LAUSR

P eer review is a key component of scientific publication. While imperfect, numerous studies have demonstrated that peer review improves the quality of manuscripts. There have been several interventions to improve peer review quality that have yielded mixed results, for example, blinding reviewers to author identity, using statistical checklists, or conducting reviewer training. Several studies have shown that reviewer ratings have low inter-rater reliability and only a limited ability to predict the eventual impact of articles. We at JGIM use a three-stage process. First, the co-editors in chief evaluate submissions for fit and quality. About 40% of articles are rejected at this stage. Those who pass this initial review are sent to one of the deputy editors for a secondary, indepth screen. Deputy editors can reject the article, accept as is, or send out for external review. At each stage, we ask, Bis it new, is it true, will it change practice and is the writing clear?^ We ask peer reviewers to identify scientific strengths and weakness and assess the likely impact on the field of general internalmedicineandif thearticle isclearlywritten.Whileweask reviewers to make recommendations regarding acceptance, we are mindful of the unreliability of this process and use these recommendations as only one element in what is ultimately an editorialdecision.Manygoodpapers are returned toauthorswith regrets, and we have rejected some important and impactful work. Peer reviewers, editors, and the editorial process are all fallible.We are grateful to authors who continue to submit to the journal, hoping that we finally get it right. The good news is that most articles rejected at JGIMare eventually accepted at another journal, so never say die! JGIM could not exist without the tireless energy and effort expended by our cadre of deputy editors. We thank them for their continued, voluntary effort to improve our journal. The debt we owe our deputy editors cannot be repaid. If you have an interest in joining this exciting, passionate group, please feel free to email any of the editors. JGIM would also like to thank the many talented people who have volunteered their time to serve as peer reviewers for the journal over the past 12 months. In 2017–2018, 958 reviewers providedatotalof1188reviewswithameanqualityscoreof4.3on a scale of 1–6 (as judged by our JGIM deputy editors). Of these, 210 provided at least two reviews, and 23 provided three ormore. We are indebted to them for their service. If you are not currently reviewing for JGIM, please consider signing up at JGIM.org. Among this group of dedicated peer reviewers, there is a group that stands out. These top peer reviewers performed at least two reviews between July 2017 and June 2018, returned all reviews within 30 days, and received a quality score of 4 or greater (out of amaximum score of 6) on all reviews. An asterisk identifies the 114 reviewers who meet these criteria. We congratulate them on their service to the academic community and thank them for their efforts on behalf of the journal. JGIM Reviewers, 2017–2018