Computational frameworks integrating deep learning and statistical models in mining multimodal omics data.

BACKGROUND: In health research, multimodal omics data analysis is widely used to address important clinical and biological questions. Traditional statistical methods rely on the strong assumptions of distribution. Statistical methods such as testing and differential expression are commonly used in omics analysis. Deep learning, on the other hand, is an advanced computer science technique that is powerful in mining high-dimensional omics data for prediction tasks. Recently, integrative frameworks or methods have been developed for omics studies that combine statistical models and deep learning algorithms. METHODS AND RESULTS: The aim of these integrative frameworks is to combine the strengths of both statistical methods and deep learning algorithms to improve prediction accuracy while also providing interpretability and explainability. This review report discusses the current state-of-the-art integrative frameworks, their limitations, and potential future directions in survival and time-to-event longitudinal analysis, dimension reduction and clustering, regression and classification, feature selection, and causal and transfer learning.

iNGNN-DTI: prediction of drug-target interaction with interpretable nested graph neural network and pretrained molecule models.

MOTIVATION: Drug-target interaction (DTI) prediction aims to identify interactions between drugs and protein targets. Deep learning can automatically learn discriminative features from drug and protein target representations for DTI prediction, but challenges remain, making it an open question. Existing approaches encode drugs and targets into features using deep learning models, but they often lack explanations for underlying interactions. Moreover, limited labeled DTIs in the chemical space can hinder model generalization. RESULTS: We propose an interpretable nested graph neural network for DTI prediction (iNGNN-DTI) using pre-trained molecule and protein models. The analysis is conducted on graph data representing drugs and targets by using a specific type of nested graph neural network, in which the target graphs are created based on 3D structures using Alphafold2. This architecture is highly expressive in capturing substructures of the graph data. We use a cross-attention module to capture interaction information between the substructures of drugs and targets. To improve feature representations, we integrate features learned by models that are pre-trained on large unlabeled small molecule and protein datasets, respectively. We evaluate our model on three benchmark datasets, and it shows a consistent improvement on all baseline models in all datasets. We also run an experiment with previously unseen drugs or targets in the test set, and our model outperforms all of the baselines. Furthermore, the iNGNN-DTI can provide more insights into the interaction by visualizing the weights learned by the cross-attention module. AVAILABILITY AND IMPLEMENTATION: The source code of the algorithm is available at https://github.com/syan1992/iNGNN-DTI.

Evaluating in vivo effectiveness of sotrovimab for the treatment of Omicron subvariant BA.2 versus BA.1: a multicentre, retrospective cohort study.

BACKGROUND: In vitro data suggested reduced neutralizing capacity of sotrovimab, a monoclonal antibody, against Omicron BA.2 subvariant. However, limited in vivo data exist regarding clinical effectiveness of sotrovimab for coronavirus disease 2019 (COVID-19) due to Omicron BA.2. METHODS: A multicentre, retrospective cohort study was conducted at three Canadian academic tertiary centres. Electronic medical records were reviewed for patients ≥ 18 years with mild COVID-19 (sequencing-confirmed Omicron BA.1 or BA.2) treated with sotrovimab between February 1 to April 1, 2022. Thirty-day co-primary outcomes included hospitalization due to moderate or severe COVID-19; all-cause intensive care unit (ICU) admission, and all-cause mortality. Risk differences (BA.2 minus BA.1 group) for co-primary outcomes were adjusted with propensity score matching (e.g., age, sex, vaccination, immunocompromised status). RESULTS: Eighty-five patients were included (15 BA.2, 70 BA.1) with similar baseline characteristics between groups. Adjusted risk differences were non-statistically significant between groups for 30-day hospitalization (- 14.3%; 95% confidence interval (CI): - 32.6 to 4.0%), ICU admission (- 7.1%; 95%CI: - 20.6 to 6.3%), and mortality (- 7.1%; 95%CI: - 20.6 to 6.3%). CONCLUSIONS: No differences were demonstrated in hospitalization, ICU admission, or mortality rates within 30 days between sotrovimab-treated patients with BA.1 versus BA.2 infection. More real-world data may be helpful to properly assess sotrovimab's effectiveness against infections due to specific emerging COVID-19 variants.

