- First Online:
- Cite this article as:
- Moskowitz, R. Medicine Studies (2010) 2: 121. doi:10.1007/s12376-010-0043-3
- 147 Views
The physician’s task is to make sense of the patient’s story in the generally accepted terms and concepts of the scientific world. Yet the two narratives, the physician’s account and the patient’s story out of which it was made, continue to exist side by side. Although the patient’s story is contained in and explicated by the physician’s, it has not been replaced by it. Nor are the two narratives simple translations of each other. They are incommensurable: neither can be comprehended in or satisfactorily reduced to the other’s terms.
The medical interpretation of the patient’s story bears great power for healing. As the location of the malady in the social universe, a diagnosis relieves suffering in itself, as well as in the guidance it provides for therapeutic action. Like all power, it must be exercised with care.
– Kathryn Montgomery Hunter, Doctors’ Stories
Illness and Disease
to help them make sense out of what is happening to them;
to help them construct a working mythology of their illness;
to help them discover and navigate a safe passage through it; and
to help them envision a life beyond it, insofar as possible.
But how do physicians accomplish these purposes, and how faithfully do our words and actions reflect what our patients actually feel and are bothered by? Like my colleagues, I take “illness” to refer to what people actually complain of, encompassing both the particular symptoms and the general malaise that prompt them to seek (or avoid) our help. But these felt complaints can only hint at or approximate the full extent of their condition, and our endeavor to understand that totality raises a still more basic problem, namely, how we know that someone is truly sick.
This is by no means a silly or rhetorical question, for if we stipulate that being sick is merely a sensation or awareness of feeling ill in some way, then we ignore the obvious possibility that the patient, the doctor, or both could be mistaken, that we might only suspect or fear an illness that does not actually materialize, as with anxious patients who keep imagining the worst, even when no other evidence turns up. Virtually every practice attracts its share of devoted followers for whom definite confirmation of a disease long feared or suspected brings a perverse yet genuine relief. Although some might well argue that obsessiveness of such high degree is itself a kind of sickness, both physicians and patients rightly search for outwardly perceptible signs to confirm the reality of what they feel.
Yet if, on the other hand, we suppose that an adequate definition is possible with nothing but such objective evidence, we are even more likely to fall into the opposite error, by singling out an abnormal but harmless variant, say, which the patient has lived with happily for years or even decades, and might never be bothered by in the future. The obvious solution would be to postulate that illness should include both elements, i.e., both an inner awareness of feeling unwell, and some manifest interference with normal functioning that could be witnessed and attested to by other observers, such as one’s spouse, family members, friends, employers, colleagues, teachers, doctors, and the like.
But what is perhaps most striking about this formula is the extent to which modern medicine has managed to ignore it, by concealing the subjective elements of illness within the vastness of the seemingly more objective and precise term “disease” wherever possible. Reducing illness to disease enables modern physicians to sidestep the metaphysical question of whether or not a person is really ill, in favor of the more practical task of organizing the myriad forms and manifestations of disease into a taxonomy or classification scheme that assists both diagnosis and treatment. After several notable attempts in ancient times, such as Greco-Roman theories of the four humors, the tripartite Hindu or Ayurvedic system, the Five-Element disciplines of China, Japan, and Tibet, and the innumerably diverse oral traditions of tribal and folk medicine, many of which are still in use, the concept of disease in contemporary Western medicine is distinctly modern, and radically different from all of these.
tracing the signs and symptoms of living patients back to anatomic and microscopic structures first identified in the cadaver;
refining the physical examination itself, on the basis of these correspondences; and
developing new diagnostic instruments and tests that could likewise be defined in wholly objective terms, independently of the patient’s lived experience.
Two early inventions of this kind, the stethoscope and mercury sphygmomanometer, or blood pressure machine, were introduced in the former and latter halves of that remarkable period, and taught physicians how to distinguish normal and abnormal sounds in the heart and lungs, such as murmurs, arrhythmias, wheezes, râles, rhonchi, and so forth; how to measure and standardize the blood pressure; and ultimately to identify diseases, complications, and sequel both old and new, like mitral stenosis, pneumonia, bronchitis, asthma, hypotension, and high blood pressure, among many others.
Flushed with the power of such discoveries, the modernists perhaps too easily phased out the “vital force” or unitary life principle of ages past, in favor of the limited but seemingly more precise unity of separate “disease processes,” such as “cancer,” “hypertension,” and “tuberculosis,” which could also be defined solely on the basis of objective, measurable abnormalities—cancer cells, high blood pressure, and TB bacilli—and thus diagnosed and studied in the abstract, as concatenated sequences of mechanical causes and effects, quite independently of the patients who happened to exhibit them. Such pathological “entities” could then be used to group and ultimately explain the clinical signs and symptoms, insofar as the diagnostic abnormalities would naturally tend to produce them, like headache and neurological deficits in the case of brain tumors, heart disease and stroke for hypertension, or cough and hemoptysis for TB.
This elegant sleight-of-hand is still famously celebrated in the weekly or monthly spectacle of “Grand Rounds,” also known as the Clinical Pathological Conference, or CPC, wherein the medical history, physical examination, and laboratory tests of a selected patient are reviewed and discussed by the attending physician to explain how the clinical diagnosis was arrived at, and then confirmed or refuted by the pathologist, who climaxes a series of anatomic and microscopic views of the surgical or autopsy specimen with the long-awaited announcement of what the patient really had. The subtext and ulterior purpose of these ceremonies is to uphold the final authority of the disease process as the most accurate version of the patient’s underlying condition, whatever the clinical signs and symptoms, and thus in every respect superior to them in explanatory value.
