In response to my posting of February 1, in which I took the position that the obstetrics malpractice cases pursued by John Edwards when he was a plaintiff's trial attorney in North Carolina were not based on "junk science" as that term is used and understood in current parlance, Walter Olson responded on February 2.
Recall that the cases in question reportedly involved claims for damages for cerebral palsy allegedly arising from perinatal hypoxia - deprivation of oxygen to the newborn, before or during delivery. The "causation hypothesis" under discussion is that cerebral palsy is or can be caused by perinatal hypoxia, sometimes also called intrapartum asphyxia.
Olson's responsive posting cited two sources from the political rather than from the medical literature:
"A two-volume report from the Institute of Medicine, entitled Medical Professional Liability and the Delivery of Obstetrical Care, in the course of exploring its subject, built a substantial case that many obstetricians were being wrongly sued. It appeared in 1989. In his widely reviewed book Galileo's Revenge, which was and remains the leading popular work assailing "junk science", my Manhattan Institute colleague Peter Huber accords a central role (and a full chapter) to cases charging obstetricians with causing cerebral palsy."
The Institute of Medicine's report had this to say:
"Until recently, cerebral palsy was thought to be linked to abnormal parturition, difficult labor, premature birth, and hypoxia or asphyxia of the infant. The committee evaluated more recent data that cast serious doubt on the correlation between presumed hypoxia and later cerebral palsy."
The publication went on to cite the many studies published by Nelson and Ellenberg (the most active researchers in this area) in the 1980s and their conclusions, including the following specific items. We will correlate the statements with the reverse and deductive conclusions that are not mentioned:
"Of the cases studied, 69% [of the cases of CP] did not have even one clinical marker of asphyxia [at birth]."
(This means that 31% did. The conclusion, one with which all will agree: other things may also cause CP.)
"Of the 21% that did [note the mathematical error], 58% had an alternative explanation."
(This means that for 42% of those which did show asphyxia, it was the only factor. The presence of two or more alternative factors complicates the analysis for the 58%, which explains why OB malpractice cases are so complex.)
It then quotes directly from Nelson and Ellenberg, including the following:
"[The] results suggest a relatively small role for factors of labor and delivery in accounting for cerebral palsy."
A "relatively small" role is not the same as "no role".
Galileo's Revenge, published in 1991, was a well-written book, but it has been subjected to some criticism of its own. A review of the author's carefully-chosen language is instructive. At p. 82, Huber cites to a well-known 1986 study published in the New England Journal of Medicine for the proposition that:
"More than half [of babies later diagnosed with cerebral palsy] who display at least one sign of asphyxia at birth also exhibit some more important risk factor, like congenital malformation, low birth weight, or microcephaly."
This language, though, means that nearly half of the CP babies who showed evidence of perinatal asphyxia did not have any other associated risk factor.
The various sources cited by Huber did not conclude that hypoxia at birth does not cause cerebral palsy. What they concluded and declared was that hypoxia at birth is not the only thing that causes cerebral palsy. That is a different thing entirely.
Huber (who is an engineer, not a physician) stated the following as a conclusion, made without any citation to medical authority:
"Yes, it is possible for asphyxia to cause cerebral palsy, but only if the asphyxia is especially severe and prolonged."
For a critical evaluation of Huber's book, see "Galileo's Retort: Peter Huber's Junk Scholarship", 42 Am. Univ. Law Rev. 1637 (Summer 1993) by Kenneth J. Chesebro, and "Junk Scholarship In Search Of Junk Science" by William F. Gallagher. In the latter, Gallagher commented:
"Huber's treatment of the cerebral palsy cases would embarrass even a first year law student. He argues that obstetric malpractice accounts for a negligible number of cerebral palsy cases, and consequently most suits brought by plaintiffs that allege medical malpractice as the cause of infant cerebral palsy are frivolous, and efforts to reduce cerebral palsy by improving obstetrical care are nothing more than a cruel and expensive hoax. Huber asserts that most cerebral palsy babies are doomed long before an obstetrician comes near them. His argument relies most heavily on an article appearing in 1986 in the New England Journal of Medicine, Karin B. Nelson and Jonas H. Ellenberg, Antecedents of Cerebral Palsy: Multivariate Analysis of Risk, 315 New Eng. J. Med. 81 (1986). Huber claims that this study brings to a definitive end the century long debate over the causes of cerebral palsy.
