In Early Diagnosis, Too, We Find That Less Is More
In Early Diagnosis, Too, We Find That Less Is More
The United States Preventive Services Task Force has recommended against several screenings, yet health plans are finding that it’s not so easy to just drop coverage
The United States Preventive Services Task Force has recommended against several screenings, yet health plans are finding that it’s not so easy to just drop coverage
Michael D. Dalzell
Senior Contributing Editor
Once upon a time, we believed that if you stay outside when it’s windy and cold, you’ll get sick; that women should rest during their periods; and that if you read in low light, you’d go blind. All simple tips for prevention — all since debunked.
Though strategies for maintaining good health have evolved with the evidence, the constants have been prevention and early detection. But excessive health screenings have led to overdiagnosis, costly treatment, and avoidable morbidity. So now, evidence is beginning to suggest that with early diagnosis, less is more.
Since 2011, the U.S. Preventive Services Task Force has discouraged the use of eight screenings that, in broad populations, are either wasteful or potentially harmful. While dropping coverage of them might seem to be an easy call for health plans, it’s not. Most of these services are popular and widely if wrongly perceived as the right thing to do.
“All payers are struggling with these new recommendations,” says Robert McDonough, MD, head of clinical policy and research at Aetna. “It’s fairly complicated to translate the recommendations into coverage policy.”
The reasons — member expectations, politics, liability, and physician attitudes — have payers taking a go-slow approach to digesting some of the USPSTF recommendations. Health plans “are right to be cautious about it, because you’re exposing yourself to criticism regardless of what you do,” says Paul Keckley, PhD, executive director of the Deloitte Center for Health Solutions. For payers — the champions of evidence-based care — less is more puts them in a quandary.
Stepping on toes
Under the auspices of the Agency for Healthcare Research and Quality, the USPSTF reviews scientific but not cost evidence for and against preventive health screenings, counseling, and medications. The panel — experts in prevention from primary care fields — transforms its reviews into recommendations. Many health care policy experts and payer decision makers consider USPSTF recommendations to be the gold standard for prevention.
The task force emphasizes transparency. It sets a research plan for each topic, conducts systematic reviews, votes on a draft recommendation, and encourages comment at each step. When hepatitis C draft recommendations were released for comment in November, “We had a webinar and 50 or 60 partners were on the phone,” says Virginia Moyer, MD, USPSTF chairwoman and a professor of pediatrics at Baylor College of Medicine. “These were advocacy organizations, others who make recommendations, lots of people. We want as much public comment as we can get.”
The panel grades its conclusions from A to D — A being its highest endorsement and D warning that the harms of a service outweigh the benefits. But if the evidence is inconclusive, the result is an “I statement.” Congress latched on to this taxonomy, and ultimately, the Affordable Care Act required that services with an A or B recommendation be made available without cost sharing.
Health plans are free to do as they wish with C, D, and I statements. Given the USPSTF’s output of the past two years, payers have had plenty to think about — most notably, using PSA tests to screen for prostate cancer. The task force’s D recommendation was a change from its 2008 I statement — an indication that evidence changes as time goes on.
But attitudes don’t shift so easily. After the PSA recommendation was finalized last May, the American Urological Association expressed outrage that the task force was “doing men a great disservice by disparaging what is now the only widely available test for prostate cancer.” Otis Brawley, MD, chief medical officer of the American Cancer Society, countered with his hope that the recommendation “shuts down mass screenings, where men are only told that this will help them, which is stretching the truth. They’re rarely told about the potential for harm.” Even Richard Ablin, MD, who discovered PSA in 1970, lamented in a New York Times op-ed that the PSA test had become a “hugely expensive public health disaster.”
The controversy was reminiscent, though on a lesser scale, of 2009, when the task force downgraded its position on mammography screenings for women in their 40s from B to C. The ensuing firestorm became the stuff of legend. People and organizations with something to lose attacked USPSTF’s statistical methodology, accused it of pandering to payers, and appealed to emotion on the floor of Congress. Within months, several professional societies issued guidelines supporting routine screenings for women in their 40s. Good science — or defensiveness?
