Dataset Information

Assessment of Consistency Between Peer-Reviewed Publications and Clinical Trial Registries.

ABSTRACT:

Importance

Clinical trial registries are intended to increase clinical research transparency by nonselectively identifying and documenting clinical trial designs and outcomes. Inconsistencies in reported data undermine the utility of such registries and have previously been noted in general medical literature.

Objective

To assess whether inconsistencies in reported data exist between ophthalmic literature and clinical trial registries.

Design, setting, and participants

In this retrospective, cross-sectional study, interventional clinical trials published from January 1, 2014, to December 31, 2014, in the American Journal of Ophthalmology, JAMA Ophthalmology, and Ophthalmology were reviewed. Observational, retrospective, uncontrolled, and post hoc reports were excluded, yielding a sample size of 106 articles. Data collection was performed from January through September 2016. Data review and adjudication continued through January 2017.

Main outcomes and measures

If possible, articles were matched to registry entries listed in the ClinicalTrials.gov database or in 1 of 16 international registries indexed by the World Health Organization International Clinical Trials Registry Platform version 3.2 search engine. Each article-registry pair was assessed for inconsistencies in design, results, and funding (each of which was further divided into subcategories) by 2 reviewers and adjudicated by a third.

Results

Of 106 trials that met the study criteria, matching registry entries were found for 68 (64.2%), whereas no matching registry entries were found for 38 (35.8%). Inconsistencies were identified in study design, study results, and funding sources, including specific interventions in 8 (11.8%), primary outcome measure (POM) designs in 32 (47.1%), and POM results in 48 (70.6%). In addition, numerous data pieces were unreported, including analysis methods in 52 (76.5%) and POM results in 38 (55.9%).

Conclusions and relevance

Clinical trial registries were underused in this sample of ophthalmology clinical trials. For studies with registry data, inconsistency rates between published and registered data were similar to those previously reported for general medical literature. In most cases, inconsistencies involved missing data, but explicit discrepancies in methods and/or data were also found. Transparency and credibility of published trials may be improved by closer attention to their registration and reporting.

SUBMITTER: Sun LW

PROVIDER: S-EPMC6512264 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Study Design: Scoping review Objective: To study the design, clinical setting and outcome measures used in spinal cord injury rehabilitation publications. Methods: A literature search on PubMed and Medline was conducted focusing on articles published between 1990-2016 and using "traumatic SCI", "functional outcomes", "rehabilitation", "work" and "return to work" as outcomes. Studies were categorized based on design (intervention, including RCTs vs. non-intervention studies), settings (inpatient vs. outpatient vs. transition), and outcome measures used (impairment vs. function vs. participation/integration vs. quality of life vs. symptoms). Work-related studies were categorized independently. Results: Five hundred forty-four articles met the inclusion criteria. Of these, 234 were interventional studies, including 23 RCTs. Studies were evenly divided among inpatient, outpatient and transition settings. Of the 234 interventional studies, 143 used functional evaluations. Sixty-one different functional instruments were used, with a predominant use of the Functional Independence Measure (61 times) and an additional use of SCI-specific measures, i.e. Spinal Cord Independence Measure and Craig Handicap Assessment and Reporting Technique (13 times each). Fifty-one studies measured mobility, while only three measured hand functions. The work-related sub-analysis revealed 32 intervention studies (no RCTs), of which 15 used functional evaluations and only three focused on tetraplegia. Conclusion: Our study revealed a paucity of intervention trials and RCTs, indicating a dearth of knowledge that would be needed to establish evidence-based practice guidelines. This is particularly true for tetraplegia. While standard measures of function were frequently used, providing valuable data, there is no consensus about what exact outcome measure to use. Using newer measurement techniques, for instance based on the application of item response theory, should be considered to enhance uniformity.

Project description:BackgroundOutdoor air pollution is a major threat to global public health that needs responsible participation of researchers at all levels. Assessing research output is an important step in highlighting national and international contribution and collaboration in a certain field. Therefore, the aim of this study was to analyze globally-published literature in outdoor air pollution - related respiratory health.MethodOutdoor air pollution documents related to respiratory health were retrieved from Scopus database. The study period was up to 2017. Mapping of author keywords was carried out using VOSviewer 1.6.6.ResultsSearch query yielded 3635 documents with an h-index of 137. There was a dramatic increase in the number of publications in the last decade of the study period. The most frequently encountered author keywords were: air pollution (835 occurrences), asthma (502 occurrences), particulate matter (198 occurrences), and children (203 occurrences). The United States of America ranked first (1082; 29.8%) followed by the United Kingdom (279; 7.7%) and Italy (198; 5.4%). Annual research productivity stratified by income and population size indicated that China ranked first (22.2) followed by the USA (18.8). Analysis of regional distribution of publications indicated that the Mediterranean, African, and South-East Asia regions had the least contribution. Harvard University (92; 2.5%) was the most active institution/organization followed the US Environmental Protection Agency (89; 2.4%). International collaboration was restricted to three regions: Northern America, Europe, and Asia. The top ten preferred journals were in the field of environmental health and respiratory health. Environmental Health Perspective was the most preferred journal for publishing documents in outdoor pollution in relation to respiratory health.ConclusionResearch on the impact of outdoor air pollution on respiratory health had accelerated lately and is receiving a lot of interest. Global research networks that include countries with high level of pollution and limited resources are highly needed to create public opinion in favor of minimizing outdoor air pollution and investing in green technologies.

Project description:ObjectivesWomen in oral health science face similar societal issues and challenges as those in other STEMM careers, and gender disparities continue to exist as evidenced by fewer women represented as first and last authors in scientific publications. Pre-prints may serve as a conduit to immediately disseminating one's work, bypassing the arduous peer review process and its associated inherent biases. Therefore, the purpose of this study was to 1] compare the gender of first and last authors in pre-print versus peer reviewed publications, 2] examine the composition of first and last author pairs as stratified by publication type, and 3] examine the correlation between woman authorship and institutional geographic location and publication metrics stratified by publication type.MethodsThe keyword "oral health" was used to search for publications in BioRxiv and Pubmed in the years 2018 and 2019. Gender of first and last authors were determined, and its frequency was considered as the primary outcome. Additionally, the geographic location of the author's associated institution and publication metrics measured by Altmetrics score were extracted. Data was descriptively summarized by frequencies and percentages. Chi-square analysis was conducted for categorical variables which included the relationship between gender and publication type as well as gender and region of author's associated institution. Binomial regression analysis was conducted to analyze the relationship between gender and Altmetrics.ResultsWoman first authors comprised 40.3% of pre-prints and 64.5% of peer reviewed publications [p<0.05]. Woman last authors comprised 31.3% of pre-prints and 61.5% of peer reviewed publications [p<0.05]. When analyzing the relationships between first and last author, the Man-Man pairing represented 47.7% of the pre-print publications and the Woman-Woman pairing comprised a majority of the of the peer review publications at 47.5%. All results were statistically significant with a p-value <0.05. No significant correlation was found between region of institution or Altmetrics and gender of first or last authors [p>0.05].ConclusionFor the first time in oral health science, it was found that women show higher representation as first and last author positions in peer reviewed publications versus pre-prints.

Project description:BackgroundQuantitative data are lacking on published malaria research. The purpose of the study is to characterize trends in malaria-related literature from 1990 to 2009 in 11 Asian-Pacific countries that are committed to malaria elimination as a national goal.MethodsA systematic search was conducted for articles published from January 1990 to December 2009 in PubMed/MEDLINE using terms for malaria and 11 target countries (Bhutan, China, North Korea, Indonesia, Malaysia, Philippines, Solomon Islands, South Korea, Sri Lanka, Thailand and Vanuatu). The references were collated and categorized according to subject, Plasmodium species, and whether they contained original or derivative data.Results2,700 articles published between 1990 and 2009 related to malaria in the target countries. The annual output of malaria-related papers increased linearly whereas the overall biomedical output from these countries grew exponentially. The percentage of malaria-related publications was nearly 3% (111/3741) of all biomedical publications in 1992 and decreased to less than 1% (118/12171; p < 0.001) in 2009. Thailand had the highest absolute output of malaria-related papers (n = 1211), followed by China (n = 609) and Indonesia (n = 346). Solomon Islands and Vanuatu had lower absolute numbers of publications, but both countries had the highest number of publications per capita (1.3 and 2.5 papers/1,000 population). The largest percentage of papers concerned the epidemiology and control of malaria (53%) followed by studies of drugs and drug resistance (47%). There was an increase in the proportion of articles relating to epidemiology, entomology, biology, molecular biology, pathophysiology and diagnostics from the first to the second decade, whereas the percentage of papers on drugs, clinical aspects of malaria, immunology, and social sciences decreased.ConclusionsThe proportion of malaria-related publications out of the overall biomedical output from the 11 target Asian-Pacific countries is decreasing. The discovery and evaluation of new, safe and effective drugs and vaccines is paramount. In addition the elimination of malaria will require operational research to implement and scale up interventions.

Project description:BackgroundThere are increasing calls for patient involvement in sharing health research results, but no evidence-based recommendations to guide such involvement. Our objectives were to: (1) conduct a systematic review of the evidence on patient involvement in results sharing, (2) propose evidence-based recommendations to help maximize benefits and minimize risks of such involvement and (3) conduct this project with patient authors.MethodsTo avoid research waste, we verified that no systematic reviews were registered or published on this topic. We co-created, with patients, a PRISMA-P-compliant protocol. We included peer-reviewed publications reporting the effects of patient involvement in preparing peer-reviewed publications or results summaries from health research studies. We searched (9/10/2017) MEDLINE, EMBASE and the Cochrane Database of Systematic Reviews, and secondary information sources (until 11/06/2018). We assessed the risk of bias in eligible publications and extracted data using standardized processes. To evaluate patient involvement in this project, we co-created a Patient Authorship Experience Tool.ResultsAll nine eligible publications reported on patient involvement in preparing publications; none on preparing results summaries. Evidence quality was moderate. A qualitative synthesis of evidence indicated the benefits of patient involvement may outweigh the risks. We have proposed 21 evidence-based recommendations to help maximize the benefits and minimize the risks when involving patients as authors of peer-reviewed publications. The recommendations focus on practical actions patient and non-patient authors can take before (10 recommendations), during (7 recommendations) and after (4 recommendations) manuscript development. Using the Patient Authorship Experience Tool, both patient and non-patient authors rated their experience highly.ConclusionsBased on a systematic review, we have proposed 21 evidence-based recommendations to help maximize the benefits and minimize the risks of involving patients as authors of peer-reviewed publications.