To the ancient aphorism, first do no harm, should be added a modern extension: compassion without competence is crap.
Michael B. Shimkin, Report from Laputa, "Some Polemics on Medical Mythology,"Archives of Environmental Health, 22:151-153 (1971).

The cancer research scientist is a chronic doubter. He is never sure what he is looking for, nor what he wants to find. The experimental cancer therapist knows what he is looking for and what he wants to find; but he needs the scientist to tell him where to look. The scientist is apt to be a doubter, the therapist a true believer. The scientist is suspicious, even angry, with the therapists's evangelism, and the therapist is impatient with the scientist's equivocation and criticism. The rare individual who does both leads a schizophrenic existence. Usually, he forsakes one calling for the other, or goes into administration (and often very effectively, since he understands the problems of both groups).
The objective critic of the experimental therapist can sometimes be impaled on the horns of a dilemma by an unscrupulous experimental therapist. If the critic asks for adequately controlled experiments, he is reminded that the therapist is a physician, i.e., he cannot withhold a possible cure. If the critic points to the basic doctrine of the humane physician "above all, do no harm (primum non nocere), he is informed that science takes precedence over the physician's art. When the contradictions in the above arguments are presented, the critic is then reprimanded for criticizing some one who is "at least trying to do something."

Experiments that harm patients deserve to be criticized regardless of the motivation. Most practitioners are humane physicians who try; and if they fail to produce a cure or alleviate symptoms, they comfort relatives, relieve suffering, and allow their patients to die with dignity. There are also sadists and cancer quacks, with and without M.D. degrees, both in and out of respectable institutions, who deserve to be tarred and feathered.

Treatments, for good or evil, are much like laws. They are easy to institute and hard to repeal. Current medical history is full of treatments that hurt and kill people. In my lifetime, I have seen the following:
1.X-ray given to shrink the thymus gland of infants. Results: It did the children no good and subsequently caused leukemia and cancer of the thyroid in some.
2. X-ray for a whole variety of lesions, including warts (which can be wished away) and ankylosing spondylitis (a disease of the spine). Results: the lesions disappeared and the leukemia rate increased.
3. High oxygen to premature infants. Results: blindness in some.
4. Removal of the tonsils. Results: in some cases, very helpful, but in most cases, unnecessary (even in pre-antibiotic days). Some operative deaths (percentagewise low; in actual numbers of deaths, considerable).
5. Antibiotics for colds. Results: does no good for colds. Can cause sickness and death in allergic patients; destroys normal bacteria needed for resistance to other diseases and can produce antibiotic-resistant strains of bacteria.
6. Antibiotics for virus infection. Results: in most cases either worthless or potentially harmful.
7. Injection of plastics for cosmetic reasons. Results: Experimental evidence indicates that the probable long-term result may be cancer in those treated.
8. Thalidomide as a tranquilizer (not many in the U.S.). Result: deformed children when used in early pregnancy.
9. Stilbestrol in early pregnancy. Result: cancer of the vagina in female children of women so treated (see chapter "Chemical Carcinogenesis.")

These are easy to document. There is a long list of probably harmful treatments and a longer list of possibles.

There is sometimes a conflict between what is preferable in the treatment of patients and what is needed to be scientific. There may be times when the appropriate application of the scientific method, to the detriment of the patient, may prove to be beneficial to a much larger number of people. I am, therefore, willing to accept an institution that practices good medicine and poor science, or one that practices poor medicine and good science. I have no patience with one that practices neither good medicine nor good science.

A physician friend of mine says that it would be better to concentrate my criticism on the out-and-out cancer quack, rather than the medical profession. I disagree: The con man is not going to improve as a consequence of public pressure, while some members of the medical profession might. It is because the physician is conscientious and trying to do the best for his patient that improvement is possible. The only thing that can be done in this book about the cancer quack is to warn the public, and this has already been done many times.