The Disease Process as Explanation
Although long accustomed to the practical advantages of this viewpoint, and to rejoicing in its accomplishments, we seldom appreciate the profound revolution in human thought that was needed to bring them about. Modern surgery, for example, presupposes the exact and detailed identification of the parts of the human body, their gross and microscopic structure, and their physiological functions, a truly monumental achievement, which would not have been possible without the uninterrupted collaboration of dedicated scientists in all parts of the world, extending over many generations, and indeed continuing right up to the present, and beyond.
What we call the immediate cause of a phenomenon is nothing but the physical and material conditions in which it exists or appears. The object consists in finding the relations that connect every phenomenon with its immediate cause, in defining the conditions necessary for the appearance of the phenomenon. When the experimenter succeeds in learning the necessary causes of a phenomenon, he is in some sense its master. He can predict its course and appearance; he can promote or prevent it at will.
As a corollary to the above, neither physiologists nor physicians must imagine it their task to seek the cause of life or the essence of disease. That would be entirely wasting one’s time in pursuing a phantom. The words “life,” “death,” “health,” and “disease” have no objective reality. When a physiologist invokes the “vital force,” he likewise does not see it: he merely pronounces a word. Only the vital phenomenon exists, with its material conditions: that is the one thing he can study and know.1
Health and disease are not two essentially different modes, as the ancient practitioners believed. These are obsolete medical ideas. In reality, between these two modes there are differences only of degree: exaggeration, disproportion, and discordance of normal phenomena constitute the diseased state.2
In the spirit of William Harvey, the Seventeenth-Century physician who discovered the extent and direction of the circulation of the blood, as well as the arterial and venous compartments and the capillary network that separates them, physicians today draw on much the same reasoning to suspect acute cholecystitis, for instance, as the most likely explanation for attacks of fever and pain in the right upper quadrant of the abdomen that radiates to the right shoulder or shoulder blade, and to confirm the diagnosis by ultrasound of the gallbladder, in conjunction with other tests. The train of reasoning proceeds “backwards,” so to speak, from the signs and symptoms of the patient to the pathognomonic features of the known disease process, which lead us to the anatomical structures that would have to be affected in order to produce them, in much the same way that a plumber or auto mechanic locates the offending valve or pipe in a closed system of human manufacture.
Diagnosing acute diseases of this sort, such as heart attack, stroke, pneumonia, yellow fever, stomach ulcer, appendicitis, and so forth, regularly proves its worth not only by identifying and confirming, but also in large part explaining the illnesses that patients are actually complaining of. Perhaps because these pathological “entities” explain the “how” or pattern, as much as the “why” or precipitating cause or circumstances of illness, acute diseases tend to exhibit a consistent natural history across the whole range of various possible reactions to them, and thus yield an approximate standard or average against which variations in severity, duration, clinical course, prognosis, and outcome can all be assessed and even predicted with some accuracy. This ever-growing body of clinical experience may then be offered to the patient as a model of the healing work that remains to be done, as well as what to expect in the aftermath. Further proof of the relevance and accuracy of these diagnostic categories lies in the remarkably high degree of congruence between the clinical and the pathological data, as recapitulated in the Grand Rounds or CPC ritual just described.
More or less the same is true of subacute and intermittent conditions, such as rheumatic fever, pleurisy, nephritis, migraine, sciatica, asthma, malaria, and the like, even if chronic traces of these illnesses remain behind, since both their defining characteristics and the patient’s overriding concerns still refer back to the acute episode or flareup, in which subjective and objective elements again coincide quite closely, and thus invite deeper levels of understanding. In other words, they too may fairly be called “diseases” in the modern sense, precisely because they are also illnesses, such that the patient’s signs and symptoms can be comprehended and at least in part explained as pathological mechanisms, which have in turn been elucidated by observation and experiment.
In relatively well-defined diseases of this kind, the patient’s lived experience of illness remains distinct from the pathological knowledge that was acquired in order to explain it, and continues to exist side by side with it, such that the “disease” perspective, which applies equally to all patients with the same diagnosis, and the “illness” perspective, which seeks out those features, which are peculiar to and characteristic of each individual, tend to complement and enrich each other, to achieve the optimal “fit.” Thus, the modern physician is trained to anticipate conduction abnormalities in heart attack patients, uses routine electrocardiography to detect them as early as possible, and can often prevent fatal arrhythmias by treating them promptly at this stage, but also knows, or should know, that the fear and depression so often encountered in these same patients can easily hamper their recovery, even if the electrical picture stabilizes. In diagnosis as well as treatment, successful practitioners learn how to use both of these perspectives idiomatically and synergistically in each case.
Vague, Uncertain, Long-Term Diagnosis: the “Nocebo” Effect
Physicians have been trained to think of themselves as scientists, and to search for a biologically detectable reason for every complaint, partly in the fear that they may be sued for missing something. The explicit teaching at many medical schools is that every clinical problem is reducible to some detectable abnormality that can be captured on a screen, be given a number by a laboratory test, or otherwise quantified. Specialists see their main task as ruling out potential structural alterations in some segment of the body. As the diagnostic approaches multiply, so do the number of tests the patient undergoes. There must be a middle ground between mindless ordering of tests and foolish ignoring of disease. Does every complaint need a workup [and] a diagnosis? We need to revise our concepts of health and disease, and begin to think about restraining the percentage of the gross national product that medical care entails. We need more public discussion, and not simply external financial restraint.3
The same quest for wholly objective and impersonal diagnostic criteria also underlies the concept of “preventive medicine” as we know it today, and earns considerable prestige for the effort, even when it fails or falls short. As chronic illnesses are reframed as ongoing disease processes, it seems feasible, and even irresistibly tempting, to try to identify their microscopic, microchemical, and microbiological precursors, to devise methods for detecting them as early as possible, and to abort them before they become real illnesses by correcting the abnormalities wherever they can be found.
