1. Add the phrase "within each stratum" to the last sentence in the section preceding the section titled "Interaction". In the PDF version this is at line 5 in the first column of page 4. The whole sentence should read: "Consequently, it is not required that F controls be linked with each case, only that the total number of controls be F times the total number of cases within each stratum."
2. In the discussion section, EGRET SIZ is mentioned, and it is suggested that input for such a program which computes sample size for logistic regression would probably be simpler than input for DESIGN or DAYSMITH. In fact, EGRET SIZ performs the calculations only for categorical confounders, and requires input for each category, so it does not currently appear to require simpler input.
Pre-publication history
The pre-publication history for this paper can be accessed here:
References
- Edwardes MD. Sample Size Requirements for Case-Control Study Designs. BMC Medical Research Methodology. 2001;1 doi: 10.1186/1471-2288-1-11. [DOI] [PMC free article] [PubMed] [Google Scholar]