“For 10 years, I’ve studied how medical management policies are created in insurance plans and how important the professional society guidelines are,” says Keckley. “There have been some interesting data to say the professional society guidelines sometimes are more protective of their own guild [versus] parsing the science.”
The PSA and mammography controversies feed into payer ambivalence about shaping coverage around the USPSTF’s recent string of “nos.”
“We recognize a number of sources for information,” says Richard L. Snyder, MD, senior vice president and chief medical officer at Independence Blue Cross in eastern Pennsylvania. “USPSTF is one, but we also look to the professional societies that have peer review in their process of developing guidelines that do not have obvious conflicts of interest.”
Why do payers pay?
To Snyder’s point, conflicting guidelines make it difficult for members — let alone chief medical officers — to be sure how to proceed. Within two months of the USPSTF’s 2009 mammography revision, screenings of women ages 40–49 dropped 7 percent, according to a study by Mayo Clinic researchers. Lead author Amy Wang, MD, believes subsequent guideline releases muted the effect.
“Without the other guidelines being issued, the drop might have been bigger,” she says. “Within a couple of months, the American Cancer Society, American College of Radiology, and American College of Surgeons released their own guidelines, which went against the USPSTF.” By the end of 2011, the drop had leveled off to 5 percent.
Still, that is a significant reduction — and it suggests that when USPSTF talks, clinicians listen. So then, if the USPSTF concludes that an intervention’s harms outweigh the benefits — as it did with the PSA — wouldn’t a decision to drop coverage for it be a slam dunk?
Not necessarily, and for a host of reasons.
There’s a need for education, says Snyder. “You don’t want to lose credibility with your customers by saying, ‘Yesterday we told you to get this. Now we’re telling you not to get it.’ There is a reason the original guideline was arrived at, and there’s a reason why new evidence leads you to a different result. That is the challenge in communicating to members.”
In part, its about managing your customer. “Members expect — especially with the ACA but even before the ACA — that the preventive services their physicians provide are covered at first dollar,” says Aetna’s McDonough. “There could be member dissatisfaction over [denials of] services that, even though they are not specifically recommended, are commonly performed as part of preventive services.”
The way many standard benefit plans are written makes it difficult to amend coverage policies, at least in the short term. “It’s probably a mistake to have undue specificity over the particular preventive services written into your benefit plan, because those could change,” says McDonough. “It takes a long time to make these changes.”
Complicating the coverage question further, most of the services the USPSTF discourages on a population basis have some clinical utility on an individual basis. Snyder says it’s hard to tell from a claim form alone whether the physician had a good reason for using the procedure.
“When we receive a code on a bill for the test that aligns with that code, it can be difficult to know if it’s intended as a preventive screening or a diagnostic test to rule out a condition. So many of these tests that are not appropriate for screening could be very appropriate if the patient has symptoms or findings that would leave you to believe that they might have that condition.”
So overuse of preventive services is costly, and a movement to address it has been building for some time. The National Physicians Alliance’s Good Stewardship Program, piloted by the American Board of Internal Medicine, developed “top 5” lists of wasteful tests and procedures in internal medicine, family medicine, and pediatrics. The program has expanded, with 10 specialty societies having issued their own top 5 lists (See “Doctor Groups Promote ‘Top 5’ Lists” below.).
Doctor groups promote ‘top 5’ lists
When the National Physicians Alliance launched its Good Stewardship project, the primary care physicians involved agreed that avoiding the interventions on the “top 5” lists — prescribing brand-name statins, ordering a complete blood count at each routine physical, imaging for low back pain, to name three — would improve quality of care. They also expected that patients would be their biggest obstacle.
“The reality is very, very different,” says Stephen R. Smith, MD, professor emeritus of family medicine at Brown University. Smith directs the Good Stewardship Project.
