Open Access

Comments on: blood product transfusion in emergency department patients: a case control study of practice patterns and impact on outcome

International Journal of Emergency Medicine201710:32

https://doi.org/10.1186/s12245-017-0158-3

Received: 11 August 2017

Accepted: 28 November 2017

Published: 6 December 2017

Abstract

Clinical decision makings according studies result require the valid and correct data collection, andanalysis. However, there are some common methodological and statistical issues which may ignore by authors. In individual matched case- control design bias arising from the unconditional analysis instead of conditional analysis. Using an unconditional logistic for matched data causes the imposition of a large number of nuisance parameters which may result in seriously biased estimates.

Keywords

Case- control studyMatched dataConditional logistic regression

Correspondence

Dear Editor:

We read with interest, the article entitled “Blood product transfusion in emergency department patients: a case-control study of practice patterns and impact on outcome” by Beyer and colleagues published in International Journal of Emergency Medicine [1]. While we congratulate the authors for their cogent thesis, we were concerned about a methodological issue that may have been overlooked in the peer review.

To carry out this study, the authors used a matched case-control design and control subjects were matched with cases on a one-to-one basis for many factors including ED diagnosis, hemoglobin value, age, and gender, such that individual matching for all factors was assured. However, the authors analyzed the data inappropriately using a regression model. For individual matched case-control studies, we need to use conditional logistic modeling instead of the ordinary logistic regression methodology [2].

In matched data, conditional logistic (partial likelihood) analysis provides a valid approximation of the rate ratio and adjusts for the sampling variability found in estimating standard error and confidence intervals. Using an unconditional logistic for matched data causes the imposition of a large number of nuisance parameters which may result in seriously biased estimates [2]. The take home message for readers is to use the appropriate statistical model in order to avoid analysis pitfalls that can be anticipated from the beginning.

Declarations

Authors’ contributions

SK wrote the firs draft of manuscript. MK reviewed and commented on first draft. Both authors read and approved the final manuscript.

Competing interests

The authors declare that they have no competing interest.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Authors’ Affiliations

(1)
Research Center for Health Sciences, Hamadan University of Medical Sciences
(2)
Department of Epidemiology, School of Public Health, Hamadan University of Medical Sciences
(3)
Department of Epidemiology and Biostatistics, School of Public Health, Tehran University of Medical Sciences

References

  1. Beyer A, Rees R, Palmer C, Wessman BT, Fuller BM. Blood product transfusion in emergency department patients: a case-control study of practice patterns and impact on outcome. Int J Emerg Med. 2017;10(1):5.View ArticlePubMedPubMed CentralGoogle Scholar
  2. Breslow NE, Day NE. Conditional logistic regression for matched sets. Stat Methods Cancer Res. 1980;1:248–79.Google Scholar

Copyright

© The Author(s). 2017

Advertisement