More Than a Decade Since the Latest CONSORT Non-inferiority Trials Extension: Do Infectious Diseases Trials Do Enough?

More than a decade after the Consolidated Standards of Reporting Trials group released a reporting items checklist for non-inferiority randomized controlled trials, the infectious diseases literature continues to underreport these items. Trialists, journals, and peer reviewers should redouble their efforts to ensure infectious diseases studies meet these minimum reporting standards.

Respiratory viral infections including COVID-19 in solid organ transplantation.

PURPOSE OF REVIEW: Respiratory viral infections are prevalent and contribute to significant morbidity and mortality among solid organ transplant (SOT) recipients. We review updates from literature on respiratory viruses, including severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), in the SOT recipient. RECENT FINDINGS: With the wider availability and use of molecular diagnostic tests, our understanding of the epidemiology and impact of respiratory viruses in the SOT population continues to expand. While considerable attention has been given to the coronavirus disease 2019 (COVID-19) pandemic, the advances in prevention and treatment strategies of SARS-CoV-2 offered valuable insights into the development of new therapeutic options for managing other respiratory viruses in both the general and SOT population. SUMMARY: Respiratory viruses can present with a diverse range of symptoms in SOT recipients, with potentially associated acute rejection and chronic lung allograft dysfunction in lung transplant recipients. The epidemiology, clinical presentations, diagnostic approaches, and treatment and preventive strategies for clinically significant RNA and DNA respiratory viruses in SOT recipients are reviewed. This review also covers novel antivirals, immunologic therapies, and vaccines in development for various community-acquired respiratory viruses.

A self-knowledge distillation-driven CNN-LSTM model for predicting disease outcomes using longitudinal microbiome data.

Motivation: Human microbiome is complex and highly dynamic in nature. Dynamic patterns of the microbiome can capture more information than single point inference as it contains the temporal changes information. However, dynamic information of the human microbiome can be hard to be captured due to the complexity of obtaining the longitudinal data with a large volume of missing data that in conjunction with heterogeneity may provide a challenge for the data analysis. Results: We propose using an efficient hybrid deep learning architecture convolutional neural network-long short-term memory, which combines with self-knowledge distillation to create highly accurate models to analyze the longitudinal microbiome profiles to predict disease outcomes. Using our proposed models, we analyzed the datasets from Predicting Response to Standardized Pediatric Colitis Therapy (PROTECT) study and DIABIMMUNE study. We showed the significant improvement in the area under the receiver operating characteristic curve scores, achieving 0.889 and 0.798 on PROTECT study and DIABIMMUNE study, respectively, compared with state-of-the-art temporal deep learning models. Our findings provide an effective artificial intelligence-based tool to predict disease outcomes using longitudinal microbiome profiles from collected patients. Availability and implementation: The data and source code can be accessed at https://github.com/darylfung96/UC-disease-TL.

Methodological and Reporting Quality of Noninferiority Randomized Controlled Trials Comparing Antiretroviral Therapies: A Systematic Review.

