunization campaigns.</summary>
		
        
                	<content type="image/png" src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/65dc2fde72246c940146e26dfc43bae8" />
		
		<published>2026-11-17T11:45:21-05:00</published>
	</entry>
	<entry>
		<id> https://www.jmir.org/2026/1/e91156 </id>
		<title>A Personalized Exercise Assistant Using Reinforcement Learning: Results From a Large-Scale Mobile Health Randomized-Controlled Trial</title>
		<updated>2026-11-17T11:30:24-05:00</updated>

					<author>
				<name>Amy Armento Lee</name>
			</author>
					<author>
				<name>Narayan Hegde</name>
			</author>
					<author>
				<name>Nina Deliu</name>
			</author>
					<author>
				<name>Emily Rosenzweig</name>
			</author>
					<author>
				<name>Arun Suggala</name>
			</author>
					<author>
				<name>Sriram Lakshminarasimhan</name>
			</author>
					<author>
				<name>Qian He</name>
			</author>
					<author>
				<name>John Hernandez</name>
			</author>
					<author>
				<name>Martin Seneviratne</name>
			</author>
					<author>
				<name>Rahul Singh</name>
			</author>
					<author>
				<name>Pradnesh Kalkar</name>
			</author>
					<author>
				<name>Karthikeyan Shanmugam</name>
			</author>
					<author>
				<name>Aravindan Raghuveer</name>
			</author>
					<author>
				<name>Abhimanyu Singh</name>
			</author>
					<author>
				<name>Hariharan Manoharan</name>
			</author>
					<author>
				<name>My Nguyen</name>
			</author>
					<author>
				<name>James Taylor</name>
			</author>
					<author>
				<name>Jatin Alla</name>
			</author>
					<author>
				<name>Sofia S Villar</name>
			</author>
					<author>
				<name>Hulya Emir-Farinas</name>
			</author>
				<link rel="alternate" href="https://www.jmir.org/2026/1/e91156" />
					<summary type="html" xml:base="https://www.jmir.org/2026/1/e91156">Background: Consistent physical inactivity among adults and adolescents poses a major global health challenge. Mobile health (mHealth) interventions, particularly just-in-time adaptive interventions (JITAIs), offer a promising avenue for scalable and personalized physical activity promotion. However, developing and evaluating such adaptive interventions at scale, while integrating robust behavioral science, presents methodological hurdles. Objective: The Personalized Exercise Assistant Using Reinforcement Learning (PEARL) study aimed to assess the feasibility and effectiveness of a reinforcement learning (RL) algorithm, informed by health behavior change theory (capability, opportunity, and motivation–behavior [COM-B]), to personalize the content and timing of physical activity nudges via the Fitbit app (Google LLC) compared to fixed and random nudging strategies, and to a control group with no nudges. Methods: We conducted a large-scale, 4-arm randomized controlled trial (RCT) enrolling 13,463 Fitbit users. Participants were randomized to: (1) control (no nudges), (2) random (random content/timing), (3) fixed (logic based on baseline COM-B survey), and (4) RL (adaptive algorithm). The primary outcome was the change in average daily step count from baseline to 2 months. Secondary outcomes included user engagement and survey responses regarding capability, opportunity, and motivation. Results: A total of 7711 participants were included in the primary analysis (mean age 42.1, SD 9.0 years; 6657/7711, 86.3% female). At 1 month, the RL group showed a significant increase in daily steps compared to control (+296 steps; &lt;.001), random (+218 steps; =.005), and fixed (+238 steps; =.002) groups. For the prespecified primary analyses at 2 months, the RL group sustained a significant increase against the control (+210 steps; =.01), but pairwise differences comparing the RL group against the random and fixed groups were not statistically significant. Generalized estimating equation (GEE) models confirmed a sustained significant increase in the RL group (+208 steps; =.002). In exit surveys, the RL group reported higher favorable responses regarding nudge customization (224 of 605, 37%) compared to other groups. Conclusions: This study demonstrates the feasibility and early efficacy of using RL to personalize digital health nudges at scale. To our knowledge, this represents one of the largest, nonincentivized randomized trials of this duration to evaluate RL for physical activity. While the magnitude of the observed step increase is modest, a sustained daily improvement of 200‐300 steps can contribute to broader public health goals by reducing sedentary behavior in large populations. The adaptive approach outperformed static behavioral rules, showcasing the promise of dynamic personalization in a real-world mHealth setting. Notably, participants receiving nudges based on fixed, theory-informed rules did not show a significant advantage over the random group, underscoring the challenge of using static baseline assessments to address nonstationary, day-to-day behavioral contexts. However, several limitations warrant consideration: the findings are specific to a 2-month follow-up, the study population was predominantly female (6657/7711, 86.3%) self-selected Fitbit users, and the observed step gains were modest in magnitude. Trial Registration: ClinicalTrials.gov NCT07783724 https://clinicaltrials.gov/study/NCT07783724 (Retrospectively Registered)</summary>
		
