[Skip to Navigation]
Comment & Response
July 8, 2021

Methodological Considerations for the Case-Control Study of Metformin and Age-Related Macular Degeneration

Author Affiliations
  • 1Department of Ophthalmology and Visual Sciences, Faculty of Medicine, The University of British Columbia, Vancouver, British Columbia, Canada
  • 2Department of Medicine, Faculty of Medicine, The University of British Columbia, Vancouver, British Columbia, Canada
JAMA Ophthalmol. 2021;139(8):918-919. doi:10.1001/jamaophthalmol.2021.2129

To the Editor We read with interest the study by Blitzer et al1 on the association between metformin and age-related macular degeneration (AMD). We would like to highlight 3 potential limitations of this study. First, individuals with AMD were matched with control participants on a 1:1 scheme by anemia, age, hypertension, geographic location, and comorbidities. However, they were not matched by calendar time, another potential source of bias. Calendar time bias occurs when changes in prescribing trends over time affect the likelihood of exposure among individuals with specific conditions. A study by Le and Lee2 showed that metformin prescribing steadily increased during the time span of the study by Blitzer et al1 (2008 to 2017). Moreover, the introduction of other antidiabetic agents during this period might also have affected metformin prescribing at different years, leading to a differential rate of metformin exposure among individuals with AMD and control individuals.

Add or change institution
×