This was an exploratory pilot study forming part of a programme of work to develop and trial an effective web-based intervention to reduce the risk of transmission of respiratory infections by promoting hand washing and other preventive behaviours in pandemic and non-pandemic contexts. The main purpose of this study was to confirm that the behavioural determinants we had identified from theory were related as predicted to intentions and to establish the validity of our measures of behavioural intentions. Participants (N = 84) completed a self-report web-delivered questionnaire measuring intentions to engage in hand washing and the hypothesised behavioural determinants of intentions, based on the theory of planned behaviour and protection motivation theory. In a factorial 2 × 2 design, half of the participants were first randomised to receive messages about potential negative consequences of pandemic flu (the "high-threat" condition) and half were assigned to receive "coping" messages describing the rationale and effectiveness of hand washing for reducing the risk of infection. A substantial proportion of variance in intentions was explained by measures of attitudes (instrumental and affective), social norms (descriptive and injunctive), perceived behavioural control (especially, access to hand gel) and perceived risk (in particular, the likelihood of catching pandemic flu). Our measures of intentions were sensitive to between-group differences, and although our design did not permit causal inference (particularly in view of selective dropout among those required to read most web pages), the pattern of differences was in the expected direction, that is, hand-washing intentions tended to be stronger in those receiving the high-threat message and coping messages. This study provided encouraging confirmation that our intervention development was proceeding correctly. Measures of intentions proved sensitive to group differences, and the behavioural determinants included in the study explained a substantial proportion of the variance in intentions. The study also provided useful indications that our high-threat message might increase hand-washing intentions, that providing hand gel might be beneficial and that it would be necessary to actively manage the risk of selective dropout in the intervention group.