XX Factor: the blog

Smells Like Rationing

We keep hearing from proponents of health care reform that government rationing of health care is a “canard.” We don’t have health care reform yet, but with the new recommendations from the U.S. Preventive Services Task Force that women shouldn’t get mammograms until the age of 50, and then only every two years, it feels like we’re getting the rationing.

The Los Angeles Times writes that “[i]nsurance companies and Medicare administrators … said they they would continue to pay for the procedure -- although it is not clear how long they can resist the panel's influence.” The LAT adds that the panel’s recommendations are “generally followed” by insurers and Medicare. (The NYT does say that Congress requires Medicare to pay for annual mammograms, which provides some measure of comfort.) The panel is made up of “health care experts” but no oncologists, and not surprisingly, oncologists and organizations like the American Cancer Society are unhappy about the new guidelines.

There are legitimate concerns to be addressed regarding mammograms. Mammograms expose women to radiation, and a false positive on a mammogram can lead to an unnecessary biopsy. But in my eyes, those concerns pale in comparison to the fact that breast cancer in younger women can be more aggressive and more resistant to treatment. What boggles my mind is that the panel worries about “anxiety” resulting from false positives and complications from a minor procedure like a biopsy, but describes as “modest” the 15 percent reduction in the death rate that has resulted from mammograms. I doubt any women who have survived breast cancer because of early detection would consider that to be a “modest” benefit.

The panel’s recommendations aren’t that different from the NHS guidelines in Great Britain, where women over age 50 are “invited” to have a mammogram every three years. That should raise a red flag: Women in the United States are more likely than their British counterparts to be diagnosed with breast cancer, but they are also more likely to survive. If the cancer is caught early, the survival rate in the United States is 97 percent, compared with 78 percent in Britain. That sounds like an argument for maintaining our current standards, not reducing them.

Tags: breast cancer, health care reform, mammograms

Rachael Larimore Slate copy chief and mother of three. Addicted to coffee, Facebook, and the Sprout channel.

Comments

Good Blog

By: VioletRose | Sun, 03/21/2010 - 01:57

It is great to hear from you.You have done a great job.
Stream Movies

Don't Worry

By: patresponse | Wed, 11/18/2009 - 18:17

The cries of "rationing!" by Rachel and her ilk have succeeded in getting HHS Secretary Sebelius to declare that U.S. mammogram policy will be unchanged, no matter what actual doctors and scientists and silly folks like that recommend. http://www.msnbc.msn.com/id/34019898/ns/health-cancer

And this is why we will never be able to bring down health care costs, if we simultaneously retain the influence of the right-wing scare machine AND give government more control over health care:
Every time the cost-benefit analysis indicates that we should redirect our resources (e.g., from mammograms at 40 that only the women who are currently insured can get, to mammograms at 50 that will cover every legally resident American), the right wing will declare that the evil socialist Obama is rationing our care, and the Democrats will back away with their hands up "Oh no no no, we're not going to change anything." It will happen over and over, and at the end of the process the right wing's prediction will have come true: we'll be spending more on health care, because we're covering more people without having made sensible, science-based decisions about what health care is most useful.

US vs UK comparison very misleading

By: Kierra | Wed, 11/18/2009 - 15:30

The "survival rate" that you give at the end of the column is a five-year survival rate according to the source link. This is very misleading in a discussion on when to start screening. Even if treatment made no difference in how long a woman survived her cancer, earlier screening and diagnosis would still show an increase in survival time after diagnosis based only on the fact that the cancer was caught sooner in its natural progression. This is called lead-time bias. (There is a good discussion of this bias as well as other factors that impact the effectiveness of early cancer screening here: http://scienceblogs.com/insolence/2007/04/detecting_cancer_early_part_1_...)

The link for the numbers you give also doesn't give the primary source of the data.

not a supporter

By: GingerR | Wed, 11/18/2009 - 13:43

How are we going to extend coverage to those without when we continue to pile requirements onto coverage that have negative cost-benefits?

The breast cancer lobby is vocal and powerful so they come down like a rock on suggestions they don't like. But not every woman is high-risk. Why screen every woman all the time?

I think it is wrong to present all these antidotal stories about breast cancers that wouldn't have been caught. How can we insist that more money be spent on breast cancer detection when some women go without.

Every woman who is ready to protest should get on her bathroom scale and see if she's got a BMI of 25 or less. If not then she should focus her efforts on doing something that has been shown to reduce her risk - that's maintaining a normal weight. Heave forbid, maybe she's ward off heart disease as well.

It's just not that simple

By: lizrichardson03 | Wed, 11/18/2009 - 07:48

Some other people have talked a bit about the reasons that the benefits of screening can sometimes outweigh the risks, but I want to address some of the points raised in this post, as some are a bit misleading.

First, false positives and anxiety (which you are dismissive of, but which has real costs and needs to be weighed against the sometimes questionable benefits of population-based screening) are by no means the only concerns around mammography. Screening for cancer also detects cancers that would never have been lethal. Harmless invasive cancers are fairly common (one study, for example, reports that 37% of women aged 40-54 who were autopsied after dying from causes other than breast cancer had lesions of invasive or non-invasive cancer at autopsy, and half were visible on radiography). It is impossible to distinguish between lethal and harmless cancer, so all detected cancers are treated, and I'm sure you're fully aware of the harsh side effects that come with cancer treatment.

