Macro vs. Micro

There are few decisions, few things in life, which are “every single time, 100% guaranteed” right courses of action. Almost every decision has risks and benefits, upsides and downsides. That’s just the way things are.

It’s a no-brainer that seat-belts save lives, right? But they also may take lives in certain instances. A few years ago on Oprah, there was a family whose young daughter was decapitated by a seat belt when the vehicle crashed. My Grandma probably survived a wreck which totaled her car because she wasn’t wearing a lap belt. She ended up with a broken pelvis, but because she only had a shoulder belt on (one of those automatic seat belts that required manually latching the lap belt, which she did not do), when she pulled out into oncoming traffic, and the car smashed into her, she had enough leeway in the shoulder-belt that the force of the wreck pushed her out of her seat and towards the other side of the car instead of crushing her in the driver’s seat. But my father was undoubtedly killed in a car wreck because he did not wear his seat belt. On balance, seat belts save many more lives than they could possibly be blamed for taking.

Often, a certain decision will be the right or best course for the majority of people, although a minority might either be hurt or harmed in some way. The recent change in recommendation for mammograms is a perfect example of this. The new recommendation is that women younger than 50 not have mammograms at all, and that women 50-74 have mammograms only every other year. ACOG disagrees, finding fault with the science or method of study on which the new recommendation was founded, reiterating their support for annual mammograms for women aged 50-74, and annual or biennial mammograms for women in their 40s. [But we’ll say that ACOG is wrong, and the USPSTF is right, for the sake of argument.] The reasoning behind the new recommendations is that a lot of women are being exposed to a lot of radiation by way of mammograms; some cancers are not detected by these mammograms anyway; some cancers that are detected are slow-growing or are actually mild or not life- or health-threatening, but there is the urge to treat even these cancers aggressively, which may cause a lot of pain and suffering; many “false positives” (or other cancer “scares”) along with much unfounded fear is being endured by healthy women; and many healthy women are undergoing painful procedures such as biopsies (or even mastectomies) because of the test results which may be wrong or inconclusive.

The first news I had of the change in recommendation was from a Christian news source that was decrying the changes, pointing to the fact that there would be some younger women who would not have these breast X-rays done, who would actually have breast cancer, and would die since they did not get early detection nor early treatment. This news organization treated the recommendations in a negative light, taking the tone that women would die from following the new as opposed to the old recommendations, and that this was not right or good, because life is precious and valuable, so every effort should be made to conserve life. Macro vs. micro — the big picture vs. the small picture. Yes, it’s possible, perhaps even probable, that a few women in their 40s might die from breast cancer that might have been detected and treated earlier, were they to have regular mammograms. But there are women who come down with breast cancer in their 30s — why not screen all 30-year-olds as well? Is it not possible for women even in their 20s to get breast cancer — why not screen them as well? Are their lives not worth saving? Is not their life as valuable?

But at what cost? How many women might be exposed to enough radiation from mammograms as to give them fatal cancer? How many women are going to be given life-altering mastectomies (particularly unnecessarily)? At what point do the above negatives and risks outweigh the positives?

A similar argument was made some time ago on some birth blog, in which the blogger pointed out that if such-and-such recommendations were carried out, something like 38,000 women would have to have a C-section in order to save the life of one baby under X circumstances or with X disease. Naturally, a woman whose baby had that condition responded (not very happy), because she took the statement to mean that the blogger was saying that her child wasn’t worth the sacrifice. Obviously, that’s not what anybody was saying. That statement was not intended to be personal — a “micro” look, but rather a “macro” look at the broad situation… but it was personal to that particular woman, because she was affected personally by it.That’s the “little” picture.

The big picture is looking at groups of people who will be affected by a certain policy or procedure. Mammographies for all women which will greatly increase milder negative outcomes (the pain of the procedure, the costs involved, the false positives, etc.) and potentially slightly increase the major negative outcomes (increased risk of cancer, etc.) vs. mammographies for fewer women, knowing that a tiny percentage of women will be severely adversely affected. As soon as the recommendations came out, women who had breast cancer at a young age came out of the woodwork to tell their stories of early detection and treatment which saved their lives. These stories are obviously compelling. But equally compelling would be stories of women who endured the hell of a false diagnosis, going through (perhaps) unnecessary mastectomies and chemotherapy with months of sickness and recovery time, and believing they were going to die, or worse, actually dying from a complication of an unnecessary surgery, or dying from an infection due to being in a weakened state from chemotherapy.