"However, Huber fails to inform the reader that the editors of the New England Journal of Medicine, in the very issue that published the Nelson and Ellenberg article, noted significant analytical flaws in the article. In an unusual step, the editors paired the publication of the Nelson and Ellenberg study, which they obviously regarded as provocative, with their own critique of the article's underlying analytical premise. The author of the editorial, Dr. Nigel Paneth, of Columbia University College of Physicians and Surgeons, stated concern that the methods used by Nelson and Ellenberg could have obscured the asphyxia. Birth and the Origins of Cerebral Palsy, 315 New Eng. J. Med. 124, 125 (1986). Dr. Paneth criticized Nelson and Ellenberg's study for confounding prediction with cause: the fact that certain pre-birth conditions unrelated to delivery may predict cerebral palsy does not mean that physicians cannot overcome these problems through state of the art birth procedures that can ensure a healthy baby. Thus, failure to use these procedures may constitute a proximate cause of cerebral palsy.
"Moreover, Huber fails to acknowledge that one of the co-authors of the Nelson and Ellenberg study essentially recanted her conclusion two years later. See, Karen B. Nelson, What Proportion of Cerebral Palsy is Related to Birth Asphyxia? 112 J. Pediatrics 572, 573 (1988). Huber's failure to even apprise the reader of either Dr. Paneth's editorial or Nelson's disavowal of her earlier study is incomprehensible. If he did it in a Connecticut court, he would probably end up before the grievance committee. Huber can hardly claim inadvertence. He was well aware of both the editorial criticism and the Nelson recantation because he cited both sources earlier in his chapter as background on the problem of cerebral palsy."
I do not venture to say whether Huber's analysis or Gallagher's is more sound. I simply point out that there are many views clamoring to be heard.
Medical sources
Each of the sources cited by Olson has some useful information, but I prefer to refer to medical sources when discussing medical issues. Edwards practiced in North Carolina from 1977 to 1998, so the medical sources which were extant during that time period should be considered.
The following is the entry from Williams Obstetrics, 17th edition, 1985, p. 794:
"Cerebral palsy may result from preterm birth complicated by asphxia in utero [i.e., cord compression] or in the newborn period, from severe hyperbilirubinemia, from cerebral and cerebellar malformations, and from infections acquired in utero. . ."
It then goes on to note an early study (dated 1982) that raised some questions about the causation issue, but still found a causal association in a certain number of cases.
"Holm has provided a review of 142 cases of cerebral palsy. One half were the consequence of events that occurred before labor and delivery. No more than 10 percent were considered to be caused by labor and the method of delivery."
Again, a 10% correlation is not the same as no correlation.
Compare later developments in the fields of obstetrics and pediatrics. The entry from the 1997 edition of Williams is much more detailed and much more reserved on this issue, but it still supports the causation hypothesis in certain specified cases. The following is language based on a study done under the joint auspices of the American College of Obstetricians and Gynecologists and the American Academy of Pediatrics in the mid-1990s:
"In order to establish that hypoxia near delivery was severe enough to cause hypoxic ischemic encephalopathy [brain damage caused by oxygen deprivation], all of the following must be met:
[The text then goes on to list highly technical names and descriptions for what we can identify in shorthand somewhat-plain English as
1. Cord pH less than 7.0
2. Apgars at 3 or less after five minutes
3. Neurologic problems
4. Dysfunction of several organs]
In the section entitled "Brain Disorders", the author notes a number of studies that call some of the previously-held conclusions into question, but the ultimate conclusion repeats the factors listed in the ACOG-AAP consensus criteria stated above.
The conclusion is that, in those cases in which the listed factors are present, the medical profession today accepts a causal relationship between intrapartum asphyxia and neonatal brain damage that can later manifest as cerebral palsy. A graphic figure taken from a 1985 study shows the following:
- For babies with mild to moderate brain damage at birth, 12% had fetal hypoxia as a risk factor.
- For those with severe brain damage at birth, 22% were associated with fetal hypoxia.
I will repeat my statement. The state of medical thought on the causation issue has seen a lot of development over the last 20-30 years. I believe that the position in favor of a causal connection was not by any means "junk science" during the 1980s and 1990s, and even today there is still widespread controversy on the topic. A conclusion that birth defects are caused by maternal use of Bendectin (as an example) can accurately be called "junk science", because (1) there is no reliable medical evidence demonstrating the connection and (2) the medical profession as a whole does not accept the connection as having been scientifically established. By contrast, the belief that there are causal connections between perinatal asphyxia and cerebral palsy has been accepted by the medical profession in the past and continues to be accepted today, though in a more narrow range of cases.
This issue is enormously complicated from both a medical and legal perspective. It is one which is also enormously important, given the amount of money which passes from the medical profession to professional liability claimants in these cases. The causation issue does not lend itself well to the simple statements and overarching accusations which tend to be used in political commentary.
4:02:31 PM 
|
