It appears that my recent post about the Lancet study has stirred things up a bit over at Dr Crippin’s blog. The trans-Atlantic discussion in the comments section is quite an interesting read, and I encourage you to head one over there for it. From what I read, we Americans have slightly different recommendations for mammogram screening than do our counterparts in the UK.
Mammograms in America
We recommend routine mammograms every 1-2 years in women age 40 and over. (Read the NCI justifications for this recommendation here) Although this screening is not universally free, it is covered by all insurances (some annually, some every 2 years). For women without insurance, there is a free national screening program, and many local screening programs exist as well. Women can self–refer, but must identify a physician to receive the mammogram report.
Mammograms in Britian
In the UK, breast cancer screening appears to be free to all for women through the NHS Screening Programme, but is not routinely offered until after age 50, and then every 3 years:
The NHS Breast Screening Programme provides free breast screening every three years for all women in the UK aged 50 and over. Around one-and-a-half million women are screened in the UK each year. Women aged between 50 and 70 are now routinely invited.
Because the programme is a rolling one which invites women from GP practices in turn, not every woman will receive an invitation as soon as she is 50. But she will receive her first invitation before her 53rd birthday. Once women reach the upper age limit for routine invitations for breast screening, they are encouraged to make their own appointment.
Higher risk women can be identified by their physicians and referred to more intensive screening programs in the UK. The women in these programs either have BRCA mutations or a 20% or higher chance of being a BRCA carrier. The intensive screening programs offer individualized risk assessment and screening, including mammography, sonograms and MRI.
It appears from the NICE guidelines (see below) that only women who raise concerns about their family history to their GP get referrals to high risk centers:
1.3.1.2 Healthcare professionals should respond to women who present with concerns but should not, in most instances, actively seek to identify women with a family history of breast cancer.
Some thoughts
Both the UK and American program recommendations are evidence-based, but appear to be relying on either different evidence, or making different conclusions based on the same evidence. Perhaps concerns about liability and very strong breast cancer awareness and concerns among American women are driving our guidelines a bit more than those in the UK.
The increase in mammogram use in the US and the UK has been associated with a decline in mortality from breast cancer in both countries. Advances in treatment, however, may be as much responsible for the decline in mortality, since in the US, the rate of new advance breast cancers has not changed, despite a marked increase in diagnosis of early stage and in-situ cancers.
One could argue that the increase in mammograms has done nothing more than diagnose early indolent cancers that would never have killed anyone. It’s a salient argument, but until we have some way of distinguishing who is going to get the aggressive vs indolent tumors, it doesn’t change screening. However, advances in proteomics and DNA testing are rapidly being made, and currently are driving treatment decisions so that women with more agressive tumors get more agressive treatments.
Someday, we’ll have a better screening than mammogram. It may be MRI, it may be a serum test. It may even be a genetic test to identify the women at risk, so that those at no risk can avoid mammograms altogether. Wouldn’t that be great?
Category: Second Opinions
Really interesting postings!
I’m also willing to guess that the difference in recommendations comes because of the ways the programs are paid for and who profits.
The NHS makes recommendations and then pays for mammograms. By doing them every third year for women over 50 (in general), they save 2/3rds the money on the every third year AND the cost of, say, 3 mammograms for women between 40 and 50. They take a higher risk of missing aggressive cancers in the off years of the 50s, but reduce the risks (and associated costs) of false positives.
In the US, the recommending bodies are either governmental or an organization made up of people who MAY profit from doing mammograms. (An MD who’s part of a multi-practice clinic that includes radiology may profit from recommending more mammograms if the clinic profits from doing them, no?) (If I were really cynical, I’d say that the clinic profits from doing the procedures to follow up on false positives, too.)
On the other hand, the payers are either a separate governmental entity, the patient, or the patient’s insurance.
That means some recommender(s) may be biased towards recommending more mammograms, while insurance companies might well try to limit the number they’ll pay for.
It’s complicated and confusing, and the best we have is really partial knowledge. And even that’s hard to put together well.
Once again, thank you for taking on a challenging issue. I am a big fan of Lillie Shockney’s “Ask An Expert” section of Hopkins website (www.hopkinsbreastcenter.org)and have observed how many younger women have breast ca, an issue which Lillie has addressed.
