Q&A: ACP's Daisy Smith on Overuse of Diagnostic Tests


Cynthia (Daisy) Smith, vice president, clinical programs at ACP

NEW YORK (360Dx) – Accelerating spending has been one of healthcare's biggest stories for many years. But while the focus has often been on drug costs, the use and overuse of tests are also a significant driver of rising healthcare costs. In January 2012, the American College of Physicians published a report in the Annals of Internal Medicine describing the problem and laying out 37 tests that the authors said clinicians "often use in a manner that does not reflect high-value, cost-conscious care."

The report put forth three principles that clinicians should consider before ordering diagnostic tests or screens for patients: a diagnostic test should not be performed if it does not change management of the disease; a test should not be performed if the pretest probability of disease is low, since this may result in a higher probability of a false-positive result than a true-positive; and lastly, the true cost of a test includes downstream costs such as follow-up tests.

Meanwhile, other efforts such as Choosing Wisely also are raising awareness about unnecessary diagnostic testing and screening. Launched in 2012 by the ABIM (American Board of Internal Medicine) Foundation, Choosing Wisely seeks to foster a dialogue between patients and healthcare providers on wasteful and unnecessary testing, treatments, and procedures. Consumer Reports is a partner in the effort.

360Dx recently spoke with Cynthia (Daisy) Smith, vice president, clinical programs at ACP, about what, if anything, has changed since the report was published, the steps that ACP has taken to address the issue, and what more needs to be done.

Below is an edited transcript of the interview.


What has changed in the five years since the article appeared?

Dr. [Amir] Qaseem who initially published … the seminal paper on … high-value cost-conscious care … launched the initiative with the original paper. What we did was create a framework … because high-value care sounds kind of vague. We really wanted to make it simple, so there are five different steps that doctors need to go through to provide high-value care for their patients.

By high-value care, we mean balancing the potential benefit of any given test or treatment with its harms and costs, with the ultimate goal of improving patient outcomes. It's really training them how to do this balancing act and how to incorporate costs into their decision-making, which had never been there before.

The steps are number 1, understanding the benefit, harms, and relative costs [of care] that you're thinking of offering to a patient. And then step 2 is really the process of decreasing or eliminating tests that you don't think provide any benefits, or [that] may be harmful.

Some of these tests do have harms, and certainly the downstream tests, if there are false positives, also have harms and costs associated with them.

Step 3 is choosing interventions and tests that maximize the benefits and minimize the harms. And [step] 4 is this process of customizing the care plan with the patient so that it incorporates their values and concerns.

One thing we didn't ever want to say [is] these are tests you never want to do, and we wanted to make sure it wasn't one size fits all. We wanted them to practice customizing and individualizing the care to the patient in front of them.

Finally, step 5, which I think is probably the most important step, is looking for system-level opportunities to improve outcomes and minimize harms and reduce healthcare wastes.

So, we launched a high-value care curriculum for internal medicine residents in 2012, which was wildly successful. It's a curriculum that anyone could download for free from our website. Since we launched it in 2012, we've had over 45,000 downloads.

Just to put that in perspective, there are only 396 residency programs in the country. Building on that, we created some curricula for medical students, also for fellows, and then finally for practicing clinicians.

And they're all really case-based teaching and learning. In addition, our high-value care task force, which is [Qaseem's] task force that generates the science and the papers, have published a total of 12 papers basically giving best practice advice.

What's come out of that, which is part of the curriculum, is encouraging residents and students to engage in step 5 along with their faculties, so really looking for system-level opportunities. And that grass-roots campaign really took off. We've probably had a reach of about 200,000 [clinicians], but where the change really starts to happen are in these grass-roots efforts of quality improvement programs.

So, looking for duplicative tests that are often ordered, re-doing order sets, so that tests don't get done on patients that don't need [them].

Have there been material differences in the way physicians practice that resulted from ACP's efforts?

So, that's the issue, it's in the measurement, particularly tests that are ordered unnecessarily. That's just a very hard thing to measure. [The University of Pennsylvania] did a wonderful project, which I helped them [with] because I'm on the faculty there … where they tried to get rid of daily blood draws. They measured their impact on gallons of patient blood that was saved by their intervention and getting rid of the whole conduct of daily labs in the hospital.

So again, on a project by project [basis, there are some positive conclusions] but on a national [level] it's really hard to measure, but the trends are a lot of these localized projects and initiatives have really been successful in decreasing unnecessary screening and diagnostic tests.