"If the cancer is caught early, the survival rate in the United States is 97 percent, compared with 78 percent in Britain."

First, it's always a good idea to check your sources with the Daily Mail, which has a pretty bad reputation in reporting on anything related to science or health. Second, even accepting this characterization as accurate, in this statement you're talking about treatment, not screening, and they aren't even remotely the same thing. I'd also add that comparing the US and the UK is a tricky business unless you control for the fact that US data doesn't tend to capture the uninsured, who don't get access to expensive cancer treatment at all, and die at a predictably higher rate. In the UK they have universal access to care, and unless you control for that difference you're going to get skewed results. It's also important to note that early detection does not necessarily lead to a higher survival rate, due in part to the problem of overdiagnosis.

And yes, a 15% reduction in breast cancer mortality is, in fact, a modest population-wide effect, because that reduction has to be weighed against the costs and risks of mammography, just as with any screening programme. Think about it this way (cited from Welch;BMJ 2009;339;b1425):

For every 1000 women who are screened annually for 10 years:

One woman will avoid dying from cancer

Between 2 and 10 women will be overdiagnosed and treated (that is, through mastectomy or radiation or chemo or some combination thereof)

10-15 women will be told they have breast cancer earlier than they would otherwise have been told, but this will not affect their prognosis (that is, early detection will make NO DIFFERENCE to their survival)

100-500 women will have at least one “false alarm” (about half of these women will undergo a biopsy)

My point here is that this decision is not nearly as easy as you lay out in this post. Breast cancer is a highly politicized disease, and part of the problem is that even experts disagree very strongly on the best policy (oncologists do not necessarily have training in epidemiology, so they are not actually the only experts that count, as you seem to imply). Unfortunately, this means that women are potentially left very confused about the best choice for them. Weighing risks and benefits can be difficult work where there is as much uncertainty as there is with cancer screening. We need to have a much more careful, reasoned, evidence-based discussion about it, and to be honest, you're not really helping here.

Ah, any excuse to rag on the health reform bill...

By: auros | Tue, 11/17/2009 - 21:21

Never mind the science, there are Democrats out there, and we must portray all government entities as scary monsters who are out to kill your mom!

Fair to ask the question

By: contango | Tue, 11/17/2009 - 20:58

I like the panel's concern about mammograms causing "anxiety." What about the "anxiety" of finding out you have late-stage breast cancer, which could have been detected earlier with a yearly mamogram?

At best, it's a paternalistic reason; at worst, it may be cover for other, more materialistic reasons.

It's not limited to women, however. The same "anxiety" was cited as one reason to reduce the frequency of prostate cancer screening. The PSA level, like an anomaly found in a mammogram or pap smear, is a preliminary piece of information indicating the need for further investigation. It's not like if you have an elevated PSA level, you run straight to the surgeon - you start a "watch and wait" regimen. Just as you would if you learned you have precancerous cells on the cervix, or something potentially troubling in a mammogram.

If people misunderstand the significance of early screening tools and consequently freak out, then they need patient education about the test in question. But you DON'T toss the preliminary screening just because it's a rough tool. That's ridiculous.

Check out Amy Tuteur on

By: Kit-Kat | Tue, 11/17/2009 - 16:33

Check out Amy Tuteur on Salon--she wrote a post about how these guidelines are not actually new, but are basically identical to guidelines that were proposed a decade ago and shot down by Congress.

By why let a little fact get in the way of a fun, totally unsupported ideological argument?

Does ANYONE edit Rachel?

By: Caerolle | Tue, 11/17/2009 - 16:26

I think they should call her column "Non Sequiter." How is it that she is continuously allowed to make such outrageous links between idealogy and random facts? OH. MY. GOD!

If it looks like and walks like a canard...

By: meimei00 | Tue, 11/17/2009 - 15:25

Healthcare in the US is _already_ rationed. It is disproportionately rationed to people who can pay for it, and dictated by insurance companies. Rationing is rational, but can we find a more equitable way to do it?

You're bringing up an important concept that every primary care doc (myself included) runs into every day: To screen or not to screen? Screening tests are notoriously tricky to devise, and the mammogram comes bedecked with a pink ribbon, making it hard to rationally (there's that word again!) evaluate it. The true measure of a good screening exam is that it significantly reduces mortality (see cervical and colon cancer screening). It's unclear that mammography does that, and in the process of doing so, it is costing resources that we hold very dear (see current health care cost control debate). There is a trope in popular thinking that catching something earlier is better, and that is often true, but in the case of minor, incidental cancers, like many breast cancers, it isn't. What might work for one person does not necessarily bear out on a population level, and that is why we're still looking for better screening tests that will _significantly_ reduce overall mortality from breast cancer.

There is a nice article from a UCSF cancer researcher on this topic in a recent JAMA: http://jama.ama-assn.org/cgi/content/abstract/302/15/1685

Read it and see what you think.