[Skip to Content]
[Skip to Content Landing]
Views 233
Citations 0
Comment & Response
March 12, 2020

Notice of Retraction and Replacement. Kang et al. Association of statin use and high serum cholesterol levels with risk of primary open-angle glaucoma. JAMA Ophthalmol. 2019;137(7):756-765

Author Affiliations
  • 1Channing Division of Network Medicine, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts
  • 2Department of Ophthalmology, Harvard Medical School, Massachusetts Eye and Ear Infirmary, Boston, Massachusetts
  • 3Department of Ophthalmology and Visual Sciences, University of Michigan Medical School, Ann Arbor
  • 4Center for Eye Policy and Innovation, University of Michigan, Ann Arbor
  • 5Department of Health Management and Policy, University of Michigan School of Public Health, Ann Arbor
  • 6National Institute for Health Research Biomedical Research Centre, Moorfields Eye Hospital National Health Service Foundation Trust, University College London Institute of Ophthalmology, London, United Kingdom
  • 7Department of Biostatistics, Harvard T. H. Chan School of Public Health, Boston, Massachusetts
  • 8now with Department of Ophthalmology, Icahn School of Medicine at Mount Sinai, New York, New York
JAMA Ophthalmol. Published online March 12, 2020. doi:10.1001/jamaophthalmol.2020.0027

To the Editor We write to explain errors that occurred in our Original Investigation “Association of Statin Use and High Serum Cholesterol Levels With Risk of Primary Open-Angle Glaucoma,”1 published online on May 2, 2019, and in the July 2019 issue of JAMA Ophthalmology. In the original article, we reported that statin use of 5 or more years was associated with a lower risk of primary open-angle glaucoma (POAG) and that every 20-mg/dL increase in total serum cholesterol was associated with an increased risk of POAG. However, because of errors in our analyses, these findings are not accurate.

Recently, we discovered a serious coding error in the analyses using combined data from 3 large cohorts included in this study. This error was discovered by the corresponding author (J.H.K.) during review of the results of the analyses for another study that used the same programming and analysis. This coding error resulted in there being no adjustment for age and 2-year period at risk, despite our reporting that we stratified for age in months and 2-year period at risk (which essentially is adjusting for confounding by age and secular time). The error occurred because when the data from a single cohort are used, the analysis macros developed by our statisticians automatically adjust and stratify by age in months and 2-year period without explicitly stating these 2 variables as parameters. However, when a third variable is needed for adjustment and stratification (ie, cohort), all 3 must be explicitly stated as parameters for adjustment and stratification to occur. We had only specified the cohort variable as a macro parameter; thus, the published analyses were only adjusted and stratified by cohort but not age or 2-year period at risk.

When we corrected our analyses to properly stratify by age in months and 2-year period at risk in addition to cohort, the results were substantially changed, and we no longer observed the originally reported significant associations between statin use and serum cholesterol levels with risk of POAG.

Specifically, following the corrected programming and reanalysis, we observed that the relative risk (RR) for statin use of 5 or more years in relation to POAG was 0.93 (95% CI, 0.75-1.15; P for linear trend = .49), which contrasts with the previously reported RR of 0.79 (95% CI, 0.65-0.97; P for linear trend = .02). Also, the RR for every 20-mg/dL increase in total serum cholesterol in relation to POAG risk was 1.02 (95% CI, 0.98-1.07; P = .32), which contrasts with the previously reported RR of 1.07 (95% CI, 1.02-1.11; P = .004). None of the other main associations were significant. Thus, our previously reported observed associations were confounded by these 2 important variables and were incorrect. These errors affected the Abstract, article text, Tables, Figure, and Supplement 1.

We sincerely apologize to the readers and editors of JAMA Ophthalmology for this inadvertent yet serious error. Our article has been retracted and replaced online. An online supplement with the original version of the article with the incorrect information highlighted and a version of the replacement article with the corrections highlighted is available in the online supplement to the article.

Back to top
Article Information

Corresponding Author: Jae H. Kang, ScD, Channing Division of Network Medicine, Brigham and Women’s Hospital, Harvard Medical School, 181 Longwood Ave, Boston, MA 02115 (nhjhk@channing.harvard.edu).

Published Online: March 12, 2020. doi:10.1001/jamaophthalmol.2020.0027

Conflict of Interest Disclosures: Dr Stein reported receiving grants from the National Institutes of Health and Research to Prevent Blindness during the conduct of the study. Dr Khawaja reported receiving personal fees from Allergan, Novartis, Thea, Grafton Optical, and Santen outside the submitted work. Dr Rosner reported receiving grants from the National Institutes of Health during the conduct of the study. Dr Kang reported receiving grants from the National Institutes of Health during the conduct of the study. Dr Wiggs reported receiving grants from the National Eye Institute during the conduct of the study and grants from the National Eye Institute outside the submitted work. Dr Pasquale reported receiving personal fees from Bausch & Lomb, Eyenovia, and Verily Inc outside the submitted work. No other disclosures were reported.

References
1.
Kang  JH, Boumenna  T, Stein  JD,  et al.  Association of statin use and high serum cholesterol levels with risk of primary open-angle glaucoma.  JAMA Ophthalmol. 2019;137(7):756-765. doi:10.1001/jamaophthalmol.2019.0900PubMedGoogle ScholarCrossref
×