OBJECTIVE: To assess the capabilities of large language models (LLMs), including Open AI (GPT-4.0) and Microsoft Bing (GPT-4), in generating structured reports, the Breast Imaging Reporting and Data System (BI-RADS) categories, and management recommendations from free-text breast ultrasound reports.
METHODS: In this retrospective study, 100 free-text breast ultrasound reports from patients who underwent surgery between January and May 2023 were gathered. The capabilities of Open AI (GPT-4.0) and Microsoft Bing (GPT-4) to convert these unstructured reports into structured ultrasound reports were studied. The quality of structured reports, BI-RADS categories, and management recommendations generated by GPT-4.0 and Bing were evaluated by senior radiologists based on the guidelines.
RESULTS: Open AI (GPT-4.0) was better than Microsoft Bing (GPT-4) in terms of performance in generating structured reports (88% vs. 55%; p < 0.001), giving correct BI-RADS categories (54% vs. 47%; p = 0.013) and providing reasonable management recommendations (81% vs. 63%; p < 0.001). As the ability to predict benign and malignant characteristics, GPT-4.0 performed significantly better than Bing (AUC, 0.9317 vs. 0.8177; p < 0.001), while both performed significantly inferior to senior radiologists (AUC, 0.9763; both p < 0.001).
CONCLUSIONS: This study highlights the potential of LLMs, specifically Open AI (GPT-4.0), in converting unstructured breast ultrasound reports into structured ones, offering accurate diagnoses and providing reasonable recommendations.

OBJECTIVE: To systematically understand the transparency of outcome measurement time point reporting in meta-analyses of acupuncture.
METHODS: We searched for meta-analyses of acupuncture published between 2013 and 2022 in PubMed, Embase, and Cochrane Library. A team of method-trained investigators screened studies for eligibility and collected data using pilot-tested standardized questionnaires. We documented in detail the reporting of outcome measurement time points in acupuncture meta-analyses.
RESULTS: A total of 224 acupuncture meta-analyses were included. Of these, 98 (43.8%) studies did not specify the time points of primary outcome. Among 126 (56.3%) meta-analyses which reported the time points of primary outcome, only 22 (17.5%) meta-analyses specified time points in corresponding protocol. Among 48 (38.1%) meta-analyses that estimated treatment effects of multiple time points, 11 (22.9%) meta-analyses used inappropriate meta-analysis method (subgroup analysis) to pool effect size, and none of the meta-analyses used advanced methods for pooling effect sizes at different time points.
CONCLUSIONS: Transparency in reporting outcome time points for acupuncture meta-analyses and appropriate methods to pool the effect size of multiple time points were lacking. For future systematic reviews, the transparency of outcome measurement time points should be emphasized in the protocols and final reports. Furthermore, advanced methods should be considered for pooling effect sizes at multiple time points.

暂无摘要。

What outcomes should be reported in all studies investigating uterus-sparing interventions for treating uterine adenomyosis?
We identified 24 specific and 26 generic core outcomes in nine domains.
Research reporting adenomyosis treatment is not patient-centred and shows wide variation in outcome selection, definition, reporting and measurement of quality.
An international consensus development process was performed between March and December 2021. Participants in round one were 150 healthcare professionals, 17 researchers and 334 individuals or partners with lived experience of adenomyosis from 48 high-, middle- and low-income countries. There were 291 participants in the second round.
Stakeholders included active researchers in the field, healthcare professionals involved in diagnosis and treatment, and people and their partners with lived experience of adenomyosis. The core component of the process was a 2-step modified Delphi electronic survey. The Steering Committee analysed the results and created the final core outcome set (COS) in a semi-structured meeting.
A total of 241 outcomes was identified and distilled into a \'long list\' of 71 potential outcomes. The final COS comprises 24 specific and 26 generic core outcomes across nine domains, including pain, uterine bleeding, reproductive outcomes, haematology, urinary system, life impact, delivery of care, adverse events and reporting items, all with definitions provided by the Steering Committee. Nineteen of these outcomes will apply only to certain study types. Although not included in the COS, the Steering Committee recommended that three health economic outcomes should be recorded.
Patients from continents other than Europe were under-represented in this survey. A lack of translation of the survey might have limited the active participation of people in non-English speaking countries. Only 58% of participants returned to round two, but analysis did not indicate attrition bias. There is a significant lack of scientific evidence regarding which symptoms are caused by adenomyosis and when they are related to other co-existent disorders such as endometriosis. As future research provides more clarity, the appropriate review and revision of the COS will be necessary.
Implementing this COS in future studies on the treatment of adenomyosis will improve the quality of reporting and aid evidence synthesis.
No specific funding was received for this work. T.T. received a grant (grant number 2020083) from the South Eastern Norwegian Health Authority during the course of this work. T.T. receives personal fees from General Electrics and Medtronic for lectures on ultrasound. E.R.L. is the chairman of the Norwegian Endometriosis Association. M.G.M. is a consultant for Abbvie Inc and Myovant, receives research funding from AbbVie and is Chair of the Women\'s Health Research Collaborative. S.-W.G. is a board member of the Asian Society of Endometriosis and Adenomyosis, on the scientific advisory board of the endometriosis foundation of America, previous congress chair for the World Endometriosis Society, for none of which he received personal fees. E.S. received outside of this work grants for two multicentre trials on endometriosis from the National Institute for Health Research UK, the Rosetrees Trust, and the Barts and the London Charity, he is a member of the Medicines and Healthcare Products Regulatory Agency (MHRA), Medicines for Women\'s Health Expert Advisory Group, he is an ambassador for the World Endometriosis Society, and he received personal fees for lectures from Hologic, Olympus, Medtronic, Johnson & Johnson, Intuitive and Karl Storz. M.H. is member of the British Society for Gynaecological Endoscopy subcommittee. No other conflict of interest was declared.
N/A.

