Lean Left

« Previous | Next »

There’s A Doctor in Chicago; I Hope She’ll Be All Right
Posted by KTK

There has been a lot of discussion of the affidavit of Dr. William P. Cheshire, brought in by Florida Governor Jeb Bush at the last minute to argue that Terri Schiavo was never really in a “persistent vegetative state” but was merely in a “minimally conscious state.” Why this matters was never discussed; Cheshire merely argues that it is true and then jumps to the conclusion that that means her feeding tube cannot be removed. Commentators have been suspicious of the fact that he has almost no experience with PVS, and is a member of an evangelical Christian bioethics institute in Illinois (ironically, the one Lean Left’s own evangelical lost boy, Joe Carter, works for). I have read Dr. Cheshire’s affidavit in light of the facts and history of the Schiavo case, and of the moral principles usually taken as governing in cases of that kind. I am convinced that Dr. Cheshire has raised issues that are not absurd on their face, and also that his findings, even if true, do not offer any reason to change the outcome of the Schiavo case. Dr. Cheshire’s affidavit ranges from parts that are intriguing and provocative to parts that are simply asinine. It betrays a considerable degree of intrusion of Cheshire’s own precommitments, including in its diagnostic discussion and especially toward the end where he oversteps even his own overblown “findings” to come to conclusions that are not supported by his prior, and highly suspect, claims. In the end, I think he has offered little to affect the outcome of this case, and much to call his own intentions, and even his competence as a physician and an ethicist, into question. It is germane to ask how this person became involved in such a case, and why someone of his proclivities was selected to consult.

Dr. Cheshire’s Mandate and Focus

Regarding his affidavit, he raises some significant and relevant concerns, but also introduces much that is irrelevant, and displays in some places extremely shoddy or simply distorted thinking.

He points out that there is a relatively new diagnostic category of “minimally conscious state” which was not known at the time of Schiavo’s diagnosis in “persistent vegetative state”. (Contrary to Joe Carter’s suggestion, “MCS” is not a subset of “PVS”, and “persistent vegetative state” is not distinguished from “permanent vegetative state”. The latter was a term used by the court’s appointed consulting neurologist during the evidentiary hearing, to emphasize that Terri Schiavo had no positive prognosis, but is otherwise identical to “persistent vegetative state.”) His implicit point - that Schiavo could be in MCS and not in PVS, is interesting. But he then makes a logical leap by discussing in detail an fMRI study (essentially a study of blood flow in the brain) of two patients in MCS who showed greater-than-expected neurological activity, from which he concludes that “there is much we do not know” about all patients with severe brain damage. First, this conclusion does not follow from the evidence. Second, the question whether Schiavo is in PVS or MCS is not dispositive of this case.

That last point is vital (no pun). Cheshire’s entire clinical discussion in his affidavit is focused on raising doubts as to whether Terri Schiavo is in PVS, or might be in MCS. There is no doubt she is severely brain damaged, and Cheshire, unlike the buffoons who testified for the Schindlers in Judge Greer’s hearing in November, 2002, does not hold out any hope that her condition can be cured or improved. The entire medically-focused substance of his affidavit is addressed to the question whether her PVS diagnosis might be mistaken, and whether she is capable of feeling pain.

These are important questions, but they are not the central questions in this case. The governing legal and moral principle is that of patient autonomy – the right of any person to choose what treatment they want under any circumstances, and to refuse any treatments they do not want, no matter the circumstances. The Supreme Court has ruled that all patients may decline any treatment for any reason – it does not matter whether the patient is “terminally ill,” or whether they are wholly, partially, or not at all conscious. No one may force any treatment – of any kind – on any patient against their will. (And arguing that food and water are “not treatments” or “not life-support,” as so many in this case have, so many times, is completely irrelevant – you may not force food on a person against their will any more than you may force insulin, Laetrile, or Viagra on a person against their will.) The Court has also made it perfectly clear that patients may indicate their wishes in advance, and those wishes must still be honored after the patient loses consciousness. The Court has given states the authority to prescribe the standards of evidence required to determine a patient’s will if the patient is unconscious; thus, all states recognize a “living will,” but some require that it can only be filled out under specified conditions, that it remains valid for longer or shorter periods of time, and so forth. Almost all states also recognize oral statements, in the absence of a living will, as evidence of a patient’s wishes if those statements can be determined in a way that convinces a court of their validity and relevance. Florida is one state that explicitly allows evidence of oral statements to determine a patient’s wishes (and, interestingly, when conservatives in the Florida legislature tried this week to invalidate oral evidence, they lost the vote – the legislature as a whole continues to believe that oral evidence of patients’ wishes is valid information – which is undoubtedly a good thing in a state with many elderly residents who lack written living wills). Note that patients do not have to be in PVS or any other specific medical condition in order to refuse food, water, or any other treatment. You can refuse to eat even if you’re perfectly fine, and nobody can force you to do so; you can also refuse if you’re disabled but mentally competent, and you can specify in advance that you are not to be fed if you suffer some degree of mental disability, which does not necessarily have to be as serious as PVS or worse.

