Key points are not available for this paper at this time.
The results suggest that the Freeman-Tukey transform is to be preferred over the logit transformation in the meta-analysis of prevalence.
Abdulmajeed et al. (Sat,) studied this question.