The point of these types of discussions is not to say that one person’s life or health is more valuable than another’s; but just to try to figure out what is best for the population as a whole. In the above example of “38,000 C-sections to save the life of one baby” — there is a cost involved in 38,000 C-sections — certainly a monetary cost, but also a cost in the lives and health of the women undergoing C-sections, as any major surgery has potentially fatal complications both in the short-term and the long-term; plus long-term health implications for the mother and potentially fatal complications for future babies. Recently on some blog, I read about a woman who died several years following a C-section, as a result of the C-section, due to some sort of bowel problem. Adhesions are a common problem after any surgery including C-sections, and they may cause bowel and other problems (including pain, sometimes severe); and C-sections certainly increase the risk of future complications to the mother and any future pregnancy she has, with the risk increasing exponentially with every additional C-section. I remember reading some time ago (though perhaps not in a medical journal or study) that the typical rate of maternal mortality associated with a vaginal birth is somewhere around 2/100,000, while the mortality rate associated with C-section is either 5 or 6 times that. Assuming that these facts are correct, the MMR for C-sections is 10-12/100,000, or about 1/10,000. So, if 38,000 C-sections are necessary to save the life of one baby (in that example), it may be that 3-4 women would die. Plus, 38,000 women would be put at risk of future complications that may threaten or take their life, or they may have a problem in a future pregnancy that could take the life of that baby, or themselves. For instance, the risk of placental problems like accreta or increta or placenta previa (and possibly placental abruption, but I can’t remember off the top of my head) increases with a single C-section and increases exponentially with every additional C-section. With placenta accreta or increta, the placenta grows into or through the wall of the uterus, often necessitating a hysterectomy; and all of these placental problems increase the risk of maternal blood loss, which can be problematic and even potentially life-threatening if steps are not quickly taken. Rare, perhaps, but possible. While the woman in the above case could say that it would certainly be worth 38,000 C-sections to save the life of her baby… is that what is really best for the population as a whole? We save the life of one baby, but take the lives of 3 women within 6 weeks of the birth; plus potentially more, years down the road, either as a direct result of the surgery like the woman with bowel problems, or indirectly like a future pregnancy problem due to a uterine scar; plus rob many other women of their possibility of future childbearing (increased risk of hysterectomy); plus the increased risk of miscarriage and fetal death due to a scarred uterus. And there is evidence to suggest that babies born by C-section unnecessarily (which would be the case in 37,999 C-sections, in this example), are more likely to have problems such as more NICU admissions, longer NICU stays, increased risk of childhood asthma, etc. Some people point the finger at C-sections for a whole boat-load of problems including things like an increased risk of learning disabilities, autism, etc., but I don’t know if that is necessarily accurate. What if 1% of babies born by unnecesarean have asthma that they wouldn’t have had, had they been born vaginally; and of that number, 1% will have such severe asthma that they will likely die from an asthma attack (or some sort of lung infection, or something) before they reach the age of 5 or 10? That’s a small number, but it’s still 3-4 children, which is far more than the number of babies saved (1) due to the level of unnecessary C-sections (38,000). And if the rate of these other complications (some life-threatening; others not) is also increased, then you have a higher health and potentially a higher life cost as a result of trying to save the life of that one baby.

Is that one baby not worth saving? Most certainly! Every baby is worth saving! Every life should be saved if possible. But… it’s not always possible, and sometimes there are unintended consequences. I’m not arguing that some lives are worth more than others; or that some people should be allowed to die so that others can live. But I’m pointing out the necessary problems that occur when you look at only one side of the equation and not the other. If there were no “human costs” involved in mammography (increased radiation, false negatives, unnecessary painful biopsies, etc.), then I would not necessarily have a problem were there to be a recommendation to screen every woman from the age of 16 or 18 or whatever. It certainly wouldn’t be “cost effective” — there would undoubtedly be millions or billions of dollars spent on mammograms to find 1 woman in her 20s and 3 women in their 30s and 100 women in their 40s who had cancer and save them. If there were no human costs involved in a C-section, it would likewise be not “cost effective” but still worth doing if it would save lives. I’m frugal, but you can’t put a cost on human life (although “bean counters” make a pretty good effort at it!). However, it is possible, as I pointed out in my above example, for there to be more lives lost by a “conservative approach” and an attempt to save every life, than there will be lives gained. This is not to diminish the value of the life lost; but to place a value on all lives.

Looking only at the “little picture,” you see that some lives will undoubtedly be lost by following one approach, and most people (including myself) draw back in horror at the thought of callously saying, “Eh… big deal — one life — it doesn’t matter.” No — that life is valuable too! But looking at the “big picture,” you see that — as things stand now — there may be a much higher cost in both human life and health, in attempting to save every life that is threatened by a certain disease or anomaly. There are still plenty of situations where there are risks and benefits both of action and non-action. We can say (sometimes) which is better in the big picture; but that does not mean that by following “the safest route for most” that everyone will be saved. Just like there are instances in which seat-belts actually take lives, instead of saving them.

