The incidence of pertussis in the Netherlands in 1989-1995 was estimated from notifications by law, hospital admissions, positive serodiagnostics (positive one-point serology and positive two-point serology) and isolations of Bordetella. After the epidemic peak in 1994 the incidence of pertussis decreased in 1995 (notifications 3.4 versus 2.2/100,000). The proportion vaccinated pertussis cases increased, especially among children aged 1-4 years. The estimated vaccine-efficacy for children aged 1-4 years amounted to 89-96% in 1989-1993, 79% in 1994 and 71% in 1995. For those children aged 5-9 years the vaccine-efficacy was also higher in 1989-1993 (83-89%) than in 1994 (72%) and in 1995 (61%). It is not possible to make reliable conclusions due to the possible biases. However, the possibility must be considered that enhancement of susceptibility for symptomatic infection of vaccinated children is caused by decreased sensitivity of circulating B. pertussis strains to vaccine-induced immunity. Due to the possible decline in vaccine efficacy and the underreporting of notifications, the expected introduction of an acellular vaccine, the planned change in case-definition for notification (inclusion of one-point serology), physicians have to be informed and surveillance should be continued. At notification the results of laboratory diagnosis should be specified. Continuation of the possibility to compare the databases of notification and serodiagnosis on the individual level is of great importance to be able to detect possible shifts in notification discipline.