Poor Follow-up After Elevated Prostate-specific Antigen Tests: A Population-based Cohort Study

Abstract

BackgroundAlthough prostate-specific antigen (PSA) testing is common, little is known about the pattern of retesting by either PSA values or subsequent prostate biopsies. Poor follow-up of high PSA values may lead to delayed diagnosis. ObjectiveTo estimate the probabilities of follow-up (including retesting, prostate biopsies, diagnosis, and cause-specific death) for men undergoing prostate cancer testing at a population level. Design, setting, and participantsCohort study design for men living in Stockholm with no previous diagnosis of prostate cancer between 2003 and 2015. Men were linked to the national health and population registries in Sweden. We report follow-up for men aged 50–79 yr at 2003 or at their index PSA test. Outcome measurements and statistical analysisState probabilities with 95% confidence intervals (CIs) were calculated using multistate Markov models. Results and limitationsAmong men not previously diagnosed with prostate cancer with an initial PSA value of ≥10ng/ml, the proportions at 1 yr with no subsequent testing or only elevated PSA test values >3ng/ml were 21.7% (95% CI: 19.5, 23.9), 25.2% (95% CI: 23.9, 26.6), and 47.7% (95% CI: 46.2, 49.1) for those aged 50–59, 60–69, and 70–79 yr, respectively. No significant changes were noticed when stratifying by comorbidities. Limitations include the lack of detail from patient medical charts. This detail would have allowed for more accurate assessment of appropriate clinical follow-up. ConclusionsRegardless of medical history, a large proportion of men with PSA≥10ng/ml were not followed appropriately at 1 yr after the index PSA test. This may partially explain why opportunistic testing is not as effective as screening within trials to reduce prostate cancer mortality. Patient summaryFor men aged 50–69 yr, who undertake a prostate-specific antigen (PSA) test, a PSA level of >10ng/ml should prompt further investigation. However, we found that one out of 10 of these men did not receive repeat testing within 1 yr of the initial test. This may partially explain why opportunistic prostate cancer testing is less effective than screening trials.