Alice Stewart and Sudden Infant Death Syndrome
—Alice Stewart, quoted in The Woman Who Knew Too Much by Gayle Greene
—Robert X. Cringely, Finding Meaning in a Lost Life, April 2002
Alice Stewart was a doctor and a scientist who was one of the first practioners of a new post-war discipline, epidemiology. For what became the Oxford Study of Childhood Cancer, she drew up questionnaires, recruited volunteer interviewers, and assembled by hand enormous amounts of data about thousands of children. Her methods were unorthodox. She started the questions from conception, not birth. She tracked forward, from a healthy population, not backwards, from the sick. She did what none of the lofty, eminent men of her day had thought to do: “I asked the mums.”
She asked the mums. She gathered data, a beginner’s mind, and one brilliant statistician. With little funding, and sometimes active hostility in place of support, she and George Kneale coaxed stories from her facts and figures. Fifty years later, that data is still talking.
The most famous of her many discoveries was the link between x-rays and cancer, though it took decades for her findings to be accepted and for shoestores to stop x-raying for fit. Her later work on radiation and cancer was systematically squashed by the nuclear industry. Epidemiology, with its thirty-year studies, is a science for the long-lived. Alice was working right into her nineties, fighting suppression by powerful lobbies. As consolation for the Nobel Prize she may have deserved, she won the Right Livelihood Award, the “Alternative Nobel” awarded in Sweden. The British Embassy in Stockholm did not even take her out to lunch.
Her theories on leukemia epidemics were grounded in her background as a working doctor, not an academic. She realised, for example, that antibiotics had unmasked the true incidence of leukemia. It was the healthiest children who succumbed to leukemia, early doctors often noted. Leukemia dramatically compromises the immune system, so long before the disease visibly manifests, these children used to die of minor infections. When these infections were cured by antibiotics, they lived through long enough to develop cancers of the blood.
Her childhood cancer data also turned up a peculiar finding: twice as many children who died of leukemia under six months old were born between January and June than between July and December. Yet these periods are climatically the same in Britain—both half summer, half winter. The difference, she noted, was whether a month-old child was surviving into warmer or colder weather.
Alice, true to her name, looked the mirror image of this information and thought about SIDS deaths, which occur more often in winter.
Sudden Infant Death Syndrome (SIDS) kills between 6,000 and 7,000 babies a year in the United States alone, or between 2 and 3 cases per 1,000 live births. Nobody knows what SIDS is, or why it’s on the rise. What is known is that it happens more often in winter than summer and that it occurs mainly between four and six months of age.
Myeloid leukemia is more acute than lymphatic leukemia. It has a shorter latency, manifesting between one and three years of age rather than two to four, and it involves the red blood cells as well as the white. Children who are incubating myeloid leukemia are—like all pre-leukemics—more infection sensitive than normal children. But they are also born with a defect in their hemoglobin, they have something wrong with their red cells as well as their white.
“While in the womb,” Alice explains, “the fetus produces fetal hemoglobin, which is geared to receiving oxygen through the placenta; but soon after birth this is replaced by adult hemoglobin, geared to receiving oxygen through breathing. At birth you have both kinds of hemoglobin present, enabling you to breathe through both the placenta and the new apparatus of the lungs; then you gradually get rid of the fetal hemoglobin. But children who are incubating this kind of leukemia don’t make the changeover from fetal to adult hemoglobin and are left with too much fetal hemoglobin. This hemoglobin fails to take up oxygen from the lungs, so that when they go into a deep sleep, or have the first effects of respiratory infection, the oxygen level falls to a fatal level and they’re liable to go into anoxia—shortage of oxygen.
“There have been studies showing that children who die of SIDS have an exceptionally high ratio of fetal to adult hemoglobin—though this is difficult to measure after death, and it’s not something all hematologists accept.”