The legal issue in the Schiavo case has always been what Terri Schiavo wants, and the court’s rulings on withdrawal of her feeding tube have always been reflective of Terri Schiavo’s wishes, not her husband’s and not based on some sort of evaluation of whether her “life is worth living” or not. The Court has heard evidence from many people and has concluded that Terri’s wishes were known convincingly and that they were to the effect that she would not want to live in the condition she currently is in. (There have been complaints that this evidence is not convincing. Rather than go through that again, I will just note that Terri’s wishes were testified to by Michael Schiavo, his sister, and her husband; they claim that Terri Schiavo made repeated remarks to the effect that she would prefer that her caregivers “let her go” and that she wanted “no tubes”, during discussions following the death of a person who had experienced long-term supportive care in in response to a TV show about a person on life support. Michael’s sister testified that she had a close friendship with Terri and had had such discussions with her a dozen or more times over many years. Terri’s parents testified that she had once made a remark to the effect that she thought Nancy Cruzan’s feeding tube should not have been discontinued [the judge at first discounted this due to a mistake over the date of the remark, then later weighed it in the balance against the other evidence]. The judge ruled that, overall, there was clear and convincing evidence that Terri Schiavo had held firm and unambiguous wishes not to be treated under the circumstances in which she has been living. Appeals of that judgment were upheld, and further appeals to the state Supreme Court were denied.) The court’s various orders to terminate treatment have always been predicated on this wish of Terri Schiavo’s, not on the personal request of Michael Schiavo or on his, the court’s, or any other person’s evaluation of her “quality of life” or belief that her “life is not worth living.” The only person who, legally and morally, can make that determination is Terri Schiavo, and it is her previously-expressed determination in that respect that the court has acted on.

So what is the role of Cheshire’s evaluation in the first place? His affidavit states that he was called in “to provide an independent and objective medical review of allegations of possible abuse, neglect, or exploitation of Ms. Theresa Marie Schiavo” (namely, that she was being denied treatment she could have benefited from because of a belief that her condition was more severe than it really is). He went on to explain that “As my charge is to investigate the possibility of abuse or neglect, it matters whether Terri would be able to recognize and feel the consequences of abuse or neglect. Some actions might even be unintentionally neglectful if performed by persons unaware of Terri’s level of awareness.” (Note: this last remark seems to presume that Terri Schiavo does have a level of awareness, but I presume he meant it hypothetically.) This assessment is reasonable: it certainly would be abusive to treat Terri Schiavo as unaware and unresponsive if she is not truly in PVS or a worse state - but it is not germane to the question whether her feeding tube can or should be withdrawn. The questions being addressed by Cheshire are not unimportant, but they are not determinative. (To put that another way: if everything he later says is true and Terri Schiavo is really in MCS, not PVS, her prior expressions of will still seem clearly to indicate she would not want to be maintained in that state. At most, this would call for providing anaesthesia while terminating her care, but has no bearing on whether that care should be discontinued.)

So, the question of Terri’s Schiavo’s exact level of mental functioning is relevant but it does not bear on the major practical question at the heart of this issue. There is no question that, whatever that level is, (a) it is not going to improve, and (b) it is not high enough to allow Terri Schiavo to communicate her “current wishes” [it seems clear she has none], meaning we are still bound by her prior wishes – to terminate treatment – as controlling. This limits the practical effect of Cheshire’s opinion, whatever it may be and even if we believe it.

The Evidence

Now, as to that opinion: Cheshire adduces seven points that he feels “cast a reasonable doubt on the prior diagnosis of PVS.” These range from vague to clearly wishful thinking to simply bizarre.

1. Her behavior is frequently context-specific. By this he means that the otherwise indecipherable movements and noises she makes sometimes seem to occur in conjunction with events around her: “her facial expression brightens and she smiles in response to the voice of familiar persons” and “Her agitation subsides and her facial demeanor softens when quiet music is played. When jublant piano music is played, her frace brightens . . . .”.

The trial judge had already addressed that specific claim by noting that the piano music played by Dr. Hammesfahr (the bogus “Nobel Prize nominee” who “couldn’t find” the unpublished records of the unproven treatment he claimed could restore Terri Schiavo’s speech) was much louder than any other noise in the room and he believed it was simply eliciting startle reflexes.

