Low-Value Diagnostic Testing for Back Pain, Sinusitis, Headache, and Syncope Is Common and Varies Across VA Medical Centers
Low-value care is defined as the use of health services whose immediate or downstream harms or costs exceed their potential benefits (i.e., cancer screening for patients with a limited lifespan). As healthcare costs rise and Veterans receive more care outside VA, quantifying the frequency and patterns of low-value care delivery is essential for VA to reduce wasteful healthcare spending without compromising quality or patient satisfaction. This study sought to determine the frequency and degree of variation in low-value diagnostic testing for four common conditions across 127 VAMCs. Investigators utilized VA data from FY2014-2015 to identify a national cohort of Veterans with any of four conditions: low-back pain (n=343,024; 31%), headache (N=79,176; 7%), sinusitis (N=52,889; 5%), and syncope (n=23,776; 2%). To identify low-value diagnostic testing within VA, investigators applied a claims-based approach originally developed for use in Medicare data. They also sought to examine the degree of correlation between the receipt of different types of low-value diagnostic tests for each condition at the VAMC level. Both sensitive and specific criteria were applied to identify low-value testing. For example, to identify low-value testing for low-back pain, the sensitive criteria captured any Veteran who received low-back imaging less than 6 weeks after receiving a low-back pain diagnosis, whereas the specific criteria further excluded Veterans from the numerator with a diagnosis code for an emergent condition in the 30 days prior to testing and those Veterans who underwent their first test greater than 6 weeks after their initial back pain diagnosis.
- Low-value diagnostic testing for four conditions was common; it affected 5-21% of Veterans, varied 2-to-5 fold across VAMCs, and was significantly correlated at the VAMC level.
- Applying sensitive criteria, the overall and VAMC-level of low-value testing frequency varied substantially across conditions: 5% (range 3-10%) for sinusitis, 13% (9-23%) for headache, 18% (11-25%) for low-back pain, and 20% (16-28%) for syncope. Applying specific criteria lowered the overall frequency and range of low-value testing across VAMCs: 2% (range 1-5%) for sinusitis, 9% (6-15%) for headache, 6% (4-8%) for low-back pain, and 13% (11-17%) for syncope.
- Findings reinforce the need to address low-value diagnostic testing, even in integrated health systems like VA, with robust utilization management practices.
- This study only evaluated Veterans’ receipt of care within VA. Nearly all Veterans aged ³ 65 years are dually enrolled in VA and Medicare, and others receive non-VA care that is paid for by VA.
- This study did not incorporate provider characteristics in its models. Such data are not as readily available, but in one prior VA study, the ratio of nurse practitioners and physician’s assistants to physicians was associated with the overuse of low-value PSA testing.
All authors, except Ms. Huang, are part of HSR&D’s Center for Health Equity, Research and Promotion (CHERP) located in Pittsburgh and Philadelphia, PA.
Radomski T, Feldman R, Huang Y, Sileanu F, Thorpe C, Thorpe J, Fine M, and Gellad W. Evaluation of Low-Value Diagnostic Testing for Four Common Conditions within the Veterans Health Administration. JAMA Network Open. September 22, 2020;3(9):e2016445.