Where these local quality improvement initiatives have been [implemented] … they definitely have been successful. And there have been some pretty low-cost interventions that have been done where even just giving a little training to a medical student on a team [to ask] … 'How is this test that we're ordering going to change what we do for this patient?' Just asking one or two very simple questions on rounds … has a really sustainable impact. And that study was done at three or four institutions.

How much of the rise in healthcare spending is attributable to unnecessary testing and screening? The focus seems to be more on drugs.

There's a lot of focus on drugs and the costs of drugs. People like to talk about the costs of drugs, but that's not a huge driver. The big ticket items actually are hospitalizations, because a day in the hospital outweighs everything else, but testing [makes up a part of rising healthcare costs]. It depends on what you include in the costs, but it's not just the cost of the initial test — and this is something that we have included in our curricula — it's the cost of all the downstream tests. If you have a false-positive result, what are all those downstream costs that you have?

[According to the National Academy of Medicine — formerly the Institute of Medicine — in 2010] in terms of the overall waste, unnecessary services — that could be tests or treatments — is attributed $210 billion, compared to prices that are too high, which is $105 billion. And there's administrative costs and other things, but the biggest ticket is unnecessary services. Unnecessary services have a lot in there because it's not just tests and treatments, but hospitalizations, that kind of thing.

How do you address defensive medicine and physicians who order tests because of a fear for being sued for malpractice?

In a lot of these things, we do have a whole thing on overcoming barriers to high-value care where we address malpractice concerns. We've addressed it with statistics and papers that actually show that … with the way our care is so fragmented, there's a much higher malpractice risk for ordering tests and not following [up with the patient].

In the end, it all comes down to the strength of your relationship and clear and concise communication with the patient, and documentation.

We give [clinicians] general guidance on how to better communicate risks, harms, and benefits with patients. And we show them that there's really very little data that supports the perception [that not ordering tests puts them at risk].

The perception is that if you order more tests, that you'll be protected. But the reality is really quite different, and there are a number of stories that we quote.

What trends have you seen with patients?

We create quite a number of patient-educational materials, and when we created these cases for practicing docs, we created take-home tools to help transform their practice. A lot of the tools were patient-education-focused tools, and we partnered with Consumer Reports to help us create and to disseminate some patient-education tools.

We did some live meetings, actually, that included patients, and I was quite surprised. I thought patients would be resistant to this idea, but … given that they're paying more out-of-pocket for expensive diagnostic testing, they really don't want the test if it's unnecessary or maybe harmful to them, or may not be accurate.

I am impressed because I still practice and I see patients, how open they are to having these conversations in an open and transparent way. Consumer Reports created posters and things for examination rooms about Choosing Wisely, and five questions that patients should ask before they do a diagnostic test.

We as an organization also have a center for patient partnership and healthcare, and a lot of patient-educational materials. I don't want to downplay what we're doing, but I think [Consumer Reports] has some pretty unique things with the posters in the room.

Have you noticed any changes in hospitals in their guidelines for staff in terms of using diagnostic tests over the past five years?

Yes, absolutely. Many of the quality improvement projects … are hospital-based because it's much easier to control that, and I think a lot of it is about changing standard order sets and revising standard order sets, and incorporating clinical decision support into the electronic health record.

For instance, the American College of Radiology has appropriate-use criteria, so many folks have incorporated those appropriate-use criteria — when you need which thing — into their electronic health records. You can't just order an MRI of the back, no questions asked. It'll say, 'What's the indication for this?' before you get the order.

So, I would say, particularly now as we're going to bundle payments and advanced alternative payment models and accountable care organizations …  some hospitals are not incentivized to do more [tests], because they're only going to be paid a certain amount, they want to cut down on unnecessary testing as well, so there's some alignment.

In terms of new technologies, like next-generation sequencing, will that continue to drive up testing and unnecessary testing?

I think new technologies may have an impact of bringing healthcare costs down as long as the results are more reliable, more accurate, and are more affordable. I'm thinking about [something] like high-risk HPV testing and patients with Pap smears. Because I can combine that with my Pap smear, I can do them in women over the age of 30 every five years, as opposed to bringing them back and doing the test more frequently.

I do think a lot of the problem we're in now is that patients and clinicians sort of have a love affair with technology, and I think technology just sort of made it so much easier to look … when maybe we didn't necessarily have to look.