OBJECTIVE: Adverse drug reaction (ADR) reporting is generally of poor quality, which may delay post-marketing regulatory actions. Here, we evaluated the quality of ADR reporting at our institution and examined the roles of clinical pharmacists in this process.
METHODS: We retrospectively reviewed ADR reports at our hospital between 2017 and 2019 to assess the number, source, drugs, and routes of administration. The quality assessment of ADR case reports form issued by the China Adverse Drug Reaction Monitoring Centre was used to assess the quality of ADR reports. Quality scores of ADR reports from pharmacists and nonpharmacists were assessed before and after review by clinical pharmacists.
CONCLUSIONS: Reports of adverse drug reaction reporting by healthcare professionals increased annually, with 59, 77 and 82 reports submitted in 2017, 2018 and 2019, respectively. The numbers of new or serious ADR reports by healthcare professionals in 2017, 2018, and 2019 were 5 (8.47%), 77 (11.69%) and 82 (10.98%), respectively. New or serious ADR reports accounted for approximately 10% (23/218) of all reported cases, and more than 70% (158/218) of the reports were from pharmacists. Systemic administration accounted for more than 80% (233/265) of adverse reactions, whereas ADRs due to topical drug use were rarely reported. The drugs that reportedly triggered ADRs were mainly antibacterial and patented Chinese medicines and accounted for more than half of all reported cases. The scores of ADR reports from pharmacists and nonpharmacists before modification by clinical pharmacists were 86.69 ± 8.12 and 68.36 ± 5.94, respectively, and the scores of ADR reports from pharmacists and nonpharmacists after modification by clinical pharmacists were 91.14 ± 6.64 and 90.02 ± 5.63, respectively.
CONCLUSIONS: In a real-world setting, pharmacists are commonly responsible for most ADR reports. The quality of ADR reports from pharmacists and nonpharmacists before review did not reach the standard of excellence. An audit of clinical pharmacists may improve the overall quality of ADR reports. However, under-reporting of adverse reactions still occurs.

The prospective registration of systematic reviews represent an effective strategy for reducing the selective reporting of outcomes. However, the relationship between registration and the reporting quality of systematic reviews on surgical interventions remains unclear.
MEDLINE was searched for relevant systematic reviews of randomized controlled trials investigating surgical interventions published in 2020. Data concerning general characteristics and registration information were independently extracted. The reporting quality was evaluated in accordance with pre-established evaluation criteria. Univariate and multivariate linear regression were performed to identify factors associated with improved reporting quality.
A total of 135 systematic reviews were analyzed, of which 50 (37%) were registered. Registered systematic reviews achieved a significantly higher compliance rate on all items compared with non-registered reviews. Registered reviews also demonstrated significantly higher proportions of the reporting of seven items. Multivariate regression analysis showed that registration status and funding support were associated with better reporting quality.
Although prospective registration associates with higher reporting quality in systematic reviews, the number of prospective registrations remains low. Therefore, prospective registration should be encouraged among authors, peer reviewers, and journal editors, as well as institutions, to enhance the value of systematic reviews in evidence-based surgical practice.

