Screening Tests in the Age of Austerity

Tension over healthcare spending is at a new high as the country prepares for the Supreme Court's imminent ruling on the Obama administration's healthcare law, the 2010 Affordable Care Act. Laboratorians have long felt the squeeze on healthcare spending, as screening and diagnostic tests have been among the primary targets over the years. Now with renewed controversy about the value of screening tests like prostate-specific antigen (PSA), it seems clinical labs will increasingly face the assumption that overuse and wasteful testing is rampant—a reality laboratorians will have to deal with even as they work to bring in new, innovative tests to improve care. However, the current evidence on overuse of lab testing is slim. And even where evidence does exist, a clear picture has yet to form of how physicians and patients could make better decisions—especially about screening tests.

Healthcare will need the expertise of the lab community because screening remains "inherently messy," according to Russell Harris, MD, MPH, a professor of medicine and director of the program on prevention in education and practice for the University of North Carolina School of Medicine in Chapel Hill. "With false positives, false negatives, and overdiagnosis, screening is not clean—that's just the dilemma we face," he said. "But I think people trained in clinical lab science have a lot to offer here, and I hope they can help us educate clinicians." Harris was a member of the United States Preventive Services Task Force (USPSTF) from 2003–2008.

Overuse Looms, But Evades Easy Measurement

Whether they be regulators, lawmakers, or pundits on the evening news, all seem to agree that the cost of healthcare is becoming unsustainable. Most recently, Medicare has projected that by 2020, national health spending could reach $4.6 trillion and comprise nearly 20% of gross domestic product. Making matters worse, it appears that the nation is addicted to profligate testing and treatments that by some estimates consume up to 30% of healthcare spending.

Despite this highly charged atmosphere, professional and consumer groups have chosen to team up and weigh in on how the nation should grapple with its healthcare spending problem. In a first-of-its-kind response to overuse, a new campaign called Choosing Wisely from the American Board of Internal Medicine Foundation has brought together nine top medical societies as well as Consumer Reports to educate both physicians and patients about common unnecessary tests and treatments. Each physician specialty society published a list of "Five Things Physicians and Patients Should Question." Many of the 45 items implicate unnecessary imaging or laboratory screening tests (See Box, below).

Two separate but related initiatives from the American College of Physicians (ACP) parallel Choosing Wisely. ACP, a Choosing Wisely participant, has announced its own partnership with Consumer Reports. The two organizations are developing patient-oriented brochures and other resources to help patients understand the benefits, harms, and costs of tests and treatments for common clinical issues. The resources will be derived from ACP's evidence-based clinical practice recommendations published in Annals of Internal Medicine and on the Consumer Reports website. This patient-centered initiative comes 2 years after ACP's other project, the High Value, Cost-Conscious Care Initiative, which was aimed at physicians.

The most recent product from the High Value, Cost-Conscious care series, published in January 2012, focused on screening and diagnostic tests. ACP convened a workgroup of physicians under a consensus-based process to identify tests that did not reflect high-value care (Ann Intern Med 2012;156:147–149). Similar to Choosing Wisely, ACP's list of 37 clinical scenarios includes many imaging tests, but about half encompass clinical lab tests (See Box, below).

Despite testing being a target, many of the group's recommendations may not provoke much controversy in the lab community. But if laboratorians do disagree with such recommendations, new research reveals that the medical literature on overuse of lab tests is extremely limited. A study published as part of the Archives of Internal Medicine's Less Is More series reviewed 114,831 publications over 21 years and found only 172 articles that addressed overuse of healthcare (Arch Intern Med 2012;172:171–178). The majority of the studies focused on four interventions: antibiotics for upper respiratory tract infections, and three cardiovascular procedures. Just a handful addressed lab tests, notably PSA and fecal occult blood testing (FOBT).

According to study coauthor Salomeh Keyhani, MD, MPH, the reason for the paucity of studies on overuse of lab tests is clear: too few definitive guidelines. "If you want to eliminate inappropriate care, you have to designate what exactly is inappropriate, which is not an easy thing," she said. "Diagnostic tests are a particular challenge in terms of establishing when it's appropriate to order them. The indications for diagnostic testing are not routinely evaluated in the same way as indications for therapeutic procedures." Keyhani is an assistant professor of medicine and of health evidence and policy at the Mount Sinai School of Medicine in New York.

Moreover, where guidelines do exist, whether for tests or treatments, they often conflict, Keyhani noted. "In the U.S., we have a free market for guidance," she said. "We have every single medical specialty society with its own emphasis and own focus putting out guidelines, and to some extent, they disagree."

Data and Decision-Making

With screening tests showing up on lists of questionable practices and under the spotlight for research on overuse, there should be no surprise that screening tests also stimulate the most public controversy. To be sure, the public has strong opinions. A seminal study in 2004 found that 74% of American adults believed that finding cancer early via screening saved lives most or all of the time, and many said that an 80-year-old who chose not to be screened was irresponsible (JAMA 2004;291:71–78). In addition, two-thirds of respondents indicated they would want to be screened for a cancer even if no treatment was available.