A familiar early example is “essential,” primary, or “idiopathic” hypertension, that is, simple high blood pressure in patients without kidney disease, pheochromocytoma, or other organ pathology to explain it. All denoting a purely statistical abnormality, these terms are applied more or less interchangeably to readings above a predetermined normal range, based on studies linking them with higher rates of stroke, heart attack, and cardiac failure years and even decades later. During my student years, pressures in the 140/90 range were considered mild or “borderline,” and earmarked for observation, while those of 150/100 or higher were said to be “moderate,” and recommended for treatment.
Hypertension is prevalent and accounts for a large proportion of cardiovascular morbidity and mortality worldwide. The level of blood pressure is related to the risk of stroke, ischemic heart disease, heart failure, and death in a continuous and consistent fashion for values as low as the optimal blood pressure of 115/75.
In 1937 the eminent cardiologist Paul Dudley White wrote that “for aught we know, hypertension [even] in advanced cases may be an important compensatory mechanism which should not be tampered with.” Our knowledge has advanced a lot since then. In 1967 treatment was shown to confer clinical benefit in patients with severe hypertension (diastolic BP 115–130). Since then clinical trials have shown benefit in treating less severe degrees of diastolic and systolic hypertension, even in older persons. An overview of such trials involving 160,000 patients indicates that any common regimen to lower pressure reduces the risk of major cardiovascular events, and larger reductions even more so. In a study of patients 80 and older, active treatment achieved a 21% lower risk of death from any cause, a 64% lower risk of heart failure, and a 30% lower risk of stroke. These results prove that it is never too late to start treatment in old people.4
What greater anxiety can a patient suffer than to be told that he is likely to develop a serious or fatal disease unless he agrees to take powerful drugs for the rest of his life, and possibly even then? What doctor is courageous or foolhardy enough not to diagnose hypertension, or not to tell his patients that they have it? What patient would be willing not to know if he had it, or be less anxious for not knowing? The diagnosis of hypertension is anxiety-provoking by its very existence, because it causes every patient to wonder if he has it, even if he never goes to the doctor. Our system rests on the validity of pathological diagnosis in providing accurate information about the objective reality of a disease (abnormal physiology, natural history, and prognosis), apart from the lived experience of the patient. The assumption is untestable within the system, and functions like a religious commandment: obey it without question, and ignore it at your peril.
Yet pathological evidence is a poor guide to whether our patients will recover, worsen, or die, and what their lives will be like, when measured against what nobody dares measure, namely, what would happen if the exam was not done and the patient sent home without treatment. When is pathological diagnosis useful? It can be life-saving in acute diseases, where it helps doctor and patient to know what they are up against and what they have to do next. It is often injurious when the patient presents with no symptoms, is told that he has a disease on the basis of some test that has no correlation with his actual experience, and must be kept on a regimen of drugs throughout life. The anxiety provoked by making these diagnoses will pressure patient and doctor to discover the disease they seek to prevent.
The placebo effect denotes the healing power of giving our patients blank pills and passing them off as real. I propose its opposite, the nocebo effect (literally, “I will harm”) to signify the damaging effect of telling a patient what he “really” has, of redefining his reality in such a way that he must forever live in its shadow, by substituting its law (i.e., ours) for his own.5
Are we entitled to expose patients who have identified risk factors but are otherwise healthy to long-term administration of drugs for prevention of a disease that most of them will never have?7
Apart from these occasional caveats, quibbles, and nitpickings, however, the preventive strategy continues to thrive as if it embodied the highest ideals of the medical enterprise, largely unchallenged by the public despite fundamental deficiencies in the concept. Apart from the risk of human error, which is significant but to some extent unavoidable, one equally important and obvious problem is the subjective bias that is inherent in the interpretation of many tests, and in some cases in the nature of the testing medium.
As an example of the second type, a retrospective study of mammograms in women previously biopsied revealed surprisingly large discrepancies in the readings of board-certified radiologists and their recommendations for treatment, discrepancies that were significant enough to cast serious doubt on the efficacy of the procedure as a tool for detecting breast cancer and its precursors.8 For 150 women diagnosed with cancer, the radiologists’ blinded recommendations for further workup varied from 74% to 96% of the total, a discrepancy of nearly one-fourth, while for 123 women with negative biopsies over a period of 3 years the variation was much larger, from 11 to 65%, depending on the radiologist.9
Fibrocystic disease is defined as a condition in which there are palpable lumps in the breast, usually associated with pain and tenderness, that fluctuate with the menstrual cycle and become progressively worse until menopause. This description applies to the majority of women of reproductive age. Most of these lumps represent a physiologic nodularity that is under hormonal control.
Physiologic nodularity is clinically important because of the difficulty of distinguishing it from a “dominant” lump, which requires aspiration and biopsy. Autopsy studies have shown that 58% of patients of all ages had histological changes associated with fibrocystic disease, and that 89% of women over 70 without breast disease also had them. Other studies have concluded that the differences between normal breasts and those with clinical features of cystic mastitis are only of degree, not of quality. Yet a diagnosis of “normal breast tissue” almost never appears on a surgical pathology report. Is it reasonable to define as a disease a process that occurs clinically in 50% and histologically in 90% of women?
The real problem is that this nebulous disease has been said to impart a twofold to fourfold increase in the risk of breast cancer. But many studies show no greater incidence of it in cancerous than in non-cancerous breasts at autopsy. In fact it was found in 58% of non-cancerous breasts at autopsy, but only 26% of breasts with cancer. What a review of previous studies does suggest is a roughly twofold increase in the risk of cancer following breast biopsy. This increase applies to all patients who have had a biopsy, whatever the reason for it, and regardless of the outcome.
