Time trends in ovarian cancer incidence in Navarra (1973-1995): effect of age, birth cohort and period of diagnosis
P. González1, M.T. Jiménez1, G. López-Abente1, M. Pollán1, E. Ardanaz2, J.J. Viñes3
1. Servicio de Epidemiología del Cáncer. Centro Nacional de Epidemiología. Instituto de Salud Carlos III. Madrid.
Background. Ovarian cancer is the sixth most common malignancy and the first killer among gynaecologic cancers in women of the European Union. Five years after diagnosis relative survival rates were less than 35%. The aims of the present study are to describe the time frends in ovarian cancer incidence in Navarra over the period 1973-1995, and to identify the age, period-of-diagnosis and birth-cohort components underlying the trend's evolution.
Methods. Standard summary indexes of incidence were calculated. Log-linear Poisson models to quantify risks of ovarian cancer and the relative annual trend were used. Age-period-cohort models were adjusted in order to ascertain the effects on time trend exerted by the respective age, period-of-diagnosis and birth-cohort components.
Results. Ovarian cancer incidence has showed annual rises of 2.1% over the period 1973-1995, with stabilisation over the period 1988-1995. A 54% increase in the risk of ovarian cancer was shown for the period 1993-1995 vs.1973-1977. The more recent generations of women born in Navarra do not show significant changes in the risk of ovarian cancer.
Conclusions. A considerable rise in ovarian cancer incidence in Navarra over the period 1973-1995 was found, less pronounced in the last decade. The risk of ovarian cancer shown in the more recent generations of women born in Navarra is stable.
Key words. Incidence. Ovarian cancer. Epidemiology. Time trends. Age-Period-Cohort modelling.
Aceptado para su publicación el 10 de enero de 2001.