I returned from vacation this past Monday to a slew of headlines and media commentary about newly released guidelines from the American College of Physicians suggesting that women in their 40’s should reconsider the routine recommendation for screening mammograms in that age group.
The headlines were very specific, such as “Benefits of Mammography For Women In 40s Challenged” (Washington Post) and “Mammograms Under 50 Optional for Many Women: Group” (Reuters).
A press release from the College noted that the guidelines were evidence-based and advised women to become part of the decision making process. The risks of mammography included “false-positive results, possible treatment for lesions that would not have become clinically significant, and radiation exposure.”
I have waded through the three articles and an editorial that comprised the report in the current issue of the Annals of Internal Medicine, and I am wondering if the media—and some of the experts who have been quoted—have actually read the articles.
On top of that, I doubt that many have taken the time to actually look at the data in the United States, which has shown a clear decline in mortality from breast cancer in women in their 40’s since the early 1990’s.
Let me note at the outset that I have what some may say is a conflict of interest in writing this particular blog.
First, and foremost, I am (obviously) employed by the American Cancer Society. What you probably do not know is that I manage the department in the Society that is responsible for developing our various guidelines on the prevention and early detection of cancer.
I am also very active—and proudly so—in the American College of Physicians, the organization that authored this guideline (although I had no role whatsoever in the publication or review of this guideline).
Let’s try to take a look at these articles and see what they say.
First, there is an article on the long-term effects on women who have a false-positive mammogram.
This issue has always been of serious concern to those who have studied the pluses and the minuses of screening for breast cancer. In fact, it is a matter of serious concern when discussing screening for any cancer where early detection has been effective, including prostate, cervical and colorectal cancer.
Very important in this discussion is the definition of a “false-positive” result. It can be as simple as having someone come back for more x-ray views of the breast after a screening mammogram. It can be as complicated as recommending a breast biopsy of a suspicious breast lesion which turns out to be negative.
The article is too long for me to summarize here, but when I read it I was struck by the fact that the authors, after having set out the potential issues surrounding a false-positive mammogram, really couldn’t substantiate a significant, large negative impact on women who had a false-positive mammogram.
The results of the various studies they reviewed were mixed, and sometimes contradictory. Some theories were advanced, but no overwhelming evidence that false-positive mammograms were serious detriments to the health and well-being of most women.
The authors conclude, “Some women with false-positive results on mammography may have differences in whether they return for mammography, occurrence of breast self-examinations, and levels of anxiety compared with women with normal results. Future research should examine how false-positive results on mammography affect other outcomes, such as trust and health care use.”
The editors’ interpretation of this article said, “False-positive mammograms may have persistent small effects on some women’s psychological well-being and behavior.”
To me, there are no earth-shattering findings here. It is about what one would expect, and is not really different from what patients deal with whenever the doctor suspects something may be wrong, and that supposition turns out (fortunately) to be incorrect.
The next article in this series was a systematic review of the scientific literature regarding screening mammography in women 40-49 years of age.
The authors note that “breast cancer is one of the most common causes of death for women in their 40s…Of the 44,000 women who die of breast cancer each year, less than one-fifth received their diagnoses between the ages of 40 and 49 years.”
They also note that “more than 98% of women will not develop breast cancer between 40 and 50 years of age.”
If one follows that logic, we should make clear that for women age 50 and over, 90% will never develop the disease. And, by the way, the highest incidence of breast cancer occurs in women who are over 70 years of age.
There is never a “good” time to be diagnosed with breast cancer. But for women in their 40’s who are diagnosed with this disease, the potential years of life lost are substantial. That is many, many years as a mother, sister, daughter, wife and friend.
So, while the numbers may be smaller, the impact may be much greater.
The authors of this paper also noted that they were going to limit their analysis to the benefits and risks of screening mammography in women age 40-49.
Why, you might ask? Why did they deliberately choose not to examine the question of whether or not mammograms in this age group saved lives?
The answer: “Currently, 8 published metanalyses discuss the effect of mammography screening in women 40-49 years of age on breast cancer mortality rates. All but 1 demonstrate a reduction in mortality rates from screening mammography.”
The researchers go on to comment about the quality of various mammography screening trials, and as has been noted by many authors previously, there are pros and cons to many of those trials.
They do note that the reduction in death rates from a screening mammography program takes several years to occur. That means that the benefit women see in reducing deaths from breast cancer in their 50s begins with screening in their 40s.
The article states, “Although the precise contributions of screening in women 40-49 years of age and screening after a woman turned 50 years of age are difficult to determine, several analyses suggest that the most benefit is attributable to screening when women are between 40 and 49 years of age.”
The authors conclude, “In summary, the body of evidence indicates that women who undergo screening mammography between 40 and 49 years of age are less likely to die of breast cancer than women who do not undergo screening mammography, although the magnitude of the effect is smaller than that among women 50 years of age or older.”
Again, to me, no surprise here either. Their conclusion is reasonably straight forward.