Almost all judgments in clinical medicine are tentative and almost all treatments are experimental. That is a pretty dogmatic statement; let me explain: The reason that I say that most judgments are tentative, is because it is still impossible to predict the future. The perceptive physician is aware of this, and continually follows his patient through the course of treatment, changing it when necessary. It is reasonably well established that penicillin is one of the best treatments for infection. However, in administering the treatment, the physician may find out that the organism that he is trying to kill appears to be resistant to the drug, or the patient may be allergic to penicillin, making the drug potentially lethal. The perceptive physician immediately changes his treatment to another antibiotic. The one who follows scriptures will continue the treatment, endangering the life of the patient.
To say that all treatment is experimental is also a valid judgment, in the same way that all automobiles are experimental, all airplanes are experimental. Unfortunately, in the process of trying to please both himself and his client, the physician may (as General Motors did with automobiles) provide a product that is far from the best. It is also far from the worst. This point was brought home to me in a discussion with a gynecologic surgeon who believed that he is able to separate two different types of cancer of the uterus; one being 100% lethal, and the other relatively susceptible to treatment. He is trying radical surgery, in an attempt to cure the lethal type. I questioned him about the untreated controls in his experiments and he said "I don't need any: this disease is 100 percent fatal, and if I save just one patient, we will have accomplished something." This made sense to me. His procedure is what we refer to, in the laboratory, as a "pilot" experiment, which is the type of experiment that you perform when you are looking for promising leads. His answer to my next question was disconcerting. I asked him, "Suppose that, out of ten patients, two get well. What are you going to do then?" His reply was, "This will become our method of treatment." At this point he and I intellectually parted company. I believe that once an experimental therapist has a procedure that works he should not rest on his laurels. The next step should be to refine the procedure to minimize the risk of injury to the patient. This would consist of a series of controlled experiments in which one group of patients received the new operation, while another group had a similar operation with some of the undesirable parts of it left out. These experiments should be continued until the best possible treatment was found; one that is lifesaving and has a few undesirable side effects as possible.
The question in my mind is not whether physicians should experiment on their patients. They do experiment, they have to, and they will continue to do so. I would like the experiments to be well controlled so that thc physicians of the future will have better and more reliable treatments than did their predecessors. Most of the experimental therapy performed today is very poorly controlled. Perhaps this is because the traditions of scientific method are new to medicine. The tradition in the healing arts consists of a belief in the dicta of "authorities." The scientific tradition, in contrast, is one of skepticism and the constant questioning of authority. Medicine (especially academic medicine) has a mixture of scientists, who often are the "authorities," and practitioners, who are, with rare and wonderful exceptions, true believers. Sometimes the scientists abandon the scientific tradition and become dogmatic authorities. There is usually a running battle in medical schools between the scientists and the practitioners, and there is nothing "scientific" about the way that it is fought. The ideal medical school professor who can be a critical scientist and a humane practitioner is rare. And one who can also teach is still rarer.

I don't know what motivates a physician to specialize in cancer therapy and I am not sure that I want to know. In the tradition in which I was raised, the noblest thing that a human being can aspire to is to help the hopeless. I do not understand how a person can do this because, even with the outer atmosphere of cheerfulness and hope, it must be a veritable internal hell for the man who has to do it. Every physician is grateful to the man who lifts the awesome burden of the incurable cancer patient from his shoulders. It must also be said that there are also therapists who thrive on human misery, whom the world would be better off without. Despite my admiration for the conscientious therapist, I cannot share his enthusiasm.
Victories of cancer therapists are small, and they come hard. The obstetrician has the joy of bringing new life into the world; the surgeon who takes out an appendix or removes a tumor successfully has the satisfaction of knowing that his patients get well. The cancer therapist, in contrast, spends most of his time giving people a little bit of added life, or making people more comfortable. He knows that most of his patients will die in a relatively short time. These are dedicated men, driven by some inner need to serve the hopeless. Most of them work in the tradition of Father Damien. It is these people who are dedicated to help ing the hopeless who make this world a more beautiful place to live in. Yet, as is often the case, highly dedicated people can be blind to the objective truths about what they are doing.
To give a person a year or two more of life is usually a desirable thing. It is desirable if that year or two of life is relatively pain-free and the person so blessed is allowed to lead a relatively normal life. To literally bring someone back from the grave, only to have him die in pain three months later, is not only not a service, but is unnecessarily cruel. Treatment with large doses of x-ray makes patients very ill for a period of time, as do some of the chemicals used in chemotherapy. The skillful therapist is often able to make the distinction between conditions where his treatment can do some good, and those where it probably cannot. To suggest to a patient, who cannot be helped, that he not be treated might be the kind thing to do. It is, unfortunately, the rare therapist who will not treat a patient, and the patient may be subjected to unnecessary discomfort. Part of the reason for his treating the patient is his enthusiasm, and part is the insistence of some patients and their relatives that the doctor "do something."
A dogma that has permeated medical practice for as long as it existed is that anything that a physician does is better than doing nothing. It is extremely difficult for someone who has devoted his life to helping people to admit to himself, much less to the world, that his patients might have been better off if he had never seen them. Some of the more cynical members of the thinking public have made this evident. Mark Twain's remark, "He has been a doctor a year now and has had two patients --no, three, I think --yes, it was three; I attended their funerals," testifies to this skepticism. It was well known by the public that the great general hospital in Vienna, in the first half of the seventeenth century, had an extremely high mortality rate in pregnant women. This was due to a condition known as puerperal fever. It was Ignaz Semmelweis who showed that the infective material that caused this disease was conveyed by the hands of the doctors and medical students from the autopsy room to the expectant mother. He was virtually laughed out of the profession, and ended his days in a lunatic asylum. Perhaps it takes a madman to keep "sane people" from doing harm. A more recent example illustrates that the attitudes in medicine are slow to change. H. J. Muller discovered that x-rays produce genetic changes (mutations) in 1927. He received the Nobel Prize for this work in 1946. By that time it was also well established that x-rays could induce leukemia in mice, and that the leukemia incidence was considerably higher in radiologists than in other medical practitioners and the population at large. This had the effect of inducing radiologists to wear lead aprons when performing fluoroscopy, but had little effect on how they treated their patients. A concerted campaign was started approximately twenty years ago by a number of people in fundamental biology and radiology to induce people using x-rays to reduce the exposure to themselves and their patients. It met with unbelievable resistance and hostility. It is not possible to equate the number of people who have been sent to their death because of the use of medical x-ray against the number of people whose lives have been saved by the same tool. There is no way of estimating either.

The blindness of the fanatic is a source of strength (he sees no obstacles) but it is the cause of intellectual sterility and emotional monotony.
Eric Hoffer, The True Believer