More recently, in 2011 a draft "D" recommendation—the strongest negative statement—from USPSTF against PSA screening at any age led to public outcry and widespread media coverage (CLN 2011;37:11). USPSTF also took a more cautious view of Pap testing in March of this year, recommending women ages 21 to 65 be screened only every 3 years. Then in April, USPSTF published draft recommendations on screening for chronic kidney disease that contains an "I" statement for insufficient evidence. In recent years, USPSTF advisories have taken on more weight as most payers, including Medicare, rely heavily on their recommendations to make decisions about coverage and reimbursement.

Evidence suggests that cancer screening tests in particular have unusual patterns of utilization: potentially significant overuse in some cases, and underuse in others. A Government Accountability Office (GAO) report released in January found that use of some screenings—for cardiovascular disease and cervical cancer—by Medicare beneficiaries generally aligned with clinical recommendations, but other cancer screening tests did not. For example, even though USPSTF recommends biannual breast cancer screening in women ages 65 to 74, only two out of three beneficiaries in this age group received a mammogram in 2008 or 2009. In the case of colorectal cancer screening, only one in four beneficiaries ages 65 to 75 received any of the recommended regimens. With PSA testing for prostate cancer, overuse was the problem: almost half of men age 75 or older were tested, contrary to USPSTF recommendations.

In a more striking example, researchers at the University of Chicago Medical Center evaluated changes in national screening rates before and after the USPSTF 2008 recommendation against PSA screening in men older than 75. They found that PSA screening rates were unchanged from 2005 to 2010 in all age groups (JAMA 2012; 307:1692–1694). Apparently, physicians and patients had completely ignored the USPSTF recommendation.

Patients' enthusiasm alone may not be to blame for such a discrepancy between recommendations and practice. There also is reason to doubt that physicians themselves can comprehend or communicate even the basic facts about evidence when it comes to screening tests. A nationally representative survey of internal medicine physicians found that a majority were easily stumped by questions about basic statistics and would recommend a screening test to save lives based on irrelevant evidence (Ann Intern Med 2012;156:340–349).

The researchers expected that physicians would understand that in screening tests, survival statistics are susceptible to lead-time and overdiagnosis biases. In fact, more than 20 years ago the National Cancer Institute concluded that reduced mortality in a randomized trial could be the only reliable evidence that a screening test saved lives. However, when presented with evidence about two hypothetical screening tests, physicians in the study overwhelmingly favored a test backed by evidence of improved 5-year survival over one with an improved mortality rate.

The inability of some physicians to interpret data about screening tests appears even more dire considering how USPSTF and other groups have leaned toward optional shared decision-making in their recommendations. This concept emphasizes the importance of doctor-patient collaboration to arrive at healthcare decisions and puts greater emphasis on patient autonomy and choice. The concept gained momentum especially with PSA testing after USPSTF's 2008 "I" recommendation for prostate cancer screening in men younger than 75, which suggested that these men "should be assisted in considering their personal preferences before deciding whether to be tested."

Many opinion leaders in healthcare expressed deep frustration with this strategy, questioning shared decision-making for controversial screening tests for which the evidence perplexes both doctors and patients. For example, Allan Brett, MD, is among those who welcomed USPSTF's more conservative 2011 draft recommendations against using PSA tests to screen for prostate cancer.

After those draft recommendations were published, Brett, a professor of medicine at the University of South Carolina School of Medicine and member of the university's Center for Bioethics and Medical Humanities, wrote an editorial praising USPSTF for giving physicians clearer guidance (N Engl J Med 2011; 365:1949–1951). "For two decades, primary care physicians have been expected to present a flawed screening test to patients, cloaking the flaws in an elaborate ritual of informed decision making. In turn, men have been expected to make sense of a confusing mix of hypothetical outcomes," he wrote. Due to the difficulty of digesting the data on screening, these patient-physician discussions about PSA testing were "essentially a charade," and their decisions "reflected their general concerns about cancer or their general inclination to accept or resist medical interventions." Brett is also the editor-in-chief of Journal Watch General Medicine.

Brett also recently wrote about coping with patient pressure for unnecessary tests and procedures more generally (JAMA 2012;307:149–150). Physicians have been overcorrecting for their traditionally paternalistic tendencies since the 1980s, he argued. Now the pendulum has swung too far, with physicians' intellectual authority routinely questioned. "Yes, there should be a partnership between doctors and patients in decision-making, but when it crosses the boundary into things that don't plausibly confer medical benefit, physicians should be able to say no," he told CLN.