“Most of the feedback that we’ve gotten ... is that when doctors try to do the right thing for their patients, it’s rare that they get pushback from patients.”
The program grew as part of the American Board of Internal Medicine’s Choosing Wisely campaign. Today, 10 specialties have joined, including the American Society of Clinical Oncology.
“ASCO established a Cost of Cancer Care task force several years ago to begin to address escalating costs in cancer care, and is developing several initiatives to help providers and patients with medical decision making as it relates not only to clinical benefits and toxicities, but also to cost and quality,” says Sandra Swain, MD, president of the American Society of Clinical Oncology. “These initiatives include ASCO’s recent publication of a ‘Top Five List in Oncology,’ and introduction of cost information into its clinical practice guidelines.”
“There is no moral obligation to do things that are unnecessary, maybe wasteful, and maybe harmful or risky to the patient,” says Stephen R. Smith, MD, professor emeritus of family medicine at Brown University. He directs the Good Stewardship Project.
NPA studied waste associated with the 15 interventions on its three primary care lists. Among them, a complete blood cell count for a general medical exam was performed at 56 percent of visits. Annual ECGs — a USPSTF D recommendation — took place at 19 percent of visits. In all, the 15 activities ring up $6.8 billion in unnecessary costs each year.
That is just the beginning. The studies on excessive testing pile up (see “Tallying the Excesses of Prevention” on the next page), each one a slice of a pie estimated to be worth $250 billion a year. That includes the expense of follow-up tests and treatments connected with false positives and incidental findings.
“When you put someone through a CT scan, the radiologist is going to read the entire CT, not just the colon. You’re going to have other blips and ditzels in the X-ray field, and those need to be followed up,” says McDonough. “You might see the bottom of the lung and some sort of nodule, and the physician has to decide — Do I follow it up, do I do a biopsy, or do we follow it over time? All of those decisions [have] costs to society.”
Keith Marton, MD, described his wife’s firsthand experience with this phenomenon on KevinMD.com. It started with a CA-125 blood test for ovarian cancer — a USPSTF D recommendation because of its false positives. CT scans were inconclusive. After the removal of her ovaries and uterus and the expenditure of $50,000, no cancer was found.
Politics of health care
The CA-125 illustrates a problem that CMOs face in eliminating coverage for nonrecommended preventive services: state mandates. No professional society advocates routine ovarian cancer screenings, but seven states require coverage of the CA-125. Thirty-six states mandate coverage of the PSA.
“The evidence changes, but mandates stay on the books,” says McDonough. “Some states still mandate coverage of bone marrow transplants for breast cancer, even though we have a body of evidence to show that this service is not beneficial compared with chemotherapy.”
The politics of health care coverage is a beast. Even as the Affordable Care Act requires coverage of USPSTF recommendations graded A and B, an amendment in the House of Representatives tacked on coverage for three breast cancer interventions that USPSTF gave a C, D, and an I — including annual screening mammograms for women in their 40s.
“Before the mammography recommendation, we spent an awful lot of time just wishing that somebody knew we existed,” says Moyer, the USPSTF chairwoman. “Then the ACA happened. When the ACA tied coverage to some of our recommendations, we suddenly became much more in the public eye.”
The task force’s higher profile didn’t immunize it from political persecution. And even though Moyer says being in the limelight taught USPSTF “some very hard lessons about how to go about communicating things” (see “Care Advised in Communicating Recommendations to Patients” on Page 22), not even Reagan or Lincoln would have been able to cut through the emotion of cancer. After the USPSTF PSA recommendations were released, one House bill sought to eliminate the task force, and another would have changed its makeup.
“Every year we get legislators who know someone who had something. They advocate on behalf of those people and draft legislation that conflicts with the evidence,” says Snyder, at Independence Blue Cross.
Part of the problem also lies in the interpretation of medical findings, says Moyer. “We’re better and better at finding things that are littler and littler and less and less important. Yet we’re placing the same importance on those things that we place on the big things. We have this mistaken idea that a single cancer cell, if unchecked, will multiply until it kills you. That is simply not true.”