The study goes on to discuss a number of other issues, including whether screening actually results in less disfiguring or less aggressive treatment, and the risks of mammography including radiation induced cancer (no conclusive evidence that mammography increases the risk of cancer), overdiagnosis of cancers that would never cause a problem for a woman during her lifetime (a known fact), false positive test results (They do occur. This study says, “Overall, these studies found that false-positive mammograms were associated with a small increase in generalized anxiety and depression during the evaluation period, which resolved quickly after the evaluation was completed.”), false reassurance that a negative mammogram might lead a woman to delay seeking medical attention for a newly found lump in the breast (in one study it did, in the other it wasn’t clear), and pain from having a mammogram (well known, and frequently discussed by women who have had a mammogram).”
The article goes on to note that women who have a family history of breast cancer are more likely to have a breast cancer detected by screening. Again, no surprise here either.
What was surprising to me was the failure to note that most cases of breast cancer—including women in their 40s—are what we call “sporadic,” that is they occur in women who do not have a strong family history of the disease.
The study also mentions the Gail model as a measure of a woman’s risk for developing breast cancer. But (and this is an important “but” as you will see later in the paper) the authors also note that this model of breast cancer risk—which is widely used by experts in this field, along with other models—is better at predicting the risk of breast cancer in a population of women as opposed to an individual woman.
Despite all of this discussion, this paper concludes that women in their 40s who undergo mammography “will increase their risks of undergoing unnecessary procedures, breast cancer-related anxiety, discomfort that the time of screening and exposure to low-dose radiation.”
But a reading of the data in the article does not necessarily support these conclusions, as you can see from the quotes provided. Perhaps the most salient comment of this paper is regarding the risk of false positive results in younger women because of their generally higher rate of dense breast tissue. Most experts agree that that is a problem.
Many of these issues are also applicable to older women as well.
But here is what I think is the most disconcerting statement in the article:
“Given this difference (in risks and benefits for screening mammograms in women between 40 and 49), a woman 40-49 years of age who had a lower-than-average risk for breast cancer and higher-than-average concerns about false-positive results might reasonably delay screening. Measuring risks and benefits accurately enough to identify these women remains a challenge.” (Emphasis mine)
I would suggest you read that last sentence very carefully. As you will see below, if you are a woman between the ages of 40 and 49, you are going to be told in these guidelines to understand the risks and benefits of mammography before you make your decision.
That, in and of itself, is not a bad thing. You should always understand the risks and benefits of any medical procedure.
The danger here is that you are being told you can “reasonably delay screening” based on a risk assessment that is, according to the article itself, imperfect. In addition, the experts don’t have the data to tell you what you should do based on that risk assessment.
In other words, right is the “right” level of risk where you should be concerned?
Now, for a discussion of the guideline itself.
The guideline makes 4 distinct recommendations:
1) In women 40-49 years of age, clinicians should periodically perform individualized assessment of risk for breast cancer to help guide decisions about screening mammography.
That would be ok, up to the point about making a recommendation for screening mammography.
Women should know their risk of breast cancer. There are preventive strategies and additional tests that might be useful if a woman is at high risk, especially if she is in her 40s. The reality is that today doctors and patients do not discuss this. Too little time, and frankly too little knowledge. This has been a known problem for many years, especially now that we have medications which can reduce the risk of breast cancer in some women at high risk.
But to determine whether or not to get a mammogram? The guideline is absolutely silent on what the authors consider an “acceptable” risk to recommend a mammogram.
Once again, they point out the limitations of the Gail model as it applies to an individual woman.
I am not aware of any study that has looked at a risk-based model to back up this recommendation. In short, there is no evidence to support the conclusion that this is an effective breast cancer screening strategy
The practical implication is that this information without evidence-based guidance leaves both you and your doctor without any basis for a recommendation whether you should have a mammogram, in accordance with these guidelines.
2) Clinicians should inform women 40 to 49 years of age about the potential benefits and harms of screening mammography.
Not a bad idea. There are benefits and harms to this procedure. There are benefits and harms associated with every screening procedure. You should know the answers.
3) For women 40-49 years of age, clinicians should base screening mammography decisions on benefits and harms of screening, as well as on a woman’s preferences and breast cancer risk profile.
Here, the authors advise women and their doctors that “women who are at substantially lower-than-average risk for breast cancer or who are concerned about potential risks of mammography may derive a less-than-average benefit from screening mammography.”
It is true that women who have lower than average risk may have a less than average benefit from mammography, if only we could accurately measure that risk. But, once again, that doesn’t mean you won’t get breast cancer since most women who develop breast cancer don’t have many risk factors.
As to the other risks of the procedure itself, there is nothing—absolutely nothing—to suggest that if you are concerned about the pain of the procedure or a false positive result that those concerns are going to reduce your risk of getting the disease.
4) We recommend further research on the net benefits and harms of breast cancer screening modalities for women 40-49 years of age.
Again, this seems somewhat common sense.
Recent reports on the use of MRI in the screening of women at high risk (and with increased breast density) address some of these concerns, and pending research on the use of ultrasound will provide additional helpful information.