        
                	<content type="image/png" src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/68518c01c30949897453175ab8a0e874" />
		
		<published>2026-11-17T11:30:24-05:00</published>
	</entry>
	<entry>
		<id> https://www.jmir.org/2026/1/e91943 </id>
		<title>Behavior Change Techniques Associated With Abstinence in a Chat-Based Smoking Cessation Intervention: Secondary Analysis of a Randomized Clinical Trial Using Machine Learning</title>
		<updated>2026-11-17T11:30:24-05:00</updated>

					<author>
				<name>Yajie Li</name>
			</author>
					<author>
				<name>Xiaoyu Su</name>
			</author>
					<author>
				<name>Yee Tak Derek Cheung</name>
			</author>
					<author>
				<name>Shengzhi Zhao</name>
			</author>
					<author>
				<name>Tai Hing Lam</name>
			</author>
					<author>
				<name>Man Ping Wang</name>
			</author>
					<author>
				<name>Tzu Tsun Luk</name>
			</author>
				<link rel="alternate" href="https://www.jmir.org/2026/1/e91943" />
					<summary type="html" xml:base="https://www.jmir.org/2026/1/e91943">Background: Isolating the candidate intervention components that drive the success of digital behavioral interventions is crucial for understanding intervention mechanisms and optimizing future interventions. Objective: This study aimed to identify behavior change techniques (BCTs) delivered in a chat-based smoking cessation intervention and determine their independent associations with biochemically validated abstinence. Methods: This study is a secondary analysis of a pragmatic, cluster-randomized clinical trial conducted in Hong Kong. From June 18 to September 30, 2017, 624 adult daily smokers were recruited from 34 community sites and randomized to the intervention group. Participants received 3 months of personalized, real-time chat-based cessation support delivered by a live counselor via a mobile instant messaging app. The analytic sample included 230 participants (n=170, 73.9% male; mean age 43.0, SD 13.6 years) who engaged with the chat support. All chat dialogues during the real-time support were retrieved and systematically coded using an established BCT taxonomy for smoking cessation (BCTTsm). Four machine learning methods, least absolute shrinkage and selection operator regression, Extreme Gradient Boosting, the Boruta algorithm, and recursive feature elimination, were used to identify the BCTs that are most predictive of biochemically validated tobacco abstinence outcomes at 3 and 6 months. A prespecified supermajority rule retained only BCTs selected by at least 3 of the 4 models for subsequent analysis. The stability of the selected BCTs was assessed using 500 bootstrap resamples for each model. Poisson regression with robust variance was subsequently used to examine the independent associations between the selected BCTs and abstinence outcomes. Covariates included sex, age, educational level, nicotine dependence, intention to quit in 30 days, and quit attempts at baseline. Results: Systematic coding identified 35 distinct BCTs from the BCTTsm as having been delivered. Three BCTs were consistently selected by machine learning as the most predictive of both 3- and 6-month abstinence, with moderate to high stability (ranging from 61% to 100%): “provide rewards contingent on successfully stopping smoking” (BM4), “provide information on withdrawal symptoms” (RC6), and “provide information on consequences of smoking and smoking cessation” (BM1). In multivariable analyses, 6-month abstinence was significantly associated with BM4 (risk ratio [RR] 7.9, 95% CI 4.2-14.6), RC6 (RR 3.3, 95% CI 1.6-6.5), and BM1 (RR 2.7, 95% CI 1.4-5.5), with consistent results for 3-month abstinence. Conclusions: Three BCTs, contingent rewards (BM4), information on withdrawal symptoms (RC6), and information on consequences (BM1), were identified as candidate intervention components associated with success in a chat-based smoking cessation intervention. This study is the first to apply machine learning for BCT identification, moving beyond simple frequency counts to data-driven feature selection. These findings provide empirical evidence for the candidate components of effective digital behavioral intervention that may warrant further testing and offer guidance for future intervention optimization. Trial Registration: ClinicalTrials.gov NCT03182790; https://clinicaltrials.gov/study/NCT03182790</summary>
		