Alice’s theory is that SIDS children have difficulty replacing passive immunity with active, and fetal hemoglobin with adult, and the two effects combined might be sufficient to cause a sudden death. SIDS children die when they’re sleeping, and the mechanism of death seems to be respiratory obstruction—purple bruises are sometimes present, tiny bleeding points called “petechiae”, perhpas resulting from the infants attempts to take deep breaths against some obstruction in the airways. [...]
“SIDS deaths are more common in winter than summer, which is when the immune-compromised child is more likely to succumb to infection. They often occur in a family situation where an older child brings an infection home, or where everyone in the family has a cold and the child goes to bed with sniffles and doesn’t wake up. You have no defense of your own, so you meet with an infection and go out like a light. [...]
“It’s also known that SIDS children have an easy delivery with a short second stage of labor. The second stage of labor is when the baby becomes dependent on its own hemoglobin for breathing and when any defect in its system could be fatal. These babies would have to have got into the world fairly easily because if they’d had a difficult labor, they’d have died.”
Alice’s theory of SIDS has been there in the literature since 1975 but no one has picked it up. This is the more remarkable, since it could so easily be tested. “There’s a blood test done on all children shortly after birth—the same test should be used to look at fetal hemoglobin. Then when the mother gets the follow-up exam at four weeks, do a second test for proportion of fetal to adult hemoglobin—then monitor the population for all causes of death in the next eleven months.
“According to me, you’d expect children who died of SIDS to have shown a high proportion of fetal hemoglobin at one month of age. You can’t test for this after death, since the blood count can only be diagnosed by flowing blood, but you could monitor children while alive—and you could easily establish whether SIDS children have a disproportionate amount of fetal hemoglobin.
“I tried to launch a study of SIDS in America through the Childhood Cancer Research Institute, but there wasn’t enough funding, and nobody in England has shown the slightest interest. I simply can’t understand why. No one knows anything about this mysterious syndrome—they’re stuck—so why not test my theory. As long as SIDS remains a mystery, my theory is as good as any other.”
Bob Cringely’s account of his son Chase’s death from SIDS two years ago is hard to forget. As a parent he desperately wants answers:
Well, here is the theory of an extraordinary scientist who doggedly proved herself right so many times while she was alive. It seems to me, as laywoman as they come, to show all Alice Stewart’s practicality, intuition, and good science. Why is nobody testing this? For her memory, and for Chase’s, I’d like to think that someday someone will dig out those journals and try.
Reference: All Alice Stewart excerpts are taken from Chapter 15 of The Woman Who Knew Too Much: Alice Stewart and the Secrets of Radiation by Gayle Greene. (Many thanks to Alice’s granddaughter, Elly, for lending me her copy. )
See also “Our Brilliant Careers”, a 1996 documentary on Alice Stewart produced by Channel 4.
Wednesday, February 18th 2004 at 11:19 am
If Alice Stewarts theories were correct, the incidence of SIDS would not have shown such a dramatic decline in recent years in the US – http://www.sids.org/annualrates.htm.
That is probably the reason for the lack of interest!
Kevin Horgan
Crescent College 1970-76
Wednesday, February 18th 2004 at 11:48 am
Alice Stewart deserves more credit. I’m reminded of a tract called “Abuse of Women” where it is written
Of all creatures women be best,
Cuius contrarium verum est.
Wednesday, February 18th 2004 at 12:30 pm
So many questions. So SIDS rates have gone down over the last 5 years in the US. I wonder what the leukemia rates have done? Also, I wonder about the role education for new parents in preventive measures — and concerted efforts to educate have indeed been made over the last several years — has played. If you alert parents to the environmental risk factors (exposure to colds, lying face down, etc) and those are then lowered, wouldn’t SIDS rates go down even if the underlying causal problem (immune system, hemoglobin or whatever it is) still was in place? Isn’t it possible that the triggering mechanism, if you will, was simply removed? Also, I’m no scientist, but couldn’t part of her theory could be correct without the rest of it being correct? What if it’s not leukemia per se, but does involve a delayed transition from fetal to adult hemoglobin, leaving the infant vulnerable during those first six months until that transition is finally made? When you’ve got 3 out 1,000 dying every year, doesn’t it make sense to gather all the data you can?