I experienced my first call-back a year ago following a Cat 3 Bi-Rads. I was followed for a year and have now been cleared to return to normal screening. I also have one first-degree relative with breast ca. The stress of receiving an abnormal letter, the inability to be quickly scheduled for a comprehensive MMG and u/s, and a referral to a breast surgeon provoked enormous stress. I do tend to agree with the statement that “mammograms tend to be over-reported” in the U.S. (based on the call-back rate). I also believe that there needs to be a better way of “lowering the recall rate.” However, those are not reasons to deny screening to women in their 40s. I am assuming that this a difference between the “public” and “private” system of the two countries.
I am curious as to your recommendations re density issues. Will insurance pay for an annual u/s for density?
Bardiac:
I believe (perhaps naively) that the folks making mammogram recommendations are driven by their concerns for women’s ehalth and not by reasons of money. However, cost-benefit analysis is always a part of the screening decissions, so in that sense, they are driven by money…
Anonymous-
No, insurers will not always pay for routine breast ultrasound, although if there is a fmaily history, a case can ususally be made. The reasons to recommend sonogram in women with dense breasts is that the sensitivity of mammography in these women is reduced.
Thansk for visiting.
The problem that the posts don’t address is that screening is not risk-free. When the benefit is small, the risks are important.
20% may sound huge but for an individual it is meaningless unless taken in the context of an individual risk. For an individual, the ARR (or NNS) is more important. As far as “actually screened” vs “invited” is concerned, there is a reason epidemiologists use “invitation to screen”. It’s been shown that women who refuse invitation are also less likely to go to the doctor when they can feel a tumor.
I also found interesting the following comment in the previous post as it seems to completely ignore the fact that considerably more women are diagnosed with cancer in the screened group:
We all know that cancers picked up earlier are more likely to be amenable to breast-conserving surgery and less likely to require chemotherapy. I don’t know about you, but if I am going to get breast cancer, I would rather have it diagnosed at a stage where I can keep my breast and avoid chemo. Also, chemo can be associated with secondary cancers years later, something a 10 year study would not pick up.
While the rate of mastectomies may be less in the screened group, the total number of mastectomies is higher – it’s been shown many times. Many DCIS may require mastectomies, and many “overdiagnosed” cases may require radiation. Radiation has also been associated with secondary malignancy, including soft tissue sarcomas – see NCR PDQ website, harms of screening section.
The inability to predict which tumor will spread is not a “salient” point. It is a vital piece of information that women need to make truly “informed” decisions, yet it is the one that doctors always “forget” to mention.
As far as chemotherapy is concerned, the recommendations for the latter are constantly revised to include more early stage cancers, so unless you can show a study that compares an absolute numbers of women treated with chemotherapy in screened and unscreened group you cannot claim benefit. What you are saying is that (chemotherapy cases)/(total diagnosed cases) is smaller in screened women. But you are forgetting to mention that the (total number of diagnosed cases) is significantly higher in screened women, it is not clear if less screened women are treated with chemotherapy than un-screened.
Recommendations notwithstanding, women should be provided accurate information about both the magnitude of benefit as it applies to them (and not the meaningless RRR) as well as risks of screening.
ANONYMOUS:
Well thought out comments, thank you for them. So I take it then, that you are arguing against routine mammograms for women between ages 30 and 50? And, if you are female, that you do not get them yourself if you are between ages 40 and 50?
Re. the US, a bit of history – please note the following document, written by a panel of experts in evaluating evidence who were selected on the basis of not having vested interests for or against screening: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=9267441&dopt=Abstract.
The National Cancer Institute, who sponsored the document, the next day overturned the panel’s findings and recommended screening for women 40-49.
And what about the Canadians?