As to whether she recognizes faces or changes affect in response to circumstances, that will be dealt with below. For now, let’s get some of Cheshire’s other “findings” on the board:

2. Although she does not seem to track or follow visual objects consistently or for long periods of time, she does fixate her gaze on colorful objects or human faces for some 15 seconds at a time and occasionally follows with her eyes at least briefly as these objects move from side to side. When I first walked into her room, she immediately tuned [sid] her head toward me and looked directly at my face. There was a look of curiosity or expectation in her expression, and she maintained eye contact for about half a minute. Later, when she again looked at me, she brought her lips together as if to pronounce the letter “O,” and although for a moment it appeared that she might be making an intentional effort to speak, her face then fell blank, and no words came out.

3. Although I did not hear Terri utter distinct words, she demonstrates emotional expressivity by her use fo single syllable vocalizations such as “ah,” making cooing sounds, or by expressing gutteral sounds of annoyance or moaning appropriate to the context of the situation. [He then quotes an affidavit of a speech pathologist claiming that some medical records – which cannot be located – state that she once said “stop” while a medical procedure was being performed. He offers no other evidence.]

4. Although Terri has not consistently followed commands, there appear to be some notable exceptions. In the taped examination by Dr. Hammesfahr from 2002, when asked to close her eyes she began to blink rpidly. Although it was unclear whether she aqueezed her grip when asked, she did appear to raise her right leg four times in succession each time she was asked to do so. Rehabilitiation notes from 1991 indicated that she tracked [objects with her eyes] inconsistently, and although did not develop a yes/no communication system, did follow some commands inconsistently and demonstrated good eye contact to family members.

5. There is a remarkable moment in the videotape of the September 3, 2002 examination by Dr. Hannesfahr that seemed to go unnoticed at the time. At 2:44 p.m., Dr. Hammesfahr had just turned Terri onto her right side to examine her back with a painful sharp stimulus (a sharp piece of wood), to which Terri had responded with signs of discomfort. Well after he ceased applying the stimulus and had returned Terri to a comfortable position, he says to her parents, “So we are going to have to roll her over….” Immediately Terri cries. She vocalizes a crying sound, “Ugh, ha, ha, ha,” presses her eyebrows together, and sadly grimaces. It is important to note that, at that moment, no on is touching Terri or causing actual pain. Rather, she appears to comprehend the meaning of Dr. Hammesfahr’s comment and signals her anticipation of pain. This response suggests some degree of language processing and interpretation at the level of the cerebral cortex. It also suggests that she may be aware of pain beyond what could be explained by simple reflex withdrawal.

Problems with the Evidence

These various “observations” all have a common theme: they rely upon subjective interpretations of the movements and sounds Terri Schiavo makes during the day. They are certainly very provocative – given the descriptions here, Dr. Cheshire’s interpretations of these incidents do not seem obviously impossible, and if those interpretations are correct they would suggest that Schiavo has more conscious awareness of her situation that is believed, or at least is capable of feeling pain. That would be a very significant observation in respect of how her care should be (should have been) managed, and how the process of withdrawing feeding should be managed. It does not, it must be emphasized again, tell against that decision in its own right (and at this point, just days from her expected demise, it hardly matters – even if these observations were all true, there is nothing that now needs to be done about them). However, it is important to note just what these observations add up to: a limited series of occasional incidents in which Terri Schiavo’s eye movements or vocalizations seemed to correlate in time with events taking place around her. Some of these same observations were already explicitly discussed at the original trial, when almost-identical claims were made about Schiavo’s level of awareness. Note, too, that Cheshire’s observations are based in part on evidence and tapes that were viewed in that trial, and that he conducted no direct physical examination of Terri Schiavo at all. Here is what Judge Greer noted about the original evidence:

At first blush, the video of Terry [sic] Schiavo appearing to smile and look lovingly at her mother seemed to represent cognition. This was also true for how she followed the Mickey Mouse balloon held by her father. The court has carefully viewed the videotapes as requested by counsel and does find that these actions were neither consistent nor reproducible. For instance, Terry [sic] Schiavo appeared to have the same look on her face when Dr. Cranford rubbed her neck. Dr. Greer testified she had a smile during his (non-videoed) examination. Also, Mr. Schlinder [sic] tried several more times to have her eyes follow the Mickey Mouse balloon but without success. Also, she clearly does not consistently respond to her mother. The court finds that based on the credible evidence, cognitive function would manifest itself in a constant response to stimuli.

