[Skip to Navigation]
Sign In
Editor's Note
April 21, 2020

Correcting the Scientific Record—Retraction and Replacement of a Report on Dialysis Ownership and Access to Kidney Transplantation

Author Affiliations
  • 1Editor in Chief, JAMA
  • 2Executive Managing Editor, JAMA
  • 3Executive Editor, JAMA
JAMA. 2020;323(15):1455. doi:10.1001/jama.2020.4368

The September 10, 2019, issue of JAMA included an Original Investigation titled “Association Between Dialysis Facility Ownership and Access to Kidney Transplantation” by Gander and colleagues,1 and an accompanying Editorial by Boulware and colleagues.2 The authors of the research report subsequently notified the editors that a major coding error had occurred in their data management and analyses, and that they discovered these errors had resulted in incorrect data in their published article. As reported in their letter of explanation,3 the authors performed a thorough reanalysis of their data to address these errors and found that the main findings and overall conclusion were consistent with the main findings and conclusion reported in the published article, although the corrected analysis revealed important differences in effect size and strength of the associations.

The article by Gander et al1 has been retracted and replaced. The replacement article includes PDF copies of the original article with the errors highlighted and the replacement article published as a new second Supplement. The replacement article includes a notice that the article has been replaced and is linked to the letter of explanation. The Editorial by Boulware et al2 and a related Editorial by Kshirsagar et al4 that was published in JAMA Surgery also have been corrected and include notices of correction.

As Gander et al report in their explanatory letter,3 the directionality of the main findings of the study was unchanged from the original article: patients who received dialysis at for-profit dialysis centers, compared with those who received dialysis at not-for-profit centers, had lower likelihood of placement on the deceased donor waiting list, receipt of a living donor kidney transplant, and receipt of a deceased donor kidney transplant. However, the magnitude of the differences in the 5-year cumulative incidence rates between for-profit and not-for-profit centers was attenuated, although the differences and the corresponding hazard ratios remained statistically significant for the 3 main outcomes. The differences in the absolute between-group 5-year cumulative incidences changed from −13.2% in the original article to −2.6% in the corrected analysis for waiting list placement, from −2.3% to −0.9% for living donor kidney transplant, and from −4.3% to −1.4% for deceased donor kidney transplant.

The authors were forthcoming in acknowledging the discovery of the errors in their original report and performed the necessary reanalyses and corrections in a thorough and scientifically responsibly manner, including enlisting an experienced analyst who had not been involved with the initial study to conduct an independent analysis to verify the corrected results. As part of the editorial evaluation, JAMA consulted with reviewers who indicated that the corrected findings merited publication, especially considering the importance of the clinical and policy issues, as well as the large proportion of patients with end-stage kidney disease who receive dialysis at for-profit centers. As Boulware et al2 explain in their corrected Editorial, the small absolute differences in outcomes between for-profit and not-for-profit dialysis facilities “could be viewed as reassuring” and “could be explained by lack of information on potential confounding factors.” However, the editorialists also emphasize that the findings of Gander et al suggest that the rates of kidney transplantation remain suboptimal, and that “heightened vigilance is needed to ensure profit motives do not unduly influence transplant access in all types of facilities.”2

JAMA is committed to ensuring an accurate scientific record. Just as errors and mistakes occur in everyday life, errors are bound to occur in scientific research, such as those that may involve data, analyses, or reporting. In both cases, honest acknowledgment of the mistake and taking necessary and thorough steps to rectify the errors are of paramount importance to ensure personal, professional, and scientific integrity. We urge authors to report errors they discover in their work, so that along with editors, the appropriate course of action can be taken to ensure the accuracy of the scientific record.

Back to top
Article Information

Corresponding Author: Howard Bauchner, MD, JAMA, 330 N Wabash, Chicago, IL 60611 (howard.bauchner@jamanetwork.org).

Conflict of Interest Disclosures: None reported.

Gander  JC, Zhang  X, Ross  K,  et al.  Association between dialysis facility ownership and access to kidney transplantation.  Retracted and replaced April 21, 2020.  JAMA. 2019;322(10):957-973. doi:10.1001/jama.2019.12803PubMedGoogle ScholarCrossref
Boulware  LE, Wang  V, Powe  NR.  Improving access to kidney transplantation: business as usual or new ways of doing business?  Corrected on April 21, 2020.  JAMA. 2019;322(10):931-933. doi:10.1001/jama.2019.12784PubMedGoogle ScholarCrossref
Gander  JC, Zhang  X, Ross  K,  et al.  Notice of retraction and replacement. Gander et al. Association between dialysis facility ownership and access to kidney transplantation. JAMA. 2019;322(10):957-973.  Published April 21, 2020.  JAMA. doi:10.1001/jama.2019.2328Google Scholar
Kshirsagar  AV, Kibbe  MR, Gerber  DA.  Transplant first, dialysis last.  Corrected on April 21, 2020.  JAMA Surg. 2019. doi:10.1001/jamasurg.2019.4286PubMedGoogle Scholar