Anaesthetists' risk assessment of placebo nerve block studies using the SHAM (Serious Harm and Morbidity) scale.

ABSTRACT

The SHAM (Serious Harm and Morbidity) scale was developed to categorise the severity of potential complications of placebo control interventions in the context of local anaesthesia research. A convenience sample of 43 anaesthetists used the SHAM scale to grade ten published randomised controlled trials investigating local anaesthesia nerve blocks. The Fleiss κ statistic assessed agreement between these anaesthetists and probability of random agreement (Pr(e)) when using the SHAM scale; a κ > 0 shows concordance between assessors above random agreement. Overall κ was 0.50 (95% CI 0.49-0.51, p < 0.001), Pr(e) = 0.21. There was moderate agreement between assessors in determining whether studies were low-risk (SHAM score 0-2) or high-risk (SHAM score 3-4) (κ 0.60 (95% CI 0.58-0.62), Pr(e) = 0.51). Compared with anaesthetists given clinical examples of interventions when applying the SHAM score, anaesthetists who were not given examples showed significantly less inter-individual agreement (κ 0.76 (95% CI 0.72-0.81), Pr(e) = 0.5 vs 0.45 (95% CI 0.41-0.49), Pr(e) = 0.52, p < 0.0001). These results suggest that the SHAM score can be successfully used to grade the severity of potential complications of placebo-controlled interventions in local anaesthesia research and represent a first step towards the score's validation.