where: a = sample size; b = remission proportion; c = follow-up
years; n = the number of studies pooled.
Assuming a beta distribution around the proportions of remitted
cases 95% confidence intervals (95%CI) were estimated with bootstrap
methods using the @RISK programme add-on for Microsoft Excel
(2000). The α1 and α2 parameters of the beta distributions were N⁎p
and N⁎(1−p), respectively where N is the total number of cases
followed up and p is the proportion remitted.