        
                	<content type="image/png" src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/907ad094e3fd073db68b03447edc9062" />
		
		<published>2026-11-17T11:30:24-05:00</published>
	</entry>
	<entry>
		<id> https://www.jmir.org/2026/1/e94936 </id>
		<title>AI Predictive Models for Discharge Disposition in Patients Following Primary Total Knee Arthroplasty: Systematic Review and Meta-Analysis</title>
		<updated>2026-11-16T17:00:31-05:00</updated>

					<author>
				<name>Dongdong Xie</name>
			</author>
					<author>
				<name>Rui Chen</name>
			</author>
					<author>
				<name>Chen Yue</name>
			</author>
					<author>
				<name>Xiaoya Liu</name>
			</author>
					<author>
				<name>Guorui Cao</name>
			</author>
					<author>
				<name>Ruina Jiao</name>
			</author>
				<link rel="alternate" href="https://www.jmir.org/2026/1/e94936" />
					<summary type="html" xml:base="https://www.jmir.org/2026/1/e94936">Background: Although a growing number of AI-based predictive models for discharge disposition in patients following primary total knee arthroplasty (TKA) have been constructed, their performance, quality, and applicability remain unclear. Objective: This study aimed to evaluate the performance, quality, and applicability of published AI-based predictive models for discharge disposition in patients following primary TKA. Methods: PubMed, Web of Science, Embase, the Cochrane Library, Scopus, China National Knowledge Infrastructure, Wanfang Database, China Science and Technology Journal Database, and SinoMed Database were searched from inception to May 26, 2025. We included original studies that developed AI-based predictive models for discharge disposition in adult patients (age ≥18 years) undergoing primary TKA. Studies involving simultaneous bilateral or revision TKA or those that did not develop a predictive model were excluded. Risk of bias and reporting quality were assessed using the PROBAST+AI (Prediction Model Risk of Bias Assessment Tool+AI) and TRIPOD+AI (Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis+AI), respectively. Pooled area under the curve values and 95% CIs were estimated using a DerSimonian-Laird random-effects model with the Hartung-Knapp-Sidik-Jonkman adjustment. Heterogeneity was assessed using the Cochran test and ² statistic. Subgroup analyses were conducted according to algorithm type, categorized as machine learning (ML)–based and deep learning (DL)–based models. Results: A total of 5 retrospective studies from the United States involving 4 ML-based models and 1 DL-based model were included, with sample sizes ranging from 175,042 to 434,550 patients. Age and sex were the most frequently used predictors. All included studies were rated as having a high risk of bias according to the PROBAST+AI checklist. The pooled area under the curve was 0.75 (95% CI 0.67‐0.82; ²=99.9%; &lt;.001) for internal validation and 0.77 (95% CI 0.51‐1.00; ²=99.8%; &lt;.001) for external validation, indicating a high degree of heterogeneity. Subgroup analysis showed no statistically significant difference between ML-based and DL-based models (&gt;.05). Conclusions: AI-based predictive models for discharge disposition following primary TKA demonstrate a certain level of discriminative performance. However, the current evidence is limited, as all included studies were at high risk of bias and showed a high degree of heterogeneity. In addition, all included studies were conducted in the United States, which may limit the generalizability of these findings to other health care settings. Future studies should follow PROBAST+AI and TRIPOD+AI checklists while developing and validating models across different countries and health care systems. Trial Registration: PROSPERO CRD420251059218; https://www.crd.york.ac.uk/PROSPERO/view/CRD420251059218</summary>
		
