We carried out and analysed a questionnaire survey on consumer attitudes to verify the effectiveness of blanket testing of US beef in Japan. Applying an ordered probit model to the data yielded the following findings: blanket testing dramatically increases purchase probabilities, so it is one of the most important tools for reviving Japanese consumers’ meat demand. The effect of blanket testing is especially large in the region where the consumers traditionally ate more beef (the Kansai area). Purchase probability is lower in households with awareness of country-of-origin of beef before the BSE crisis than households without it. Purchase probability of the households that had eaten US beef before the BSE outbreak is higher than those that had not. Purchase probability is not affected by family structure.