Dr. Hammesfahr testified that he felt that he was able to get Terry Schiavo to reproduce repeatedly to his commands. However, by the court’s count, he gave 105 commands to Terry Schiavo and, at his direction, Mrs. Schindler gave an additional 6 commands. Again, by the court’s count, he asked her 61 questions and Mrs. Schindler, at his direction, asked her an additional 11 questions. The court saw few actions that could be considered responsive to either those commands or those questions. The videographer focused on her hands when Dr. Hammesfahr was asking her to squeeze. While Dr. Hammesfahr testified that she squeezed his finger on command, the video would not appear to support that and his reaction on the video likewise would not appear to support that testimony. . . .

Another issue involved the piano music played via cassette tape in her room during Dr. Hammesfahr examination. Dr. Maxfield testified she related to it and “tried to sing”. However, this music was played markedly louder than any other music or voice commands of the doctors. It was probably louder than the handclasp or dropped objects that always seemed to produce a startle reflex. Dr. Greer testified that the length of the reflex depends on the direction of the stimulation.

So, the original testimony based on this evidence was flawed because it depended on carefully selected incidents from among (what Cheshier himself describes as) “many hours of videotapes”. The few incidents Hammesfahr cited of “purposeful” movement were a ludicrously unrepresentative sample of almost 200 express trial stimuli as well as other incidents in Schiavo’s room. It is apparent that Cheshire has employed the same procedure. He has based his opinion in part on the same videotapes Hammesfahr submitted as evidence, as well as case notes and his own visits with her; from this he notes:

• two instances in which she turned toward him or “looked” at him, and one in which she pursed her lips while looking at him without speaking
• repeated instances of incoherent noises which he claims “demonstrate emotional expressivity”
• one third-hand report, for which the documentation is somehow unavailable, of one instance in which she used one coherent word at least 12 years ago if not more
• one instance of blinking that was not what Schiavo had actually been asked to do but was kind of similar to it
• one instance in which she did not respond as commanded
• one or a few instances of leg-raising which he says “appear” to be on command
• “inconsistent” eye-tracking, “inconsistent” responsiveness to command, and “good” eye contact with family
• one vocalization of “Ug, ha, ha, ha” given before a stimulus was applied, which he interprets as anticipatory.

This is the cumulative result of hours of direct observation, 4 ½ hours at least of videotape of other doctors’ examinations, and a voluminous case record spanning 15 years.

Some things should be noted at once: Terri Schiavo, like many patients in PVS, is certainly capable of movement and noise-making. In PVS, these are the result of brainstem reflexes that cause random muscle activity (including in the larynx); PVS patients also often exhibit “affective release” – meaning reflexive actions such as laughing or smiling that appear to convey emotion but are non-conscious. Dr. Cheshire’s affidavit includes a detailed statement acknowledging and describing these facts. Terri Schiavo clearly exhibits a great deal of muscular activity – even the selectively-edited videotapes that her parents have released show her almost constantly in motion, with continuous eye movements spanning almost her whole visual field. She is also capable of responding to stimulus in a reflexive way: she can withdraw from painful stimuli, turn toward pleasant stimuli, and move or turn in response to sound – all through brainstem reflexes that involve no conscious awareness. Thus, one would expect that at any given time, Terri Schiavo would at least momentarily turn her eyes toward any particular person or object in her room, and that her eyes would randomly follow a moving object in front of her face at least occasionally, just by chance. And this in fact is what is seen, even on the edited videotape her parents use as evidence of her “consciousness,” during most of which she is not looking at anything at all. (Judge Greer’s quantitative observation makes clear what a tiny percentage of the time her movements actually coincide with any event around her.) One would also expect that she could reflexively respond to a person entering her room, or to loud music. In short, one would expect that exactly the types of incidents and “responses” that Cheshire claims to observe in Terri Schiavo would occur as the result of random chance even if she were completely unconscious and unresponsive.

Dr. Cheshire’s Interpretive Malady

How to we interpret Cheshire’s observations, then? Most obviously, the almost pathetic paucity of incidents suggests that the events he reports are, in fact, the random occurrences one would otherwise expect. Further, it is strking that in the direct, personal observations he reports many cases are qualified as “appears to” or “seems to”. Not only does he not know that these incidents were really conscious behavior, but he is not sure the incidents themselves even occurred. As for the rest, some of it is hearsay, and much of it is single incidents picked out of 15 years of daily recordkeeping, or reported as “inconsistent.” There is, in short, no reliable information at all that Terri Schiavo behaves in any way other than would be expected by from the random, uncoordinated movements and sounds of a completely unconscious person in PVS. There is nothing more than an extremely thin record of isolated incidents which Dr. Cheshire himself describes as being in many cases ambiguous. It is possible that a conscious person could make the kinds of movements and sounds that she makes, but the fact that she makes them is not proof that she is conscious.

