The objective was to explore the feasibility of prospective recruitment of general practice patients to a future trial on the effects of paracetamol on the symptoms of the common cold. In October 2000, recruitment letters were sent by a general practitioner to 276 patients aged between 18 and 75. 65 patients returned signed consent forms. Postal packs were sent to the first 60 of these patients to reply. They contained a daily diary, symptom score questionnaires, an EQ-5D questionnaire, a pack of paracetamol tablets, and detailed instructions. When a cold developed and symptoms reached a specified level, one gram of paracetamol was taken four times daily for two days and questionnaires were completed for four days. Telephone interviews and crosschecking of data were used to identify difficulties in completing the study. The overall recruitment rate was 24% (65/276) with 95% confidence interval 19% to 29%. Recruitment was significantly higher (p < 0.001) amongst patients registered with the main researcher (43% (29/68)) than with other doctors in the practice. The response rate, i.e., those who returned completed study packs, was 77% (46/60). Thirty participants experienced a cold and returned their data during the six months of the study, representing 50% (30/60, 95% Cl 37% to 63%) of participants and 11% (30/276, 95% Cl 7% to 15%) of those initially approached. The mean cumulative four-day symptom score was 34.6, standard deviation 14.7. The EQ-5D health status on day four was 0.89. We conclude that a prospective postal recruitment method in general practice is a potentially useful method of conducting research on the common cold. A better response is obtained when the initial letter is signed by the patient's own doctor.