Article Title

The Impact of Clinical Guidelines on Prostate Cancer Screening Practices in a Population-Based Setting, 2000–2013

Publication Date



prostate cancer, screening


Background/Aims: Controversial clinical guidelines have recommended against widespread prostate-specific antigen (PSA) testing, despite evidence suggesting a benefit in certain populations. This study evaluates the impact of the publication of prostate cancer screening trial results in 2009 and changes to the U.S. Preventive Services Task Force (USPSTF) guidelines in 2012 on temporal trends in the use of PSA tests among men 40–80 years old, with follow-up through 2013.

Methods: Men aged 40–80 without a history of prostate cancer who sought care at Fallon Health (Meyers Primary Care Institute [MPCI]) or Henry Ford Health System (HFHS) contributed to this analysis. We counted one PSA test per person per calendar year, with rates defined per 1,000 person-years. Men were censored based on prostate cancer diagnosis, PSA test results ≥ 4 ng/mL, disenrollment, death or end of follow-up (Dec. 31, 2014). We also examined trends in PSA testing among high-risk men (African-American, family history of prostate cancer). Test rates were compared among three time periods: 2000–2008, 2009–2012 and 2013–2014.

Results: On average, 17,144 men at MPCI and 16,733 men at HFHS had a PSA test each year. Mean age at PSA test was 57 years, which increased over time at both sites. PSA test rates at MPCI declined 33% over the study period from an average of 474 tests/1,000 person-years (2000–2008) to 391/1,000 person-years (2009–2012) to 317/1,000 person-years (2013). At HFHS, PSA test rates rose gradually from 2000 to 2008, with between 30% and 52% of the eligible population (20,838 of 69,550 men in 2000 and 19,366 of 37,434 men in 2008) undergoing ≥ 1 test, followed by a 21% decline from 2008 to 2013 (7,151 of 23,749 men in 2013). At both sites, this decline was attenuated among high-risk men.

Conclusion: This analysis of two population-based electronic health datasets provides evidence of decreasing use of PSA testing over time, although high-risk populations experienced a lesser decline. Although we are unable to determine causality, it is plausible that results of recent screening trials and/or restrictive changes to the USPSTF guidelines have impacted PSA testing practices over the past 14 years. As a next step, we will investigate trends in medical follow-up to elevated PSA test results.




June 27th, 2016


August 12th, 2016