Mark Rabnett recently wrote about his experience trying to apply the Cochrane filter (as discussed in the International Journal of Epidemiology) to retrieve reports of controlled trials in PubMed. Not only did he discover an error in the filter's search strategy, but he also discovered that recent changes to publication types and subject descriptors made by the National Library of Medicine can cause your previously working filters to flounder.

Mark's post reminded that there are many of us who have our tried and true filters on PubMed or Ovid and that we need to check them at the beginning of the year to see if they still up to par.