Author + information
- Received October 29, 2018
- Revision received December 30, 2018
- Accepted January 7, 2019
- Published online April 1, 2019.
- Division of Cardiovascular Medicine, Department of Medicine, Vanderbilt University Medical Center, Nashville, Tennessee
- ↵∗Address for correspondence:
Dr. Joshua A. Beckman, 1215 21st Avenue South, MCE 5472, Nashville, Tennessee 37215.
The vast expansion of medical research has prompted clinician-facing organizations to create tools to translate science into practice. In the cardiovascular space, the American College of Cardiology, in collaboration with the American Heart Association, published its first guideline in 1984 and has published >20 guidelines in the years since. The U.S. Preventive Services Task Force (USPSTF) was created by Congress in 1984 “to improve the health of all Americans by making evidence-based recommendations about clinical preventive services.” In 1998, the Agency for Healthcare Research and Quality (AHRQ) was authorized by Congress to organize and support the Task Force.
Recommendations of the USPSTF are translated into practice through 2 avenues. First, legislation requires Medicare to pay for an entire service, including the copay, if the service is already covered by Medicare and the Task Force gives an A or B rating. The second avenue by which recommendations are translated into practice is through influence on both internal medicine and family practice professional societies. These groups use USPSTF recommendations as the starting point in their guidance to clinicians. They do so because, as former American Academy of Family Physicians President John Meigs Jr. notes, the USPSTF performs an “objective, rigorous assessment of scientific and medical evidence.” Thus, the perception of the USPSTF as an objective, neutral, science-based arbiter of evidence has enabled a multipronged system to organize and define useful clinical preventive services in the United States. I have served as a USPSTF-selected external reviewer for >5 years and have come to believe that this perception is flawed.
Recently, the USPSTF re-evaluated peripheral artery disease (PAD) screening with an ankle brachial index (ABI). The Task Force concluded that the evidence is insufficient (I) to assess the balance of benefits and harms of ABI screening for PAD in asymptomatic adults (1). The document’s conclusions have been supported and opposed (2,3). This commentary focuses on inconsistencies with the USPSTF review process, highlights opportunities for improvement, and reveals the effects of a governmental organization’s subjective to adherence to its own rules. This commentary will use the PAD Screening document as a vehicle to highlight 4 key areas of concern where the document deviates from self-created USPSTF standards: incorrect definition of the condition studied, arbitrary population selection, discretionary exclusion of evidence, and inconsistency of benefit required for recommendation.
Nonstandard Diagnostic Criteria Minimizes Evidence Base
The USPSTF writes, “While the term ‘abnormal ABI’ is often used interchangeably with ‘PAD’ in clinical practice and research, this review will differentiate an abnormal ABI from PAD diagnosed by a confirmatory imaging study” (4). This definition is a fundamental misinterpretation of the disease. This requirement for arterial imaging to diagnose PAD contravenes the currently accepted standard definition of PAD and places the USPSTF outside of mainstream medicine. There is no organization in the United States that uses plaque imaging for PAD diagnosis, including the USPSTF parent organization, AHRQ.
As a result of this spurious choice, the USPSTF discards the extensive evidence base showing the validated test characteristics of the ABI in both diagnosis and link to cardiovascular outcomes (5). We oppose the baseless disposal of a significant body of evidence arising from a nonstandard PAD definition. Its exclusion vitiates the value of the recommendation.
Arbitrary Population Selection Reduces Screening Efficiency
The USPSTF Methods Framework states that risk stratification should be used to “accurately target screening, counseling, and prevention toward those persons who are likely to experience greater benefit than harm” (6). However, in the PAD document, the USPSTF considers all “generally asymptomatic adults”—a selection that increases screening inefficiency. PAD does not occur in persons from 21 to 50 years of age with important frequency. Large epidemiological studies, cited in the PAD screening evidence review, establish a high-risk cohort of patients as persons above the age of 65 years. In their response to public comment, the USPSTF explains that “patients in higher-risk groups…would likely already be candidates for interventions based on their global CVD [cardiovascular disease] risk, raising concern about the clinical significance of screening” (1). First, this strawman argument draws equivalence between patient eligibility for treatment and actual physician prescription. Fewer than one-half of persons screened in recent trials for PAD take appropriate medical therapy despite elevated cardiovascular risk. Second, this concern is unrelated to the question: does PAD screening in an age-appropriate group with an ABI provide benefit? Third, recent data show that aspirin therapy does not provide benefit in this age group in the absence of atherosclerosis. The USPSTF’s rules of review suggest that they should refine the screening target to improve efficiency and study the population at risk. The current document does not evaluate PAD screening for those persons at high risk for PAD.
Arbitrary Requirements for Evidence
The USPSTF noted in its 2013 PAD screening recommendations that a significant limitation in the evidence base was the lack of a large screening study to answer their Key Question 1: “Is screening for PAD in generally asymptomatic adults with the ABI effective in reducing CVD or PAD morbidity…or mortality?” In the 2013 PAD Screening Systematic Evidence Review, they highlighted the in-progress VIVA (Viborg Vascular) screening study. The VIVA study, published in 2017, studied vascular screening in 50,000 Danish men 65 to 75 years of age and showed a mortality benefit (7). Vascular screening included abdominal aortic ultrasound and ABI. The authors reported discovery rates of 3%, 11%, and 10% for abdominal aortic aneurysm, PAD, and hypertension, respectively. The number needed to invite for vascular screening was 169 persons to save 1 life over 4.4 years of follow-up. This compares favorably to the ∼1,300 women 50 to 59 years of age needed to screen with biennial mammography to prevent 1 breast cancer–related death at a decade. It would be only the second screening service to reduce all-cause mortality among the >25 USPSTF-recommended services.
