Developing Appropriate Quality Measures in Oncology

Leonard Lichtenfeld, MD: Quality measures are interesting because they’ve been around for a while, and there have been a lot of discussions about how to do those best. I will share with you that I serve on the Cancer Standing Committee (CMS). It’s one of the quality committees from the NQF, or the National Quality Folks. They vet very carefully in a public process, in a very detailed process—and having sat through the meetings I can attest to that—whether or not a quality measure is ready for prime time. And then CMS takes that quality measure, and sometimes CMS takes it without the NQF seal of approval.

I actually will go back to the day I sat in a fairly large meeting a number of years ago, when a quality measure was as basic as taking blood pressure when a patient walked into the office. We’ve come a long way. When do you do a skin exam for a patient who has a diagnosis of melanoma? Those are the kinds of early quality measures.

So, the NQF does it, the AMA does it, and professional societies do it, and ultimately we try to be as evidence-based as we possibly can. And as I said, the NQF process is very rigorous in whether or not it accepts the measure, and there’s a lot of opportunity for public comment. The AMA has a process where they have room for public comment.

But I’ll tell you where the problem is from my perspective. The problem—and I’m not alone on this—is that it’s not just about the point in time. Did a woman get her mammogram at an appropriate interval? It’s a process of care. Is that woman at high risk or average risk? If she’s at average risk, did she get a mammogram consistent with recognized guidelines? And then there are various guidelines. The American Cancer Society has some, and the Service Task Force has others. But I’m not getting into that discussion.

What I am saying is, did she get her mammogram? If she had an abnormal mammogram, what was the outcome of that? If it turned out that she had cancer, did she get the right follow-up test? Did she get it in a timely way? If she had cancer and then the treatment, was the treatment appropriate for what was happening? And then did the radiation therapy start at the right time? Did the medication, the chemotherapy, or the hormonal therapy start at the right time? And was she followed up appropriately? That’s a process. That’s what we should be doing.

When we talk about all these EMR systems, or electronic medical record systems, why can’t we really do quality? Why can’t we make sure that people have access to care? Why can’t we make sure that if they have a problem, the care is the right care? And then why can’t we know what the outcome is for that person? That’s a big change. With everybody talking, you hear all of this about health information technology, electronic medical records, artificial intelligence, and all those things, and yet we can’t know the outcome—now, some institutions may be able to, but many patients go from place A to place B to place C to place D, and nobody links it together. And who’s worse off for it? The person getting the care. We can do much better.

Value frameworks are being established by organizations like the National Competence of Care Network, or NCCN, ASCO, and others. They’re a work in progress. Sometimes you see them, and I will share that when we looked at some of the original ASCO work, we did have some problems with it. We shared our opinion, and when I say “we,” I mean the American Cancer Society. But we had to start somewhere.

Now the question becomes, What’s the value, and who establishes that rule? And some will say that ICER (Institute for Clinical and Economic Review) has had some criticism, say that perhaps they’re too strict. Others say ICER is doing a great job. I’m not taking one side or the other on that, but one person’s value may be another person’s life. We’ve had a long history of medicine, in this country, of letting people make decisions for themselves without that restriction. And so there is no easy answer, but at some point, we’re going to have to figure out how to make this work. It’s not just opening the spigot more and more, because that spigot is endless.

Let’s put aside ASCO, let’s put aside ICER and NCCN, for the moment—what we don’t do a good job at is necessarily following the evidence. And it’s not just in oncology, in cancer care, but it’s in many other areas of medicine. Why are we still doing some of the things we do when the evidence doesn’t support the fact we should be doing it? Because we’ve always done it that way? That’s not where we should be, and there has been a slow transition. No medicine changes over decades—sometimes it changes over time, but sometimes when the evidence comes out, people change right away.

The best example that I use is hormonal therapy in postmenopausal women. Evidence came out that it didn’t work as well. Utilization dropped immediately. Breast cancer incidence dropped pretty rapidly at that point. But the gynecologists were very upset about the way that was handled, and there has been a balance over time. That was an example of when the evidence made an immediate change. Generally, evidence takes a long time. The PSA test is an example in oncology. People came out and said, “The studies were never done to show that it made a difference.” There are some people who still feel it does, but then the studies came out that it went in a different direction.

Now we have reached a balance. We have treatment where we don’t treat people, where we follow them, and some people do need treatment. And more and more, the discussion has become “What’s right for you as an individual?” as opposed to “I’m a doctor; I think this works.” We’ve had a long history of patients, consumers, and doctors thinking something works, and when the evidence shows up to say it doesn’t work the way we thought, we need to become more responsive in understanding that evidence.