These pressures on physicians are one reason that Harris advocates the use of outcomes tables that do a better job of capturing all of the potential benefits and harms of a particular screening test. "If you don't lay it out for people, it's hard for them to see what is meant by harms," he said. "We need to help people see that the benign blood test that you agree to when you have a doctor visit can end up being the first step in a cascade of events that, after a while, you won't be able to control. And that cascade can end up with your being helped, but it might also result in your being hurt."

But is it possible that giving patients more choices can boost screening where underutilization is a problem? New evidence suggests that in the case of colon cancer screening, offering patients choices about the type of test boosted compliance (Arch Intern Med 2012;172:575–582). The researchers viewed these results as particularly significant because patients frequently avoid recommended colon cancer screening. They found that patients for whom colonoscopy was recommended were less likely to complete colorectal cancer screening than either those for whom FOBT was recommended or those who were given a choice between FOBT or colonoscopy. Only 38.2% of the participants in the colonoscopy group completed that procedure, compared with 67.2% in the FOBT group and 68.8% of those allowed to choose their own screening method.

In an invited commentary on the study, Theodore Levin, MD, urged physicians to embrace the study's findings. "If having too many choices leads to confusion, the study … demonstrates that not having enough choice may lead to inaction when the only choice is colonoscopy," he wrote. "When it comes to colorectal cancer screening, providing an option other than colonoscopy for our patients is not overwhelming, but necessary." Levin is a research scientist at the Kaiser Permanente Northern California Division of Research and a gastroenterologist at Kaiser Permanente Medical Center, Walnut Creek.

When Evidence Agrees, the Lab Can Shine

When physician groups and others in healthcare talk about overuse of lab tests, it can sound like bad news for the lab. However, even when the quest for high-value care is driven by forces outside the lab, sometimes the results promote lab testing over other options. For example, high-sensitivity D-dimer testing is a clear winner in the Choosing Wisely campaign, as well as the ACP High-Value, Cost-Conscious Care Initiative. Both lists emphasize that physicians should chose a high-sensitivity D-dimer assay instead of imaging studies for patients with low pretest probability of venous thromboembolism (VTE) (See Box, above).

D-dimer testing is an example of how clear evidence for clinical utility of an assay can boost recognition of the value of the lab's contribution to patient care. Not coincidentally, it also is an area where laboratorians have worked with physicians on clinical practice guidelines. For example, a representative from AACC contributed to the American College of Chest Physicians' (ACCP) 9th edition of its authoritative clinical practice guidelines for prevention and treatment of VTE, and AACC was invited to review and endorse the guidelines (Chest 2012;141:S). The collaboration with ACCP grew out of the involvement of AACC's Evidence-Based Laboratory Medicine Committee with the Agency for Healthcare Research and Quality.

AACC endorsed the guidelines out of a desire to emphasize the utility of high-sensitivity D-dimer testing versus more costly and complex interventions, according to Stephen Kahn, PhD, chair of the AACC committee. "We were pleased to work with the American College of Chest Physicians. They are really a top-notch association when it comes to guideline development and are the premier experts in this area, recognized internationally by people in evidence-based medicine," Kahn said. "In addition, one of AACC's long-term goals is to work more closely with clinical organizations and be recognized as an important player when it comes to issues involving laboratory medicine." Kahn is a professor of pathology and director of laboratories at Loyola University Medical Center in Maywood, Ill.

Considering the difficulty physicians have with understanding and communicating complex information about screening tests, the healthcare community needs the help of laboratorians, said USPSTF Chair, Virginia Moyer, MD, MPH. "People in the lab should be thinking about how they can best communicate with their customers, but that obviously goes both ways. If the people ordering the test can be clear about why and in whom they've ordered it, then the people reporting the results can be much more helpful in the interpretation of the results," she said. Moyer is also a professor of pediatrics at Baylor College of Medicine in Houston.

In an editorial that accompanied the study on physician's understanding of screening statistics, Moyer argued that the research underscored the need for high-quality, evidence-based guidelines, and noted that many organizations have moved to align with the Institute of Medicine's (IOM) standards for guideline development (Ann Intern Med. 2012;156:392–393). AACC's National Academy of Clinical Biochemistry is among the guideline development bodies that have embraced the IOM recommendations.

Moyer said that she also looks to researchers in the lab community for continued improvements in the tests themselves. "There are many, many screening tests which we sure wish were better," she said. "Part of it is that right now, many of them are a little bit like x-rays—they're shadows. PSA is an example. It rises not only in prostate cancer, but also in anything that irritates the prostate. It says something is going on, but not what. And it doesn't even always rise when something is wrong."

According to Kahn, the emphasis for laboratorians will not be in terms of more testing or less, but making sure the right tests are used at the right time. This means a focus on evidence-based laboratory medicine, and more interdisciplinary collaborations like AACC's contribution to the ACCP guidelines. Laboratorians need to be objective about their own lab services and try to keep abreast of the key clinical practice guideline recommendations to be considered in making changes in clinical practice at their own institutions.