In their study, Brugha et al ^{Reference Brugha, Matthews, Morgan, Hill, Alonso and Jones1} discussed the search strategies employed by the compilers of the systematic reviews and meta-analyses that they analysed. We wish that they had pursued this issue in more detail.

Brugha et al wrote that ‘Authors generally gave comprehensive details of search strategies employed, including details of electronic databases searched, exact search terms, dates covered by search and other methods used’ (p. 447). In examining many systematic reviews and meta-analyses of psychiatric literature in the course of our work with the PILOTS Database, an online index to the worldwide literature of post-traumatic stress disorder (PTSD) that we produce at the National Center for PTSD, we have often observed the inadequacy of the search strategies described by their authors. It is evident that few of these studies have made proper use of the controlled indexing vocabularies used by databases such as MEDLINE and PsycINFO or displayed evidence that the thesauri in which these controlled vocabularies are published have been consulted. The reader familiar with these tools will often have reason to question the reliance that can be placed on systematic reviews and meta-analyses whose authors have not consulted them.

In Lerner & Hamblen, ^{Reference Lerner and Hamblen2} we explain in detail the importance of properly using controlled vocabularies in the compilation of systematic reviews and meta-analyses, demonstrate problems that may arise from not doing so, and offer suggestions for improving the literature searches underlying these compilations.