Context: Current data from randomized trials differ in the benefit to harms from screening for prostate cancer (PC) using prostate specific antigen (PSA) testing and it is unclear if decreases in PC mortality in the US can be attributed to introduction of screening or to changes in treatment
Objective: To investigate if rapidly increased incidence of PC diagnosis due to early and rapidly increased rate of PSA testing, is associated with decreased incidence of lethal PC.
Design: Incidence-based analysis of lethal prostate cancer using highly representative data from nation-wide population-based PC registries.
Setting: Eight Swedish counties with the most rapid and largest increased incidence of PC diagnosis (high incidence counties) from 1995 through 2002, and six counties with the smallest increased PC incidence (low incidence counties).
Participants: Men aged 50-74 years during 2000 to 2009 in the high versus low incidence counties contributing 4,528,134 versus 2,471,373 person-years at risk, respectively. There were 33,780 incident PC diagnoses and 1,577 PC deaths in high incidence counties and 16,377 PC cases and 985 PC deaths in low incidence counties.
Main Outcome Measures: Rate ratio (RR) in high vs. low incidence counties of cases with metastatic PC at diagnosis, (positive bone scan or PSA above 100 ng/ml), PC-specific mortality, and excess mortality (death among PC cases regardless of cause).