Although it could be argued that surgery itself stimulates a neoplastic change, the cancer is often on the side opposite to the biopsy specimen, so any carcinogenicity must be systemic. A more obvious explanation is the change in the criteria used to make a histologic diagnosis. The decision to perform a biopsy is subjective, based on clinical judgment of the relative risk of cancer. Since having had a biopsy makes one more likely to have another, closer surveillance may simply increase detection without increasing the actual incidence of breast cancer.10
In this case, the universal dread of the disease was aggravated by both the screening program and the biopsy that it increasingly led to, which further magnified the doctors’ already considerable fears, even when the biopsy itself was negative, and was therefore likely to prejudice their subsequent readings, thus adding to the already large total of false-positive diagnoses in women who were not ill, and might never have become so if the biopsy had not been performed in the first place.
Such ambiguities are unavoidable, even with actual illness and confirmatory laboratory or X-ray findings, because no test by itself is sufficient to explain what patients feel or experience. Every practicing physician knows substantial numbers of patients with spinal films that look like Swiss cheese, indicating degenerative osteoarthritis, advanced osteoporosis, and the like, who nevertheless experience no pain, stiffness, or functional impairment to speak of, as well as another group at least as large who suffer excruciating pain and/or serious disability with no detectable pathology whatsoever, while by far the largest group, and maybe the hardest to be clear about, inhabit the vast, unexplored territory in between. Yet no matter which point along this spectrum a given patient happens to occupy at a given moment, crucial life decisions, such as whether or not to undergo surgery, chemotherapy, or still more invasive tests and punishing treatments, continue to be made largely on the basis of X-ray shadows, statistical averages, what biopsied cells look like under a microscope, and so forth, despite the inconsistent and often unreliable correlation between any test and how well or badly a particular patient actually feels or functions, either then or later.
But these already major difficulties are compounded exponentially when the patient is not ill, and the “disease” to be tested or screened for is merely a technical abnormality, a hypothetical risk of developing a serious illness perhaps years or decades in the future. Even if the test makes sense, and is performed competently, read accurately, interpreted according to the best available standards, and improbably faithful to the natural history of the disease process, it will necessarily include substantial quotas of both false-positive and false-negative results, and still leave us in the dark about what we really need to know, namely, which cases of the disease so defined will prove to be serious clinically, and which will not. Thoroughly steeped in the exalted scientific ideals of objectivity and precision that undoubtedly inspire such exercises, most physicians tend to accept and shrug off these ambiguities as unavoidable consequences of medical practice, of doing no more nor less than what we were trained to do, as indeed they are.
an incidental finding, with no symptoms whatsoever;
spontaneous remission or cure of an actual illness;
widespread regional, lymphatic, or blood-borne metastasis;
severe, life-threatening systemic disease; and
Such disparate reactions are easier to accommodate within our general conception of acute, subacute, and intermittent diseases, as we saw, because they occur now and then, as discrete episodes or flare-ups, scattered throughout the course of patients’ lives. In contrast, for a disease to be defined as “chronic” means that it has become essentially co-extensive with the “normal” life of the patient, such that it can no longer be readily distinguished from it, and therefore cannot and must not be thought of as a thing or “entity” existing on its own, with a well-defined path or natural history that is intelligible and makes useful sense apart from the patient’s own experience of it.
This conundrum is well illustrated by the debate surrounding prostate cancer, which is defined by microscopic changes that almost all adult males will eventually develop if they live long enough, and that in the vast majority of cases will produce no symptoms and no functional impairment, and has no clinical significance whatsoever. It is nevertheless a major problem for the imagination (1) because some men will become seriously ill and die with it; (2) because everyone dreads suffering and death; and (3) because nobody knows or can know who will fall ill, who will recover, who will live, and who will not.
Resolving these genuine and intractable dilemmas is the estimable mission of the prostate-specific antigen or PSA test, which measures a glycoprotein constituent of seminal fluid that is secreted by the prostate, spills over into the blood to some extent, and is used to detect prostate cancer at its earliest and presumably most curable stage. The PSA level rises substantially with age, after ejaculation, and in men with benign prostatic hypertrophy (BPH) and prostatitis, tends to fall in parallel with low testosterone levels, and unfortunately also varies significantly from day to day in normal life.11
According to the Harvard Medical School Guide to Men’s Health, blood levels of 0-4 ng. per ml. are generally read as normal, levels of 4-10 ng. as “borderline,” and those above 10 ng. as unambiguously abnormal.12 But as is widely acknowledged, these thresholds are arbitrary and inexact, with no clear or fixed boundary between levels that are clinically unimportant and those that accompany or predict serious illness, as well as including the inevitable percentage of false positives and false negatives, as above, which in this instance tend to vary inversely with each other, no matter which threshold is chosen. The current rule of thumb considers levels below 10 ng.to indicate of early disease confined to the gland, levels of 10–20 ng. to signify local spread beyond the capsule, and levels above 50 ng. as highly suggestive of metastatic cancer involving lymph nodes, bones, and other tissues.13
There is some limited consensus that the PSA provides a quick, easy, safe, and economical way to monitor the progress of the disease and the effectiveness of the treatment in men who are significantly ill, with a biopsy-confirmed diagnosis. But routine PSA screening for all men above a certain age is still hotly debated, and rightly so, since even if it were 80% effective in detecting cancer, as alleged, it would still miss 20% of the cases, as well as subjecting millions of healthy men to expensive and potentially harmful biopsies and treatment that they didn’t need, and would never have needed. While advocates of the test point to a steady reduction in the death rate from prostate cancer since 1992,14 that result cannot be attributed to PSA screening, since a comparable decline has been reported in England and Wales, where the test has not been in general use.15 Just as with hypertension and breast cancer, once the abnormality is permitted to define the ailment, the same set of facts leads to exactly the opposite conclusion, that there is no minimum threshold that guarantees freedom from disease, just as Claude Bernard foresaw so long ago.
