Premium
Low‐Value Prostate Cancer Screening Among Older Men Within the Veterans Health Administration
Author(s) -
Radomski Thomas R.,
Huang Yan,
Park Seo Young,
Sileanu Florentina E.,
Thorpe Carolyn T.,
Thorpe Joshua M.,
Fine Michael J.,
Gellad Walid F.
Publication year - 2019
Publication title -
journal of the american geriatrics society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.992
H-Index - 232
eISSN - 1532-5415
pISSN - 0002-8614
DOI - 10.1111/jgs.16057
Subject(s) - medicine , veterans affairs , confidence interval , odds ratio , prostate cancer screening , prostate specific antigen , retrospective cohort study , prostatectomy , prostate cancer , gerontology , gynecology , cancer
BACKGROUND/OBJECTIVES Prostate‐specific antigen (PSA) screening can be of low value in older adults. Our objective was to quantify the prevalence and variation of low‐value PSA screening across the Veterans Health Administration (VA), which has instituted programs to reduce low‐value care. DESIGN Retrospective cohort. SETTING VA administrative data, 2014 to 2015. PARTICIPANTS National random sample (N = 214 480) of male veterans, aged 75 years or older. MEASUREMENTS We defined PSA screening in men aged 75 years or older without a history of prostate cancer as low value, per established definitions in Medicare. We calculated screening rates overall and by VA Medical Center (VAMC), adjusting for patient and VAMC‐level factors. We characterized variation across VAMCs using the adjusted median odds ratio (OR) and compared the adjusted OR of screening between VAMCs in different deciles of low‐value screening rates. In separate sensitivity analyses, we assessed screening in veterans at greatest risk of 1‐year mortality and among veterans after excluding those who underwent prostatectomy, had a prior PSA elevation, or had a clinical indication for testing. RESULTS Overall, 37 867 (17.7%) of veterans underwent low‐value PSA screening (VAMC range = 3.3%‐38.2%). The adjusted median OR was 1.88, meaning the median odds of screening would increase by 88% were a veteran to transfer his care to a VAMC with higher screening rates. Veterans at VAMCs in the top decile had an adjusted OR of 12.9 (95% confidence interval = 11.0‐15.2) compared to those veterans in the lowest decile. Among veterans with the greatest mortality risk (n = 23 377), 3496 (15.0%) underwent screening (VAMC range = 1.7%‐46.3%). After excluding veterans with a prior prostatectomy, PSA elevation, or a potential clinical indication, 31 556 (14.7%) underwent screening (VAMC range = 2.0%‐49.9%). CONCLUSIONS In a national cohort of older veterans, more than one in six received low‐value PSA screening, with greater than 10‐fold variation across VAMCs and high rates of screening among those with the greatest mortality risk. J Am Geriatr Soc 67:1922–1927, 2019