![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
walkitoutSubtitled: Breast Cancer and American Society
ETA: I failed to include a summary of my opinion of the book as a whole. If you have any interest in medical decision making and/or breast cancer, or just have a thing for detailed social history, this is a fantastic book. It would be a bit of a slog for someone looking for a little light reading, so while everyone would probably benefit from reading it, that's asking a little bit much.
I bought this about a year ago, and read a little of the beginning, then put it aside. It was tough reading a description of a woman suffering from breast cancer in the early 19th century. It's tough reading a description of a woman dying after years of breast cancer in any century.
When I was digging around in my blog attempting to figure out when I put the bike lights on, I stumbled across a reference to this. I thought, hey, I'll read that, and it'll make up for the nasty, unpleasant effects of reading Conner's book. More effectively than the enjoyable, but frothy, Kris Longknife novel (note to self: should probably review that, too).
Aronowitz compares the details of the life experience of Susan Emlen, Mary Cope, Rachel Carson (yup, the one you've heard of) and the treatments they chose for breast cancer. I liked a lot of what Aronowitz did with these stories: he used the women's voices whenever possible, and contemporary voices (from letters and so forth) of family and friends as a next choice. He also spent some time describing Halsted (innovator of the "radical mastectomy") and his contemporaries.
Aronowitz' thesis is an admirable one: there's a whole lot more continuity than change or progress. Breast cancer as experienced by these women (invasive, progressive and ultimately fatal) hasn't really changed. Today, we call a lot of things breast cancer that we're really good at treating, and women are really good at surviving (localized, slow to grow, nonrecurring and ultimately, something else kills you first). Aronowitz does a nice job of summarizing how and why the definition changed over time. He also does a really great job of describing the never-ending debate (which fundamentally has _not_ changed) about whether it is really possible to do anything meaningful about the old kind of cancer.
I have some complaints, which I'll mention first, then I'm going to do something I don't often do, which is supply quotes and basically say, yeah! What he said! The complaints are simple and probably predictable.
(1) While he mentioned delayed childrearing and reduced parity, he didn't mention breastfeeding in any detail. Breast cancer (including of the particularly awful kind) increased in about the same time frame that breastfeeding practices changed dramatically. I understand Not Wanting to Go There, but I wish he had anyway.
(2) While he slings it pretty good at asking women and their doctors to make impossible decisions about ill-understood interventions whose effect is only visible across populations, and draws the obvious parallels (drug treatments for hyperlipidemia, HRT, etc.), there's no indication he's read McKinley's excellent 1981 article in Milbank Quarterly, "From "promising report" to "standard procedure", which seems highly relevant to _me_. You can find a scanned copy here:
http://www.econ.canterbury.ac.nz/personal_pages/john_fountain/Teaching/HealthEcon/reading/mackinlay from promising report to.pdf
Or you can get a legit copy at JSTOR if you have access.
Calling McKinley a bit of an idealist (he really asks public policy to be rational and scientific, and for us to get something measurably effective from public funded activities) is an understatement, so his policy suggestions are kinda pie in the sky. But the article is readable, timeless and provides a great framework for making sense of medical trends, which is otherwise nearly impossible. That framework might have brought added clarity to history Aronowitz ably tells.
(3) Foucault. And not the pendulum guy, either. But really, asking for that is asking for way, way, way too much. You start thinking about Foucault as an MD and you won't be practicing medicine for much longer.
p 170 ""A prominent surgeon says that when a patient comes to him and tells him that she is worried to death about a lump which appeared in her breast a short time ago and is causing severe pain, he is fairly sure she does not have cancer," a medical journalist in 1945 [!!!!] reported. "But when a patient comes in apologizing for taking up his time and saying that she is coming only because her daughter or some friend insists on her doing so...he immediately gets another grey hair, for he is fairly sure the patient has cancer.""
The "Do Not Delay" campaign had had _that_ strong an impact, that early.
p 238
"Malcolm Pike, a professor of Community Medicine and Pediatrics at the University of Southern California, wrote the chief of NCI's Cancer Biology and Diagnosis division in 1974 that "giving a woman under the age of 50 a mammogram on a routine basis is close to unethical.""
!!!
If you've been thinking at any point in your life that the debate about breast cancer has changed or gone in any particular direction, well, maybe not entirely correct.
Those are interesting quotes that made this reader sit right up and pay attention. Aronowitz rallies a lot of evidence in service of his overall thesis, but these are pithy and punchy and unforgettable.
This longer segment, however, I found painful to read on many levels:
p 272-3
"To take one example at the blurred boundary of disease and risk, one of my patients had a very small invasive breast cancer and had to decide whether to add adjuvant chemotherapy to her existing regiment of surgery, radiation, and Tamoxifen. Her oncologist had suggested this course, but she was torn...This skepticism arose partly from the hedged way her oncologist framed his advice...She had also read conflicting advice on web sites and lay-oriented magazines and books. [New para here.] I referred her to a breast cancer expert who was active in clinical trials. He told my patient that based on stage and grade and receptor status of her particular tumor and previous treatment that her risk of a breast cancer recurrence in the next 10 years was 12 percent. If she added adjunctive [no, I really typed it right; different adjective] chemotherapy, her risk would be reduced by one tenth -- that is 10.8 percent. But such adjunctive chemotherapy carried its own risks. In addition to side effects such as fatigue and nausea, my patient also faced very low odds, less than 1 percent, of serious complications such as heart failure, a second cancer, and major infection. Based on this information, my patient agreed to undergo adjunctive chemotherapy."