What proof, then, does Cheshire offer that these incidents are evidence of consciousness? His own subjective reactions to them. This affidavit is bursting with subjective and emotional language descrbing his own feelings or his inferences about what Terri Schiavo might be feeling, from which he concludes that she must be having those feelings: a look of curiosity or expectation in her expression, . . . for a moment it appeared that she might be making an intentional effort to speak . . . she demonstrates emotional expressivity . . sounds of annoyance or moaning . . . appears to comprehend the meaning . . . signals her anticipation. Note that none of these “observations” of feelings or intentions is possible without knowing the mental state of the patient, which Cheshire acknowledges is impossible, and further that to conclude that she is having these feelings assumes the conclusion Cheshire was hired to make (unless, of course, he was hired because he had already made that assumption – something we’ll get to later). This language is logically absurd in the context of an attempt to determine whether or not the patient was experiencing the emotions described by that language! When he claims that:

She appears to comprehend the meaning of Dr. Hammesfahr’s comment and signals her anticipation of pain. This response suggests some degree of language processing and interpretation at the level of the cerebral cortex. It also suggests that she may be aware of pain beyond what could be explained by simple reflex withdrawal.

[emphasis added]

he is assuming that the “anticipatory” response really is a conscious action – if it were not, it could not suggest anything of the kind. This is a particularly remarkable claim, because he is here imputing language recognition,, conscious thought formation and will, imagination of the future, and conscious motor activity to someone whom he has immediately previously described as so far debilitated that she only “appears” to move her leg and cannot respond to commands sufficiently to indicate yes/no responses to questions.

Again and again in this case, doctors with the most superficial knowledge of the patient make grandiose diagnostic claims on the basis of single incidents that could easily have been random, and which are otherwise identical to other incidents occuring regularly on a daily basis. It is not impossible that that one, single incident – which no other doctor present or reviewing the tapes thought was significant – was in fact a conscious action taken in anticipation of the future, but there seems little reason to imagine so on the basis of one event of such dubious character. I notice as well that Dr. Cheshire here accepts incidents that occur before a stimulus as conscious, as well as incidents that only “appear” to be correlated to stimulus, as well as incidents that follow stimuli - all of which are nearly unique examples of their kind, taken from a massive patient record - while dismissing out of hand incidents in which a stimulus is not followed by a response. Given this interpretation of “evidence,” it is hard to imagine what would not count in favor of his conclusion, or what level of evidence would be considered too low to be definitive, or what could possibly count against it.

At this point it is fair to acknowledge the one reasonable point Cheshire makes. In his sixth piece of evidence, he notes that conscious responses to pain are incompatible with a diagnosis of PVS. He cites two pieces of evidence that Terri Schiavo may feel pain: she appears agitated during menstruation and this agitation seems to be relieved with ibuprofen, and “some of the records document moaning, crying, and other painful [sic] behavior in the setting of urinary tract infections.” It’s not at all clear how to interpret this: it is well documented that she makes moaning sounds and uncoordinated muscular movements frequently, so the fact that she does so while she is ill or menstruating is not surprising. A subjective impression that she does so to a greater degree at those times, and especially that that behavior is reduced by analgesics, is suggestive, but no more so than that. Dr. Cheshire claims these incidents “have been context-specific, sustained, and, in the taped example I cited [the “anticipation” action in example 5], in response to a spoken sentence, all suggest the possibility that she may be at some level consciously aware of pain.” In this, I think, he goes too far. His “anticipation” example (he gets a “two-fer” on this one, having used it in his previous argument as well) seems to me to be very likely a hasty conclusion on his part; the other examples may be “context-relevant” in the sense that she makes sounds indicative of pain when experiencing possibly painful conditions, but they are obviously not “context-specific”, since she makes the same sounds at other times as well. So it is not clear that she really is having reactions to painful situations, or especially that she is feeling them consciously, but there is suggestive evidence to that effect. This bears directly on the issue of whether she is in PVS or MCS – which, again, is not the same as the question whether her treatment should be discontinued – and some clarification is warranted.

Ah, Sweet Mystery of Life

Dr. Cheshire’s proclivity for subjective and emotional interpretations of data only gets worse as his affidavit continues, though. In his last clinical “observation,” he says:

To enter the room of Terri Schiavo is nothing like entering the room of a patient who is comatose or brain-dead or in some neurological sense no longer there. Although Terri did not demonstrate during our 90 minute visit compelling evidence of verbalization, conscious awareness, or volitional behavior, yet the visitor has the distinct sense of the presence of a living human being who seems at some level to be aware of some things around her.