You might say that after a career in medical education, Smith is involved in a new sort of education now: changing physician practices that have been ingrained through training.
“When I was a medical student on rounds, the attending physician would ask me, ‘Have you gotten this test or that test?’ The attending physician can always extend the questioning to a point where I would say ‘No, I didn’t get that test,’ and then I would be publicly humiliated, with the attending saying how I was jeopardizing the life of the patient.”
Aetna reinforces evidence-based prevention through its quality programs, reminding physicians to provide members with appropriate preventive services. “We limit those kinds of reminders to situations where there’s consensus that a service should be provided,” says McDonough. “We wouldn’t want to have in place a program that sends reminders to physicians to do PSA testing if there’s no consensus that the service should be provided.”
Sometimes, subtlety works. Sometimes, says Keckley, it doesn’t. “Every time a plan writes a recommendation, it’s going to get from a certain group of doctors an in-your-face ‘But my patients…’ or ‘I can tell you in case after case after case’ — which means they’ve seen three.”
So health plans are caught in the middle — and Keckley doesn’t blame CMOs for being averse to risk. “They don’t get a lot of ‘Attaboys!’ They get a lot of kicks in the rear, whether it’s internal for raising costs or a denial that ended up in a lawsuit.”
If anything, he says, the ACA gave plans an evidence-based defense against assertions that its decisions are arbitrary. Qualified plans will consistently embrace sets of recommendations. With lower-level plans facing premium pressures, “It created an unintended consequence that some employers and the individual market will say, ‘For my basic silver- or bronze-level product, I don’t need all that stuff.’ That will be an interesting discussion.”
“We’re going to wait for the professionals in our community to help us decide when it is appropriate to change our guidelines,” says Snyder.
USPSTF screening recommendations, 2011–2012
What the letter grades mean
Over the last two years, the U.S. Preventive Services Task Force has weighed the clinical evidence for and against adult screenings for 12 conditions.* USPSTF assigns a letter grade to each:
A = service has a substantial benefit
B = service has at least a moderate benefit
C = professional judgment warranted; service may be beneficial, depending on patient characteristics
D = harms of the service outweigh its benefits
I = evidence to assess the balance of benefits and harms is inconclusive
Current and past USPSTF screening recommendations
|Recommendations in force|
|Bladder cancer||I||Asymptomatic patients||D (2004)|
|Cervical cancer||A||Women ages 21–65 every 3 years with Pap smear or every 5 years with Pap smear and HPV test if age between 30 and 65||
A for sexually active women;
I for HPV test (2003)
|D||Women with hysterectomy|
|D||HPV testing in women ages <30|
|Chronic kidney disease||I||Asymptomatic patients||New recommendation|
|ECG for CHD prediction||D||Low-risk patients||D for low-risk patients (2004)|
|I||Asymptomatic patients at moderate to high risk|
|Hearing loss||I||Asymptomatic patients ages ≥50||B for periodically asking about hearing C for audiometric testing (1996)|
|Obesity||B||All adults||B for obese adults (2003)|
|Osteoporosis||B||Women ages ≥65 and high-risk postmenopausal women||B for women ages ≥65 and for those at high risk ages 60–64 (2002)|
|Ovarian cancer||D||All women||D (2008)|
|Prostate cancer, using PSA||D||All men||I for men age <75;
D for men ≥75 (2008)
|Testicular cancer||D||Adolescents and young men||D (2002)|
|Draft updates to screening recommendations|
|Hepatitis C||B||High-risk adults||
I for high-risk adults;
D for asymptomatic adults (2004)
|C||All adults born 1945–1965|
|HIV||A||All people age 15–65||A for high-risk populations;
C for general population (2005)
* In addition to recommendations for preventive screenings, the task force has issued recommendations (not shown) for counseling, medication use, and behavioral interventions. CHD=coronary heart disease, ECG=electrocardiogram, HPV=human papillomavirus, PSA=prostate-specific antigen.
