An observational tool was developed to record distraction and interruption in the operating theatre during surgery. Observed events were assigned to pre-defined categories and rated in relation to the level of team involvement - the sum of which was treated as a measure of intra-operative interference. Many events (0.29 +/- 0.02 per min) were observed and rated in 50 general operations sampled from a single operating theatre. The rating of individual events (r(s) = 0.65) and of cases (r(s) = 0.89) correlated between independent observers. Interference levels (1.04 +/- 0.07/min) also correlated with door opening frequency (0.68 +/- 0.03/min) (r = 0.47, p < 0.001). Some sources of interference were intrinsic to the work of the surgical team, including equipment, procedure and environment, while others were extraneous, including bleepers, phone calls and external staff. The findings highlight the need to further develop measures of interference, to assess its variation, intensity and its effect on surgical team performance.