As I looked at Terri, and she gazed directly back at me, I asked myself whether, if I were her attending physician, I could in good conscience withdraw her feeding and hydration. No, I could not. I could not withdraw life support if I were asked. I could not withhold life-sustaining nutrition and hydration from this beautiful lady whose face brightens in the presence of others.

This is unsettling for several reasons. Not only is it creepy (“this beautiful lady whose face brightens . . .”? – I submit that this is grossly unprofessional language at the very least – he sounds like Strom Thurmond in the last vestiges of his own mental deterioration), but it is breathtakingly illogical in ways Cheshire seems completely oblivious to.

He says explicitly “Terri did not demonstrate . . . compelling evidence of verbalization, conscious awareness, or volitional behavior, then in the same sentence concludes that she “seems at some level to be aware” (obviously, on the basis of evidence that is not “compelling”)! This is what Majikthise has described so aptly as “diagnostic séance.” It’s shocking in someone in his position. He does not have compelling evidence for a conclusion he then asserts is true! More than that, he offers his own “sense” of a “presence” as overriding his own self-described lack of compelling evidence! This is simple incompetence. Arguably, it rises to the level of malpractice. (Point V of the “AMA Principles of Medical Ethics” holds that physicians must “study, apply, and advance scientific knowledge.” To my mind, this is incompatible with Dr. Cheshire’s behavior in this case.) A doctor who acknowledges a complete lack of confirming clinical data but the asserts his subjective belief – contrary to the clinically-established diagnoses of better-experienced professionals - as grounds for patient-care decisions has simply abandoned any standard of professional behavior.

He goes further to say: “I asked myself whether, if I were her attending physician, I could in good conscience withdraw her feeding and hydration. No, I could not. I could not withdraw life support if I were asked.” It’s not clear what he means by this. If he means that he would not do so if he were asked by the patient herself, or by means of her pre-expressed wishes, then he is standing directly against the principle of patient autonomy in withdrawal of care – a principle upheld by the AMA, the Supreme Court, and virtually the entire medical and medical-ethics community with the exception of the far-right absolutists who refuse to acknowledge patient rights in that respect. To my mind, again, that’s grossly immoral; at the least it’s evidence of his unwillingness to grant Terri Schiavo her right to have her wishes acted on. (At the beginning of the affidavit, he states that he does believe withdrawal of care is appropriate in some cases of PVS – but he does not mention patient autonomy. It is never quite clear exactly what he means, but his language seems to imply that it is a physician prerogative to decide whether to terminate treatment – a position that the entire patient autonomy movement has been fighting for more than 20 years.) Possibly, by saying he “could not withdraw life support if I were asked” he merely meant that he did not think it was the best course in this case. But that’s not his decision to make. The whole point to this case is that it is Terri Schiavo’s wishes that are in dispute. What Dr. Cheshire thinks of her decision has nothing to do with anything. If her wishes are accurately known – as the court has held they are – then they are to be followed. So Dr. Cheshire’s panting remarks about “this beautiful lady” and his intentions toward her are either a statement of his opposition to seeing her wishes carried out because he disapproves of them – which would be grossly immoral – or a pointless and unnecessary expression of his personal preferences in a case in which his treatment preferences are irrelevant.

Bioethics, Patient Autonomy, and Physician Hegemony

But these remarks betray an even more significant problem. At this point in his affidavit, Dr. Cheshire has clearly mistaken, or abandoned, the role he was hired to fill. He himself had previously pointed out that his job was to determine whether Terri Schiavo’s PVS diagnosis was accurate, for the purpose of determining whether she was being “abused” by being subjected to pain her caregivers were not aware of. None of that has anything to do with whether her treatment should be discontinued. He was not asked to offer an opinion on that, and his opinion on that is logically irrelevant to answering the question whether she can feel pain. Yet he has taken it upon himself to inject himself into the question of what treatment decision should be made for her, on the basis of his role in assessing her pain-awareness. This is not his proper role, and it is an invasion of a moral decision which does not belong to him.

Shortly therafter, he says “This situation differs fundamentally from end-of-life scenarios where it is appropriate to withdraw life-sustaining medical interventions that no longer benefit or are burdensome to patients in the terminal stages of an illness. Terri’s feeding tube is not a burden to her. . . [and] Terri cannot be considered medically terminal.” Here he demonstrates conclusively that he is either immovably hostile to patient autonomy, or simply grossly confused about decision-making practices for unconscious patients (a fundamental concept for a bioethicist, not to mention someone who claims to be an expert and a professor in that field).