        
                	<content type="image/png" src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/fdb35265deb769558a6eababc77a1ed4" />
		
		<published>2026-11-16T17:00:31-05:00</published>
	</entry>
	<entry>
		<id> https://www.jmir.org/2026/1/e105737 </id>
		<title>Cross-Domain Transfer of Depression Voice Biomarkers Depends on the Outcome Instrument: Leakage-Controlled Cross-Sectional Evaluation Study</title>
		<updated>2026-11-16T17:00:31-05:00</updated>

					<author>
				<name>Rachel Wiley</name>
			</author>
					<author>
				<name>James Schwoebel</name>
			</author>
					<author>
				<name>Matias Caccia</name>
			</author>
					<author>
				<name>Joel Shor</name>
			</author>
					<author>
				<name>Adolfo M Garcia</name>
			</author>
					<author>
				<name>Sheehan D Fisher</name>
			</author>
					<author>
				<name>Martin Gerbert Frasch</name>
			</author>
				<link rel="alternate" href="https://www.jmir.org/2026/1/e105737" />
					<summary type="html" xml:base="https://www.jmir.org/2026/1/e105737">Background: Voice-based depression screening is commercializing rapidly, yet whether voice biomarkers generalize across clinical settings is largely untested. Generalization is usually framed as a question about populations, but it is also a question about the outcome instrument a model is scored against, an effect confounded with other between-study differences in the literature. Objective: We tested whether depression voice biomarkers transfer between perinatal and general-psychiatric speech in both directions under strict leakage control and whether transfer depends on the screening instrument. Methods: In a cross-sectional, online US study, 390 pregnant participants at a mean of 22 (SD 1) weeks’ gestation (analytical sample n=316) participated in 446 sessions; each participant provided 4 voice recordings and completed the modified 8-item Patient Health Questionnaire (PHQ-8) and modified 9-item Edinburgh Postnatal Depression Scale (mEPDS-9). We extracted 375 acoustic descriptors and 330-dimensional Population Dynamics Foundation Model community-health embeddings. Discrimination area under the curve (AUC) was assessed under leakage-controlled cross-validation, with feature and classifier selection inside training folds only and a permutation negative-control harness, across a prespecified grid of 4 speech tasks×4 outcome definitions. The prenatal-trained model was applied to 2 general-psychiatric corpora (Distress Analysis Interview Corpus–Wizard-of-Oz, n=189; Extended Distress Analysis Interview Corpus, n=219) under pure held-out inference, and the open-source Kintsugi Depression/Anxiety Model (DAM), trained on 35,000 individuals, was applied to all 3 cohorts without refitting, with Benjamini-Hochberg multiplicity control. Results: In the prenatal cohort, the preregistered outcome was at chance (AUC 0.494, 95% CI 0.431-0.560); no cell in the 16-cell grid survived adjustment under either modeling paradigm. The prenatal-trained model did not transfer to either general-psychiatric corpus (AUCs 0.505 and 0.478). Reverse transfer depended on the screening instrument. The DAM reached an AUC of 0.706 to 0.708 on general-psychiatric speech. In the same participants, it performed at chance against the PHQ-8 (AUC 0.510, 95% CI 0.411-0.609) but discriminated against the mEPDS-9 (AUC 0.645, 95% CI 0.541-0.744), a paired difference of 0.135 (95% CI 0.019-0.248; unadjusted post hoc =0.02). Exploratory item-level analysis suggests a mechanism: the DAM tracked affective-distress items (crying =0.26, sadness =0.21, and overwhelmed =0.20) but not somatic PHQ-8 items; only crying survived adjustment across 17 tests. Transfer varied by elicitation task (pregnancy narrative 0.653, 95% CI 0.550-0.753; open-ended speech 0.597, 95% CI 0.486-0.699; scripted reading 0.544, 95% CI 0.434-0.651), though these were not formally compared. Conclusions: Cross-domain comparison showed a compound shift: population, elicitation task, and outcome instrument differed together, while only the instrument could be varied on its own. Whether a depression voice biomarker transfers may depend not only on the population but also on whether the outcome instrument captures what the model can detect. Validation should state the population, elicitation task, and outcome instrument together; a tool validated against one instrument should not be assumed to work against another. Trial Registration: OSF Registries osf.io/k387t; https://osf.io/k387t/overview</summary>
		