Source: U.S. Preventive Services Task Force
USPSTF — Why no specialists?
The U.S. Preventive Services Task Force is a panel of 16 volunteers from primary care fields, including internal medicine, family practice, pediatrics, geriatrics, and OB-GYN. The task force also includes two epidemiologists and a medical director for preventive care at Group Health Cooperative.
USPSTF Chairwoman Virginia Moyer, MD, says the panel has no specialists because “We speak to primary care doctors.” There are no oncologists, for instance, on the task force, because “Oncologists don’t screen for cancer; they manage it when it’s been diagnosed.”
Putting specialists on the panel, she says, would be problematic for several reasons. “If we put a urologist on the panel, would we also need to put an ophthalmologist on the panel? And a gastroenterologist? And a cardiologist? Pretty soon, the panel would have 50 people on it — each of whom is representing a specialty and each of whom would have to be recused from the discussion of that particular topic because they would have a conflict of interest.”
Tallying the excesses of prevention
|By the numbers||What?||USPSTF recommendation|
|21%||Asymptomatic people ages 75–79 who have unnecessary colonoscopiesa||C (ages 76–85)|
|28%||Physicians who routinely screen for ovarian cancerb||D|
|122||Cases of early-stage breast cancer per 100,000 detected through screening mammographyc||
B (ages 50–74, biennially)
C (ages 40–49)
|8||Of the 122, number who would have progressed to advanced diseasec|
|$5,227,306||Amount spent to prevent 1 death from PSA-detected prostate cancerd||D|
aGoodwin GS, et al. Arch Intern Med. 2011;171:1335–1343
bBaldwin LM, et al. Ann Intern Med. 2012;156:182–194
cBleyer A, Welch HG. N Engl J Med. 2012; 367:1998–2005
dShteynshlyuger A, Andriole GL. J Urol. 2011;185:828–832
Care advised in communicating recommendations to patients
After the political circus that followed the U.S. Preventive Services Task Force’s 2009 revision of its breast cancer recommendations, the panel’s work was put on ice. Kenneth Lin, MD, a former USPSTF staffer who wrote several evidence summaries, wrote on his “Common Sense Family Doctor” blog that the Obama administration would “tolerate no further threats to the passage of health care reform legislation” and that controversial statements would have to wait until after the 2010 midterm elections. USPSTF produced no recommendations in 2010.
What the task force has learned since then could be instructive for health plans. “Communicating D recommendations [harms of the service outweigh its benefits] has been particularly challenging,” says USPSTF Chairwoman Virginia Moyer, MD.
“In an effort to improve our communication, we are partnering with many more organizations than we have in the past,” she says, noting that more than 75 people from various health care entities are in the room when the task force meets. “Those people are welcome to speak, and we value their input highly.”
For medical directors who have been around awhile, the beating USPSTF took in 2009 and 2010 may remind them of the managed care backlash of the 1990s.
“We’re going back there; everybody knows it,” says Paul Keckley, PhD, executive director of the Deloitte Center for Health Solutions. “The difference this time is there will be substantially more data, and it will be passed through various clinical information technologies. So Joe Sixpack can pull it up on his mobile phone and know the strength of evidence for this is A2.”
Though the information that feeds payer decisions will be more transparent, it matters little if the consumer’s interpretation differs from the intended message.
Amy Wang, MD, of the Mayo Clinic, researched the effect of the mammography guideline and documented a 5 percent drop in screening mammography two years after USPSTF issued it. Did USPSTF give women in their 40s the message that I don’t have to do this now? The guideline says women over 40 should have a discussion with their physicians to see if it’s right for them. So if the message people are getting is, ‘I’m 40 and I don’t need a mammogram,’ we’ll have a problem. But if the message comes out as the USPSTF had intended, then I don’t think there should be a major public health problem.”