Long-standing, long-accepted practice for unconscious patients centers on a multi-level path for decision-making that seeks to incorporate the patient’s own values as much as possible. First, and preferably, where the patient’s values and preferences are known, through a living will or other evidence, those preferences are to be followed. This is near-universal consensus among ethicists and in keeping with Supreme Court rulings. Acting on a patient’s expressed instructions is called acting on an “advance directive” – where written advance directives take the form of “living wills” or “patient-care proxy designations,” and where oral statements can sometimes substitute for written directives. Next, when the patient’s explicit preferences are not known, but the patient’s general values and personality are known by someone who is close enough to the patient to “speak for them,” that person acts as “guardian” and gives advice to caregivers in the way they understand the patient themselves would have wanted it. This is known as “substituted judgment” – another person acts as the voice of the patient using their knowledge and judgment as to what the patient would have wanted. (The guardian can be appointed by a proxy desgnation, in which case the patient usually makes certain specific requests and the guardian acts by substituted judgment for non-specified issues.) Finally, if there is no information at all what the patient would have wanted, then it is necessary to make a decision what would be best for the patient using the mythical “average reasonable person” standard to guess what such a person would want in the situation at hand. This standard holds up the “best interests” of the patient as the overriding constraint on treatment. In this three-level scheme, in other words, the patient’s own wishes are governing to the degree to which they are known, either through explicit advance directives or indirectly through substituted judgement; decisionmaking defaults to the “best interests” standard only when the paritcular patient at hand has not expressed any known wishes. The best-interests standard is usually seen as rather conservative; patients can request whatever they want in their own names, but we tend to be more cautious without direct evidence of patient wishes. (It may seem odd that we only act “in the best interests of the patient” as a last resort, but it makes sense: we assume that what patients have thought about and expressed in advance is in their own best interests, and that intimate proxy decisionmakers who know the patient well will have their best interests at heart – so a third-party judgment of the patient’s best interests really is the least-desirable option.) Furthermore, it is long-settled that patients may refuse treatments whether or not they are “terminally ill,” or have a life-threatening condition. Patients have autonomy over all their medical care, as long as they make their wishes clearly known.

How does this bear on Cheshire and Schiavo? The Schiavo case is clearly one of decisionmaking by reference to the patient’s expressed wishes. The court has always, without exception, made the decisions that it believed had been proven to be what Terri Schiavo had clearly said she herself wanted. This has never been a case of substituted judgment: Michael Schiavo has been acting legally only as the agent of Terri Schiavo’s expressed wishes, not his own preferences and not even his inferred judgment of what she “would have” wanted, but his direct knowledge of what she did want. (You can argue with that determination all you want, but the court reviewed it and it was repeatedly upheld. Legally, this is a case of decisionmaking under advance [oral] directive.) But Cheshire’s language, in his unsolicited opinion as to the proper decision in this case, is expressed entirely in the language of terminal illness and “best interests” decisionmaking. He seems to be under the simultaneous impressions, first, that treatment can only be withdrawn if she is “terminal,” and second, that the question whether to withdraw her treatment hinges on whether those treatments are “burdensome” - in his judgment. He appears to be completely ignorant of decades of work, and the longstanding ethical and legal consensus, regarding the appropriate locus of authority in medical decisionmaking, or at the very least to be so hostile to patient autonomy that he dismisses it as a relevant consideration. Either one is shocking for someone in his position. Aside from questions that have been rasied about his religious commitments or prejudices, the fact that anyone calling himself a “bioethicist” could enter this case and, completely unasked, blurt out the claim that care cannot be terminated based only on whether the patient has a “terminal”diagnosis and her care seems “burdensome” to him suggests gross incompetence in that field. Nobody with knowledge of these cases would use that language, especially since the entire fight for the last 7 years has been over the question what Terri herself has said, and given as well that the patient’s husband is acting as guardian while her family waits in the wings. Those two facts scream “advance instructions” and “substituted judgment” – two forms of decisionmaking that preclude third-party evaluations of “best interests” and make Cheshire’s maunderings about “I could not withhold life-sustaining nutrition and hydration from this beautiful lady” not merely weird and confused but idiotically off-base in this context. He does not appear to understand the facts of this case or the way these decisions are handled. His unwelcome and unsolicited opinion about the resolution of the case is not just beyond the bounds of his assignment, but is incompetent.