Wednesday, February 18th 2004 at 3:26 pm
leukemia rates have been stable in recent years while SIDS incidence has significantly decreased.
gathering data is easy – asking appropriate questions, and rigorously evaluating the data to see if it is consistent, is much harder.
Children with abnormal hemoglobin are not at increased risk of infection as a direct consequence of this!
Wednesday, February 18th 2004 at 4:40 pm
Hi Kevin,
Greetings, fellow Crescent grad! Thanks for commenting. I take it you’re a scientist, so I want to take the chance to ask some lay questions to make this clearer for myself. Feel free to address or ignore depending on your time/expertise.
1. What are the theories on why SIDS is dropping? Is it behavioural? (All the sleeping position stuff?)
2. Is myeloid leukemia also dropping with more common forms?
3. Alice didn’t claim that abnormal hemoglobin leads to infections, rather that it was an indicator of underlying latent leukemia, which might/would compromise the immune system. Is that not correct?
4. As far as I understand it, as an epidemiologist rather than a hematologist, she looked for broad, statistically significant correlations rather than clinical diagnoses in individual cases. Her x-ray research, for example, showed that fetal exposure to x-rays increased the likelihood of developing cancer, but not that it inevitably did, or indeed that all cancers came from x-rays. Might not SIDS have several possible causes, of which latent myeloid leukemia could be one? And might not the behavioural changes be helping protect these children from further respiratory stresses (which, I guess, would lead to them living to develop leukemia if her theory were correct)?
5. Do you know if any research has been done on the proportion of fetal hemoglobin in the bloodstream for healthy children versus (later) SIDS victims? Has myeloid leukemia been definitively ruled out, or are you saying that it’s so far off-track that it’s not worth the research $$$?
I ask because Alice seems to have been widely respected precisely for asking good questions and rigorously looking at the data, so I’d like to understand this better!
Saturday, February 21st 2004 at 12:50 pm
It seems hard to find an objective view of Alice Stewart’s work, but this review of “Woman Who Knew Too Much” seems open-minded:
http://www.iop.org/EJ/abstract/0952-4746/20/4/703
Monday, February 23rd 2004 at 11:14 am
I think it’s fair to say that Gayle Green was in awe of Alice and that this does not make for a 100% objective view. However Alice was an awe inspiring woman and a large part of the book was based on face to face interviews with her.
It’s also fair to suggest that some of her theories do not or will not hold water in the future. Few scientists have obtained this in a lifetimes work.
However to focus on her theories about SIDS we risk dismissing her importance by missing much of her other groundbreaking work.
For those who are interested and are in the UK at the time there is a conference commemorating her life and work on the 3rd March this year.
http://medweb5.bham.ac.uk/histmed/alicestewart
I of course have a vested interest in promoting her but I’d urge anyone with an interest to read Gayle’s book and see for themselves.
Monday, July 19th 2004 at 2:32 pm
I followed up on one part of Alice Stewart’s hypothesis, namely, as posted by Dervala, “the proportion of fetal hemoglobin in the bloodstream for healthy children versus (later) SIDS victims.”
Using measurements of newborn hemoglobin concentration for roughly 3 million infants, I noted a positive relationship with risk of subsequent sudden infant death. For more information see (David B. Richardson, Steve Wing, Fred Lorey, and Irva Hertz-Picciotto. Adult Hemoglobin Levels at Birth and Risk of Sudden Infant Death Syndrome. Arch Pediatr Adolesc Med, Apr 2004; 158: 366 – 371.
I would be interested in the basis for Kevin’s assertion that people (infants or otherwise) with abnormally elevated fetal hemoglobins are not at increased risk of infection.