The incidence and mortality of cervical cancer has been increasing among young women in New Zealand. This increase has occurred despite the availability of cervical screening services. Organized cervical screening programmes can reduced the mortality of cervical cancer and an assessment of the epidemiology of cervical cancer and the organization of cervical screening in New Zealand has been necessary. Age-period-cohort modelling and a graphical cohort analysis have been used to describe the trends in cervical cancer and the registrations of carcinoma in situ in New Zealand. Ethnic differences in these trends have also been examined. The age-period-cohort model has been extended to estimate, under different assumptions, the future incidence and mortality of cervical cancer in New Zealand until after the year 2000 if the effectiveness of cervical screening is not improved.The history of cervical screening in New Zealand, the trends in the registration rates of carcinoma in situ and the application of mathematical models of the natural history of the disease have provided estimates of the possible impact of past screening activity on cervical cancer incidence in New Zealand. A small pilot study attempted to incoporate some of the important features of successful organized cervical screening programmes within the constraints imposed by the New Zealand health service.The costs and benefits of different cervical screening strategies in New Zealand were assessed using the simulation methods of operations research.An increased risk of cervical cancer among successive generations of women born since about 1936 and an absence of an obvious alteration in the observed mortality or incidence of cervical cancer from past cervical screening activity was found (however, past screening activity may have reduced the incidence of cervical cancer by about 40%by 1983). Projections of the mortality and incidence of cervical cancer emphasize that the emerging epidemic of cervical cancer will occur among women already with pervasive disease and screening is the only available method of reducing their risk of cervical cancer. Changes in sexual behaviour are unlikely to alter this epidemic of invasive cervical cancer over the next15to 20years. Also, no reduction in the ratio of Maori to non-Maori cervical cancer mortality rates in the 30-year time period examined was observed. A very high incidence rate of cervical cancer among New Zealand resident Pacific island Polynesian women requires further investigation.From the pilot study for a cervical screening programme it was concluded that an organized cervical screening programme maybe difficult to implement because of the structure of the existing health service in New Zealand. Also, the participation of the majority of women in an organized cervical screening programme maybe difficult to achieve.Three-yearly screening would be a preferred screening frequency for a national cervical screening programme in New Zealand. A three-yearly screening frequency for asymptomatic women, unless contra-indicated on clinical grounds, would significantly increase the demands on the screening service. Screening more frequently than this could not be justified from the cost effectiveness analysis performed and would be likely to place unbearable demands on the cervical screening service.It was also thought that an organized cervical screening service requires operational objectives and guidelines that are understood by the management of the screening programme and guarantee its effectiveness and efficiency.Any causal relationship between human papillomavirus infections and the preinvasive stages of cervical cancer has not yet been established and well design studies are needed. It is important that the natural history of human papillomavirus infections is more clearly understood as clinical decisions are becoming influenced by the presence of these infections.