However, the USPSTF excluded this study from consideration. They report that “Applicability of such findings in the context of current treatment thresholds for pharmacotherapy based on global CVD risk is uncertain, as nearly all participants would have a 10% or greater 10-year CVD risk based on age and male sex alone and thus would already be candidates for consideration of statins or aspirin” (1). Moreover, “the multicomponent nature of these trials will likely preclude definitive conclusions about the effect of ABI screening alone.” The downgrading and exclusion of this study is subjective and flawed:
• As the USPSTF reports, abdominal aortic aneurysm screening has not been shown to reduce mortality (8). With the addition of ABI screening, a significant mortality reduction was found. There is no proscription against multimodality screening in the USPSTF Methods. Indeed, multicomponent testing was considered for cervical cancer screening this year (9). Excluding the VIVA study on the basis of multimodality screening is arbitrary.
• A positive screening test was deemed important by treating clinicians, significantly changed prescribing behavior within weeks of a positive test, and increased antiplatelet and statin drug prescriptions. The screening identified a significant number of patients for whom the risk factor burden did not meet the treating clinician’s threshold for therapy before consequential testing.
• Multicomponent vascular testing was associated with a decreased number of days in hospital for PAD and myocardial infarction, demonstrating consistency of results.
• The VIVA study was specifically called out by the 2018 senior author in the 2013 PAD Screening Evidence Review draft using the methods paper (10) as a study to watch to answer Key Question 1. The question of study validity has been answered—by the 2013 PAD document authors. There have been no changes to the trial since 2013, and its exclusion is discretionary.
Subjective exclusion of dispositive research published in a high-impact journal subverts the claim of rigor and objectivity.
The USPSTF is Inconsistent in the Benefit from Screening Necessary for Recommendation
As noted in the preceding text, services that receive a grade of A or B by the USPSTF are deemed important because there is a high certainty of substantial or moderate benefit. However, many recommended screening services, per se, have no evidence of direct benefit at all by the USPSTF’s own evidence reviews (Table 1). This is not to say that the conditions discussed do not merit attention and treatment. However, it is evidence of malleability in determining the value of a particular screening service. Recommendations carry weight because they are considered to be objective, solid, and abide by a set of rules. The current set of recommendations for screening do not follow the rules set by the USPSTF itself, diminishing the reputation of impartiality and trust in the organization. Moreover, the USPSTF has been entrusted by Congress to identify services that merit coverage by Medicare. Advocating for a screening service in the absence of benefit subverts this mission. The USPSTF needs to adhere to its published processes to ensure trust in recommended screening services by patients, clinicians, and payers or risk irrelevance.
The USPSTF Screening for PAD Recommendations and Evidence Review have fundamental, but correctable, flaws. ABI screening, as part of a multicomponent vascular screening, adopted as a result of following the rules as outlined by the USPSTF, would prevent more deaths than all currently recommended screening services combined. The addition of ABI screening to recommended screening services is deserved, and the recommendation grade should be revised.
The USPSTF should more strictly adhere to its processes of evaluation and recommendation creation to ensure a perception of detachment and impartiality. Adhering to these rules will improve the health of our patients while avoiding excess cost to the health care system. The USPSTF was empowered by Congress and trusted by physicians because of its neutrality, objectivity, and rigor. A corrective action plan should be put in place to restore faith in this governmental organization.
Dr. Beckman has been a consultant for Amgen, Antidote Pharmaceuticals, AstraZeneca, Bristol-Myers Squibb, Boehringer Ingelheim, Janssen, Merck, Novo Nordisk, and Sanofi; and has served on data and safety monitoring boards for and received research funding from Bayer and Novartis.
Listen to this manuscript's audio summary by Editor-in-Chief Dr. Valentin Fuster on JACC.org.
- Abbreviations and Acronyms
- ankle brachial index
- cardiovascular disease
- peripheral artery disease
- U.S. Preventive Services Task Force
- Received October 29, 2018.
- Revision received December 30, 2018.
- Accepted January 7, 2019.
- 2019 American College of Cardiology Foundation
- U.S. Preventive Services Task Force,
- Curry S.J.,
- Krist A.H.,
- Owens D.K.,
- et al.
- McDermott M.M.,
- Criqui M.H.
- Olin J.W.,
- Halperin J.L.
- Guirguis-Blake J.,
- Evans C.V.,
- Redmond N.,
- Lin J.S.
- Aboyans V.,
- Criqui M.H.,
- Abraham P.,
- et al.
- Lin J.S.,
- Evans C.V.,
- Grossman D.C.,
- Tseng C.W.,
- Krist A.H.
- Lindholt J.S.,
- Sogaard R.
- Guirguis-Blake J.M.,
- Beil T.L.,
- Whitlock E.P.
- Melnikow J.,
- Henderson J.T.,
- Burda B.U.,
- et al.