        
                	<content type="image/png" src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/212e2e835877e10f246df9d9ffb6782c" />
		
		<published>2026-11-16T17:00:31-05:00</published>
	</entry>
	<entry>
		<id> https://www.jmir.org/2026/1/e103235 </id>
		<title>The Association Between Platform-Specific Social Media Use and Suicidal Ideation in Japan: Cross-Sectional Web-Based Survey Study</title>
		<updated>2026-11-16T16:45:20-05:00</updated>

					<author>
				<name>Yusei Taniguchi</name>
			</author>
					<author>
				<name>Mizuki Hino</name>
			</author>
					<author>
				<name>Yumiko Hamaie</name>
			</author>
					<author>
				<name>Hiroaki Tomita</name>
			</author>
					<author>
				<name>Takahiro Tabuchi</name>
			</author>
					<author>
				<name>Yasuto Kunii</name>
			</author>
				<link rel="alternate" href="https://www.jmir.org/2026/1/e103235" />
					<summary type="html" xml:base="https://www.jmir.org/2026/1/e103235">Background: In Japan, suicides among young women have recently increased; however, this trend’s specific causes remain unclear. Concurrently, rising social media use has sparked concerns regarding its potential impact. However, findings on the association between social media use and mental health have been mixed, and platform-specific differences remain unclear. Objective: This study investigated the cross-sectional association between current platform-specific social media use and retrospectively reported past-year suicidal ideation in Japan, as well as age- and sex-based differences. Methods: This cross-sectional study analyzed 2 linked nationwide internet surveys: the Japan COVID-19 and Society Internet Survey (December 2024-January 2025) and the Japan Society and New Tobacco Internet Survey (February 2025-March 2025). The analysis included 19,542 individuals aged 15 to 84 years who responded to both surveys. Multivariable logistic regression assessed the associations between the use of specific social media platforms (Facebook, X [formerly known as Twitter], Instagram, YouTube, and TikTok) and suicidal ideation, adjusting for covariates. Results: Suicidal ideation was reported by 10.8% (2118/19,542) of the participants. In the adjusted models, suicidal ideation was associated with the use of X (adjusted odds ratio [aOR] 1.44, 95% CI 1.27‐1.63), Facebook (aOR 1.33, 95% CI 1.18‐1.49), and TikTok (aOR 1.33, 95% CI 1.19‐1.50; &lt;.001 in all cases) but not Instagram (aOR 0.93, 95% CI 0.82‐1.06) or YouTube (aOR 1.09, 95% CI 0.94‐1.27). Exploratory stratified analyses (interpreted with caution given multiple comparisons) suggested that the association with X use was consistent across participants younger than 45 years. Conclusions: The association between social media use and suicidal ideation varied across platforms and demographic groups. After adjusting for confounders, only instances of the use of X, Facebook, and TikTok were associated with higher odds of suicidal ideation. Thus, the relationship between social media use and suicidal ideation depends on platform characteristics and user attributes, thereby underscoring the need for platform-specific suicide prevention strategies. These cross-sectional findings do not permit directional inference and should be regarded as hypothesis generating.</summary>
		
        
                	<content type="image/png" src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/3a5e64f876bc9ea4e6cdb969eff4bd08" />
		
		<published>2026-11-16T16:45:20-05:00</published>
	</entry>
	<entry>
		<id> https://www.jmir.org/2026/1/e100446 </id>
		<title>A Social Media Mining and Questionnaire Survey Hybrid Approach to Characterizing Caregiver Knowledge, Attitudes, and Practices on House Dust Mite Control: Cross-Sectional Study</title>
		<updated>2026-11-16T16:45:20-05:00</updated>

