-
NPJ Digital Medicine Dec 2023Conversational artificial intelligence (AI), particularly AI-based conversational agents (CAs), is gaining traction in mental health care. Despite their growing usage,... (Review)
Review
Conversational artificial intelligence (AI), particularly AI-based conversational agents (CAs), is gaining traction in mental health care. Despite their growing usage, there is a scarcity of comprehensive evaluations of their impact on mental health and well-being. This systematic review and meta-analysis aims to fill this gap by synthesizing evidence on the effectiveness of AI-based CAs in improving mental health and factors influencing their effectiveness and user experience. Twelve databases were searched for experimental studies of AI-based CAs' effects on mental illnesses and psychological well-being published before May 26, 2023. Out of 7834 records, 35 eligible studies were identified for systematic review, out of which 15 randomized controlled trials were included for meta-analysis. The meta-analysis revealed that AI-based CAs significantly reduce symptoms of depression (Hedge's g 0.64 [95% CI 0.17-1.12]) and distress (Hedge's g 0.7 [95% CI 0.18-1.22]). These effects were more pronounced in CAs that are multimodal, generative AI-based, integrated with mobile/instant messaging apps, and targeting clinical/subclinical and elderly populations. However, CA-based interventions showed no significant improvement in overall psychological well-being (Hedge's g 0.32 [95% CI -0.13 to 0.78]). User experience with AI-based CAs was largely shaped by the quality of human-AI therapeutic relationships, content engagement, and effective communication. These findings underscore the potential of AI-based CAs in addressing mental health issues. Future research should investigate the underlying mechanisms of their effectiveness, assess long-term effects across various mental health outcomes, and evaluate the safe integration of large language models (LLMs) in mental health care.
PubMed: 38114588
DOI: 10.1038/s41746-023-00979-5 -
Environment International Aug 2023The World Health Organization (WHO) and the International Labour Organization (ILO) are developing joint estimates of the work-related burden of disease and injury... (Meta-Analysis)
Meta-Analysis
The prevalences and levels of occupational exposure to dusts and/or fibres (silica, asbestos and coal): A systematic review and meta-analysis from the WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury.
BACKGROUND
The World Health Organization (WHO) and the International Labour Organization (ILO) are developing joint estimates of the work-related burden of disease and injury (WHO/ILO Joint Estimates), with contributions from a large number of individual experts. Evidence from human, animal and mechanistic data suggests that occupational exposure to dusts and/or fibres (silica, asbestos and coal dust) causes pneumoconiosis. In this paper, we present a systematic review and meta-analysis of the prevalences and levels of occupational exposure to silica, asbestos and coal dust. These estimates of prevalences and levels will serve as input data for estimating (if feasible) the number of deaths and disability-adjusted life years that are attributable to occupational exposure to silica, asbestos and coal dust, for the development of the WHO/ILO Joint Estimates.
OBJECTIVES
We aimed to systematically review and meta-analyse estimates of the prevalences and levels of occupational exposure to silica, asbestos and coal dust among working-age (≥ 15 years) workers.
DATA SOURCES
We searched electronic academic databases for potentially relevant records from published and unpublished studies, including Ovid Medline, PubMed, EMBASE, and CISDOC. We also searched electronic grey literature databases, Internet search engines and organizational websites; hand-searched reference lists of previous systematic reviews and included study records; and consulted additional experts.
STUDY ELIGIBILITY AND CRITERIA
We included working-age (≥ 15 years) workers in the formal and informal economy in any WHO and/or ILO Member State but excluded children (< 15 years) and unpaid domestic workers. We included all study types with objective dust or fibre measurements, published between 1960 and 2018, that directly or indirectly reported an estimate of the prevalence and/or level of occupational exposure to silica, asbestos and/or coal dust.
STUDY APPRAISAL AND SYNTHESIS METHODS
At least two review authors independently screened titles and abstracts against the eligibility criteria at a first stage and full texts of potentially eligible records at a second stage, then data were extracted from qualifying studies. We combined prevalence estimates by industrial sector (ISIC-4 2-digit level with additional merging within Mining, Manufacturing and Construction) using random-effects meta-analysis. Two or more review authors assessed the risk of bias and all available authors assessed the quality of evidence, using the ROB-SPEO tool and QoE-SPEO approach developed specifically for the WHO/ILO Joint Estimates.
RESULTS
Eighty-eight studies (82 cross-sectional studies and 6 longitudinal studies) met the inclusion criteria, comprising > 2.4 million measurements covering 23 countries from all WHO regions (Africa, Americas, Eastern Mediterranean, South-East Asia, Europe, and Western Pacific). The target population in all 88 included studies was from major ISCO groups 3 (Technicians and Associate Professionals), 6 (Skilled Agricultural, Forestry and Fishery Workers), 7 (Craft and Related Trades Workers), 8 (Plant and Machine Operators and Assemblers), and 9 (Elementary Occupations), hereafter called manual workers. Most studies were performed in Construction, Manufacturing and Mining. For occupational exposure to silica, 65 studies (61 cross-sectional studies and 4 longitudinal studies) were included with > 2.3 million measurements collected in 22 countries in all six WHO regions. For occupational exposure to asbestos, 18 studies (17 cross-sectional studies and 1 longitudinal) were included with > 20,000 measurements collected in eight countries in five WHO regions (no data for Africa). For occupational exposure to coal dust, eight studies (all cross-sectional) were included comprising > 100,000 samples in six countries in five WHO regions (no data for Eastern Mediterranean). Occupational exposure to silica, asbestos and coal dust was assessed with personal or stationary active filter sampling; for silica and asbestos, gravimetric assessment was followed by technical analysis. Risk of bias profiles varied between the bodies of evidence looking at asbestos, silica and coal dust, as well as between industrial sectors. However, risk of bias was generally highest for the domain of selection of participants into the studies. The largest bodies of evidence for silica related to the industrial sectors of Construction (ISIC 41-43), Manufacturing (ISIC 20, 23-25, 27, 31-32) and Mining (ISIC 05, 07, 08). For Construction, the pooled prevalence estimate was 0.89 (95% CI 0.84 to 0.93, 17 studies, I 91%, moderate quality of evidence) and the level estimate was rated as of very low quality of evidence. For Manufacturing, the pooled prevalence estimate was 0.85 (95% CI 0.78 to 0.91, 24 studies, I 100%, moderate quality of evidence) and the pooled level estimate was rated as of very low quality of evidence. The pooled prevalence estimate for Mining was 0.75 (95% CI 0.68 to 0.82, 20 studies, I 100%, moderate quality of evidence) and the pooled level estimate was 0.04 mg/m (95% CI 0.03 to 0.05, 17 studies, I 100%, low quality of evidence). Smaller bodies of evidence were identified for Crop and animal production (ISIC 01; very low quality of evidence for both prevalence and level); Professional, scientific and technical activities (ISIC 71, 74; very low quality of evidence for both prevalence and level); and Electricity, gas, steam and air conditioning supply (ISIC 35; very low quality of evidence for both prevalence and level). For asbestos, the pooled prevalence estimate for Construction (ISIC 41, 43, 45,) was 0.77 (95% CI 0.65 to 0.87, six studies, I 99%, low quality of evidence) and the level estimate was rated as of very low quality of evidence. For Manufacturing (ISIC 13, 23-24, 29-30), the pooled prevalence and level estimates were rated as being of very low quality of evidence. Smaller bodies of evidence were identified for Other mining and quarrying (ISIC 08; very low quality of evidence for both prevalence and level); Electricity, gas, steam and air conditioning supply (ISIC 35; very low quality of evidence for both prevalence and level); and Water supply, sewerage, waste management and remediation (ISIC 37; very low quality of evidence for levels). For coal dust, the pooled prevalence estimate for Mining of coal and lignite (ISIC 05), was 1.00 (95% CI 1.00 to 1.00, six studies, I 16%, moderate quality of evidence) and the pooled level estimate was 0.77 mg/m (95% CI 0.68 to 0.86, three studies, I 100%, low quality of evidence). A small body of evidence was identified for Electricity, gas, steam and air conditioning supply (ISIC 35); with very low quality of evidence for prevalence, and the pooled level estimate being 0.60 mg/m (95% CI -6.95 to 8.14, one study, low quality of evidence).
CONCLUSIONS
Overall, we judged the bodies of evidence for occupational exposure to silica to vary by industrial sector between very low and moderate quality of evidence for prevalence, and very low and low for level. For occupational exposure to asbestos, the bodies of evidence varied by industrial sector between very low and low quality of evidence for prevalence and were of very low quality of evidence for level. For occupational exposure to coal dust, the bodies of evidence were of very low or moderate quality of evidence for prevalence, and low for level. None of the included studies were population-based studies (i.e., covered the entire workers' population in the industrial sector), which we judged to present serious concern for indirectness, except for occupational exposure to coal dust within the industrial sector of mining of coal and lignite. Selected estimates of the prevalences and levels of occupational exposure to silica by industrial sector are considered suitable as input data for the WHO/ILO Joint Estimates, and selected estimates of the prevalences and levels of occupational exposure to asbestos and coal dust may perhaps also be suitable for estimation purposes. Protocol identifier: https://doi.org/10.1016/j.envint.2018.06.005. PROSPERO registration number: CRD42018084131.
Topics: Humans; Adolescent; Occupational Diseases; Dust; Prevalence; Silicon Dioxide; Cross-Sectional Studies; Coal; Steam; Asbestos; Occupational Exposure; World Health Organization; Cost of Illness
PubMed: 37487377
DOI: 10.1016/j.envint.2023.107980 -
Sleep Medicine: X Dec 2023Insomnia is a common disease, and the application of various types of sleeping pills for cognitive impairment is controversial, especially as different doses can lead to... (Review)
Review
BACKGROUND
Insomnia is a common disease, and the application of various types of sleeping pills for cognitive impairment is controversial, especially as different doses can lead to different effects. Therefore, it is necessary to evaluate the cognitive impairment caused by different sleeping pills to provide a theoretical basis for guiding clinicians in the selection of medication regimens.
OBJECTIVE
To evaluate whether various different doses (low, medium and high) of anti-insomnia drugs, such as the dual-orexin receptor antagonist (DORA), zopiclone, eszopiclone and zolpidem, induce cognitive impairment.
METHODS
The PubMed, Embase, Scopus, Cochrane Library, and Google Scholar databases were searched from inception to September 20th, 2022 for keywords in randomized controlled trials (RCTs) to evaluate the therapeutic effects of DORA, eszopiclone, zopiclone and zolpidem on sleep and cognitive function. The primary outcomes were indicators related to cognitive characteristics, including scores on the Digit Symbol Substitution Test (DSST) and daytime alertness. The secondary outcomes were the indicators associated with sleep and adverse events. Continuous variables were expressed as the standard mean difference (SMD). Data were obtained through GetData 2.26 and analyzed by Stata v.15.0.
RESULTS
A total of 8702 subjects were included in 29 studies. Eszopiclone significantly increased the daytime alertness score (SMD = 3.00, 95 % CI: 1.86 to 4.13) compared with the placebo, and eszopiclone significantly increased the daytime alertness score (SMD = 4.21, 95 % CI: 1.65 to 6.77; SMD = 3.95, 95 % CI: 1.38 to 6.51; SMD = 3.26, 95 % CI: 0.38 to 6.15; and SMD = 3.23, 95 % CI: 0.34 to 6.11) compared with zolpidem, zolpidem, DORA, and eszopiclone, respectively. Compared with the placebo, zopiclone, zolpidem, and eszopiclone, DORA significantly increased the TST (SMD = 2.39, 95 % CI: 1.11 to 3.67; SMD = 6.00, 95 % CI: 2.73 to 9.27; SMD = 1.89, 95 % CI: 0.90 to 2.88; and SMD = 1.70, 95 % CI: 0.42 to 2.99, respectively).
CONCLUSION
We recommend DORA as the best intervention for insomnia because it was highly effective in inducing and maintaining sleep without impairing cognition. Although zolpidem had a more pronounced effect on sleep maintenance, this drug is better for short-term use. Eszopiclone and zopiclone improved sleep, but their cognitive effects have yet to be verified.
PubMed: 38149178
DOI: 10.1016/j.sleepx.2023.100094 -
The Lancet. Digital Health Dec 2023Machine learning and deep learning models have been increasingly used to predict long-term disease progression in patients with chronic obstructive pulmonary disease... (Meta-Analysis)
Meta-Analysis
Machine learning and deep learning predictive models for long-term prognosis in patients with chronic obstructive pulmonary disease: a systematic review and meta-analysis.