In the case of a 37-year-old woman who dies of breast cancer, we can say, “If she had just had a mammogram, it might have been caught in time,” and that might be true. But for every younger woman whose life might be saved, it’s possible that two other lives would be lost and ten or a hundred or a thousand might undergo painful and unnecessary procedures — procedures which might end up harming her health and well-being for the rest of her life. And these women’s lives and health are also important.

To sum up, this is why informed consent and patient choice is so important. You don’t want a person coming back and saying, “If I had only known that that was a possibility, I would have made a different choice”; nor do you want someone coming back and saying, “I was forced to undergo a C-section, or I was not allowed to choose a C-section, and look what happened.” Although these recommendations on mammography have been made, younger women can still choose to have mammograms, if they want to, so if they want to expose themselves to the risks which the USPSTF has said are unnecessary, nobody’s stopping them. But we also need to keep trying to further refine our technology and procedures and knowledge so that we can reduce the number of unnecessary procedures while maximizing the number of lives saved. For instance, fingerprint patterns can indicate a higher risk of developing breast cancer, with no exposure to radiation. Maybe other advances will be made in the near future, that will reduce the risks, but keep the benefits.

12 Responses

  1. Beautifully put.🙂

  2. You explained this so well. Really, really, wonderful post. I can tell I’ll be referring back to it frequently.

  3. You might be interested in this post (the author is a breast surgeon, who writes on science literacy, research findings etc.): this post is about the risk of radiation from CT scans –

    “the possibility that as many as 3% of adult cancers might be due to radiation from medical imaging studies is a problem that should sober even the most gung ho advocate of using such studies, particularly considering that the risk tends to be higher in younger people….All of medicine is a balancing of risks versus benefits. One reason I was so disturbed by the proliferation of whole-body imaging studies being marketed by unscrupulous companies on a cash basis is because, in an asymptomatic patient, the risks from radiation from such studies on average probably outweigh any conceivable benefit, especially if we take the risks of false positives leading to invasive tests such as biopsies into account.
    http://scienceblogs.com/insolence/2009/12/radiation_from_ct_scans_balancing_risks.php

    He also wrote compellingly on the breast cancer screening recommendations.

  4. “I remember reading some time ago (though perhaps not in a medical journal or study) that the typical rate of maternal mortality associated with a vaginal birth is somewhere around 2/100,000, while the mortality rate associated with C-section is either 5 or 6 times that. Assuming that these facts are correct, the MMR for C-sections is 10-12/100,000, or about 1/10,000. So, if 38,000 C-sections are necessary to save the life of one baby (in that example), it may be that 3-4 women would die.”

    No, this is faulty reasoning. It is based on the assumption that women who have a C-section are exactly the same as women who have a vaginal delivery except for the surgery. However, we know that is not the case.

    Women who die in childbirth are more likely to have a C-section as part of the effort to save their lives. Women who die are more likely to be older, to have pre-existing medical problems, to have complications of pregnancy, etc. All those things are risk factors for C-section as well.

    For example, a woman who ultimately dies of eclampsia is much more likely to have had a C-section (in an attempt to save her life) than a woman who doesn’t have a C-section. That doesn’t mean that the C-section killed her; it was the eclampsia that killed her.

    Similarly, women with pre-existing heart problems who die of heart complications are more likely to have had a C-section (to spare them the cardiac stress associated with labor) than women who don’t have heart problems. That doesn’t mean that the C-section killed her; it was the underlying heart problem that killed her.

    In other words, C-sections are correlated with maternal deaths, but correlation does not equal causation. That is especially the case when the data is not corrected for confounders.

    • Women who die in childbirth are more likely to have a C-section as part of the effort to save their lives. Women who die are more likely to be older, to have pre-existing medical problems, to have complications of pregnancy, etc. All those things are risk factors for C-section as well.

      These things may be true, but there is also an increase in the risk of death due to surgery alone. I’m sure you’ve heard of this case, since it happened in your neck of the woods. Apparently, the woman bled to death internally following a C-section, and the staff missed the signs of it until it was too late.

      However, this study from the BMJ says that even excluding emergency C-sections and adjusting for confounding variables, there is still a higher risk of death for elective C-sections than for vaginal births. “Compared with vaginal deliveries, the risk was three to five times higher for maternal death, four times higher for hysterectomy, and twice as high for being admitted to intensive care and hospital stay more than seven days (table 2)” Here’s the link to the maternal mortality/morbidity chart, for both vaginal and Cesarean births.

      • “However, this study from the BMJ says that even excluding emergency C-sections and adjusting for confounding variables, there is still a higher risk of death for elective C-sections than for vaginal births.”

        It says nothing of the kind. You need to read the paper to find out that Villar compares women who had vaginal deliveries with women who had medically indicated caesarean sections, both non-emergency and emergency. The paper never investigated elective caesareans and therefore it reaches no conclusions about them.