6.6% of men with PSA’s of 0.5 ng. or lower;
10.1% of men with PSA’s of 0.6-1.0 ng.;
17.0% of men with PSA’s of 1.1-2.0 ng.;
23.9% of men with PSA’s of 2.1-3.0 ng.; and
26.9% of men with PSA’s of 3.1-4.0 ng.,16
12.5% of men with PSA’s of 0.5 ng. or less, and
25.0% of men with PSA’s of 3.1-4.0 ng.17
While the authors of the study spoke out against doing so, these data lend added support for lowering the upper limit of normal still further, to catch these large numbers of false negatives as well. Just as with mammograms, breast biopsies, “high blood pressure,” and many other tests, the net effect of preventive screening for specific diseases is often increasingly hypochondriacal versions of the same basic dilemma that we started with, always in the name of greater precision, which in this case clearly measures something relevant to prostate function, yet in the end still leaves us hanging on nothing more than a risky wager on the outcome for you or me. Once again, our genuine and inevitable fear, doubt, and uncertainty are simply papered over with statistics and a blind faith in “preventive medicine” for its own sake, and bullied by our own obsessive need and ambition to diagnose as many diseases and abnormalities as possible, as far in advance of actual illness as possible, and regardless of the significant probability that they will never materialize or become actual health problems at all.
Endless Diagnoses and the Need for Restraint
Screening for subclinical abnormalities can be perfectly legitimate and useful in some cases, because medicine involves art as well as science, and requires ad hoc decisions on behalf of real people in concrete, here-and-now situations, while the range of our sense-perception is limited and therefore always amenable to extension and improvement. As we saw, the PSA test can be helpful in monitoring the treatment of patients who are actually sick with prostate cancer. Screening tests may also be useful in retrospect, as in an update of the celebrated Framingham Study, in which routine electrocardiograms taken at six-month intervals uncovered a surprisingly high percentage of subclinical myocardial infarctions, roughly half of which were completely asymptomatic, half exhibited symptoms but remained undiagnosed, and the total of both groups, amounting to almost 25% of the total number of heart attacks in the study, were just as likely as the rest to result in heart failure, stroke, or other major complications, including death.18 Nor would I hesitate to perform targeted screening of selected individuals at their own request, or with strongly positive family histories, or on an ad hoc basis for any reason, even a hunch. Introducing subjective or personal factors, like the patient’s fear of illness, or the doctor’s gut-level intuition of unseen forces at work, lends real meaning to the procedure, by providing an unequivocal, up-or-down criterion for interpreting it.
As a general rule, then, I propose nothing more radical or outlandish than simply taking seriously what every medical student is piously taught, and what experienced clinicians observe every day, that pathological diagnosis, the identification and classification of disease, works best when it includes both subjective and objective elements side by side, as we saw, that is, a Gestalt or ensemble of symptoms complained of by a patient, together with manifest signs elicited by the physician, such that the clinical diagnosis based on the history and physical examination can then be confirmed or refuted by laboratory evidence that would be imperceptible otherwise.
When pathology is diagnosed solely on the basis of an abnormality in a laboratory test, an X-ray shadow, or a biopsy specimen, with no clinical information to complement it, narrowly technical questions as to whether the test is relevant or its interpretation accurate tend to upstage the well-nigh irresistible tendency in either case to bully both doctor and patient into ever more numerous diagnostic and treatment procedures in the future, all of which conspire to bring about the reality, the suspicion, or at least the fear, that in retrospect appears to justify the concern. Always in the name of greater precision, obsessive reliance on diagnostic testing in reality tends to propagate an unending cycle of further doubt, uncertainty, fear, and confusion in everyone concerned that multiplies exponentially and across the board both the probability and the risk of still further interventions within the system.
This preconscious and virtually automatic response also proceeds in tandem with the proliferation of medical specialties and subspecialties, each with its own full complement of diseases, some old, some new, and many overlapping with those of various colleagues, as the finite amount of physiological space is assigned and parceled out among them. Once fields are mapped out, and a limited degree of mastery attained, disproportionate attention is similarly paid to early signs and harbingers of disease, as we have seen, and our seemingly limitless fixation on normalizing them.
While screening for thyroid disease, physicians often find increased thyroid-stimulating hormone (TSH) levels in patients whose free thyroxine (T4) levels are normal. This state, termed “subclinical hypothyroidism,” is most often an early stage of hypothyroidism. Although the condition may resolve or remain unchanged, in some patients overt disease develops within a few years. The likelihood that this will happen increases with greater TSH levels and detectable antithyroid antibodies. Because patients with subclinical hypothyroidism sometimes have subtle symptoms and mild abnormalities of cardiac function and serum lipoproteins, patients with definite and persistent TSH elevation should be considered for thyroid treatment.19
the identification of a subclinical abnormality as a hypothetical precursor of overt disease in the future;
the adoption of a preventive strategy of close surveillance and aggressive treatment of the abnormality; and
the long-delayed finding, requiring the most painstaking scholarship and the courage to speak out against the tide, that the abnormality in question represents essentially a normal variant, such that the true risk factor must be a concomitant variable, either invisible or in some cases an artifact of the surveillance itself.
A Disease for Every Pill
It is no longer surprising, mysterious, or unknown to the public that the relentless pressure and at times fanatical zeal for preventive screening and the early detection of subclinical abnormalities has been assiduously cultivated by the pharmaceutical industry, has proceeded in parallel with the development of new drugs designed to correct them, and thus provides a built-in marketing strategy both before and after the fact. This was already evident in the decades-old crusade against high blood pressure, beginning with the introduction of thiazide drugs on a large scale in the 1950s. As several new generations of drugs were developed, the disease itself has been redefined, including lower and lower thresholds of abnormal diastolic pressure, as we saw. New and ever more aggressive treatment protocols, directed against even mildly elevated or borderline pressures, have required subjecting an ever increasing proportion of the entire adult population to lifetime drug maintenance, with the result, according to some studies, of somewhat lower incidence and morbidity of heart attacks and strokes, but also the exact opposite in many cases, where the drop in blood pressure proved excessive,22 and with no clear or fixed boundary between therapeutic and dangerously low levels. Once again, the felt need for more and more aggressive treatment, in this case largely generated by the drug industry itself, subliminally persuaded doctors to broaden their definition of the disease itself.
Of the leading risk factors for coronary heart disease, 25% of children have total serum cholesterol levels above 170 mg/dl. Autopsies of children dying from other causes have found that aortic fatty streaks correlate with total cholesterol and LDL levels. Children with high cholesterol are also at risk of having high levels as adults. Many children with high cholesterol do not maintain high levels as adults, and the safety and effectiveness of treating high cholesterol has not been established. We therefore do not recommend universal cholesterol screening in children. But children older than 2 years with a parent whose total cholesterol is 240 or higher should be screened, as should those with a parent or grandparent who had a documented MI, angina, peripheral or cerebrovascular disease, sudden cardiac death, or atherosclerosis at 55 or younger, or had coronary bypass surgery or angioplasty. Screening may also be indicated if the family history is unobtainable and risk factors are present. The Canadian Task Force holds that there is insufficient evidence to recommend routine screening in children and adolescents and that individual judgment should be exercised.23
Some leading heart researchers are calling for a change in the nation’s cholesterol policy, including a retreat from universal screening and treatment of high cholesterol to prevent heart disease. In an editorial in the journal Circulation, Dr. Stephen Hulley and his colleagues at UCSF wrote that the 6% of middle-aged adults with cholesterol below 160 mg./dl. are at increased risk of dying from lung and other non-colon cancers, respiratory and digestive diseases, trauma, hemorrhagic stroke, and other causes, and concluded, “The overriding ethical obligation is to do no harm. Especially when considering the long-term use of drugs for people in good health, the burden of proof falls on the proponents of intervention.24
As diseases are broadened, multiplied, and redefined in accordance with the drug industry’s commercial priorities, even normal physiological processes become fair game for manipulation, as new abnormalities are identified, new drugs developed to correct them, and new “diseases” created after the fact as an appropriate marketing strategy. One of the best-known examples is the exploitation of menopause, beginning innocuously with the short-term use of estrogens to relieve hot flashes and other symptoms, but quickly progressing to X-ray detection of osteoporosis and “osteopenia,” or low bone density at an earlier stage, and ultimately to prescribing estrogenic hormones routinely and for many years to countless millions of women, in order to prevent hip and spinal fractures, cardiovascular disease, dementia, and a host of other common ailments of middle and later life.
The pharmaceutical industry and the medical profession have discovered a new disease: menopause, or estrogen-deficiency disease. Women with hysterectomies may want to take hormones until the natural age of menopause, while others have troubling symptoms like hot flashes and insomnia that warrant treatment. No one argues that short-term use is dangerous. But the push is onto use these drugs long-term, in the name of “disease prevention.” The American College of Obstetrics and Gynecology (ACOG) recommends that every post-menopausal woman should be on hormones for life unless she has a compelling reason not to be.
This sweeping recommendation is based on inadequate evidence. Menopause is no disease, but a normal part of life. A woman’s ovaries don’t shut down: they continue to produce hormones well into her 80s. Synthetic hormones don’t replace something that is missing, but add something that is not naturally there. Pharmaceutical companies realize that for marketing it is smarter to emphasize diseases than hormones. In this they are helped by the medical profession, which has recently redefined osteoporosis. The term once referred only to actual fractures caused by the thin bones of old women, but now it is defined as low bone density. This is like telling someone with high cholesterol that they have heart disease.
Women are encouraged to have bone density tests just the same as mammograms or Pap smears. The result is an epidemic of healthy 50-year-old women being diagnosed with osteoporosis, even though women on average don’t have hip fractures until they turn 79. Someone once said that if you’re healthy, you haven’t had enough tests done yet. We need accurate information, and must be on guard, lest vested interests sell us a bill of goods.25
It is notable that Dr. Love wrote this article solely on methodological grounds, years before the extra cases of breast cancer finally discredited long-term estrogen therapy and confirmed what many had long suspected but did not oppose publicly. Unfortunately, her arguments went unheeded for so long that the manufacturers’ enormous profits were unaffected, and when hormone-replacement therapy finally did go out of fashion, the conquest of menopause continued uninterruptedly, according to a thoroughly predictable sequence. By then the industry had already produced and marketed a new generation of even more potent and dangerous drugs, the biphosphonates, with the aid of which it similarly claims to prevent fractures by increasing the bone density, but this time by playing with the critical and still poorly understood mechanism of bone formation, and with calcium and phosphorus metabolism in particular, which influence and therefore potentially threaten the structural integrity of the body as a whole. It would of course be unfair to deny or fail to mention that much of scientific value is learned along the way.
Medicine has expanded into almost all facets of human existence, including conditions that do not cause symptoms or impair life, but indicate potential illness in the future, such as high blood pressure, blood sugar, cholesterol, osteoporosis, colon polyps, heart murmurs, carotid artery narrowing, memory loss, and sun exposure: the list is constantly expanding. One may reasonably anticipate that such risk factors will be recognized even earlier in life, at birth, in utero, or even before conception. Should everyone be screened? Everyone is tied to the medical establishment from birth, resulting in preoccupation with survival, rather than the challenge of creative living.26
One unavoidable by-product of this tendency is some kind of compendium of officially recognized diseases and subtypes, such as the prodigious ICD-10 classification, a book and software database, which runs to hundreds of densely packed pages, lists tens of thousands of pathological diagnoses, each one numerically coded, and is updated annually, as new diseases are identified, and old ones renamed or eliminated. Even the most cursory examination of this monstrosity leaves no doubt that most of the entries are merely technical abnormalities, adopted purely for administrative and taxonomic purposes, without the slightest pretense of referring to a dynamic process that could help us understand the lived experience of our patients.
