Association of prior prostate cancer screening with stage at diagnosis in a US EHR cohort, 2018–2025.

e17071 Background: The US Preventative Service Task Force recommends prostate-specific antigen (PSA)-based screening for men aged 55–69 years and generally discourages routine screening over the age of 70. In addition, the American Cancer Society recommended discussing screening beginning at 40 for patients at high-risk of developing prostate cancer. We evaluated whether evidence of prior PSA screening is associated with earlier stage at first prostate cancer diagnosis. Methods: Using a subset of electronic health records from Truveta Data, we identified adults aged ≥40 years with a first-recorded prostate cancer diagnosis in 2018–2025 (index date). Stage at diagnosis was extracted from clinical notes within 90 days of the index date and categorized as early (I–II) versus late (III–IV). To capture prior screening behavior rather than diagnostic workup for the index diagnosis, we defined prior PSA screening as any PSA screening recorded three years to six months before the index. Patients were required to have ≥1 outpatient visit in the year before the index date and ≥1 outpatient visit in years two or three before index. Patients were excluded if they had any chemotherapy or cancer treatment before index. We compared stage distributions overall and by age group (40–54, 55–69, ≥70). Multivariable logistic regression estimated the association between prior PSA screening and early-stage diagnosis, adjusting for age group, race, ethnicity, rural/urban, Elixhauser comorbidity index, and diagnosis year. Additional logistic regression models were conducted for each age group. Results: Among 1,038,594 patients with prostate cancer, 7,497 met the strict criteria to be included (mean age 73 years). Prior PSA screening was present in 2,163 (55.3%) patients. Early-stage disease was more common among those with prior PSA screening (43.8% vs 39.3%, p < 0.001). Prior PSA testing was associated with higher odds of early-stage diagnosis (adjusted OR 1.49, 95% CI 1.34-1.66). Across age strata, prior PSA testing was consistently associated with higher early-stage proportions (ages 40–54: OR 2.06, 95% CI 0.84-5.03; ages 55–69: OR 1.37, 95% CI 1.14-1.65; ages ≥70: OR 1.54, 95% CI 1.35-1.76). Conclusions: Prior PSA screening was associated with higher odds of earlier-stage at first diagnosis across age strata. The association was observed even among men in age ranges in which routine PSA screening is not generally recommended. Prior routine PSA screening may reflect greater engagement with preventive care or early evaluation of prostate health. The findings of this study support policies and practice workflows that promote timely, risk-informed screening discussions and equitable access in ages 55–69, and individualized decision-making for younger (40–54) and older (≥70) adults to balance the potential benefit of early-stage diagnosis with harms of overdiagnosis.