R. and I are not doctors. We are people who use a lot of numbers for a lot of purposes. And when we looked at those numbers, neither of us could imagine any scenario in which a person previously on the fence would be convinced by the numbers to go with the therapy. Strictly speaking, we think that numbers like that should make it bordering on criminal to even _offer_ the therapy: it isn't worth it. I don't know what the evidence is in support of those numbers, but if I know anything about medicine, the quality of the evidence is nothing like what you would want if you're hearing a number with three significant digits.
Aronowitz goes on to describe his feelings about the numbers, and what the clinician and the client did with them. What he has to say makes, surprisingly, even less sense than what the clinician and client did.
"The expert's numbers proved very helpful to my patient." ??? "They were certainly more useful, in the sense of allowing her to measure risks and benefits of alternative decisions, than the more subjective language of "low" risks and "small" benefits"." !!!!! "On the other hand, these numbers potentially presented a falsely solid picture of what was known and not known, since they were extrapolated from a small number of studies, which, while well designed, were not without problems and were not designed to answer this particular question for women just like my patient." He also notes the error bars issue.
I don't even know what to do with this, other than to say that if _this_ is the quality of medical decision making, we maybe need to work a lot harder on this than I realized. Back to chemistry people. Start with "significant digits". That's simple, and will get rid of that 10.8% thing. I'd like to know if those "well designed" studies were RCTs or something else. And then I'd like to have a little discussion about human psychology, and how we perceive words like low/small vs. numbers like 12 and 10.8. The meat of the debate would involve whether what we are measuring ("risk" of "recurrence") is measurable, and whether the measurement we chose measures what we think we are measuring. The debate will end when we all agree that what we are measuring is not measurable and even if it were measurable, the way we are measuring it wouldn't work anyway. Wrap it all up with a side order of, hey, just because it leads to a definite decision does _not_ mean it was helpful.
I'm so glad he included this awful, painful story, particularly after describing the whole Crile/Carson thing. Knowing this is the kind of crap that goes on does not make you immune to it, if you are in the context in which it all goes down.
ETA: I failed to include a summary of my opinion of the book as a whole. If you have any interest in medical decision making and/or breast cancer, or just have a thing for detailed social history, this is a fantastic book. It would be a bit of a slog for someone looking for a little light reading, so while everyone would probably benefit from reading it, that's asking a little bit much.
I bought this about a year ago, and read a little of the beginning, then put it aside. It was tough reading a description of a woman suffering from breast cancer in the early 19th century. It's tough reading a description of a woman dying after years of breast cancer in any century.
When I was digging around in my blog attempting to figure out when I put the bike lights on, I stumbled across a reference to this. I thought, hey, I'll read that, and it'll make up for the nasty, unpleasant effects of reading Conner's book. More effectively than the enjoyable, but frothy, Kris Longknife novel (note to self: should probably review that, too).
Aronowitz compares the details of the life experience of Susan Emlen, Mary Cope, Rachel Carson (yup, the one you've heard of) and the treatments they chose for breast cancer. I liked a lot of what Aronowitz did with these stories: he used the women's voices whenever possible, and contemporary voices (from letters and so forth) of family and friends as a next choice. He also spent some time describing Halsted (innovator of the "radical mastectomy") and his contemporaries.
Aronowitz' thesis is an admirable one: there's a whole lot more continuity than change or progress. Breast cancer as experienced by these women (invasive, progressive and ultimately fatal) hasn't really changed. Today, we call a lot of things breast cancer that we're really good at treating, and women are really good at surviving (localized, slow to grow, nonrecurring and ultimately, something else kills you first). Aronowitz does a nice job of summarizing how and why the definition changed over time. He also does a really great job of describing the never-ending debate (which fundamentally has _not_ changed) about whether it is really possible to do anything meaningful about the old kind of cancer.
I have some complaints, which I'll mention first, then I'm going to do something I don't often do, which is supply quotes and basically say, yeah! What he said! The complaints are simple and probably predictable.
(1) While he mentioned delayed childrearing and reduced parity, he didn't mention breastfeeding in any detail. Breast cancer (including of the particularly awful kind) increased in about the same time frame that breastfeeding practices changed dramatically. I understand Not Wanting to Go There, but I wish he had anyway.
