In a randomised, double-blind, 3-way cross-over trial, the incidence of nausea associated with 2 doxycycline 100 mg formulations (Doryx' and Vibramycin') were compared. The original study cohort comprised 103 healthy male volunteers, with 97 subjects completing the trial. Subjects were randomly allocated to 1 of 3 treatment sequences and received a single dose of Doryx', Vibramycin' or placebo, with a 7-day washout prior to cross-over. At half-hourly intervals, from 0 to 2 h post-dose, subjects completed questionnaires to indicate if they felt nauseous. Data were analysed according to a log-linear method for the analysis of cross-over trials with categorical responses. Seventeen, 29 and 11 subjects experienced nausea with Doryx', Vibramycin' and placebo, respectively. A significantly greater number of volunteers indicated a positive response with Vibramycin' vs Doryx' and vs placebo; the positive response frequency was not significantly different for the Doryx' vs the placebo regimen. Treatment sequence had no significant effect on response, although a marked first-dose effect was noted; the first (vs the second and vs the third) regimen was 1.5-2 times more likely to induce a positive response.