A Conversation with Dr. Virginia Moyer, Chair of the U.S. Preventive Services Task Force ▲ NCI Cancer Bulletin for November 27, 2012 - National Cancer Institute

NCI Cancer Bulletin for November 27, 2012 - National Cancer Institute

A Conversation with Dr. Virginia Moyer, Chair of the U.S. Preventive Services Task Force

Dr. Virginia Moyer

The United States Preventive Services Task Force (USPSTF) generates evidence-based recommendations about preventive health services, including cancer screening. Mandated by Congress, the task force consists of 16 volunteer experts from the fields of preventive medicine and primary care. Dr. Virginia Moyer, chair of the USPSTF and professor of pediatrics at Baylor College of Medicine, spoke with the NCI Cancer Bulletin about the task force’s mandate, challenges, and lessons learned.
Why was the USPSTF formed in 1984?
The task force was created in response to a perception that physicians in primary care didn’t really focus much on prevention. So, the idea was to look at what works in prevention and encourage primary care physicians to do those things.
Not everything you might do that sounds like prevention actually works, so the goal was to get effective prevention into practice.
How does the task force come up with its recommendations?
Once we’ve decided that we’re going to address a topic—say, a particular cancer screening test—then we lay out a logic model that asks: if you screen, what harms might come from screening, and what benefits might come from screening? Screening leads to detection and treatment, so what benefits come from treatment, and what harms come from treatment?
We very carefully look at all the available research. And then, using an explicit system to grade the evidence , we determine how sure we are that there is a net benefit and how big that potential net benefit is. Using that model, we come up with a recommendation.
The process is clear and explicit, and anyone is welcome to look at our process .
Are the task force members’ opinions independent, or do the members represent their institutions or other organizations?
We’re all volunteers, so we do not represent our organizations. We all have day jobs, but, for example, I don’t go to the task force meetings and say, “I’m representing Baylor College of Medicine.” I’m representing my own expertise.
So we are independent. We are also not federal employees, though the task force is federally supported. The USPSTF is congressionally mandated, and the Agency for Healthcare Research and Quality provides logistical support.
When is screening beneficial?
The only time screening for cancer works is when a test can distinguish people with and without a type of cancer for which we have treatment that works, but not all that well.
If treatment works fantastically well, you don’t need to screen. Testicular cancer is an example; such a high proportion of men do well regardless of stage at diagnosis that screening cannot affect outcome.
Screening is also pointless when no treatment works. With pancreatic cancer, for example, we simply don’t seem to be able to come up with a treatment that works, so there’s no point in screening yet.
But the ones in the middle are the ones where screening has the potential to be beneficial: where treatment does work, but seems to work better if you catch the cancer earlier.
The better treatment gets, the less useful screening is. In breast cancer, for example, what’s happened is that mammography was much more clearly beneficial when treatment wasn’t quite as good. And the apparent benefit of mammography is decreasing, because treatment is getting better.
So there’s this sort of “sweet spot” for cancer screening that sits somewhere in between incredibly treatable and incredibly untreatable. And that’s where we have to have studies, and we have to keep doing studies, because cancer treatment is changing.
Does the task force consider cost when grading a screening test?
We do not do formal cost-effectiveness analyses. We do think about the resource implications of a recommendation, because those are important in any medical decision. For example, our depression screening recommendation says that you should screen when treatment is available. There’s really no point in screening if treatment resources aren’t available.
The one thing we don’t do is look for ways to cut costs—that is explicitly not something we do.
What has been the task force’s biggest challenge in devising cancer screening recommendations?
I think the biggest challenge is that the general public does not understand that cancer is not a linear disease. You don’t start with one cancer cell that, if it isn’t stopped, will inevitably grow and then kill you. That simply is not how cancer works. If cancer worked that way, you would never overdiagnose people because, if you found a cancer, it would be destined to progress. And if you had treatment available that worked better earlier in the course of the disease, screening would always work.
But that’s not how cancer is. It’s a highly varied disease, even within a single cancer type. For example, even within breast cancer, there are some cells that look like cancer under a microscope but never progress and probably regress. Other cells that look exactly the same under the microscope progress very rapidly and don’t seem to be stoppable no matter what you do.
So given that we have that much variability, you can pretty much predict that screening isn’t going to be perfect. What we’re trying to determine is, how good is screening? Is it good enough that its benefits outweigh its harms? Because there are always harms, sometimes real physical harms.
The toughest experience was the 2009 mammography screening recommendation, but the challenge was not really the science. The challenge was communicating with the public.
What did you learn from that experience?
What we learned from the breast cancer screening recommendation experience is that we need to do a far better job of making sure people understand who we are and how we get to our decisions. Since then, we’ve opened up pieces of our process for public comment, which has been incredibly helpful.
So, now, our research plans go out for public comment. We call everybody we can think of who might be interested; we call every advocacy group; we call all of our partners—federal agencies, physician organizations, and other kinds of provider organizations. We want to make sure we’re asking the right questions.
Once we’ve voted on a recommendation and it’s in draft form, the draft goes out for public comment as well. The chance that someone will bring information to light that we didn’t know about is pretty small at that point, but we quickly find out what we have communicated badly and how we can improve our communication.
As painful as the mammography controversy was, it has turned out to be something that probably helped us move years ahead in terms of the openness of our process.
— Interviewed by Sharon Reynolds