BACKGROUND: It is unclear whether the reporting quality of antiretroviral (ARV) noninferiority (NI) randomized controlled trials (RCTs) has improved since the CONSORT guideline release in 2006. The primary objective of this systematic review was assessing the methodological and reporting quality of ARV NI-RCTs. We also assessed reporting quality by funding source and publication year. METHODS: We searched Medline, Embase, and Cochrane Central from inception to 14 November 2022. We included NI-RCTs comparing ≥2 ARV regimens used for human immunodeficiency virus treatment or prophylaxis. We used the Cochrane Risk of Bias 2.0 tool to assess risk of bias. Screening and data extraction were performed blinded and in duplicate. Descriptive statistics were used to summarize data; statistical tests were 2 sided, with significance defined as P < .05. The systematic review was prospectively registered (PROSPERO CRD42022328586), and not funded. RESULTS: We included 160 articles reporting 171 trials. Of these articles, 101 (63.1%) did not justify the NI margin used, and 28 (17.5%) did not provide sufficient information for sample size calculation. Eighty-nine of 160 (55.6%) reported both intention-to-treat and per-protocol analyses, while 118 (73.8%) described missing data handling. Ten of 171 trials (5.9%) reported potentially misleading results. Pharmaceutical industry-funded trials were more likely to be double-blinded (28.1% vs 10.3%; P = .03) and to describe missing data handling (78.5% vs 59.0%; P = .02). The overall risk of bias was low in 96 of 160 studies (60.0%). CONCLUSIONS: ARV NI-RCTs should improve NI margin justification, reporting of intention-to-treat and per-protocol analyses, and missing data handling to increase CONSORT adherence.

Segmentation-Assisted Fully Convolutional Neural Network Enhances Deep Learning Performance to Identify Proliferative Diabetic Retinopathy.

With the progression of diabetic retinopathy (DR) from the non-proliferative (NPDR) to proliferative (PDR) stage, the possibility of vision impairment increases significantly. Therefore, it is clinically important to detect the progression to PDR stage for proper intervention. We propose a segmentation-assisted DR classification methodology, that builds on (and improves) current methods by using a fully convolutional network (FCN) to segment retinal neovascularizations (NV) in retinal images prior to image classification. This study utilizes the Kaggle EyePacs dataset, containing retinal photographs from patients with varying degrees of DR (mild, moderate, severe NPDR and PDR. Two graders annotated the NV (a board-certified ophthalmologist and a trained medical student). Segmentation was performed by training an FCN to locate neovascularization on 669 retinal fundus photographs labeled with PDR status according to NV presence. The trained segmentation model was used to locate probable NV in images from the classification dataset. Finally, a CNN was trained to classify the combined images and probability maps into categories of PDR. The mean accuracy of segmentation-assisted classification was 87.71% on the test set (SD = 7.71%). Segmentation-assisted classification of PDR achieved accuracy that was 7.74% better than classification alone. Our study shows that segmentation assistance improves identification of the most severe stage of diabetic retinopathy and has the potential to improve deep learning performance in other imaging problems with limited data availability.

Comparison of Preprint Postings of Randomized Clinical Trials on COVID-19 and Corresponding Published Journal Articles: A Systematic Review.

Importance: Randomized clinical trials (RCTs) on COVID-19 are increasingly being posted as preprints before publication in a scientific, peer-reviewed journal. Objective: To assess time to journal publication for COVID-19 RCT preprints and to compare differences between pairs of preprints and corresponding journal articles. Evidence Review: This systematic review used a meta-epidemiologic approach to conduct a literature search using the World Health Organization COVID-19 database and Embase to identify preprints published between January 1 and December 31, 2021. This review included RCTs with human participants and research questions regarding the treatment or prevention of COVID-19. For each preprint, a literature search was done to locate the corresponding journal article. Two independent reviewers read the full text, extracted data, and assessed risk of bias using the Cochrane Risk of Bias 2 tool. Time to publication was analyzed using a Cox proportional hazards regression model. Differences between preprint and journal article pairs in terms of outcomes, analyses, results, or conclusions were described. Statistical analysis was performed on October 17, 2022. Findings: This study included 152 preprints. As of October 1, 2022, 119 of 152 preprints (78.3%) had been published in journals. The median time to publication was 186 days (range, 17-407 days). In a multivariable model, larger sample size and low risk of bias were associated with journal publication. With a sample size of less than 200 as the reference, sample sizes of 201 to 1000 and greater than 1000 had hazard ratios (HRs) of 1.23 (95% CI, 0.80-1.91) and 2.19 (95% CI, 1.36-3.53) for publication, respectively. With high risk of bias as the reference, medium-risk articles with some concerns for bias had an HR of 1.77 (95% CI, 1.02-3.09); those with a low risk of bias had an HR of 3.01 (95% CI, 1.71-5.30). Of the 119 published preprints, there were differences in terms of outcomes, analyses, results, or conclusions in 65 studies (54.6%). The main conclusion in the preprint contradicted the conclusion in the journal article for 2 studies (1.7%). Conclusions and Relevance: These findings suggest that there is a substantial time lag from preprint posting to journal publication. Preprints with smaller sample sizes and high risk of bias were less likely to be published. Finally, although differences in terms of outcomes, analyses, results, or conclusions were observed for preprint and journal article pairs in most studies, the main conclusion remained consistent for the majority of studies.