        My comment on this point was published by the BMJ both as a rapid response and in the print edition of December 15, 2007.

        • Talking with you is oftentimes reminiscent of this hilarious Monty Python sketch, only not as funny.

          In your initial comment you blamed “confounding factors” and “complications of pregnancy” as the cause of maternal deaths during C-sections. I pointed to a study that looked at “elective cesareans” see chart, and also intrapartum cesareans.

          “We excluded emergency caesarean delivery without labour, which denoted women referred for a caesarean before onset of labour with the diagnosis of acute severe fetal distress, severe vaginal bleeding, uterine rupture, maternal death with a living fetus, eclampsia, or any other diagnosis considered by the attending staff to require emergency elective caesarean delivery.”

          Also, “For intrapartum caesarean delivery the most common indications were cephalopelvic disproportion (35%), fetal distress (26%), and previous caesarean delivery (32%).” However, in a comment some time ago in which you were decrying the current high C-section rate, you said, “The C-section rate is rapidly approaching 30%. This is an almost 50% increase since I started OB training 20 years ago. There is no medical reason for this. Babies have not increased in size by 50% and fetal distress has not increased in incidence by 50%. The increased C-section rate is not medically necessary, it is a response to the legal realities.” Ok, so many of these C-sections were “medically indicated” but not “medically necessary,” I guess.

          In another similar comment, promoting VBACs and decrying forced repeat C-sections, you said, “The end result is that many women who want a VBAC, even women who have had a successful VBAC in the past, cannot have one and are forced to have a medically unnecessary surgical procedure. These women are denied appropriate medical care because of legal concerns.” So a C-section due simply to previous C-section is “medically unnecessary”.

          I’ll grant that this study has weaknesses, in that it does not reliably distinguish between C-sections done in the middle of labor for real reasons and those done for defensive medicine reasons, and the retrospective nature of the study has limitations on it. But it does look at the mortality rate from elective non-emergency C-sections. Do you happen to have a study that says what you want it to say?

          • You wrote that this was a study about elective C-sections. It isn’t.

            • I said it excluded [pre-labor] emergency C-sections and adjusted for confounders, and gave the link to the full study for anyone to read it more in-depth (perhaps a free subscription is required to view it).

              However, I failed Dr. Amy 101, which is, when you say anything, to require you to provide proof — I think your favorite phrases include things like, “I require facts, not assertions,” and “you have no proof, only anecdotes.” What proof do you have that what you have said is true? Where is the study that shows that, “Women who die in childbirth are more likely to have a C-section as part of the effort to save their lives. Women who die are more likely to be older, to have pre-existing medical problems, to have complications of pregnancy, etc.”?

  5. I wrote a post on my blog with an attitude similar to the one you are describing from the article you read. My mom is one of those few people who was saved by early detection. She was in her fifties, and at her yearly mammogram they found cancer. I guess there’s a chance she may have survived another year without detection, but the doctor said it was fast-growing and it had not been visible the year before, so I’m glad she wasn’t in the practice of only getting one every other year.

    I had a mammogram at 19 because of a lump my gynecologist found at what I think was my first visit ever. The mammo doesn’t work well on dense breast tissue, and they knew it wouldn’t do much good on someone my age, but they said they had to do it to show the insurance company an ultrasound was necessary. They did the mostly useless, rather scary mammo and then did the ultrasound and got the info they needed. My question is why don’t we just do breast ultrasounds on the younger women if all we’re really trying to avoid is excess radiation? I also don’t like that women are being discouraged from performing breast self-exams, because it seems to send the message that women needn’t bother trusting themselves or knowing their own bodies.

    I had a homebirth and we rarely ever go to the doctor. I hate unnecessary medical procedures, vaccinations, tests, etc. I am concerned about the harm caused by too much radiation exposure, and I’m not sure when I’ll end up getting another mammogram because I’ll probably be either pregnant or breastfeeding for a long, long time (I like to nurse for at least two years and I nursed halfway through my last pregnancy). But my MOM. I’m glad she had that mammo. My main concern is that this new recommendation will translate into insurance companies refusing coverage. I think if a woman says she needs a test because her gut tells her she does, she should get it.

    • I think if a woman says she needs a test because her gut tells her she does, she should get it.
      I agree — there is something about intuition I find very hard to denigrate.

      My main concern is that this new recommendation will translate into insurance companies refusing coverage.
      That’s possible, but so far insurance companies haven’t started refusing coverage of other elective procedures, like elective inductions and C-sections.

      I’ve heard about breast ultrasounds — my mom has had one or two, rather than mammograms (concerns about the radiation, as well as the squishing of the breast being damaging); I think if I needed breast imaging done, I’d go that route — not sure why they’re not more widely used.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: