Importance Large language models (LLMs) may facilitate the labor-intensive process of systematic reviews. However, the exact methods and reliability remain uncertain. Objective To explore the feasibility and reliability of using LLMs to assess risk of bias (ROB) in randomized clinical trials (RCTs). Design, Setting, and Participants A survey study was conducted between August 10, 2023, and October 30, 2023. Thirty RCTs were selected from published systematic reviews. Main Outcomes and Measures A structured prompt was developed to guide ChatGPT (LLM 1) and Claude (LLM 2) in assessing the ROB in these RCTs using a modified version of the Cochrane ROB tool developed by the CLARITY group at McMaster University. Each RCT was assessed twice by both models, and the results were documented. The results were compared with an assessment by 3 experts, which was considered a criterion standard. Correct assessment rates, sensitivity, specificity, and F1 scores were calculated to reflect accuracy, both overall and for each domain of the Cochrane ROB tool; consistent assessment rates and Cohen kappa were calculated to gauge consistency; and assessment time was calculated to measure efficiency. Performance between the 2 models was compared using risk differences. Results Both models demonstrated high correct assessment rates. LLM 1 reached a mean correct assessment rate of 84.5% (95% CI, 81.5%-87.3%), and LLM 2 reached a significantly higher rate of 89.5% (95% CI, 87.0%-91.8%). The risk difference between the 2 models was 0.05 (95% CI, 0.01-0.09). In most domains, domain-specific correct rates were around 80% to 90%; however, sensitivity below 0.80 was observed in domains 1 (random sequence generation), 2 (allocation concealment), and 6 (other concerns). Domains 4 (missing outcome data), 5 (selective outcome reporting), and 6 had F1 scores below 0.50. The consistent rates between the 2 assessments were 84.0% for LLM 1 and 87.3% for LLM 2. LLM 1's kappa exceeded 0.80 in 7 and LLM 2's in 8 domains. The mean (SD) time needed for assessment was 77 (16) seconds for LLM 1 and 53 (12) seconds for LLM 2. Conclusions In this survey study of applying LLMs for ROB assessment, LLM 1 and LLM 2 demonstrated substantial accuracy and consistency in evaluating RCTs, suggesting their potential as supportive tools in systematic review processes.

Background: Conflicts of interest (COIs) of contributors to a guideline project and the funding of that project can influence the development of the guideline. Comprehensive reporting of information on COIs and funding is essential for the transparency and credibility of guidelines. Objective: To develop an extension of the Reporting Items for practice Guidelines in HealThcare (RIGHT) statement for the reporting of COIs and funding in policy documents of guideline organizations and in guidelines: the RIGHT-COI&F checklist. Design: The recommendations of the Enhancing the QUAlity and Transparency Of health Research (EQUATOR) network were followed. The process consisted of registration of the project and setting up working groups, generation of the initial list of items, achieving consensus on the items, and formulating and testing the final checklist. Setting: International collaboration. Participants: 44 experts. Measurements: Consensus on checklist items. Results: The checklist contains 27 items: 18 about the COIs of contributors and 9 about the funding of the guideline project. Of the 27 items, 16 are labeled as policy related because they address the reporting of COI and funding policies that apply across an organization's guideline projects. These items should be described ideally in the organization's policy documents, otherwise in the specific guideline. The remaining 11 items are labeled as implementation related and they address the reporting of COIs and funding of the specific guideline. Limitation: The RIGHT-COI&F checklist requires testing in real-life use. Conclusion: The RIGHT-COI&F checklist can be used to guide the reporting of COIs and funding in guideline development and to assess the completeness of reporting in published guidelines and policy documents. Primary Funding Source: The Fundamental Research Funds for the Central Universities of China.

