The placebo effect has not been characterised in pregnant women suffering from nausea and vomiting of pregnancy (NVP). Our aim was to characterise determinants of the placebo effect in women treated with placebo for NVP. We analysed data from a multicentre, double blind randomised controlled trial of Diclectin (delayed release doxylamine and pyridoxine) vs placebo for the treatment of NVP. A total of 127 women in the placebo arm and 130 in the active arm provided evaluable data for this analysis. Women who chose to continue placebo on a compassionate basis (n = 41) had significantly better improvement in symptoms of NVP and higher Wellbeing scores than those who did not ask to continue compassionate use. Results were similar in the active drug arm. The request to continue compassionate use of either placebo or active drug could be predicted by greater improvement in symptoms of NVP during the trial period.