Journal of the Korean Society of Radiology (대한영상의학회지)

The Korean Society of Radiology (대한영상의학회)
권호 표지
DOI QR코드

DOI QR Code

Statistical Mistakes Commonly Made When Writing Medical Articles

의학 논문 작성 시 발생하는 흔한 통계적 오류

  • Soyoung Jeon (Biostatistics Collaboration Unit, Yonsei University College of Medicine) ;
  • Juyeon Yang (Biostatistics Collaboration Unit, Yonsei University College of Medicine) ;
  • Hye Sun Lee (Biostatistics Collaboration Unit, Yonsei University College of Medicine)
  • 전소영 (연세대학교 의과대학 의학통계실) ;
  • 양주연 (연세대학교 의과대학 의학통계실) ;
  • 이혜선 (연세대학교 의과대학 의학통계실)
  • Received : 2022.07.29
  • Accepted : 2023.02.26
  • Published : 2023.07.01
Download PDF
⟨ Previous Next ⟩

Abstract

Statistical analysis is an essential component of the medical writing process for research-related articles. Although the importance of statistical testing is emphasized, statistical mistakes continue to appear in journal articles. Major statistical mistakes can occur in any of the three different stages of medical writing, including in the design stage, analysis stage, and interpretation stage. In the design stage, mistakes occur if there is a lack of specificity regarding the research hypothesis or data collection and analysis plans. Discrepancies in the analysis stage occur if the purpose of the study and characteristics of the data are not sufficiently considered, or when an inappropriate analytic procedure is followed. After performing the analysis, the results are interpreted, and an article is written. Statistical analysis mistakes can occur if the underlying methods are incorrectly written or if the results are misinterpreted. In this paper, we describe the statistical mistakes that commonly occur in medical research-related articles and provide advice with the aim to help readers reduce, resolve, and avoid these mistakes in the future.

의학 논문을 작성할 때 통계학은 필수적인 요소로 알려져 있고 중요성이 강조되고 있지만 많은 논문에서 통계적 오류가 발생하고 있다. 의학 논문에서 발생할 수 있는 통계적 오류는 설계 단계에서의 오류, 분석 단계에서의 오류, 작성과 해석 단계에서의 오류로 분류할 수 있다. 설계 단계에서는 연구의 가설이나 자료의 수집 및 분석 계획이 명확하지 않으면 오류가 발생한다. 분석 단계에서는 연구의 목적과 자료의 특성을 충분히 고려하지 않고 올바른 분석 방법을 적용하지 않으면 오류가 발생한다. 분석을 수행한 후에는 결과를 해석하여 논문을 작성하게 되고, 이 단계에서 분석 방법을 잘못 작성하거나 결과를 올바르게 해석하지 못하면 오류가 발생한다. 본 논문에서는 의학 논문에서 흔히 발생하는 통계적 오류에 대해 고찰하고 오류를 줄이는데 기여하고자 한다.

Keywords

References

  1. Winters R, Winters A, Amedee RG. Statistics: a brief overview. Ochsner J 2010;10:213-216 
  2. Ali Z, Bhaskar SB. Basic statistical tools in research and data analysis. Indian J Anaesth 2016;60:662-669  https://doi.org/10.4103/0019-5049.190623
  3. Petra G. Common statistical errors in medical publications. Available at. https://www.slideshare.net/AustralianNationalDataService/common-statistical-errors-in-medical-publications. Published 2017. Accessed June 15, 2022 
  4. Montgomery DC. Design and analysis of experiments. 10th ed. Hoboken, NJ: John Wiley & Sons Inc 2017:14-15 
  5. Shin IH, Yim HW. Calculation of sample size using excel in clinical research. 1st ed. Paju: Koonja 2009:13-14 
  6. Altman DG, Bland JM. Absence of evidence is not evidence of absence. BMJ 1995;311:485 
  7. Yoo KY, Kang DH, Ko KP, Kwak J, Kim YJ, Kim Y, et al. Research methodology in medicine. Seoul: Seoul National University Publishing & Cultural Center 2014:206-207 
  8. Kang SH. Medical statistics necessary for new drug development. 2nd ed. Paju: Free Academy 2013:198-199 
  9. Harrison E, Pius R. Should I convert a continuous variable to a categorical variable? Available at. https://argoshare.is.ed.ac.uk/healthyr_book/should-i-convert-a-continuous-variable-to-a-categorical-variable.html. Published 2020. Accessed January 15, 2021 
  10. Sibbald B, Roland M. Understanding controlled trials. Why are randomised controlled trials important? BMJ 1998;316:201 
  11. de Boer MR, Waterlander WE, Kuijper LD, Steenhuis IH, Twisk JW. Testing for baseline differences in randomized controlled trials: an unhealthy research behavior that is hard to eradicate. Int J Behav Nutr Phys Act 2015;12:4 
  12. Austin PC. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res 2011;46:399-424  https://doi.org/10.1080/00273171.2011.568786
  13. Linden A. Graphical displays for assessing covariate balance in matching studies. J Eval Clin Pract 2015;21:242-247  https://doi.org/10.1111/jep.12297
  14. Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med 2009;28:3083-3107  https://doi.org/10.1002/sim.3697
  15. Ranganathan P, Pramesh CS, Aggarwal R. Common pitfalls in statistical analysis: intention-to-treat versus per-protocol analysis. Perspect Clin Res 2016;7:144-146  https://doi.org/10.4103/2229-3485.184823
  16. Kirkham EM, Weaver EM. A review of multiple hypothesis testing in otolaryngology literature. Laryngoscope 2015;125:599-603  https://doi.org/10.1002/lary.24857
  17. Shaffer JP. Multiple hypothesis testing. Annu Rev Psychol 1995;46:561-584  https://doi.org/10.1146/annurev.ps.46.020195.003021
  18. Makin TR, Orban de Xivry JJ. Ten common statistical mistakes to watch out for when writing or reviewing a manuscript. Elife 2019;8:e48175 
  19. Pepe MS, Janes H, Longton G, Leisenring W, Newcomb P. Limitations of the odds ratio in gauging the performance of a diagnostic, prognostic, or screening marker. Am J Epidemiol 2004;159:882-890  https://doi.org/10.1093/aje/kwh101
  20. Naeger DM, Kohi MP, Webb EM, Phelps A, Ordovas KG, Newman TB. Correctly using sensitivity, specificity, and predictive values in clinical practice: how to avoid three common pitfalls. AJR Am J Roentgenol 2013;200:W566-W570  https://doi.org/10.2214/AJR.12.9888
  21. Khamis HJ. An application of Bayes' rule to diagnostic test evaluation. J Diagn Med Sonogr 1990;6:212-218  https://doi.org/10.1177/875647939000600403
  22. Efron B, Tibshirani R. An introduction to the bootstrap. London: Chapman & Hall 1993:203-204