BACKGROUND
Machine learning and deep learning models have been increasingly used to predict long-term disease progression in patients with chronic obstructive pulmonary disease (COPD). We aimed to summarise the performance of such prognostic models for COPD, compare their relative performances, and identify key research gaps.
METHODS
We conducted a systematic review and meta-analysis to compare the performance of machine learning and deep learning prognostic models and identify pathways for future research. We searched PubMed, Embase, the Cochrane Library, ProQuest, Scopus, and Web of Science from database inception to April 6, 2023, for studies in English using machine learning or deep learning to predict patient outcomes at least 6 months after initial clinical presentation in those with COPD. We included studies comprising human adults aged 18-90 years and allowed for any input modalities. We reported area under the receiver operator characteristic curve (AUC) with 95% CI for predictions of mortality, exacerbation, and decline in forced expiratory volume in 1 s (FEV). We reported the degree of interstudy heterogeneity using Cochran's Q test (significant heterogeneity was defined as p≤0·10 or I>50%). Reporting quality was assessed using the TRIPOD checklist and a risk-of-bias assessment was done using the PROBAST checklist. This study was registered with PROSPERO (CRD42022323052).
FINDINGS
We identified 3620 studies in the initial search. 18 studies were eligible, and, of these, 12 used conventional machine learning and six used deep learning models. Seven models analysed exacerbation risk, with only six reporting AUC and 95% CI on internal validation datasets (pooled AUC 0·77 [95% CI 0·69-0·85]) and there was significant heterogeneity (I 97%, p<0·0001). 11 models analysed mortality risk, with only six reporting AUC and 95% CI on internal validation datasets (pooled AUC 0·77 [95% CI 0·74-0·80]) with significant degrees of heterogeneity (I 60%, p=0·027). Two studies assessed decline in lung function and were unable to be pooled. Machine learning and deep learning models did not show significant improvement over pre-existing disease severity scores in predicting exacerbations (p=0·24). Three studies directly compared machine learning models against pre-existing severity scores for predicting mortality and pooled performance did not differ (p=0·57). Of the five studies that performed external validation, performance was worse than or equal to regression models. Incorrect handling of missing data, not reporting model uncertainty, and use of datasets that were too small relative to the number of predictive features included provided the largest risks of bias.
INTERPRETATION
There is limited evidence that conventional machine learning and deep learning prognostic models demonstrate superior performance to pre-existing disease severity scores. More rigorous adherence to reporting guidelines would reduce the risk of bias in future studies and aid study reproducibility.
FUNDING
None.
Topics: Adult; Humans; Reproducibility of Results; Deep Learning; Quality of Life; Pulmonary Disease, Chronic Obstructive; Prognosis
PubMed: 38000872
DOI: 10.1016/S2589-7500(23)00177-2 -
Plastic and Reconstructive Surgery.... Dec 2023Children have been suggested to benefit from digit replantation due to a greater neurogenerative capacity. We aimed to conduct a systematic review on digit replantation...
BACKGROUND
Children have been suggested to benefit from digit replantation due to a greater neurogenerative capacity. We aimed to conduct a systematic review on digit replantation in children to provide a comprehensive overview of survival rates and functional outcomes.
METHODS
A systematic literature search was conducted on Ovid MEDLINE, Embase, and the Cochrane Controlled Register of Trials for studies published between 1980 and 2023. We included peer-reviewed studies reporting on digit survival rates in pediatric patients under the age of 18 years who underwent single or multiple digit replantations distal to the metacarpophalangeal joint. Preoperative, intraoperative, and postoperative outcomes were extracted, and pooled estimates were derived using univariable analysis.
RESULTS
Twenty-two studies reporting on 761 patients and 814 digit replantations were included in our study. Most replantations occurred in the index (n = 74), Tamai zone I (n = 168), and from clean-cut injuries (n = 190). The mean survival rate was 76% (n = 618/814), with a mean range of motion at the distal interphalangeal joint ranging from 64 degrees to 90 degrees and two-point discrimination ranging from 3.8 mm to 6.4 mm. Compared with clean-cut injuries, digit replantations from avulsion [odds ratio (OR), 0.81; 95% confidence interval (CI), 0.74-0.89] or crush (OR, 0.71; 95% CI, 0.59-0.82) injuries were associated with a lower odds of survival. Digit replantations performed with two venous (OR, 1.43, 95% CI; 1.28-1.59) or arterial anastomoses (OR, 1.65; 95% CI, 1.48-1.81) were associated with a higher odds of survival.
CONCLUSIONS
Our systematic review suggests that digit replantation may be a viable option in children. Further research is required to explore functionality after digit replantation in diverse pediatric populations.
PubMed: 38098954
DOI: 10.1097/GOX.0000000000005482 -
NPJ Digital Medicine Aug 2023Global Human papillomavirus (HPV) vaccination rates remain low despite available WHO-approved vaccines. Digital interventions for promoting vaccination uptake offer a...
Global Human papillomavirus (HPV) vaccination rates remain low despite available WHO-approved vaccines. Digital interventions for promoting vaccination uptake offer a scalable and accessible solution to this issue. Here we report a systematic review and meta-analysis examining the efficacy of digital interventions, comparing educational and reminder approaches, for promoting HPV vaccination uptake (HVU). This study also identifies factors influencing the effectiveness of these interventions. We searched PubMed, PsycInfo, Web of Science, and the Cochrane Library from each database's inception to January 2023. Three raters independently evaluate the studies using a systematic and blinded method for resolving disagreements. From 1929 references, 34 unique studies (281,280 unique participants) have sufficient data. Client reminder (OR, 1.41; 95% CI, 1.23-1.63; P < 0.001), provider reminder (OR, 1.39; 95% CI, 1.11-1.75; P = 0.005), provider education (OR, 1.18; 95% CI, 1.05-1.34; P = 0.007), and client education plus reminder interventions (OR, 1.29; 95% CI, 1.04-1.59; P = 0.007) increase HVU, whereas client education interventions do not (OR, 1.08; 95% CI, 0.92-1.28; P = 0.35). Digital intervention effectiveness varies based on participants' gender and the digital platform used. Interventions targeting male or mixed-gender participants demonstrate greater benefit, and reminder platforms (SMS, preference reminders, or electronic health record alerts) are more effective in increasing HVU. Digital interventions, particularly client and provider reminders, along with provider education, prove significantly more effective than client education alone. Incorporating digital interventions into healthcare systems can effectively promote HPV vaccination uptake. Reminder interventions should be prioritized for promoting HVU.
PubMed: 37644090
DOI: 10.1038/s41746-023-00912-w -
High incidence of trigger finger after carpal tunnel release: a systematic review and meta-analysis.International Journal of Surgery... Aug 2023Trigger finger (TF) often occurs after carpal tunnel release (CTR), but the mechanism and outcomes remain inconsistent. This study evaluated the incidence of TF after... (Meta-Analysis)
Meta-Analysis
INTRODUCTION
Trigger finger (TF) often occurs after carpal tunnel release (CTR), but the mechanism and outcomes remain inconsistent. This study evaluated the incidence of TF after CTR and its related risk factors.
MATERIALS AND METHODS
PubMed, Embase, and Scopus databases were searched up to 27 August 2022, with the following keywords: "carpal tunnel release" and "trigger finger". Studies with complete data on the incidence of TF after CTR and published full text. The primary outcome was the association between CTR and the subsequent occurrence of the TF and to calculate the pooled incidence of post-CTR TF. The secondary outcomes included the potential risk factors among patients with and without post-CTR TF as well as the prevalence of the post-CTR TF on the affected digits.
RESULTS
Ten studies with total 10,399 participants in 9 studies and 875 operated hands in one article were included for meta-analysis. CTR significantly increases the risk of following TF occurrence (odds ratio=2.67; 95% CI 2.344-3.043; P <0.001). The pooled incidence of TF development after CTR was 7.7%. Women were more likely to develop a TF after CTR surgery (odds ratio=2.02; 95% CI 1.054-3.873; P =0.034). Finally, the thumb was the most susceptible fingers, followed by middle and ring fingers.