The Patient as Specimen
Reducing illness to disease also harms patients in an indirect but even more fundamental way, which I have not directly addressed, but is implicit in what I have already said. Without regard for the individuals who suffer from it, studying a disease process in the abstract tends to reduce the patient to a mere specimen, and his or her actual experience to an automatism, a self-propelling chain of necessary causes, which also seems pre-programmed to worsen, because the corresponding tendency of every patient with every illness to recover depends upon idiosyncratic variables within the individual that are rendered invisible, mysterious, or simply irrelevant by the disease concept, and simply fall through the cracks, rarely if ever to be seen or thought of again.
By thus ignoring and indeed obscuring the self-healing capacity of the patient, modern medicine repeatedly undermines and defeats the seamless integrity, resilience, and restorative power of life itself, whatever we choose to call it, precisely because it is global, indivisible, and therefore resistant to definition, quantification, or analysis into separate parts, even when we remember to look for it. Once we fall ill, or are labeled as a specimen of a disease, whether in the present or the future, it is difficult not to feel overawed, intimidated, and even compelled to accept such pathology as given, to surrender ourselves to what we suppose or are told to be its laws, and to forget that every illness must also be received by the patient, and expressed by each of us in our own way, such that whatever factors may have influenced our getting sick in the first place are just as likely to help us recover in the future. Because the reality and even the concept of self-healing have been largely hidden from view, and are therefore unfamiliar to most people in most situations, the idea of chronic disease, in particular, includes nothing but its tendency to persist and get worse, and hence seems to require an aggressive strategy of treatment, to control it insofar as possible.
The concept of objectified or reified disease processes similarly distorts and trivializes both case management and prognosis, by reducing the difficult art of treatment to the technology of forcibly correcting the abnormalities used to define them, by killing the TB bacilli, selectively destroying the cancer cells, normalizing the blood pressure, and so forth. In the same way that histologically “cured” cancer patients who develop cachexia and aplastic anemia as a result of their chemotherapy may be at least as sick as before, these crude oversimplifications leave a dangerous ambiguity in the assessment of improvement and worsening, indeed in the definition of the disease itself.
Thus, patients who develop a severe or intractable illness following apparently successful treatment of another illness pose a critical dilemma that cannot be solved or even meaningfully stated within the concept of the disease process. For whether or not the two conditions are said to be related to each other, the patient will still require a separate diagnosis and treatment for each of them, and there is still no meaningful way to address the condition of the individual as a living whole, an integrated bioenergetic system that evolves and develops through time. In either case, the net effect of medical science as a philosophy or conceptual system is always simply to multiply, both exponentially and across the board, the number of diseases, abnormalities, and diagnostic and treatment procedures to be comprehended within it, through its awesome power to subdivide the living patient into ever more numerous identifiable and potentially controllable phenomena.
Diagnosis as Truth and Falsehood
It would of course be unjust to dwell on the authentic defects and limitations of routine screening tests without also acknowledging the numerous instances where our patients would have died or suffered far worse had they not been done. Just such a patient came to see me recently, as I was finishing up this article, and congratulating myself on having demolished the established concept of preventing various specific diseases in healthy patients without illness or symptoms. An attractive divorcée with two young children who had been my patient for several years, this time she arrived quite distraught and sobbing uncontrollably, after a routine Pap smear found a large number of what her gynecologist described as “highly abnormal and suspicious cells,” and for which the latter recommended complete hysterectomy—removal of the uterus, ovaries, Fallopian tubes, and even the vagina and regional lymph nodes—even though and also precisely because she felt entirely well, had no other complaints or symptoms, and ultrasound, colposcopy, and endometrial biopsy had all failed to detect a localized tumor mass to account for them. Consulting another gynecologist for a second opinion at my request, she learned that the original Pap smear actually read “adenocarcinoma, type undetermined,” which explained why immediate action was called for, as well as why her own doctor had lied that she did not have cancer, and had urgently insisted on the surgery as a 100% effective way of preventing the near certainty of it in the future.
In short, hers seemed to be precisely the test case I had envisioned earlier, a life-changing decision made solely on the basis of what cells looked like under a microscope, with as yet no symptoms or illness for them to explain. Yet whatever the differences among us as to the exact interpretation of these findings, or what we thought the wisest course for her to follow, all of us, her two specialists, the patient herself, and even I, seemingly contrary to what I had just written, accepted the provisional truth of this report as signifying a very high probability of serious, life-threatening illness in the near future, and agreed on the urgency for some kind of active and timely intervention to try to save her life, or at least to buy her as much time as possible for continuing to parent her twins as she sees fit. By that calculus, neither she nor I had the slightest doubt that her diagnosis was accurate, useful, and indeed supremely valuable to her, however, grim and terrible the news it brought, and although the action it called for will almost certainly be disfiguring, or even worse.
As this by no means rare example makes clear, preventive screening tests and procedures continue to earn a central place in modern diagnosis, and cannot be dismissed a priori on purely ideological grounds, even when no signs and symptoms of illness are manifest. As to when they are most likely to be helpful, harmful, or merely distracting, the doubts and scruples I have just cited may help to redefine the circumstances under which the added information they provide may be judged as true, false, or uncertain, and even what we mean by these heavily loaded words.