Prognostic factors for streptococcal toxic shock syndrome: systematic review and meta-analysis.

OBJECTIVES: To quantify the prognostic effects of demographic and modifiable factors in streptococcal toxic shock syndrome (STSS). DESIGN: Systematic review and meta-analysis. DATA SOURCES: MEDLINE, EMBASE and CINAHL from inception to 19 September 2022, along with citations of included studies. ELIGIBILITY CRITERIA: Pairs of reviewers independently screened potentially eligible studies of patients with Group A Streptococcus-induced STSS that quantified the association between at least one prognostic factor and outcome of interest. DATA EXTRACTION AND SYNTHESIS: We performed random-effects meta-analysis after duplicate data extraction and risk of bias assessments. We rated the certainty of evidence using the Grading of Recommendations, Assessment, Development and Evaluation approach. RESULTS: One randomised trial and 40 observational studies were eligible (n=1918 patients). We found a statistically significant association between clindamycin treatment and mortality (n=144; OR 0.14, 95% CI 0.06 to 0.37), but the certainty of evidence was low. Within clindamycin-treated STSS patients, we found a statistically significant association between intravenous Ig treatment and mortality (n=188; OR 0.34, 95% CI 0.15 to 0.75), but the certainty of evidence was also low. The odds of mortality may increase in patients ≥65 years when compared with patients 18-64 years (n=396; OR 2.37, 95% CI 1.47 to 3.84), but the certainty of evidence was low. We are uncertain whether non-steroidal anti-inflammatory drugs increase the odds of mortality (n=50; OR 4.14, 95% CI 1.13 to 15.14; very low certainty). Results failed to show a significant association between any other prognostic factor and outcome combination (very low to low certainty evidence) and no studies quantified the association between a prognostic factor and morbidity post-infection in STSS survivors. CONCLUSIONS: Treatment with clindamycin and within clindamycin-treated patients, IVIG, was each significantly associated with mortality, but the certainty of evidence was low. Future research should focus on morbidity post-infection in STSS survivors. PROSPERO REGISTRATION NUMBER: CRD42020166961.

The effect of disease co-occurrence measurement on multimorbidity networks: a population-based study.

BACKGROUND: Network analysis, a technique for describing relationships, can provide insights into patterns of co-occurring chronic health conditions. The effect that co-occurrence measurement has on disease network structure and resulting inferences has not been well studied. The purpose of the study was to compare structural differences among multimorbidity networks constructed using different co-occurrence measures. METHODS: A retrospective cohort study was conducted using four fiscal years of administrative health data (2015/16 - 2018/19) from the province of Manitoba, Canada (population 1.5 million). Chronic conditions were identified using diagnosis codes from electronic records of physician visits, surgeries, and inpatient hospitalizations, and grouped into categories using the Johns Hopkins Adjusted Clinical Group (ACG) System. Pairwise disease networks were separately constructed using each of seven co-occurrence measures: lift, relative risk, phi, Jaccard, cosine, Kulczynski, and joint prevalence. Centrality analysis was limited to the top 20 central nodes, with degree centrality used to identify potentially influential chronic conditions. Community detection was used to identify disease clusters. Similarities in community structure between networks was measured using the adjusted Rand index (ARI). Network edges were described using disease prevalence categorized as low (< 1%), moderate (1 to < 7%), and high (≥7%). Network complexity was measured using network density and frequencies of nodes and edges. RESULTS: Relative risk and lift highlighted co-occurrences between pairs of low prevalence health conditions. Kulczynski emphasized relationships between high and low prevalence conditions. Joint prevalence focused on highly-prevalent conditions. Phi, Jaccard, and cosine emphasized associations involving moderately prevalent conditions. Co-occurrence measurement differences significantly affected the number and structure of identified disease clusters. When limiting the number of edges to produce visually interpretable graphs, networks had significant dissimilarity in the percentage of co-occurrence relationships in common, and in their selection of the highest-degree nodes. CONCLUSIONS: Multimorbidity network analyses are sensitive to disease co-occurrence measurement. Co-occurrence measures should be selected considering their intrinsic properties, research objectives, and the health condition prevalence relationships of greatest interest. Researchers should consider conducting sensitivity analyses using different co-occurrence measures.

What Is the Optimal Follow-up Length for Mortality in Staphylococcus aureus Bacteremia? Observations From a Systematic Review of Attributable Mortality.

Background: Deaths following Staphylococcus aureus bacteremia (SAB) may be related or unrelated to the infection. In SAB therapeutics research, the length of follow-up should be optimized to capture most attributable deaths and minimize nonattributable deaths. We performed a secondary analysis of a systematic review to describe attributable mortality in SAB over time. Methods: We systematically searched Medline, Embase, and Cochrane Database of Systematic Reviews from 1 January 1991 to 7 May 2021 for human observational studies of SAB. To be included in this secondary analysis, the study must have reported attributable mortality. Two reviewers extracted study data and assessed risk of bias independently. Pooling of study estimates was not performed due to heterogeneity in the definition of attributable deaths. Results: Twenty-four observational cohort studies were included. The median proportion of all-cause deaths that were attributable to SAB was 77% (interquartile range [IQR], 72%-89%) at 1 month and 62% (IQR, 58%-75%) at 3 months. At 1 year, this proportion was 57% in 1 study. In 2 studies that described the rate of increase in mortality over time, 2-week follow-up captured 68 of 79 (86%) and 48 of 57 (84%) attributable deaths that occurred by 3 months. By comparison, 1-month follow-up captured 54 of 57 (95%) and 56 of 60 (93%) attributable deaths that occurred by 3 months in 2 studies. Conclusions: The proportion of deaths that are attributable to SAB decreases as follow-up lengthens. Follow-up duration between 1 and 3 months seems optimal if evaluating processes of care that impact SAB mortality. Clinical Trials Registration: PROSPERO CRD42021253891.

How Generalizable Are Randomized Controlled Trials (RCTs) in Staphylococcus aureus Bacteremia? A Description of the Mortality Gap Between RCTs and Observational Studies.

In Staphylococcus aureus bacteremia, mortality rates in randomized controlled trials (RCTs) are consistently lower than observational studies. Stringent eligibility criteria and omission of early deaths in RCTs contribute to this mortality gap. Clinicians should acknowledge the possibility of a lower treatment effect when applying RCT results to bedside care.

Staphylococcus aureus bacteraemia mortality: a systematic review and meta-analysis.

BACKGROUND: Precise estimates of mortality in Staphylococcus aureus bacteraemia (SAB) are important to convey prognosis and guide the design of interventional studies. OBJECTIVES: We performed a systematic review and meta-analysis to estimate all-cause mortality in SAB and explore mortality change over time. DATA SOURCES: The MEDLINE and Embase databases, as well as the Cochrane Database of Systematic Reviews, were searched from January 1, 1991 to May 7, 2021. STUDY ELIGIBILITY CRITERIA: Human observational studies on patients with S. aureus bloodstream infection were included. PARTICIPANTS: The study analyzed data of patients with a positive blood culture for S. aureus. METHODS: Two independent reviewers extracted study data and assessed risk of bias using the Newcastle-Ottawa Scale. A generalized, linear, mixed random effects model was used to pool estimates. RESULTS: A total of 341 studies were included, describing a total of 536,791 patients. From 2011 onward, the estimated mortality was 10.4% (95% CI, 9.0%-12.1%) at 7 days, 13.3% (95% CI, 11.1%-15.8%) at 2 weeks, 18.1% (95% CI, 16.3%-20.0%) at 1 month, 27.0% (95% CI, 21.5%-33.3%) at 3 months, and 30.2% (95% CI, 22.4%-39.3%) at 1 year. In a meta-regression model of 1-month mortality, methicillin-resistant S. aureus had a higher mortality rate (adjusted OR (aOR): 1.04; 95% CI, 1.02-1.06 per 10% increase in methicillin-resistant S. aureus proportion). Compared with prior to 2001, more recent time periods had a lower mortality rate (aOR: 0.88; 95% CI, 0.75-1.03 for 2001-2010; aOR: 0.82; 95% CI, 0.69-0.97 for 2011 onward). CONCLUSIONS: SAB mortality has decreased over the last 3 decades. However, more than one in four patients will die within 3 months, and continuous improvement in care remains necessary.

Mogamulizumab efficacy is underscored by its associated rash that mimics cutaneous T-cell lymphoma: a retrospective single-centre case series.

BACKGROUND: Mogamulizumab is a humanized antibody against chemokine receptor type 4. It was recently approved by the US Food and Drug Administration for relapsed or refractory mycosis fungoides (MF) and Sézary syndrome (SS). The most commonly reported adverse event in the phase III licensing trial was drug eruption (28%), now termed mogamulizumab-associated rash (MAR). Clinical recommendations about MAR and its treatment differ between the current package insert and postapproval insights reported from two single-centre studies that focused on its characterization, but less so on outcomes and clinicopathological differentiation from cutaneous T-cell lymphoma (CTCL). OBJECTIVES: To describe our experience in the diagnosis of MAR and treatment of patients with CTCL with mogamulizumab. METHODS: This is a single-centre retrospective case series study. RESULTS: We found a higher incidence of MAR in patients with CTCL (17 of 24, 68%) than previously reported. MAR development is associated with complete (11 of 17) or partial (four of 17) responses, with an overall response rate of 88%, compared with 29% (two of seven) in patients without MAR. Diagnosis of MAR may be obscured by its ability to mimic key CTCL features both clinically and histologically, but an absence of T-cell-receptor clonality and relatively decreased CD4 : CD8 ratio compared with baseline lesions strongly favour MAR over recurrent disease. CONCLUSIONS: MAR has the potential to create a significant management problem for patients on mogamulizumab. Misidentification of MAR as recurrent CTCL may detrimentally result in the premature discontinuation of mogamulizumab in patients whose disease is historically hard to treat. Thorough clinicopathological investigation of new lesions during treatment with mogamulizumab is required to inform ideal treatment decisions and achieve better outcomes.

A scoping review of preprocessing methods for unstructured text data to assess data quality.

Introduction: Unstructured text data (UTD) are increasingly found in many databases that were never intended to be used for research, including electronic medical record (EMR) databases. Data quality can impact the usefulness of UTD for research. UTD are typically prepared for analysis (i.e., preprocessed) and analyzed using natural language processing (NLP) techniques. Different NLP methods are used to preprocess UTD and may affect data quality. Objective: Our objective was to systematically document current research and practices about NLP preprocessing methods to describe or improve the quality of UTD, including UTD found in EMR databases. Methods: A scoping review was undertaken of peer-reviewed studies published between December 2002 and January 2021. Scopus, Web of Science, ProQuest, and EBSCOhost were searched for literature relevant to the study objective. Information extracted from the studies included article characteristics (i.e., year of publication, journal discipline), data characteristics, types of preprocessing methods, and data quality topics. Study data were presented using a narrative synthesis. Results: A total of 41 articles were included in the scoping review; over 50% were published between 2016 and 2021. Almost 20% of the articles were published in health science journals. Common preprocessing methods included removal of extraneous text elements such as stop words, punctuation, and numbers, word tokenization, and parts of speech tagging. Data quality topics for articles about EMR data included misspelled words, security (i.e., de-identification), word variability, sources of noise, quality of annotations, and ambiguity of abbreviations. Conclusions: Multiple NLP techniques have been proposed to preprocess UTD, with some differences in techniques applied to EMR data. There are similarities in the data quality dimensions used to characterize structured data and UTD. While a few general-purpose measures of data quality that do not require external data; most of these focus on the measurement of noise.

Secondary Analysis of a Systematic Review: Are Antifungal Noninferiority Trials at Risk of Eroding Effectiveness Because of Biocreep?

Noninferiority randomized controlled trial (RCT) effectiveness may erode when results favor the active control over time and when a decreasingly effective control arm is used in serial trials. We analyzed 32 antifungal noninferiority RCTs (NI-RCTs) for these scenarios in this secondary analysis of a systematic review. Our exploratory analysis suggests that the erosion risk in the effectiveness of antifungal noninferiority trials is uncommon. Findings are limited by small sample size and overall risk of bias.

Methodological and reporting quality of non-inferiority randomized controlled trials comparing antifungal therapies: a systematic review.

BACKGROUND: Detailed reporting is essential in non-inferiority randomized controlled trials (NI-RCTs) to assess evidence quality, as these trials inform standards of care. OBJECTIVES: The primary objective was to evaluate the methodological and reporting quality of antifungal NI-RCTs. DATA SOURCES: Medline, EMBASE, the Cochrane CENTRAL and the United States Federal Drug Administration (FDA) drugs database were searched to 9 September 2020. STUDY ELIGIBILITY CRITERIA: NI-RCTs differing by antifungal formulation, type, dose, administration and/or duration were included. Articles were independently assessed in duplicate using quality indicators developed by the Consolidated Standards of Reporting Trials (CONSORT) group. PARTICIPANTS: Patients enrolled in antifungal trials for prophylactic and therapeutic use. METHODS: The Cochrane RoB 2.0 tool was used to assess risk of bias. Descriptive statistics were used; all statistical tests were two sided. RESULTS: Of 32 included studies, 22 (68.7%) did not justify the NIM. Handling of missing data was not described in 20 (62.5%). Intention-to-treat (ITT) and per-protocol (PP) analyses were both reported in 12/32 (37.5%) studies. Eleven of 32 studies (34.3%) reported potentially misleading conclusions. Industry-financed studies were more likely to report only the ITT analysis (n = 14/27, 51.9%). Methodological and reporting quality was unaffected by publication year; risk of bias from missing data changed over time. Overall risk of bias across included studies was moderate to high, with high risk in randomization process (n = 8/32, 25%), missing outcome data (n = 5/32, 15.6%), and selection of reported result (n = 9/32, 28.1%). CONCLUSIONS: Justification of the non-inferiority margin, reporting of ITT and PP analyses, missing data handling description, and ensuring conclusions are consistent with reported data is necessary to improve CONSORT adherence. Small sample size and overall risk of bias are study limitations. (Systematic Review Registration Number PROSPERO CRD42020219497).

In patients hospitalized for COVID-19, tocilizumab reduces mortality at 28 d.

SOURCE CITATION: WHO Rapid Evidence Appraisal for COVID-19 Therapies (REACT) Working Group; Shankar-Hari M, Vale CL, Godolphin PJ, et al. Association between administration of IL-6 antagonists and mortality among patients hospitalized for COVID-19: a meta-analysis. JAMA. 2021;326:499-518. 34228774.