Defining low-value PSA testing in a large retrospective cohort: Finding common ground between discordant guidelines

Abstract

Background: Reports of low-value prostate-specific antigen (PSA) testing (testing in which the harms outweigh the benefits) generally employ population level data sources. While such results may be generalizable, they often lack the detail necessary to understand provider clinical decision making and guideline concordance. Using a retrospective study of PSA testing at our institution we intend to characterize the frequency and patterns associated with low-value PSA testing.

Methods: We leveraged the electronic health record to determine guideline-defined low-value testing in our health system from 07/01/2012 to 06/30/2017. Secondarily, we measured the between-testing interval for repeat tests and the rates of prostate cancer risk factors and comorbidities among men receiving screening.

Results: Overall, 21,145 PSA tests were performed on 12,303 men. The rate of low-value testing ranged from 23.4 to 56.8%, depending upon the specific guideline. For repeat tests, the median between-testing interval was 12.6 months. Risk factors for prostate cancer were uncommon, but more frequent in men age <55 years compared to men age 55-69 years (17.6% vs. 13.5%, p < 0.001). Screened older men (age >70 years) were more likely to have a Charlson Comorbidity Index ≥ 3, compared to the 55-69 reference group (31.4% vs. 17.3%, p < 0.001).

Conclusion: Low-value prostate cancer testing is prevalent. Between-testing intervals were often times shorter than recommended. Screening among younger men was frequent despite low rates of risk factors. High rates of comorbidity may limit life expectancy among older men receiving screening. These findings highlight the need for improved guidance with prostate cancer screening.

Keywords: Biomarker; Oncology; Public health; Risk factors; Social medicine; Tumor markers.