Improving levels of evidence in studies published in spinal journals from 1983 to 2011.

INTRODUCTION: The evidence basis for spinal surgical practice is closely related to the quality of clinical studies published. In the hierarchy of study designs, the quality of evidence from a randomised controlled trial (RCT) and meta-analysis are considered better than other types of prospective and retrospective study designs. We assessed the level of evidence in studies published from 1983 to 2011 in 4 journals, publishing predominantly on spine related topics. METHODS: Computerised search of the Medline database was undertaken to evaluate the articles published in 4 'spinal' journals (Spine, European Spine Journal, Journal of Spinal Disorders and Techniques and Spinal Cord) from 1983 to 2011. The numbers of RCTs, meta-analyses and other clinical trials (non-RCT) over the time periods 1983-7, 1988-92, 1993-7, 1998-2002, 2003-7 and 2008-11 were compared. RESULTS: From a total number of 21775 articles evaluated, there were 888 (4.1%) RCTs, 93 (0.4%) meta-analyses and 1355 (6.2%) other clinical trials. Overall, comparing the time periods 1983-7 with 2008-11, there were increases in the proportion of RCTs (1.6% to 5.1%), meta-analysis (0% to 0.6%) and other clinical trials (0.9% to 6.3%) and a decrease in the proportion of other articles published (97.6% to 88.1%). These changes appear to have plateaued after 2003-7. CONCLUSIONS: Although RCTs, meta-analysis and other clinical trials form a small proportion of the studies published in leading 'spinal' journals, there have been an encouraging increase in their proportion over the years. This is comparable with other larger surgical specialities and provides a better evidence basis for clinical practice in spinal disorders.