The example of my patient is especially instructive in an additional sense, because the Pap smear is widely regarded as inadequate for detecting frank or invasive cancer, so that the presence of unequivocally malignant cells was already sufficiently unusual to alarm every doctor and technician who saw them, while the absence of well-differentiated cell types left open the possibility of ovarian cancer, statistically the most dangerous kind, and called for immediate action just to rule it out. Under these ominous circumstances, not least because she felt well and had no symptoms, the test helped to convince my patient that she had or would soon develop a life-threatening illness, and that drastic action was imperative to forestall it. Finally, the rather high degree of integrity of the various possible disease processes, that is, of vaginal, cervical, endometrial, tubal, ovarian, and bladder cancer, respectively, was also an important component of the urgency for further diagnostic and surgical procedures to differentiate them, so that the only disagreements between the various players involved the ever-present issue of how best to balance the need for effective diagnosis and treatment with the patient’s optimal comfort and quality of life. In short, the only ambiguity remaining in her case was the standard policy of screening all adult women.
But that is the whole point. Pap smears are done routinely to screen the entire adult female population, involving the detection of cervical dysplasia and other cellular abnormalities as risk factors for cancer of the cervix in the relatively distant future. Even its most zealous proponents would hesitate to advocate the procedure solely to catch the very few exceptional cases like my patient, and it is now widely known, as we saw, that mild and moderate dysplasia, while sometimes progressing to cancer after periods of months or years, most often revert to normal if simply left alone and allowed to do so. Charting a path somewhere between these two essentially uncontested truths suggests the reasonable and increasingly popular compromise of doing the test, yes, but then simply watching, waiting, and taking no further action unless the morphology spreads and worsens.
At least on the face of it, a similar strategy seems equally applicable to mass screening for cholesterol, blood pressure, bone density, and so forth, which are likewise designed to assess risk factors for particular diseases and illnesses, for the most part in the relatively distant future. As with the Pap smear, the main problem with routine screening is not so much the discomfort of the procedure itself, but rather the pain, suffering, disability, and expense incurred by the further diagnostic tests and treatment procedures that they seem to make necessary. No competent physician would hesitate to recommend further testing and corrective action for those with abnormalities lying well outside the normal range, and thus indicative of something in the present, some illness as yet unsuspected and unperceived by either the patient or the doctor, but which a more detailed and extensive workup would probably reveal, and which the patient would want, expect, and be grateful to know about. Given a blood pressure of 180/110, a cholesterol over 300, a TSH of 15 or 20, a liver transaminase in the 400s, or a PSA of 20, few doctors would be content to dole out the usual drug without first trying to get to the bottom of these findings, to establish a more definite and comprehensive diagnosis, because all of them indicate authentic disease processes and indeed actual illnesses that are already under way, so that the pathology is very likely to illuminate, explain, and indeed to predict them. In this situation, such striking abnormalities and the diseases that they represent may both be considered real and true, in the same sense and to the same extent as the acute, subacute, and intermittent diseases we discussed earlier.
With “borderline” result, on the other hand, a slight deviation with at most possible significance in the distant future, it seems completely excessive and amounts to cruel and unusual punishment to inflict these largely arbitrary and ever-changing standards on otherwise healthy patients for years and decades of their lives, artificially forcing their test results back into line and holding them there, without compelling evidence of benefit to their general health and well-being from doing so, let alone due regard for the predictably adverse effects of the coercion itself. Simply being less zealous about enforcing such dubious diagnoses would save enormous sums in follow-up visits, drug costs, and the tests to monitor them, to say nothing of the anxiety that sustains them all, and would thereby help to reduce the size and weight of our famously bloated medical enterprise to a more nearly human scale.
Meanwhile, on the receiving end, so to speak, confronted by their own fear and hesitation on the one hand, and the vast knowledge and formidable power of the medical system on the other, our patients themselves provide a fitting epilogue in their profound desire, need, curiosity, and fascination to match up and integrate their own unique lived experience with the independent and profoundly alien version of the body as a machine, so blissfully or horribly anonymous and neutral. That is why it makes good sense, for medical as well as social, political, and ethical reasons, to order such tests for those patients who want them, whether we recommend them or not, and to build a patient-centered criterion into the system, even if it costs a little more, to allow patients to screen themselves, to allow their physicians to try to dissuade them when it seems appropriate, and thus to facilitate an ongoing negotiation about fine-tuning their care.
The high value and awesome power of a proper diagnosis for individual patients are nowhere more evident than in the remarkable growth and diversity of charities and support groups for those afflicted with diseases of every description, many recruiting informally through the Internet, and achieving major status and political clout through conferences, publications, fund drives, and research sponsorship. Although many individuals find it distasteful to be thus lumped together with people having little else in common, or to dwell on their diseases to the exclusion of pleasanter or more edifying topics, many others who suffer most intensely tend to find comfort, solidarity, and even personal truth in the company and fellowship of others similarly affected, at times attaining a level of peace and equanimity that would have hardly been possible otherwise. Still others are moved to social action, such as lobbying Congress to call attention to diseases that are new, controversial, or poorly understood, like Autism Spectrum and Asperger’s Syndrome disorders, or to right a perceived wrong, such as the side effects of drugs, vaccines, toxins, chemicals, and additives. Precisely because they empower patients to find and speak their truth, however unreasonable or unpalatable it may seem to others, and thereby to participate more actively in a health care system that seems increasingly disinclined to listen, such groups deserve encouragement and support from physicians, no less than the public at large.
Declare the past, diagnose the present, and foretell the future;
But as to diseases, let us try to help, or at least to do no harm.27
Ibid., p. 365.
Ibid., p. 366.
Hippocrates, Epidemics, Book I, Section 2, par.xi.