Association of Prostate-Specific Antigen Screening Rates and Subsequent Metastatic Prostate Cancer Incidence in a National Healthcare System

Abstract

<h3>Purpose/Objective(s)</h3> There is substantial controversy about the benefit of prostate-specific antigen (PSA) screening. PSA screening rates have declined since 2008 due to changing practice guidelines and conflicting clinical trial results regarding the benefit of screening. Since then, the incidence of metastatic prostate cancer (mPCa) has risen; however, there is no direct epidemiologic evidence of a correlation between population PSA screening rates and subsequent mPCa rates. <h3>Materials/Methods</h3> We used data from the United States Veterans Healthcare Administration (VHA) to calculate yearly PSA screening rates, long-term non-screening rates (no PSA measured in prior three years), and age-adjusted mPCa incidence rates from 2005 to 2019. Rates are reported per 100,000 men and were aggregated by race, age group, and VHA facility (128 facilities). We used random effects linear regression to examine the influence of yearly facility-level PSA screening rates from 2010-2013 on subsequent facility-level mPCa incidence from 2014-2019, adjusting for population demographics. <h3>Results</h3> The cohort included 4.7 million men in 2005 and increased to 5.4 million in 2019. PSA screening rates declined from a high of 50.8% in 2008 to 37% in 2019, with declines in all age and race groups; the long-term non-screening rate similarly increased in this period. mPCa incidence increased from a low of 4.6 in 2008 to 7.9 in 2019, driven by increases in the 55-69 and 70+ year old age groups. Lower facility-level PSA screening rates were associated with higher subsequent mPCa incidence (0.87 unit increase in 2014-2019 mPCa incidence rate per 10% decrease in 2010-2013 PSA screening rate, 95% confidence interval [CI] 0.58 to 1.15, p<0.001). Similarly, higher long-term non-screening rates were associated with higher subsequent mPCa incidence (1.61 unit increase in 2014-2019 mPCa incidence rate per 10% increase in 2010-2013 long-term non-screening rate, 95% CI 1.14 to 2.08, p<0.001). <h3>Conclusion</h3> From 2008 to 2019, PSA screening rates declined and long-term non-screening rates increased in the national VHA system. Lower facility-level PSA screening rates were associated with higher subsequent rates of mPCa. These data can be used to inform shared decision making about the potential benefits of PSA screening in men who wish to reduce their risk of prostate cancer metastases.