Publication and non-publication of clinical trials: longitudinal study of applications submitted to a research ethics committee

Elm, Erik von; Röllin, Alexandra; Blümle, Anette; Huwiler, Karin; Witschi, Mark; Egger, Matthias

doi:10.4414/smw.2008.12027

Cited by 43 publications

(12 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By contrast, some features such as the type of observational study (retrospective, prospective, or cross-sectional) and the class of funding agency (US National Institutes of Health, other US Federal agencies, industry, or other) can hardly be associated with publication status. The latter example is particularly surprising as most previous works have reported that the source of funding is a strong indicator of publication status [ 8 , 23 , 50 ], with the exception of Gandhi et al [ 14 ].…”

Section: Resultsmentioning

confidence: 98%

“…To analyze the relationship between a feature and the publication outcome, we applied the chi-square test (in line with the related literature [ 8 , 9 , 14 , 16 , 23 , 49 , 50 ]) but, because of its sensitivity to the sample size [ 51 , 52 ], we also carried out the Cramér V association test for discrete variables. In this analysis, we followed the related work and focused on categorical features only.…”

Section: Resultsmentioning

confidence: 99%

“…The most significant numerical feature is the number of enrolled participants, with a possible explanation being that it may affect the reliability of the results (thus ultimately increasing the odds of publication). Similarly, a larger number of facilities has been linked to higher publication rates [ 8 ]. The number of outcomes indicates the size and complexity of the study, which may in turn also affect publishability.…”

Section: Resultsmentioning

confidence: 99%

“…Estimates of the publication rate of trials vary depending on the medical area and length of the follow-up period. Broadly, publication rates are in the range of 52% to 77% [5][6][7][8]. On the basis of a shorter follow-up period of 30 months from clinical trial completion, the rates tend to be lower, at approximately 11% to 46% [3,6,9].…”

Section: Introductionmentioning

confidence: 99%

“…A variety of factors have been identified as influencing publication outcome, which can be summarized as follows: (1) large clinical trials and those with noncommercial funding are more likely to be published [ 8 , 13 , 23 ]; (2) industry-funded clinical trials are less likely to appear as publications [ 7 ]; (3) the likelihood of publication is associated with the direction and significance of study findings [ 11 , 24 ], although whether to assign this publication bias to rejection by journals or the lack of time and interest by the investigators has been disputed [ 7 ]; (4) place of conduct of the research may affect the odds of publication [ 23 ]; (5) some fields have higher publication rates, for example, neurology and psychiatry [ 13 ] (this may in certain cases be related to the existence of subareas, eg, vascular neurology, with niche journals allowing for easier dissemination [ 25 ]); and (6) lack of time and resources by the authors, and even disagreement between coauthors, have been mentioned as potential factors in the literature [ 26 ] but are not captured directly in the description of clinical trials and, therefore, are difficult to quantify.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Predicting Publication of Clinical Trials Using Structured and Unstructured Data: Model Development and Validation Study

Wang¹,

Šuster²,

Baldwin³

et al. 2022

J Med Internet Res

View full text Add to dashboard Cite

Background Publication of registered clinical trials is a critical step in the timely dissemination of trial findings. However, a significant proportion of completed clinical trials are never published, motivating the need to analyze the factors behind success or failure to publish. This could inform study design, help regulatory decision-making, and improve resource allocation. It could also enhance our understanding of bias in the publication of trials and publication trends based on the research direction or strength of the findings. Although the publication of clinical trials has been addressed in several descriptive studies at an aggregate level, there is a lack of research on the predictive analysis of a trial’s publishability given an individual (planned) clinical trial description. Objective We aimed to conduct a study that combined structured and unstructured features relevant to publication status in a single predictive approach. Established natural language processing techniques as well as recent pretrained language models enabled us to incorporate information from the textual descriptions of clinical trials into a machine learning approach. We were particularly interested in whether and which textual features could improve the classification accuracy for publication outcomes. Methods In this study, we used metadata from ClinicalTrials.gov (a registry of clinical trials) and MEDLINE (a database of academic journal articles) to build a data set of clinical trials (N=76,950) that contained the description of a registered trial and its publication outcome (27,702/76,950, 36% published and 49,248/76,950, 64% unpublished). This is the largest data set of its kind, which we released as part of this work. The publication outcome in the data set was identified from MEDLINE based on clinical trial identifiers. We carried out a descriptive analysis and predicted the publication outcome using 2 approaches: a neural network with a large domain-specific language model and a random forest classifier using a weighted bag-of-words representation of text. Results First, our analysis of the newly created data set corroborates several findings from the existing literature regarding attributes associated with a higher publication rate. Second, a crucial observation from our predictive modeling was that the addition of textual features (eg, eligibility criteria) offers consistent improvements over using only structured data (F1-score=0.62-0.64 vs F1-score=0.61 without textual features). Both pretrained language models and more basic word-based representations provide high-utility text representations, with no significant empirical difference between the two. Conclusions Different factors affect the publication of a registered clinical trial. Our approach to predictive modeling combines heterogeneous features, both structured and unstructured. We show that methods from natural language processing can provide effective textual features to enable more accurate prediction of publication success, which has not been explored for this task previously.

show abstract

Section: Resultsmentioning

confidence: 98%

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Predicting Publication of Clinical Trials Using Structured and Unstructured Data: Model Development and Validation Study

Wang¹,

Šuster²,

Baldwin³

et al. 2022

J Med Internet Res

View full text Add to dashboard Cite

show abstract

Prevalence, Characteristics, and Publication of Discontinued Randomized Trials

Kasenda

Elm

You

et al. 2014

JAMA

314

334

View full text Add to dashboard Cite

show abstract

Sharing Clinical Trial Data

2015

JAMA

149

View full text Add to dashboard Cite

All patients and their relatives want the best information possible regarding the effectiveness and safety of therapies. Responsible sharing of clinical trial data serves this public interest by strengthening the science that is the foundation of safe and effective clinical care. Sharing also fosters sound regulatory decisions, generates new research hypotheses, and increases the scientific knowledge gained from the contributions of clinical trial participants, the efforts of clinical trial investigators, and the resources of clinical trial funders. However, results from about onethird of clinical trials remain unpublished 4 years after trial completion, 1 and much data from trials are never analyzed. Several large pharmaceutical companies 2 and some academic investigators are already sharing clinical trial data, and the European Medicines Agency will do so beginning in 2015. 3 There are compelling reasons to share clinical trial data. The issue now is no longer whether to share clinical trial data but instead what specific data to share, when the data should be shared, and with what controls and

show abstract

Publication and non-publication of clinical trials: longitudinal study of applications submitted to a research ethics committee

Cited by 43 publications

References 42 publications

Predicting Publication of Clinical Trials Using Structured and Unstructured Data: Model Development and Validation Study

Predicting Publication of Clinical Trials Using Structured and Unstructured Data: Model Development and Validation Study

Prevalence, Characteristics, and Publication of Discontinued Randomized Trials

Sharing Clinical Trial Data

Contact Info

Product

Resources

About