					<author>
				<name>Peixin Yang</name>
			</author>
					<author>
				<name>Ruoling Xian</name>
			</author>
					<author>
				<name>Juan Li</name>
			</author>
					<author>
				<name>Jing Zhao</name>
			</author>
					<author>
				<name>Hangjiang Xie</name>
			</author>
					<author>
				<name>Zhenghong Jiang</name>
			</author>
					<author>
				<name>Yan Hu</name>
			</author>
					<author>
				<name>Ximing Xu</name>
			</author>
					<author>
				<name>Na Zang</name>
			</author>
					<author>
				<name>Yan Zhao</name>
			</author>
					<author>
				<name>Heping Fang</name>
			</author>
					<author>
				<name>Enmei Liu</name>
			</author>
				<link rel="alternate" href="https://www.jmir.org/2026/1/e100446" />
					<summary type="html" xml:base="https://www.jmir.org/2026/1/e100446">Background: House dust mite (HDM) sensitization commonly begins in early life and contributes to persistent allergic airway inflammation and asthma chronicity. Primary prevention via early-life environmental control is a potential pathway to reduce HDM sensitization and asthma risk. Objective: This study aims to characterize child caregivers’ knowledge, attitudes, and practices (KAP) regarding pediatric HDM control using a hybrid literature/expert-driven and social media–driven approach and to examine associations between KAP levels, child age, and caregiver HDM-related online engagement. Methods: This cross-sectional study comprised two interconnected components: (1) mining of content published between August 2023 and July 2025 from 5 major Chinese social media platforms, analyzed via latent Dirichlet allocation (LDA), and (2) a social media–enhanced web-based KAP survey administered in November 2025 to child caregivers in Chongqing, China. In total, 132,341 social media documents and 2275 caregivers of children younger than 18 years were included in the analysis. The main outcomes included social media discourse patterns and domain-specific KAP levels across 5 dimensions: foundational knowledge (K1), recommended control knowledge (K2), attitude toward social media topics (A1), attitude toward recommended methods (A2), and control practices (P). Stratified analysis was conducted by 2 exposure variables: child age (≤3 years vs &gt;3 years) and caregiver HDM-related online engagement (engaged vs nonengaged). Results: LDA topic modeling identified 5 distinct topic clusters in the social media content. Commercial, emotional, and misleading content collectively dominated the information landscape, accounting for 83.3% (n=110,254) of the included documents (65.8% of the tokens), with commercial content often systematically conflating the concepts of “disinfection” and “mite elimination.” Only 16.7% (n=22,087) of the documents (34.1% of the tokens) were classified as health educational content focusing on HDM allergy prevention, reflecting that this valid educational content is substantially longer and more information-dense than other social media posts. The average KAP levels of K1, K2, A1, A2, and P domains were 62.9%, 84.7%, 57.0%, 37.8%, and 25.8%, respectively. Social media emerged as the primary knowledge source (80.7%, n=1836), with methodological knowledge gaps (47.5%) being the top implementation barrier. Caregivers of children ≤3 years had significantly lower self-rated knowledge (23.5% vs 28.3%; =.01), stronger endorsement of recommended methods, but also greater information overload (odds ratio [OR] 1.39, 95% CI 1.15‐1.67; &lt;.001) and efficacy uncertainty (OR 1.23, 95% CI 1.01‐1.52; =.04). Caregivers with HDM-related online engagement showed better performance across multiple items in 5 domains, but also increased traditional practice with uncertain evidence (ultraviolet irradiation: OR 1.85, 95% CI 1.35‐2.53; &lt;.001) and misconception acceptance (allergy impact exaggeration: OR 1.39, 95% CI 1.04‐1.87; =.03). Conclusions: Complex and suboptimal KAP levels exist, particularly among caregivers of young children (≤3 years). HDM-related online engagement is associated with both enhanced implementation of control practices and elevated misconception endorsement. These findings reveal critical educational gaps and the necessity of social media intervention.</summary>
		
        
                	<content type="image/png" src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/b651be99acf70d1fd8db5451d1381b27" />
		
		<published>2026-11-16T16:45:20-05:00</published>
	</entry>
	<entry>
		<id> https://www.jmir.org/2026/1/e116560 </id>
		<title>Retraction: A Social Media Organizational Productivity Model: Insights From Public Health Professionals</title>
		<updated>2026-11-16T15:00:26-05:00</updated>

					<author>
				<name>JMIR Publications Editorial Office</name>
			</author>
				<link rel="alternate" href="https://www.jmir.org/2026/1/e116560" />
		