BACKGROUND: The results and data availability of vaccine trials directly affect the decisions of healthcare providers, the public, and policymakers as to whether the vaccine should be applied. However, the reporting and data sharing level of COVID-19 vaccine studies are not clear.
METHODS: A cross-sectional study was conducted. A systematic search up to 9 May 2021 in 12 databases and an updated search to 6 July 2021 were conducted in the Cochrane Living Systematic Review and Network Meta-Analysis database to identify COVID-19 vaccine trials. The basic characteristics of included trials were summarized. The reporting level was assessed according to the CONSORT checklist. The data sharing level was assessed by open science practices. Types of incomplete reporting including protocol deviation, lack of primary outcomes clarity, and the omission of harms were analyzed.
RESULTS: Finally, thirty-six COVID-19 vaccine articles reporting on 40 randomized controlled trials were included in this analysis. Based on the CONSORT checklist, the mean reporting score was 29.7 [95% confidence interval 28.7, 30.7]. Thirty-one articles (31/36, 86.1%) had data sharing statements, twenty-five articles (25/36, 69.4%) provided access to the source data. Twenty-seven articles (27/36, 75.0%) had protocol deviation, lack of primary outcomes clarity, or the omission of harms.
CONCLUSIONS: The reporting and data sharing level of COVID-19 vaccine trials were not optimal. We hope that the reporting and data sharing of future trials will be improved. We recommend establishing a comprehensive, accurate data sharing system for future vaccine trials.
BACKGROUND: This work was supported by the National Key R&D Program of China (2019YFC1710400; 2019YFC1710403).

OBJECTIVE: We aim to describe and synthesize the scientific literature on nurses\' formal reporting of workplace violence.
BACKGROUND: Research on reporting workplace violence among nurses is increasing annually, but the evidence from such reports has not been summarized or analysed.
RESULTS: A scoping review was conducted using electronic databases including PubMed, Web of Science, Scopus, CINAHL and Ovid-Embase.
RESULTS: Forty-nine studies were included in this review, and four significant issues pertinent to reporting of workplace violence was identified: (1) the reporting rate is generally low, and oral report is the most popular medium; (2) nurses are often not satisfied with how the organisation handles their report; (3) the reasons affecting reporting are complex and diverse; and (4) few studies have proposed countermeasures to promote reporting formally.
CONCLUSIONS: Scholars are paying increasing attention to workplace violence; however, the reporting of workplace violence in real world situations is not being facilitated to inform improvements in managing violence in the workplace. Therefore, more countermeasures conducive to reporting urgently need to be put forward.
CONCLUSIONS: Nurse managers are in a pivotal position to encourage reporting by their staff, respond pro-actively and advocate for more streamlined and accessible processes. The findings of this review can provide a basis for nursing managers to better manage workplace violence and improve the reporting rate among nurses.

There has been a long-standing controversy among scientists regarding the appropriate use of P-values and statistical significance in clinical research. This debate has resurfaced through recent calls to modify the threshold of P-value required to declare significance, or to retire statistical significance entirely. In this article, we revisit the issue by discussing: i) the connection between statistical thinking and evidence-based practice; ii) some history of statistical significance and P-values; iii) some practical challenges with statistical significance or P-value thresholds in clinical research; iv) the on-going debate on what to do with statistical significance; v) suggestions to shift the focus away from binary thinking of statistical significance and towards education for key stakeholders on research essentials including statistical thinking, critical thinking, good reporting, basic clinical research concepts and methods, and more. We then conclude with remarks and illustrations of the potential deleterious public health consequences of poor methods including selective choice of analysis approach and misguided reliance on binary use of P-values to report and interpret scientific findings.

OBJECTIVE: Studies that evaluate the effectiveness of concussion laws often use only a single variable (ie, presence of the law), failing to account for law complexity. We examined the association between multiple design elements of state concussion laws and rates of sports-related concussion reporting among US high school athletes.
METHODS: We derived 3 design elements of concussion laws from the 2009-2017 LawAtlas database: (1) strength of law, (2) number of law revisions, and (3) speed of law adoption. We examined the association between these design elements and rates of new and recurrent sports-related concussion reporting from the 2009-2010 through 2016-2017 academic years in a nationally representative sample of high school athletes participating in High School Report Information Online, an online data collection tool.
RESULTS: A total of 7064 sports-related concussions (6332 [89.6%] new and 732 [10.4%] recurrent concussions) were reported during the study period, with an overall rate of 39.7 sports-related concussions per 100 000 athletic exposures (eg, game or practice). Rates of new concussion reporting were higher among high schools in states with medium- or high-strength concussion laws than in states with low-strength concussion laws and in states with at least 1 concussion law revision than in states with no concussion law revisions. Rates of recurrent concussion reporting were lower among high schools in states with ≥2 concussion law revisions than in states with <2 concussion law revisions. Early law adoption was associated with higher rates of new and recurrent concussion reporting, and late law adoption was associated with lower rates of new and recurrent concussion reporting.
CONCLUSIONS: Our findings may help inform legislators of the public health effect of concussion laws.