Changes in prostate cancer screening practice by blood PSA testing between 2011 and 2017, a French population-based study

Abstract

Objective The aim of the study was to determine the trend of first blood prostate-specific antigen (PSA) test prescription in France between 2011 and 2017, based on the assumption that prostate cancer (PCa) screening is expected to decline over the years. Method Using a representative sample of the French population from the French Health Insurance database, we identified 50–52-year-old men without PCa and without any blood PSA test in the five years before 2011, 2014 and 2017 (January 1–December 31 of each year). For each of these three years, the primary outcome was the first reimbursement of a blood PSA test. We used a logistic regression model with first blood PSA test as the outcome and year as the main explanatory variable. As secondary objectives, we also identified the prescriber’s specialty, the urological consultation frequency, and the number of prostate biopsies in the year after the first blood PSA test reimbursement (only for 2011 and 2014). Results In 2011, 2014 and 2017, 5 275, 5 792 and 5 887 50–52-year-old men, respectively, were included. The percentage of patients with a first blood PSA test prescription decreased linearly from 2011 to 2017: 15.7% in 2011, 13.2% in 2014, and 12.4% in 2017 (p < .001). Blood PSA testing was mainly prescribed by general practitioners (>95%). The median interval between PSA tests was 13 months in 2011 and 14 months in 2014. Fewer than 10% of men had ≥1 consultation with an urologist during the year after the first blood PSA test. After the first blood PSA test, eight prostate biopsies were performed in 2011 and two in 2014. Conclusion Our results suggest that in France, PCa screening is a primary care issue. Although PCa screening remains controversial and confusion exists about the best practice, our study showed a linear decrease of blood PSA test prescriptions for 50–52-year-old men between 2011 and 2017, although the reason for screening was unknown. As clinical information was not available, additional evidence is needed to determine the real impact of this decrease on the cancer-specific and overall mortality.