Shift in Social Media App Usage During COVID-19 Lockdown and Clinical Anxiety Symptoms: Machine Learning-Based Ecological Momentary Assessment Study

Artificial Intelligence in Psychiatry: A Review of Biological and Behavioral Data Analyses

Artificial intelligence (AI) has emerged as a transformative force in psychiatry, improving diagnostic precision, treatment personalization, and early intervention through advanced data analysis techniques. This review explores recent advancements in AI applications within psychiatry, focusing on EEG and ECG data analysis, speech analysis, natural language processing (NLP), blood biomarker integration, and social media data utilization. EEG-based models have significantly enhanced the detection of disorders such as depression and schizophrenia through spectral and connectivity analyses. ECG-based approaches have provided insights into emotional regulation and stress-related conditions using heart rate variability. Speech analysis frameworks, leveraging large language models (LLMs), have improved the detection of cognitive impairments and psychiatric symptoms through nuanced linguistic feature extraction. Meanwhile, blood biomarker analyses have deepened our understanding of the molecular underpinnings of mental health disorders, and social media analytics have demonstrated the potential for real-time mental health surveillance. Despite these advancements, challenges such as data heterogeneity, interpretability, and ethical considerations remain barriers to widespread clinical adoption. Future research must prioritize the development of explainable AI models, regulatory compliance, and the integration of diverse datasets to maximize the impact of AI in psychiatric care.

A daily examination of perceived stress and coping strategies used among college students during COVID-19

This study examined inter- and intra-person level associations between stress and overwhelm and adaptive coping strategies used by college students during COVID-19. Participants were 55 college students (Mage = 23.31, SD = 4.49; 87% White, 85% Female), who completed a 21-day daily diary study. The common coping strategies endorsed were media use, relaxation techniques, and exercise. Variability among feelings of stress (ICC = .520) and overwhelm (ICC = .530) were similarly explained by between and within-person differences. Daily stress and overwhelm were positively associated with relaxation techniques, media, and total coping strategies. Further, daily stress was positively associated with connection with others to cope. At the between-person level, average stress was associated with avoidance strategies such as taking a break from news to cope. These findings may prove useful in the development of interventions to increase the practice of adaptive coping strategies to reduce stress and overwhelm.

Evaluating the Quality and Reliability of Total Knee Arthroplasty Rehabilitation Exercises on the Social Media Platform TikTok

Background

The utilization of social media for health-related purposes has surged, especially during the COVID-19 pandemic. TikTok, a short-form video platform, has seen substantial growth, becoming a prominent medium for health information dissemination. However, the lack of regulation poses challenges in evaluating the validity of TikTok content.

Methods

This cross-sectional study assesses TikTok videos related to total knee arthroplasty rehabilitation exercises. Search terms identified 84 videos, with 64 meeting the inclusion criteria. Engagement metrics and quality scores were analyzed, utilizing the DISCERN tool and the Total Knee Replacement Exercises Education Score.

Results

The analyzed videos accumulated nearly 6 million views, with a median of 10,293.5 (interquartile range = 4139.3-26,100.0). Health-care professionals contributed 48% of the content. Despite higher engagement metrics for health-care professional videos, the overall quality, as indicated by DISCERN and Total Knee Replacement Exercises Education scores, remained poor. No videos achieved an "excellent" rating, with the majority categorized as "poor."

Conclusions

This study underscores TikTok's substantial role in total knee arthroplasty rehabilitation information dissemination but reveals a critical deficit in content quality and reliability. Health-care professionals marginally outperformed general users but displayed overall inadequacy. The study emphasizes the necessity for improving the quality of health-related content on emerging social media platforms, especially within the realm of orthopaedic surgery.

Level of Evidence

Level III, Cross-sectional study.

Daily routine disruptions and psychiatric symptoms amid COVID-19: a systematic review and meta-analysis of data from 0.9 million individuals in 32 countries

BACKGROUND

There is currently a deficit of knowledge about how to define, quantify, and measure different aspects of daily routine disruptions amid large-scale disasters like COVID-19, and which psychiatric symptoms were more related to the disruptions. This study aims to conduct a systematic review and meta-analysis on the probable positive associations between daily routine disruptions and mental disorders amid the COVID-19 pandemic and factors that moderated the associations.

METHODS

PsycINFO, Web of Science, PubMed, and MEDLINE were systematically searched up to April 2023 (PROSPERO: CRD42023356846). Independent variables included regularity, change in frequency, and change in capability of different daily routines (i.e., physical activity, diet, sleep, social activities, leisure activities, work and studies, home activities, smoking, alcohol, combined multiple routines, unspecified generic routines). Dependent variables included symptoms and/or diagnoses of mental disorders (i.e., depression, anxiety, post-traumatic stress disorder, and general psychological distress).

RESULTS

Fifty-three eligible studies (51 independent samples, 910,503 respondents) were conducted in five continents. Daily routine disruptions were positively associated with depressive symptoms (r = 0.13, 95% CI = [0.06; 0.20], p < 0.001), anxiety symptoms (r = 0.12, 95% CI = [0.06; 0.17], p < 0.001), and general psychological distress (r = 0.09, 95% CI = [0.02; 0.16], p = 0.02). The routine-symptom associations were significant for physical activity, eating, sleep, and smoking (i.e., type), routines that were defined and assessed on regularity and change in capability (i.e., definition and assessment), and routines that were not internet-based. While the positive associations remained consistent across different sociodemographics, they were stronger in geo-temporal contexts with greater pandemic severity, lower governmental economic support, and when the routine-symptom link was examined prospectively.