A large RCT in women 40-50, with excellent mammography, showed that mammo plus high-quality (>5 minutes) physical breast exam, and instruction in breast self-exam, did not lower breast cancer deaths compared to physical exam and BSE training alone: “RESULTS: The 105 breast cancer deaths in the mammography group and 108 breast cancer deaths in the usual care group yielded a cumulative rate ratio, adjusted for mammography done outside the study, of 1.06 (95% CI, 0.80 to 1.40). A total of 592 cases of invasive breast cancer and 71 cases of in situ breast cancer were diagnosed by 31 December 1993 in the mammography group compared with 552 and 29 cases, respectively, in the usual care group. The expected proportions of nonpalpable and small invasive tumors were detected on mammography.” (http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=12204013)
Anecdotally, my doc here in the US recently admitted to me that she recommends mammography to me out of fear of liability (I’m 42). I suppose I could always promise not to sue her…
Dr Emily- Thanks for your comments, very interesting re the NCI report. I wonder whether the group would make any different conclusions noe (some 8 years and a few more studies later…)
I read the Canadian study, and it has some limitations that the researchers feel do not impact their results, but which still concern me, especially since they state that their study is not powered to detect a difference in the range of 20-25%, which is where it appears to be in the studies that do show a benefit.
They did see a difference in tumor size (smaller in the screened group), but no differnces in nodal mets. However the screened group had higher nodal involvement at baseline, and the randomization has been called into question. Additionally, close to a quarter of the non-screened group was screened anyway.
The researchers perfromed ad hoc analysis that suggests that none of the individual limitations of the study appear to impact the results, but when one considers them all, I still have my questions.
As breast cancer treatment becomes increasingly effective at preventing mortality, getting the power to distinguish the impact of mammograms on mortality becomes harder. This was a problem in the Lancet study.
The Canadian researchers did not address the impact of screening on breast conservation, which I think would be animportant outcome variable to women with breast cancer.
I think the benefit to mammograms before age 50 may be most marked in women over age 45, and perhaps a compromise would be to start the screening then rather than at 40.
Thanks again for commenting – I am finding this back and forth quite interesting.
So I take it then, that you are arguing against routine mammograms for women between ages 30 and 50? And, if you are female, that you do not get them yourself if you are between ages 40 and 50?
First, sorry it took me so long to answer – I thought that as I was too late posting to the thread nobody was reading it anymore…
To answer your questions. For high-risk women in their 30s — I don’t know enough about evidence for these women to form an opinion, I just asked about evidence for this group and mentioned some issues. Hopefully other people have better information here.
For average-risk women in their 40s – I am not as much arguing against mammograms as for accurate information and personal choice. I believe there likely is some small mortality reduction (from breast cancer – this is an interesting article on how it correlates with all-cause mortality) but it comes with a price. I think women should be aware of both benefits and risks when they make a choice and that this decision is really a choice and not an obligation or something “responsible people do”.
I think that information about the benefits should be expressed accurately using NNS instead of RRR, and that the information about harms shouldn’t be sugar-coated or hidden. As far as less treatment is concerned, I believe women need to be told that yes, in some cases they may get less treatment – if they get exactly the type of tumor that is destined to spread but that is detected before it does – but in other cases they might be subjected to unnecessary treatment and might even get unnecessary mastectomies or radiation.
As for myself, yes I am in my 40s, average risk. For the moment I am not getting mammograms, and I haven’t made a decision for later recommendations notwithstanding. I know all arguments for- and against- as I’ve been pretty thorough in researching information about recommendations, trials, criticism of the trials, author replies, conflicting estimates of overdiagnosis, rapid responses, etc (in cases where the information is free, though, I only read abstracts where it is not). To me it really comes down to what is more important for me: 1/1000 chance (1/2000, 1/500 – depending on age and trial interpretation) that my life will be saved vs a slightly higher chance of being converted into a cancer patient unnecessarily; the level of anxiety I feel before mammograms and that I would feel in case of false positives. I reserve the right to change my mind at any time.
Diora- I appreciate your point of view, and howwell thought out it is. I also must say that this discussion, and that over at NHS Blog doctor, has truly been the first time I have ever really spoken with anyone who truly questioned the recommendations that we follow here in the US.
It’s pretty much just an intellectual exercize, however, since the medico-legal and media climate in which I practice precludes me from doing anything but following those guidelines. In truth, I spend most of my time convincing my patients that they should wait until they are 40 to start routine mammograms…