CONCLUSIONS
High incidence of TF comes after CTR, and women were more susceptible than man. Clinicians were suggested to notice the potential risk of TF after CTR in clinical practice.
LEVEL OF EVIDENCE
Level III, meta-analysis.
Topics: Male; Humans; Female; Incidence; Carpal Tunnel Syndrome; Risk Factors; Trigger Finger Disorder; Thumb
PubMed: 37161585
DOI: 10.1097/JS9.0000000000000450 -
Frontiers in Digital Health 2023This systematic review aims to assess the effectiveness of virtual reality (VR) and gamification interventions in addressing anxiety and depression. The review also... (Review)
Review
This systematic review aims to assess the effectiveness of virtual reality (VR) and gamification interventions in addressing anxiety and depression. The review also seeks to identify gaps in the current VR treatment landscape and provide guidelines for future research and development. A systematic literature search was conducted using Scopus, Web of Science, and PubMed databases, focusing on studies that utilized VR and gamification technology to address anxiety and depression disorders. A total of 2,664 studies were initially identified, 15 of those studies fulfilled the inclusion criteria for this systematic review. The efficacy of VR in addressing anxiety and depression was evident across all included studies. However, the diversity among VR interventions highlights the need for further investigation. It is advised to incorporate more diverse participant samples and larger cohorts and explore a broader spectrum of therapeutic approaches within VR interventions for addressing anxiety and depression to enhance the credibility of future research. Additionally, conducting studies in varying socioeconomic contexts would contribute to a more comprehensive understanding of their real-world applicability.
PubMed: 38026832
DOI: 10.3389/fdgth.2023.1239435 -
Digital Health 2023Artificial intelligence (AI) has been increasingly applied in various fields of science and technology. In line with the current research, medicine involves an... (Review)
Review
OBJECTIVE
Artificial intelligence (AI) has been increasingly applied in various fields of science and technology. In line with the current research, medicine involves an increasing number of artificial intelligence technologies. The introduction of rapid AI can lead to positive and negative effects. This is a multilateral analytical literature review aimed at identifying the main branches and trends in the use of using artificial intelligence in medical technologies.
METHODS
The total number of literature sources reviewed is n = 89, and they are analyzed based on the literature reporting evidence-based guideline PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) for a systematic review.
RESULTS
As a result, from the initially selected 198 references, 155 references were obtained from the databases and the remaining 43 sources were found on open internet as direct links to publications. Finally, 89 literature sources were evaluated after exclusion of unsuitable references based on the duplicated and generalized information without focusing on the users.
CONCLUSIONS
This article is identifying the current state of artificial intelligence in medicine and prospects for future use. The findings of this review will be useful for healthcare and AI professionals for improving the circulation and use of medical AI from design to implementation stage.
PubMed: 37485326
DOI: 10.1177/20552076231189331 -
Digital Health 2023Artificial intelligence (AI) technologies are transforming medicine and healthcare. Scholars and practitioners have debated the philosophical, ethical, legal, and... (Review)
Review
BACKGROUND
Artificial intelligence (AI) technologies are transforming medicine and healthcare. Scholars and practitioners have debated the philosophical, ethical, legal, and regulatory implications of medical AI, and empirical research on stakeholders' knowledge, attitude, and practices has started to emerge. This study is a systematic review of published empirical studies of medical AI ethics with the goal of mapping the main approaches, findings, and limitations of scholarship to inform future practice considerations.
METHODS
We searched seven databases for published peer-reviewed empirical studies on medical AI ethics and evaluated them in terms of types of technologies studied, geographic locations, stakeholders involved, research methods used, ethical principles studied, and major findings.
FINDINGS
Thirty-six studies were included (published 2013-2022). They typically belonged to one of the three topics: exploratory studies of stakeholder knowledge and attitude toward medical AI, theory-building studies testing hypotheses regarding factors contributing to stakeholders' acceptance of medical AI, and studies identifying and correcting bias in medical AI.
INTERPRETATION
There is a disconnect between high-level ethical principles and guidelines developed by ethicists and empirical research on the topic and a need to embed ethicists in tandem with AI developers, clinicians, patients, and scholars of innovation and technology adoption in studying medical AI ethics.
PubMed: 37434728
DOI: 10.1177/20552076231186064