Letter to the editor: Sample size considerations are needed for the causal analyses of existing databases

Hern án provides reasonable arguments against power calculations for the causal analyses of existing databases, and concluded “In summary, for an important causal question, analyze your data no matter how imprecise you expect your estimate to be, publish your estimate, encourage others to do the same, and then meta-analyze.” [1]. However, the arguments of the paper do not eliminate the need to consider sample size: First, ratio measures such as risk ratios, odds ratios, and rate ratios estimated from small-sample observational studies may suffer from sparse-data bias [2], which carries over th e pooled estimates from meta-analysis of those studies [3, 4].
Source: Journal of Clinical Epidemiology - Category: Epidemiology Authors: Source Type: research