Is the Quality of Neurosurgical Literature Improving?

Introduction: The advent of evidence-based medicine has resulted in higher quality journal manuscripts in numerous medical disciplines. However, the impact in the neurosurgical literature has not been reported. Objective: To quantify the impact of evidence-based medicine on the quality of articles published in the Neurosurgery literature. Methods: Articles published in the journal Neurosurgery (founded in 1977) were reviewed for 1978, 1988, 1998, and 2008. Each decade’s sample was classified as therapeutic, diagnostic and prognostic based on a published system for determining level of evidence. Results: 438 articles were reviewed. Articles not considered included any published under the heading “Case Report” (automatically Level IV evidence) and articles which otherwise did not directly look at patient outcome (i.e, cadaver or animal studies). The rate of Level I studies held steady at 4.5-6.0%. Level II evidence increased steadily from no articles in 1978 to 40.6% in 2008. The increases in Level I and II article publications was statistically significant (p \u3c 0.001). Concurrently, Level IV articles decreased in rate (81.8% in 1978 to 42.4% in 2008), while Level III articles remained fairly constant (9.8%–13.6%). The largest category of Level II studies was prognostic, and the largest category for both Level III and IV studies was therapeutic. Among study types, the most dramatic increase was in the rate of prognostic studies (15.8% to 43.6%). Only 1% of all articles were economic analyses. Conclusion: The quality of neurosurgical literature has progressively improved over the last several decades. It is unclear how much of that is due to expanded activity in randomized, clinical trials or other Level I evidence as no significant increases were observed in Level I articles during the study period (1978–2008). Much of the literature improvement may be explained by the increase in retrospective, prognostic studies as neurosurgeons take advantage of years of accumulated data. The lack of any articles on economic and decision analyses suggests that the neurosurgical community has not yet studied the effect of costs in detail