Objective: To summarize the evidence of various exercise modalities on population with insomnia disorders. Method: PubMed, Embase, Cochrane Library, and Web of Science were searched for eligible studies published from inception to October 2022 and updated on September 2023. Systematic reviews with meta -analyses and randomized controlled trials designed to investigate the effect of various exercise modalities on population with insomnia were eligible. Results: A total of 4 SRs with (very) low methodological quality and 1034 participants in 10 network metaanalyses explored the association between different types and intensity exercise modalities with insomnia disorders. Various exercise modalities could significantly improve total sleep time and sleep quality and alleviate insomnia severity. Compared to passive control, moderate aerobic exercise, moderate aerobic exercise combined with light intensity strength and mind -body exercise can improve sleep efficiency and reduce wake after sleep onset by objectively measured. Moderate intensity strength, light intensity strength and mind -body exercise can improve sleep efficiency subjectively measured; mind -body exercise can reduce sleep onset latency and wake time after sleep onset, and increase total sleep time; moderate aerobic exercise can reduce sleep onset latency. Moderate intensity strength, light intensity strength, mind body exercise and moderate aerobic exercise combined with light intensity strength can the severity of insomnia and improv sleep quality. Conclusion: Exercise had a positive effect on relief insomnia and improve sleep quality. Moderate aerobic exercise, mind -body exercise and moderate aerobic exercise combined with light intensity strength play an important role in improving the sleep quality in people with insomnia disorders.

Objective: This study aimed to systematically review the methodological and reporting quality of clinical practice guidelines (CPGs) developed in China and published in medical journals between 2014 and 2018. Study design and setting: We conducted a comprehensive search in multiple databases: MEDLINE (via PubMed), Embase, CBM (China Biology Medicine), CNKI (China National Knowledge Infrastructure) and Wanfang Data. We included all clinical practice guidelines developed in China between 2014 and 2018. The AGREE II tool and the RIGHT checklist were used to appraise the methodological quality and reporting quality of the included guidelines, respectively. Results: We identified 17,188 records, and included finally 573 CPGs. Most (n=507, 88.5%) were published in Chinese, and 508 (88.7%) were about Western medicine. Only 62 (10.8%) of the guidelines used the GRADE approach. The mean overall score of methodological quality over all guidelines was 19.4%, and the mean scores for the AGREE II domains were 28.6% (Scope and purpose), 17.0% (Stakeholder involvement), 11.7% (Rigor of development), 32.2% (Clarity of presentation), 14.2% (Applicability) and 12.8% (Editorial independence). The mean overall score for reporting quality over all guidelines was 30.2%, with the following mean scores for each RIGHT domain: 55.6% (Basic information), 43.8% (Background), 14.5% (Evidence), 29.2% (Recommendations), 10.7% (Review and quality assurance), 12.6% (Funding and declaration of interest) and 8.4% (Other information). Subgroup analyses found that both the methodological and reporting quality were generally higher among CPGs that used evidence grading systems or reported receiving funding. Conclusion: Both the methodological quality and the reporting quality of CPGs developed in China have improved over time, but are still below the international average.

Background Acupuncture is widely used worldwide, and systematic reviews on acupuncture are increasingly being published. Although acupuncture systematic reviews share several essential elements with other systematic reviews, some essential information for the application of acupuncture is not covered by the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) statement. Considering this, we aimed to develop an extension of the PRISMA statement for acupuncture systematic reviews. Methods We used the PRISMA statement as a starting point, and conducted this study referring to the development strategy recommended by the EQUATOR network. The initial items were collected through a wide survey among evidence users and a review of relevant studies. We conducted a three-round Delphi survey and one-day face-to-face meeting to select items and formulate the checklist. After the consensus meeting, we drafted the manuscript (including the checklist) and sent it to our advisory experts for comments, following which the checklist was refined and circulated to a group of acupuncture systematic review authors for pilot test. We also selected a sample of acupuncture systematic reviews published in 2017 to test the checklist. Results A checklist of five new sub-items (including sub items) and six modified items was formulated, involving content related to title, rationale, eligibility criteria, literature search, data extraction, and study characteristics. We clarified the rationales of the items and provided examples for each item for additional guidance. Conclusion The PRISMA for Acupuncture checklist is developed for improving the reporting of systematic reviews of acupuncture interventions.