(2) While he slings it pretty good at asking women and their doctors to make impossible decisions about ill-understood interventions whose effect is only visible across populations, and draws the obvious parallels (drug treatments for hyperlipidemia, HRT, etc.), there's no indication he's read McKinley's excellent 1981 article in Milbank Quarterly, "From "promising report" to "standard procedure", which seems highly relevant to _me_. You can find a scanned copy here:
http://www.econ.canterbury.ac.nz/personal_pages/john_fountain/Teaching/HealthEcon/reading/mackinlay from promising report to.pdf
Or you can get a legit copy at JSTOR if you have access.
Calling McKinley a bit of an idealist (he really asks public policy to be rational and scientific, and for us to get something measurably effective from public funded activities) is an understatement, so his policy suggestions are kinda pie in the sky. But the article is readable, timeless and provides a great framework for making sense of medical trends, which is otherwise nearly impossible. That framework might have brought added clarity to history Aronowitz ably tells.
(3) Foucault. And not the pendulum guy, either. But really, asking for that is asking for way, way, way too much. You start thinking about Foucault as an MD and you won't be practicing medicine for much longer.
p 170 ""A prominent surgeon says that when a patient comes to him and tells him that she is worried to death about a lump which appeared in her breast a short time ago and is causing severe pain, he is fairly sure she does not have cancer," a medical journalist in 1945 [!!!!] reported. "But when a patient comes in apologizing for taking up his time and saying that she is coming only because her daughter or some friend insists on her doing so...he immediately gets another grey hair, for he is fairly sure the patient has cancer.""
The "Do Not Delay" campaign had had _that_ strong an impact, that early.
p 238
"Malcolm Pike, a professor of Community Medicine and Pediatrics at the University of Southern California, wrote the chief of NCI's Cancer Biology and Diagnosis division in 1974 that "giving a woman under the age of 50 a mammogram on a routine basis is close to unethical.""
!!!
If you've been thinking at any point in your life that the debate about breast cancer has changed or gone in any particular direction, well, maybe not entirely correct.
Those are interesting quotes that made this reader sit right up and pay attention. Aronowitz rallies a lot of evidence in service of his overall thesis, but these are pithy and punchy and unforgettable.
This longer segment, however, I found painful to read on many levels:
p 272-3
"To take one example at the blurred boundary of disease and risk, one of my patients had a very small invasive breast cancer and had to decide whether to add adjuvant chemotherapy to her existing regiment of surgery, radiation, and Tamoxifen. Her oncologist had suggested this course, but she was torn...This skepticism arose partly from the hedged way her oncologist framed his advice...She had also read conflicting advice on web sites and lay-oriented magazines and books. [New para here.] I referred her to a breast cancer expert who was active in clinical trials. He told my patient that based on stage and grade and receptor status of her particular tumor and previous treatment that her risk of a breast cancer recurrence in the next 10 years was 12 percent. If she added adjunctive [no, I really typed it right; different adjective] chemotherapy, her risk would be reduced by one tenth -- that is 10.8 percent. But such adjunctive chemotherapy carried its own risks. In addition to side effects such as fatigue and nausea, my patient also faced very low odds, less than 1 percent, of serious complications such as heart failure, a second cancer, and major infection. Based on this information, my patient agreed to undergo adjunctive chemotherapy."
R. and I are not doctors. We are people who use a lot of numbers for a lot of purposes. And when we looked at those numbers, neither of us could imagine any scenario in which a person previously on the fence would be convinced by the numbers to go with the therapy. Strictly speaking, we think that numbers like that should make it bordering on criminal to even _offer_ the therapy: it isn't worth it. I don't know what the evidence is in support of those numbers, but if I know anything about medicine, the quality of the evidence is nothing like what you would want if you're hearing a number with three significant digits.
Aronowitz goes on to describe his feelings about the numbers, and what the clinician and the client did with them. What he has to say makes, surprisingly, even less sense than what the clinician and client did.
"The expert's numbers proved very helpful to my patient." ??? "They were certainly more useful, in the sense of allowing her to measure risks and benefits of alternative decisions, than the more subjective language of "low" risks and "small" benefits"." !!!!! "On the other hand, these numbers potentially presented a falsely solid picture of what was known and not known, since they were extrapolated from a small number of studies, which, while well designed, were not without problems and were not designed to answer this particular question for women just like my patient." He also notes the error bars issue.
I don't even know what to do with this, other than to say that if _this_ is the quality of medical decision making, we maybe need to work a lot harder on this than I realized. Back to chemistry people. Start with "significant digits". That's simple, and will get rid of that 10.8% thing. I'd like to know if those "well designed" studies were RCTs or something else. And then I'd like to have a little discussion about human psychology, and how we perceive words like low/small vs. numbers like 12 and 10.8. The meat of the debate would involve whether what we are measuring ("risk" of "recurrence") is measurable, and whether the measurement we chose measures what we think we are measuring. The debate will end when we all agree that what we are measuring is not measurable and even if it were measurable, the way we are measuring it wouldn't work anyway. Wrap it all up with a side order of, hey, just because it leads to a definite decision does _not_ mean it was helpful.
I'm so glad he included this awful, painful story, particularly after describing the whole Crile/Carson thing. Knowing this is the kind of crap that goes on does not make you immune to it, if you are in the context in which it all goes down.