3 years ago

Unfounded conclusions of equivalence in diagnostic accuracy studies: a pervasive fallacy of inference in pediatric radiology scientific abstracts

Lauren A. May, Robert C. Orth, R. Paul Guillerman, Wei Zhang

Abstract

Background

In studies of diagnostic performance that fail to detect a statistically significant difference between compared techniques, investigators often declare evidence of equivalence or similarity without having actually tested that assertion due to incorrect methodology or insufficient statistical power.

Objective

The purpose of our investigation is to measure the prevalence of unfounded assertions of equivalence or similarity in comparison studies presented at the International Pediatric Radiology (IPR) meeting of 2016 and promote awareness of this fallacy of inference to the pediatric radiology community.

Materials and methods

Two pediatric radiologists independently reviewed the methodology and reporting quality of the 194 scientific paper abstracts from the 2016 IPR meeting. All comparison studies were identified and those failing to detect a statistically significance difference and making a claim of equivalence or similarity in the results or conclusion were assessed for a description of the study design type, statistical power and sample size estimator calculation.

Results

Of 194 scientific paper abstracts, 112 (58%) were comparison studies. Of these, 36/112 (32%) made unfounded inferences of equivalence or similarity in diagnostic imaging performance. No study had an equivalence or non-inferiority design. No abstract specified the statistical power of the study, and only one abstract acknowledged a small sample size as a limitation in detecting a statistically significant difference.

Conclusion

Inadequate reporting and unfounded inferences of equivalence or similarity were common in diagnostic performance comparison studies presented at IPR 2016. Failure to recognize these limitations could have adverse consequences by leading to the adoption of unvalidated imaging techniques.

Publisher URL: https://link.springer.com/article/10.1007/s00247-018-4222-9

DOI: 10.1007/s00247-018-4222-9

You might also like
Discover & Discuss Important Research

Keeping up-to-date with research can feel impossible, with papers being published faster than you'll ever be able to read them. That's where Researcher comes in: we're simplifying discovery and making important discussions happen. With over 19,000 sources, including peer-reviewed journals, preprints, blogs, universities, podcasts and Live events across 10 research areas, you'll never miss what's important to you. It's like social media, but better. Oh, and we should mention - it's free.

  • Download from Google Play
  • Download from App Store
  • Download from AppInChina

Researcher displays publicly available abstracts and doesn’t host any full article content. If the content is open access, we will direct clicks from the abstracts to the publisher website and display the PDF copy on our platform. Clicks to view the full text will be directed to the publisher website, where only users with subscriptions or access through their institution are able to view the full article.