CONCLUSIONS

This is one of the first meta-analytic evidence to show the positive association between daily routine disruptions and symptoms of mental disorders among large populations as COVID-19 dynamically unfolded across different geo-temporal contexts. Our findings highlight the priority of behavioral adjustment for enhancing population mental health in future large-scale disasters like COVID-19.

Detecting Longitudinal Trends between Passively Collected Phone Use and Anxiety among College Students

Introduction

Existing theories and empirical works link phone use with anxiety; however, most leverage subjective self-reports of phone use (e.g., validated questionnaires) that may not correspond well with true behavior. Moreover, most works linking phone use with anxiety do not interrogate associations within a temporal framework. Accordingly, the present study sought to investigate the utility of passively sensed phone use as a longitudinal predictor of anxiety symptomatology within a population particularly vulnerable to experiencing anxiety.

Methods

Using data from the GLOBEM study, which continuously collected longitudinal behavioral data from a college cohort of N = 330 students, weekly PHQ-4 anxiety subscale scores across 3 years (2019-2021) were paired with median daily phone use records from the 2 weeks prior to anxiety self-report completion. Phone use was operationalized through unlock duration which was passively curated via Apple's "Screen Time" feature. GPS-tracked location data was further utilized to specify whether an individual's phone use was at home or away from home. Within-individual and temporal associations between phone use and anxiety were modeled within an ordinal mixed-effects logistic regression framework.

Results

While there was no significant association between anxiety levels and either median total phone use or median phone use at home, participants in the top quartile of median phone use away from home were predicted to exhibit clinically significant anxiety levels 20% more frequently than participants in the bottom quartile during the first study year; however, this association weakened across successive years. Importantly, these associations remained after controlling for age, physical activity, sleep, and baseline anxiety levels and were not recapitulated when operationalizing phone use with unlock frequency.

Conclusions

These findings suggest that phone use may be leveraged as a means of mitigating or coping with anxiety in social situations outside the home, while pandemic-related developments may also have attenuated this behavior later in the study. Nevertheless, the present results suggest promise in interrogating a larger suite of objectively measured phone use behaviors within the context of social anxiety.

Smartphone Usage before and during COVID-19: A Comparative Study Based on Objective Recording of Usage Data

Most studies that claimed changes in smartphone usage during COVID-19 were based on self-reported usage data, e.g., that collected through a questionnaire. These studies were also limited to reporting the overall smartphone usage, with no detailed investigation of distinct types of apps. The current study investigated smartphone usage before and during COVID-19. Our study used a dataset from a smartphone app that objectively logged users’ activities, including apps accessed and each app session start and end time. These were collected during two periods: pre-COVID-19 (161 individuals with 77 females) and during COVID-19 (251 individuals with 159 females). We report on the top 15 apps used in both periods. The Mann–Whitney U test was used for the inferential analysis. The results revealed that the time spent on smartphones has increased since COVID-19. During both periods, emerging adults were found to spend more time on smartphones compared to adults. The time spent on social media apps has also increased since COVID-19. Females were found to spend more time on social media than males. Females were also found to be more likely to launch social media apps than males. There has also been an increase in the number of people who use gaming apps since the pandemic. The use of objectively collected data is a methodological strength of our study. Additionally, we draw parallels with the usage of smartphones in contexts similar to the COVID-19 period, especially concerning the limitations on social gatherings, including working from home for extended periods. Our dataset is made available to other researchers for benchmarking and future comparisons.

Artificial intelligence-based analytics for impacts of COVID-19 and online learning on college students' mental health

COVID-19, the disease caused by the novel coronavirus (SARS-CoV-2), first emerged in Wuhan, China late in December 2019. Not long after, the virus spread worldwide and was declared a pandemic by the World Health Organization in March 2020. This caused many changes around the world and in the United States, including an educational shift towards online learning. In this paper, we seek to understand how the COVID-19 pandemic and the increase in online learning impact college students' emotional wellbeing. We use several machine learning and statistical models to analyze data collected by the Faculty of Public Administration at the University of Ljubljana, Slovenia in conjunction with an international consortium of universities, other higher education institutions, and students' associations. Our results indicate that features related to students' academic life have the largest impact on their emotional wellbeing. Other important factors include students' satisfaction with their university's and government's handling of the pandemic as well as students' financial security.

Is Social Media a New Type of Social Support? Social Media Use in Spain during the COVID-19 Pandemic: A Mixed Methods Study

This study examines Spanish adults’ social media use during the COVID-19 pandemic using mixed-methods to assess and understand frequency, context, and changes in social media use during two critical time points in Spain. We conducted semi-structured interviews in April 2020, and two waves of surveys (April 2020, April 2021) among Spanish adults. We coded and analyzed qualitative data related to social media use during the first lockdown period in Spain using Dedoose software; and ran descriptive statistics and chi-square tests to assess changes in social media use over the two survey waves related to perceived social support and loneliness. Participants ranged in age from 18–92 and were representative of the Spanish population’s sociodemographics. Interview data show that WhatsApp was most commonly used, and that social media allowed for social support and engaging in healthy behaviors. Survey data show that women and individuals aged 18–34 had the greatest increases in social media use. Statistically significant associations were found between social support and loneliness with social media use. Our results show that promoting social media use as an emotional resource for social support in times of crisis or isolation can minimize loneliness and can be a beneficial tool for general worldwide crises.