He compounds this error as he goes on. He declares that Terri Schiavo “has demonstrated behaviors that are context-specifi, sustained, and indicative of cerebral cortical processing that . . . would not be expected in a persistent vegetative state. Based on this evidence, I believe that, within a reasonable degree of medical certainty, there is a greater likelihood that Terri is in a minimally conscious state than a persistent vegetative state.” The conclusion about MCS is what he was hired to determine, and, as discussed above, if this conclusion is reliable it suggests something important about how Schiavo’s case should have been handled - but not anything about whether she has the right to decline treatment. Yet notice how he reaches even this conclusion: on the basis of firm declarative statements about observed behaviors – behaviors that he himself had previous described with such terms as “seems to,” “appears to,” “on some occasions,” and “inconsistent.” His confidence in his evidence has grown dramatically in just 1 ½ pages.

He ties it up with another unsolicited opinion that goes beyond his mandate and betrays gross ignorance of end-of-life decisionmaking: “If Terri is sufficiently aware of her surroundings that she can feel pleasure and suffer, if she is capable of understanding to some degree how she is being treated, then in my judgment it would be wrong to bring about her death by withdrawing food and water.” Once again – and this cannot be repeated too much – that “judgment” is not his to make, and it does not follow logically from the observation that she is feeling pain. Dr. Cheshire continually conflates the clinical status of consciousness of pain with the moral judgment whether to continue treatment; he continually assumes, explicilty and implicitly, that the fact that the patient is (in even minimal degree) aware or conscious means that treatment cannot be discontinued; and he continually assumes that the decision whether to do so belongs to him or some other third-party evaluator of the patient’s “best interests.” In all this he demonstrates apparent ignorance of the history of this case, the accepted practices for dealing with these cases, and the relevant moral principles that commonly apply to them.

Who Are You Who Are So Wise in the Ways of Science?

Given all this, it is worth asking how such a boob got on this case in the first place – and especially so given that, though he is a certified neurologist, he has little clinical experience with PVS. Is Dr. Cheshire a raving Christian winger brought in to torpedo this case under cover of medical authority? I don’t know. It’s interesting to note that his own clinical facility has issued an all-but-explicit repudiation of his actions:

Dr. William Cheshire was asked by the Florida Department of Children and Families to share his expertise related to potentially vulnerable adults. As part of this commitment, he was asked to render his professional opinion for the state of Florida in the case of Terri Schiavo. He observed the patient at her bedside and conducted a review of her medical history but did not conduct an examination.

Mayo Clinic recognizes that the standard of care for the evaluation of a comatose patient includes a detailed review of the patient’s history and previous evaluations as well as the performance of a comprehensive neurological examination. In some instances, electrophysiological and imaging studies may be used to establish a diagnosis.

The views expressed by Dr. Cheshire in the case of Terri Schiavo do not represent the opinion of Mayo Clinic or its Departments of Neurology. The Mayo Clinic Departments of Neurology do not have opinions regarding the diagnosis of Terri Schiavo because they have not performed an evaluation as described above.

It hardly stretches the imagination to consider that an aggressively right-wing evangelical Christian governor did not just stumble upon a member of a pro-life, evangelical Christian bioethics center by accident. If Jeb Bush brought in a hack for the explicit purpose of promoting an especially restrictive and reactionary view of patient autonomy and arguing against termination of treatment in this case, it would hardly be the worst thing that has happened in the case, or the worst thing Jeb Bush has ever done. And Dr. Cheshire was brought in to argue against termination – Bush had openly, vocally, sought every possible avenue to prevent Terri Schiavo’s wishes being carried out in this case, and Dr. Cheshire was hired specifically for that purpose. His conclusion was certainly known in advance (it was known to Jeb Bush, at least – though Cheshire claims it wasn’t known to Cheshire himself!).

Cheshire is certainly inclined to go very far on very thin evidence; to make unambiguous statements on the basis of what he himself holds to be ambiguous evidence, to accept individual isolated incidents from within a huge 15-year case record as probative of mental states that can only be inferred even on good evidence, to uncritically accept undocumented and third-hand evidence, and to see intentionality in isolated and unrepeated movements or sounds that can easily be explained as coming from the same random sources as the overwhelming majority of the patient’s other behaviors. He has also exceeded his authority to inject himself into the moral decisionmaking aspects of the case – authority that properly belongs to the patient or the patient’s surrogate, and an issue he was not asked to comment on. He has demonstrated either concentrated hostility toward or a shocking ignorance of the basic principle of patient autonomy and its application in end-of-life decisionmaking. And he has done all this through a shoddy and logically untenable train of reasoning that by itself would invalidate most of his conclusions even if the evidence supported them and he had analyzed it competently.

I think some suspicion of his motives and his actions is warranted. More importantly, I don’t think he or his conclusions deserve much regard in respect of this case.

March 25th, 2005 General, Politics, Legal Issues, Church & State, Religion, Culture, Science, Health, Privacy | 109 comments