We estimated the incidence of pertussis in patients consulting general practitioners (GPs). Between July 2009 and April 2011, we conducted a prospective cohort study of patients attending 78 general practices (158 863 persons overall). We included patients aged ⩾3 years, with cough lasting 2–15 weeks, who gave informed consent. GPs interviewed eligible patients, collected a blood specimen, and a nasopharyngeal swab. At follow-up 30–60 days after the initial visit, physicians collected a second blood specimen and conducted patient interview. Cases were confirmed by specific IgA and/or IgG antibody titre exceeding significantly the general population background level or detection of bacterial DNA by real-time PCR. During the study period, 3864 patients with prolonged cough consulted the participating GPs, of those 1852 met the inclusion criteria, 1232 were recruited, and 288 were confirmed as pertussis cases (4% by PCR, 96% by serology). The adjusted incidence rate was 201·1/100 000 person-years [95% confidence interval (CI) 133·9–302·0], ranging from 456·5 (95% CI 239·3–870·8) in the 15–19 years group to 94·0 (95% CI 33·4–264·5) in the 25–29 years group. The reporting ratio was 61, ranging from 4 in those aged 3–5 years, to 167 in those aged 65–69 years. The study confirmed high incidence of pertussis in all age groups in the general population, in particular in adults, not appropriately documented by the existing surveillance system.