Background: Prostate cancer occurrence and stage distribution changed dramatically during the end of the 20th century. This study aimed to quantify and explain trends in incidence, stage distribution, survival and mortality in the Netherlands between 1989 and 2006. Methods: Population-based data from the nationwide Netherlands Cancer Registry and Causes of Death Registry were used. Annual incidence and mortality rates were calculated and age-adjusted to the European Standard Population. Trends in rates were evaluated by age, clinical stage and differentiation grade. Results: 120,965 men were newly diagnosed with prostate cancer between 1989 and 2006. Age-adjusted incidence rates increased from 63 to 104 per 100,000 person-years in this period. Two periods of increasing incidence rates could be distinguished with increases predominantly in cT2-tumours between 1989 and 1995 and predominantly in cT1c-tumours since 2001. cT4/N+/M+-tumour incidence rates decreased from 23 in 1993 to 18 in 2006. The trend towards earlier detection was accompanied by a lower mean age at diagnosis (from 74 in 1989 to 70 in 2006), increased frequency of treatment with curative intent and improved 5-year relative survival. Mortality rates decreased from 34 in 1996 to 26 in 2007. Conclusions: The increase of prostate cancer incidence in the early 1990s was probably caused by increased prostate cancer awareness combined with diagnostic improvements (transrectal ultrasound, (thin) needle biopsies), but not PSA testing. The subsequent peak since 2001 is probably attributable to PSA testing. The decline in prostate cancer mortality from 1996 onwards may be the consequence of increased detection of cT2-tumours between 1989 and 1995. Unfortunately, data on the use of PSA tests and other prostate cancer diagnostics to support these conclusions are lacking.