Responses
Other responses
Jump to comment:
- Published on: 11 April 2017
- Published on: 11 April 2017Incorrect study design and statistical analysis
We read with interest the recently published study by Kim et al, which the authors described as a cross-sectional, observational, case-control study. As a single study such a design is not possible since cross-sectional and case-control studies are two distinct types of study designs. The authors compared the percent with lower lid epiblepharon between those with and without congenital glaucoma and reported that controls were matched on age and date of outpatient visit to the cases, which would suggest this is a matched case-control study. However, the statistical analysis employed did not account for the matched nature of the study design and therefore was not appropriate. Statistical procedures that account for the matched nature of the study should have been employed. The authors are urged to conduct a reanalysis of their study, amending their interpretation as warranted.
Conflict of Interest:
None declared.