All leading medical journals use independent experts, known as 'peer reviewers,' to decide whether research submitted to the journals is of good quality and worthy of publication. Journal editors know that some people are better at the job of peer-reviewing than others but it is hard to predict who will make a good reviewer.
n a recent survey, experienced reviewers were asked about training they had received in peer review and about other aspects of their background. The results, published in the latest issue of PLoS Medicine, show there are no easily identifiable types of formal training and experience that predict reviewer performance. The implications of these findings are discussed by the PLoS Medicine editors in an editorial to be published in the same issue.
source - Medical News Today