Pertussis is a vaccine-preventable respiratory infection caused by Bordetella pertussis which can be fatal in infants. Although high vaccine coverage led to prolonged disease control in England, a national outbreak of pertussis in 2011 led to the largest increase in over two decades, including a marked increase in cases aged ⩾15 years. A case-control study in four regions of England was undertaken to investigate risk factors for pertussis in adolescents and adults, specifically employment type and professional and household contact with children. Pertussis cases were laboratory-confirmed and aged ⩾15 years. Controls were recruited through general practitioner nomination. Demographic and risk factor information were collected using an online survey. Multivariable logistic regression was used to estimate independent associations with outcome. Two hundred and thirty-one cases and 190 controls were recruited. None of the four employment variables (social care, education, health sector, patient contact) were significantly associated with pertussis. Professional contact with children aged < 1 year was associated with a significantly reduced odds of pertussis [odds ratio (OR) 0·25, 95% confidence interval (CI) 0·08–0·78, P = 0·017]. Household contact with ⩾1 child aged 10–14 years was associated with significantly increased odds of pertussis (OR 2·61, 95% CI 1·47–4·64, P = 0·001). Occupational contact with very young children was associated with reduced odds of pertussis, probably due to immune boosting by low-level exposures to B. pertussis. Sharing a household with a young adolescent was a significant risk factor for pertussis in adults and older teenagers. The primary focus of the childhood pertussis vaccination programmes is to prevent infant disease. Although evidence is emerging that adolescent vaccination does not provide indirect protection to infants, our results highlight the importance of children aged 10–14 years in pertussis transmission to older adolescents and adults.