Ultimately, we need to develop even better methods of diagnosing breast cancer earlier.
The authors state that they agree that the data show a reduction in mortality, but there remains a statistical possibility that the benefit may not be much in this age group.
Respectfully, I would note that others disagree.
Highly regarded organizations including the American Cancer Society, The American College of Obstetrics and Gynecology, and the United States Preventive Services Task
Force all recommend screening mammography for women ages 40-49.
The reality is that much of the research that forms the basis of these articles was performed many years ago, and no longer reflects current state of the art practice.
And, unfortunately, the research on which these guidelines are based cannot take into account the role of our current medical legal system as it influences the recommendations that doctors make about follow-up studies if they see anything suspicious on a mammogram.
And they don’t take into account the access to care issues and availability of radiologists who dedicate their practice to mammography which in turn influences the quality of the interpretation of mammograms.
The final comment by the authors of the guideline?
“Assessment of an individual woman’s risk for breast cancer is important because the balance of harms and benefits will shift to net benefit as a woman’s baseline risk for breast cancer increases, all other factors being equal. For many women, the potential reduction in risk for death due to breast cancer associated with screening mammography will outweigh other considerations.”
How your doctor is supposed to interpret that statement and make a rational recommendation to you as a woman seeking advice is, regrettably, not obvious to me.
The editorial which accompanied these articles echoes some of the same comments I made previously.
Here are some examples:
“To implement the ACP guideline in clinical practice requires negotiating major challenges. First, the guideline recommends individualized assessment of breast cancer risk, yet the science of predicting breast cancer risk is extremely inexact. Current tools to quantify risk (such as the Gail model) can distinguish among large groups of women with different levels of risk, but they do not help to distinguish an individual woman who will develop breast cancer from a woman who will not. While such factors such as family history of breast cancer, breast density, and genetic mutations may help identify women at increased risk, most women with diagnosed breast cancer have none of these risk factors.”
“Outpatient encounters are typically too brief for an adequate assessment of individual risk or for a complete discussion of the benefits and potential harms of mammography. Moreover, numerical literacy is often insufficient for meaningful communication of risk.”
Having delved into these articles and these discussions, I found myself asking what to me was the logical question:
“If experts have had these raging discussions about the potential benefits and risks of screening mammography in women between 40 and 49 years of age, what does the data show?”
After all, mammography has been around for a while. We should know whether or not it works.
Maybe that is a simplistic question, and maybe I am just not smart enough to be able to accurately interpret the numbers. But let me take a stab at answering the question.
The SEER database has information on the incidence and mortality of cancer in this country, beginning in the 1970s.
According to SEER, the mortality rates from breast cancer in women between age 0 and 49 from 1994 to 2003 declined an average of 3.4% a year (white women fared better than black women, for reasons that have been discussed many times in other blog postings).
During this same time frame, the incidence of breast cancer declined by 0.13% per year.
Obviously, deaths declined much more than new cases.
Five year survival for women ages “0” to 44 was 75% in 1975-79; it was 85.1% in 1995-2002.
Five year survival for women ages 45-54 in 1975-79 was 76.6%. In 1995-2002, it was 88.9%.
I don’t think this happened by chance, and I don’t think it happened by luck. Research has shown that mammography and improved treatments have contributed to this decline.
But my (honest) bias is that mammography has played a significant role in the decline of breast cancer deaths in this population of women at risk.
I remember the days before mammography. We diagnosed “early” breast cancer that wasn’t early. We diagnosed breast cancer that, in 1969, involved the axillary lymph nodes 40-50% of the time, and cancers diagnosed less than 2 centimeters in size were the exception and not the rule.
Mammography has changed much of that terrible scenario.
When expert organizations make pronouncements to the public, the public is looking for agreement and consistency. They are looking for clarity of recommendations. They want to know what they should do.
When we fail to deliver precise messages, we fail those who look to us for guidance.
When I reviewed these papers, as I have attempted to outline above, I tried to make some sense out of what we know, what we don’t know, what is different, and what is the basis of the difference. I tried to seek some personal level of clarity that would help me understand the reasoning behind the headlines.
I didn’t find that clarity, and I didn’t find that guidance. And, if I didn’t find it, I have difficulty believing the doctors in this country, much less the women who rely on them, will find it.
Perhaps worst of all, I don’t think the media found it. And they write the headlines that will be remembered.
We are already seeing a decline in mammography in this country. As I have noted previously, in my opinion, this may explain part of the decrease in the number of breast cancers being diagnosed in the United States. If you don’t look for it, you won’t find it.
I don’t know for sure how all of this is going to turn out, but I have written previously that there is a possibility that several years from now we will start to see the fallout from this decline in mammograms. We will start to see an increase in the number of cancers diagnosed at a more advanced stage, and possibly an increase in deaths from this disease.
One of the factors I believe may be responsible for this turn of events is the confusing messaging that has been created as a result of certain studies.
I can only hope that the events of this week don’t add to that confusion, and provide some women an excuse not to do what needs to be done when it comes to their health.