I think there is a balance whenever you get a new technology. Sometimes you can get a degree of overutilization … and testing, in general, can beget more testing. The high incidence of incidentaloma on an abdominal CT scan can be up to 50 percent of the time. So, you really shouldn't be doing an [abdominal] CT, because then you maybe end up following up on something that's abnormal but not pathologic and having to do … more scans, or more complicated and more expensive scanning.

This concept of false-positives is, I think, a huge driver in testing.

What we're doing right now is identifying and disseminating best practices in decreasing unnecessary testing and then continuing to do systematic review and do clinical guidelines and best practice advice papers. We've had 12 of those come out of our high-value task force … and in the longer term, I think the real issue is measurement.

We really have to identify and validate high-quality measures for practices to track their performance in this area beyond just claims data.

I think diagnostic firms really need to be cost-sensitive and really transparent about the operabilities, the reliabilities, and validity of the test … what the false-positive rate is, and the harms that are associated with the test, and then really do a better job of educating clinicians how to use the test.

What would be the motivation for them to cut down on testing? They're trying to make a profit.

You should ask me what their responsibilities [are]. It's sort of like asking me what incentives does a pharmaceutical company have to make their medicines affordable. And I'm not a big fan of direct-to-consumer advertising because I think that drives utilization, which I know is the whole point of it, but it also really crowds the clinician with requests for stuff that aren't even appropriate for the patient.

Recently, a study came out about Angelina Jolie and how her experience with BRCA testing drove up the numbers of unnecessary BRCA testing. What was your reaction to that?

Unfortunately, in our society we have celebrities acting as role models, and some of them are not the best role models for high-quality care, but that's where having a clinician [matters]. And by clinician, I don't mean just a physician — high-quality genetic counseling is a real issue in this country because I don't think we have access, given genomic medicine and personalized medicine and where it's going — I don't think we have a sufficiently educated workforce to really help counsel patients about that.

I think it's one thing if patients are asking for BRCA testing, but the number of patients that undergo BRCA testing, or how they interpret and act on those results is really a result of the quality of the genetic counseling and the clinical counseling that they get in that context.

Healthcare isn't provided in a vacuum, tests shouldn't be provided in a vacuum without the adequate clinical, and in this case, the genetic counseling that needs to be wrapped around there.

What is a doctor is supposed to do, though, if a patient wants to have a certain test and is insistent about it? How is the doctor supposed to act if he or she doesn't believe the test is necessary?

We have in our educational portfolio all sorts of communication tools, including videos, annotated videos, training doctors on how to address it when a patient asks for something that's not indicated. A lot of the foundational things is figuring out what the patient is really most afraid about, or is most concerned about, because once you know where the request is coming from, it's much easier to allay their fears.

You don't have to test them to allay their fears, you just have to talk to them.

Now, there are some patients who won't take no for an answer … but I think that most people don't realize that it's not just the financial cost, but there are major costs that come along with overdiagnosis, which means picking up something and labeling somebody with a diagnosis, which may never cause them a problem as long as they live, or picking up an incidental finding. You're taking a well person and turning them into a sick person, at least psychologically.

I remember I had this professor in medical school who actually discovered the gene for Huntington's disease, and she discussed all her work. And she started with a clinical story, which of course turned out to be herself and her own father, [who] died of Huntington's disease.

Here is the clinical researcher who discovered the gene and created the test, and she said 'So, what would you do if you were me? I have not tested myself.' Because why would you want to know that you have a disease which is fatal [when] there's no treatment?

Just because you can do the test and the test exists, doesn't mean that you should do the test, and you certainly shouldn't do the test in a vacuum until you have someone really counseling the patient through that process.

Do you see any changes with the various government task forces that issue recommendations for screening or testing of diseases? About a month ago, a task force recommended against screening for the herpes simplex virus, for example.

Yes, we have a whole clinical guidelines process … and we're doing a lot of our own systematic reviews in house … but it really is all about the evidence and the evolving evidence.

Overall, the conversation is really changing in the sense that I think professionals are becoming more conservative about endorsing new technologies fresh out of the gate and making sure that there's clear evidence, particularly in the screening world.

I think the public and clinicians have a tendency to really overestimate the benefits associated with many screening tests. So, I do think the tide is shifting and the burden of proof is going to be higher to adopt new screening tests because we learn.

If we talk again in another five years, where will we be?

Hopefully, we'll have accurate and better ways to measure unnecessary test utilization and a more thoroughly educated public and clinician population about how to balance benefits, harms, and costs, and better inform their decisions. And then get rid of direct-to-consumer advertising for both tests and treatments.

A girl can dream, right?