LEVELS OF EVIDENCE IN RESEARCH PRESENTED AT THE COMBINED SERVICES ORTHOPAEDIC SOCIETY ANNUAL MEETING

Abstract

The role of Evidence Based Medicine in modern surgical practice is to provide a framework for the integration of expertise, evidence and the biology of the individual patient. The research presented at the Combined Services Orthopaedic Society (CSOS) annual meeting is an important source of evidence which is used to support clinical decisions made about patients on military operations and in the NHS. The purpose of this study is to review the levels of evidence presented at this meeting since 2001.

We reviewed all of the abstracts presented at the annual meetings of the CSOS between 2001 and 2008, and a single meeting of both the Society of Military Orthopaedic Surgeons (SOMOS) and the British Trauma Society (BTS). Basic science studies, animal studies, cadaveric studies, surveys and guest lectures were excluded. The research abstracts were coded by the lead author (TJB), according to the Oxford Centre for Evidence-based Medicine Levels of Evidence. A second author (AM) reviewed the coding of all abstracts to provide inter-observer reliability. Statistical analysis included a chi-squared test to compare the percentages of each level of evidence between the meetings and between each year of the CSOS meeting.

We identified 140, 51 and 96 abstracts in the CSOS, SOMOS and BTS group respectively, which met the inclusion criteria. Level 1 evidence accounted for 73.8%, 64.7% and 68.8% in the CSOS, SOMOS and BTS groups respectively. Level 1 evidence was uncommon at all three meetings representing 4.1%, 5.9% and 8.3% in the CSOS, SOMOS and BTS groups respectively. The chi-squared test did not demonstrate any statistical difference in the evidence levels between the three groups (X2=11.63 (8df), p=0.17). There was no significant difference in the levels of evidence between years during the study period at the CSOS meeting.

The average level of evidence presented at the CSOS annual meeting compares favourably with other trauma meetings. The high proportion of level 4 evidence presented at these meetings reflects the challenging task of performing research in trauma surgery. This challenge is further exacerbated in the military environment where operational commitments must be the priority. However, simple methods to strengthen research may involve the inclusion of control groups, prospective patient enrolment, standard treatment protocols, well-defined outcome measures, logistical support for good follow-up and use of patient-focused assessment tools. Co-ordination and focus of military orthopaedic research effort may advance the quality of research produced.