LAUSR

The US National Library of Medicine (NLM) and National Institutes of Health (NIH) manage Medline, considered the flagship bibliographic database of medical research publications. Something has happened because both straightforward and refined search strategies provide low search yields with poor sensitivity and specificity. A straightforward search limited to “dentistry[mesh]”, i.e., dentistry used as a medical subject heading term (MESH), identifies 198 clinical trials, 160 randomized controlled trials, and 183meta‐analyses (MAs) + systematic reviews (SRs) out of a modest 2783 articles published in 2017 (Search on Medline via Pubmed, 2018a). Moreover, the number of articles is lower than before, i.e., the number of references published in, e.g., 2012 are 1090 clinical trials, 882 RCTs and 330 MAs + SRs (Figure 1) , giving the wrong impression that the volume of literature in dental research domains has declined over the last five years. An example of a refined search is the Cochrane highly sensitive search strategy for identifying randomized controlled trials (RCTs) in PubMed (Lefebvre, Manheimer, & Glanville, 2011), which identifies merely 324 RCTs published in 2017 within dentistry. Undersigned observed the limited search yield from Medline in early 2017, while running dedicated search strategies to update personal databases of clinical trials in various dental domains. I attributed this to a likely temporal delay of article indexing caused by, e.g., the change of directorship of NLM and possible new internal restructuring or perhaps to budget cut introduced by the new presidential administration. Anyhow, one could expect that NLM at some point would correct the backlog of non‐indexed articles in dentistry. The situation one year later seems to remain unchanged, which prompted me to examine possible reasons. A search for RCTs in Medline by combining “Dentistry[Mesh]” with the Cochrane highly sensitive search strategy for identifying RCTs in Pubmed (Search on Medline via Pubmed, 2018b) present a different search yield compared to a similar search in the Cochrane Central Register of Controlled Trials (Cochrane Central Register of Controlled Trials, 2018). The search yields are respectively for 2015: 640:484, 2016: 480:505 and 2017: 153:322 RCTs. The Pubmed RCT list reveals an inclusion from a wide spectrum of journals, including some “esoteric research” journals, as well as some incorrectly indexed SRs and commentaries. The RCTs identified in the Cochrane Central Register of Controlled Trials seems biased towards publications in journals managed by a handful of major international publishers. Surprisingly, all