The limited number of cases and diagnostic problems make accurate measurement difficult The number of cases of variant Creutzfeldt-Jakob disease (vCJD) in the United Kingdom has decreased since 2000,1 but controversy remains about how many people carry the infectious agent and will eventually develop disease. In the linked study (doi:10.1136/bmj.b1442), Clewley and colleagues add to the debate by assessing 63 007 pairs of tonsils for the only available marker of prion disease—the pathological, partially protease resistant, prion protein.2 Although more than half of the samples came from people born between 1961 and 1995, when the risk of exposure to bovine spongiform encephalopathy (BSE) infection was high, no convincingly positive tonsil specimens were detected. The study estimates that the prevalence of vCJD in the British population is zero, but with a large confidence interval of 0 to 113 per million. This result agrees with one UK survey of 2000 tonsil specimens,3 but it differs from another survey of 1427 tonsils and 11 247appendices, which found that more than 10 000 people were incubating the disease.4 However, despite the discrepancy, the 95% confidence intervals …