        
        
		<published>2026-11-16T15:00:26-05:00</published>
	</entry>
	<entry>
		<id> https://www.jmir.org/2026/1/e102896 </id>
		<title>How Message Construal Level and Outcome Framing Relate to Likes on TikTok Videos About Alcohol Reduction and Cessation: Computational Content Analysis</title>
		<updated>2026-11-16T14:30:03-05:00</updated>

					<author>
				<name>Han Li</name>
			</author>
					<author>
				<name>Drew Margolin</name>
			</author>
					<author>
				<name>Jeff Niederdeppe</name>
			</author>
					<author>
				<name>Janghee Lee</name>
			</author>
					<author>
				<name>Rumi Chunara</name>
			</author>
					<author>
				<name>Rohit Kanrar</name>
			</author>
					<author>
				<name>Dan Nettleton</name>
			</author>
					<author>
				<name>Andy J King</name>
			</author>
				<link rel="alternate" href="https://www.jmir.org/2026/1/e102896" />
					<summary type="html" xml:base="https://www.jmir.org/2026/1/e102896">&lt;strong&gt;Background:&lt;/strong&gt; Social media has transformed the public health communication landscape, giving rise to competing narratives around alcohol use and creating new opportunities for health messaging. On TikTok, short-form videos promoting alcohol reduction and cessation through hashtags such as #sobercurious and #dryjanuary have gained considerable traction. However, existing research has primarily described alcohol-related social media discourse, leaving limited theory-driven evidence about which message characteristics and communication strategies are associated with positive audience engagement, and whether these associations differ between expert and lay creators. Addressing this gap can inform the development of public health messages suited to algorithmically curated social media platforms. &lt;strong&gt;Objective:&lt;/strong&gt; Drawing on construal level theory and outcome expectancy theory, this study examined how message construal level and alcohol-related outcome framing were associated with positive audience engagement, operationalized as video likes, and whether these associations varied between experts and lay content creators. &lt;strong&gt;Methods:&lt;/strong&gt; We collected and analyzed a stratified sample of 1950 US-based TikTok videos about alcohol reduction and cessation posted between 2018 and 2025. Videos were identified through systematic hashtag curation and large language model (LLM)–assisted relevance screening. Audio was transcribed using the Whisper API, and video metadata and creator information were obtained through the TikTok Research API. Message construal level was measured using the Medical Research Council psycholinguistic dictionary, which quantifies language concreteness, while AI-assisted text annotation (GPT-4o) was used to identify the presence and valence of alcohol-related outcomes and creators’ presented professional expertise. Linear mixed-effects regression models examined associations between these characteristics and log-transformed video likes, adjusting for video views, creator follower counts, and repeated observations within creators. &lt;strong&gt;Results:&lt;/strong&gt; Greater language concreteness (ie, lower construal level) was associated with more video likes (b=0.04, 95% CI 0.01-0.08; &lt;i&gt;P&lt;/i&gt;=.005). This association was stronger among expert creators than lay creators, as indicated by a significant interaction (b=0.15, 95% CI 0.01-0.28; &lt;i&gt;P&lt;/i&gt;=.03). Videos mentioning alcohol-related outcomes received more likes than those without such mentions (b=0.07, 95% CI 0.00-0.14; &lt;i&gt;P&lt;/i&gt;=.04). Among videos mentioning outcomes, those focusing exclusively on the harms of alcohol consumption received fewer likes than those highlighting the benefits of alcohol reduction or cessation (b=–0.18, 95% CI –0.30 to –0.07; &lt;i&gt;P&lt;/i&gt;=.002). The interaction between creator type and the presence of outcome framing was not statistically significant. &lt;strong&gt;Conclusions:&lt;/strong&gt; This study extends established communication and behavioral theories to observable audience engagement with alcohol reduction and cessation content on TikTok. Our findings offer practical insights for designing public health messages suited to algorithmically curated platforms. The study also demonstrates the potential of LLM-assisted annotation for scalable health communication research. </summary>
		
        
                	<content type="image/png" src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/5630e735b44d2ec5aea759dfad4c25fc" />
		
		<published>2026-11-16T14:30:03-05:00</published>
	</entry>
</feed>