International trends in prostate cancer incidence and mortality in 20 countries from 1980 to 2002
Objective
Worldwide, since the 1980s, there have been major changes in prostate cancer detection and treatment with the availability of the prostate specific antigen (PSA) test and the introduction of hormone therapy and radical prostatectomy. To explore the effect of these advances on the burden of disease, we investigated time trends in prostate cancer incidence and mortality in 20 industrialised countries.
Data Source
Data on prostate cancer incidence during 1980 to 2002 was extracted from volumes 6 to 9 of Cancer Incidence in Five Continents. Mortality data from 1990 to 2002 were obtained from the WHO mortality database.
Methods
Age-standardised rates (ASR), based on the male population aged ≥50, were calculated for each year in each country using the European Standard Population. Annual percentage change (APC) in incidence and mortality rates, and the points in time when trends changed, were estimated by fitting join point regression models using Join point software (3.1). Overall percentage change (OPC) during the study period was defined as OPC = (1+APC)number of years −1. Trends for men aged 50–74 and ≥75 years were also analysed.
Results
In terms of incidence, in 1980, the overall ASR varied widely across countries, ranging from 18 per 100 000 in Belarus to 309 per 100 000 in Switzerland. Between 1980 and 2002, prostate cancer incidence increased in all 20 countries. The OPC was the smallest in Denmark (+40%) and highest in Italy (+336%). Generally, countries with lower incidence in early years had the highest OPCs. In North-America and Australia incidence peaked around 1994, whereas in most European countries rates rose throughout the study period. The increase was most pronounced among men aged 50–74, and in a few countries, the OPC for men aged ≥75 years was less than zero. Mortality rates decreased in North-America and some western European countries (eg France, England, Italy, Switzerland), remained stable in others (eg Scotland, Sweden, Denmark) and increased in Eastern Europe. In countries where rates fell, the decline was more pronounced among younger, than older, men. In some countries (eg France), mortality began to fall before incidence peaked.
Conclusions
International variations in prostate cancer incidence and mortality were observed. The different trend in incidence by age suggests an impact of earlier diagnosis/PSA testing. While the decreases in mortality observed in some countries might be a result of improvements in treatment or earlier detection, they could also be affected by changes in death certificate coding or competing causes of death.
