An enhanced pertussis surveillance and laboratory diagnosis program was initiated in the Halifax metropolitan area of Nova Scotia to better delineate the epidemiology of pertussis. During the 28 months of the study, 526 cases of pertussis were identified (overall yearly incidence: 74 cases per 100,000 population). Laboratory confirmation was obtained in 168 (32%) cases, including 111 (21%) by culture. Peak incidence occurred among children 2 to 5 years of age; the highest morbidity rate was seen in children less than 1 year of age. Hospitalization was required for 22 (4.2%) patients; 14 (64%) of those hospitalized were less than 1 year of age. Most (91%) patients had received at least three doses of pertussis vaccine; vaccine efficacy was estimated at 45%. The surveillance program demonstrated that the incidence of pertussis in Nova Scotia, although among the highest in North America, is still underestimated. A ninefold increase in cases was identified over the comparable period of the previous year, largely because patients meeting clinical criteria were reported. By supplementing culture techniques with immunofluorescent staining and serologic methods, we increased the rate of laboratory confirmation from 17% to 65%, suggesting that strict clinical criteria accurately reflect accurately reflect incidence. We conclude that pertussis remains a significant health problem in Nova Scotia, despite nearly universal vaccination.