The Dogs That Did Not Bark

Errors and Omissions in the New England Journal of Medicine

Theodore Dalrymple

• Publisher: Encounter Books
• Year: 2019
• ISBN:
  9781641770477
• Topic: Medicine
• Subtopic: Epidemiology

The book is a running commentary, week by week, on the New England Journal of Medicine, one of the most important general medical journals in the world, during the year 2017. It demonstrates that the conclusions of many of the papers in it are not only flawed, but obviously flawed - though for lack of time, many doctors will not examine them closely and will therefore be taken as authoritative. In some cases there is the suspicion of actual corruption. There are errors of reasoning and obvious omissions, apparently undetected by the editors. In addition, in its examination of social questions the Journal is solidly politically correct, and practically no debate on these matters appears in its printed version. Highly debatable points of view go completely undebated. The Journal reads as if there is only one possible point of view, its own, though the American medical profession, to say nothing of its extensive foreign readership, is more than a millions strong and cannot possibly be universally in agreement with the stances it takes. It is thus more megaphone than sounding board - as, in the opinion of the author, it should be. The book is not intended to be destructive, however: praise and admiration are given where the author thinks they are due. However, it should encourage the general reader to be wary of the latest medical doctrines, and to take a constructively critical stand to the latest medical news.

January 12, 2017

Not everyone, I suspect, would turn beating heart to an article with the title “Eliminating Cholera Transmission in Haiti,” the first in the Journal this week. Haiti is of small account in the world, an infamously hopeless case of political pathology; but as I have twice visited the country, nothing Haitian is alien to me. The country grips the imagination of the visitor as no other; your passion for it, though it may fade a little with time, is rekindled at the first mention of its name. Its tragic destiny is like a highly compressed history of humanity, or a memento mori for the whole of mankind.

Nevertheless, there was good news in the article: two effective oral vaccines against cholera have been produced, one in India and one in South Korea (a sign of the shift of the world’s scientific center of gravity, perhaps). I remember the days when cholera vaccines had to be injected in the upper arm and caused considerable soreness, though not necessarily much immunity to cholera. The new oral vaccines are stable for a month at 37 degrees Celsius, and therefore require no chain of refrigeration to be used successfully in remote rural areas. In Haiti, electricity is less reliable than the weather.

The second paragraph of an otherwise excellent article was almost as interesting for what it did not say as for what it did. Its first sentence informs us: “Cholera had not been recorded in Haiti until it was introduced in 2010.” We then learn that the epidemic involved 800,000 cases, which is about 8 percent of the population, and that it resulted in 10,000 deaths. (A death rate of one in eighty cases demonstrates that even in Haiti modern medical care has a long reach and a beneficial effect, for the death rate of cholera when it first appeared in Western countries in the nineteenth century was about 50 percent. We lament the state of the world, but progress, however uneven, has been patent.)¹

But notice the dog that did not bark: cholera was introduced. By what, or by whom, was it introduced? Surely this is a matter of some interest and worth a word or two. Is the omission a sign of embarrassment?

The scientific evidence strongly suggests that cholera was introduced into Haiti by Nepali troops of the United Nations peacekeeping mission there, known by its initials MINUSTAH (Mission des Nations Unies pour la stabilisation en Haiti). Just before a group of them arrived in the country in October 2010, there had been an outbreak of cholera in Nepal, but no one thought to take precautionary measures to prevent the troops spreading the disease to Haiti, a place very vulnerable to any epidemic disease favored by poverty and poor or nonexistent sanitation. It seems that the sanitary arrangements for peacekeeping troops were so poor that soon after their arrival at a camp near the Artibonite River (Haiti’s largest), a truck disposed of the soldiers’ waste straight into the river. It provoked an epidemic that has not yet ended.

Needless to say, no one intended to start an epidemic of cholera in Haiti; but once it had started, a considerable effort went into covering up its source. The American Centers for Disease Control, the World Health Organization, the United Nations and various supposedly independent researchers went in for obfuscation until the evidence became almost undeniable that the Nepali peacekeeping troops had brought it with them.² Interestingly, the obfuscators were assisted in their obfuscation by the two most influential general medical journals in the world, the Lancet and the New England Journal of Medicine, both of which refused to publish scientific papers that made a very strong case that the Nepali soldiers were the source of the epidemic. The Lancet did, however, publish an article to the effect that speculating on the source of the epidemic was unnecessary and even counterproductive.

Why all the obfuscation? One can only guess. For one thing, the Haitian population, victim already of so many disasters both natural and political, was understandably angry when it first came to suspect that the Nepali troops were the source of the epidemic. There were some anti-Nepali riots, and the authorities feared that an admission of MINUSTAH’s responsibility might lead to an explosion. Then, the longer they obscured the truth, the more difficult it was to admit. Who, in any case, likes to own up to having caused a catastrophe, even inadvertently? There was probably also a reluctance to believe that those who intended good could actually do harm, and on such a major scale. No one likes to think that the best of intentions can result in disaster, and so truth was avoided and untruth propagated.³

The NEJM certainly did not cover itself in glory during the whole episode, perhaps because it toed official lines. I surmise that the failure to mention who introduced cholera into Haiti (which after all would have taken no more than four words) was a sign of unease or even bad conscience among the editors, who normally lose no opportunity to put the Journal’s humanitarian heart upon its sleeve.

Telling the truth would threaten not just a few individuals and institutions, but an entire worldview that was more difficult and painful to give up than a bad habit. No one could be against the United Nations; it would be like favoring cruelty over kindness. Alas, this was not the first time that an agency of the United Nations caused a catastrophe and then tried to cover it up.

In the 1970s, UNICEF and the World Bank attempted to reduce the very high infant mortality rate in Bangladesh by drilling millions of tube wells to provide bacteriologically clean groundwater. (A high percentage of infant deaths were caused by fecally contaminated surface water containing the agents of fatal gastroenteritis.) As far as reducing the infant mortality rate is concerned, the wells were a great success, but unfortunately the groundwater contained a high concentration of arsenic, a carcinogen. The result was that tens of millions of Bangladeshis now face the prospect of various kinds of cancer, from which thousands have already died. When this became evident, UNICEF tried for a long time to conceal it—a normal human reaction, no doubt, but not one to be expected of a humanitarian organization with no thought of self-interest.

A review article titled “Screening for Colorectal Cancer” was of some personal interest to me because mine is a family in which this disease is frequent. In essence I read it to find some reason, or rather a rationalization, not to undergo the regular screening colonoscopy whose discomfort I fear. My eyes alighted at once upon a passage that fed my wishes:

Although not all trials have shown a significant benefit with respect to reducing mortality … several large, randomized controlled trials have confirmed the effectiveness of one-time and periodic … sigmoidoscopy, with a 26 to 31% lower mortality from colorectal cancer.…

So a regular colonoscopy appears to lower the death rate from colorectal cancer, but the paper never mentions the all-cause death rate. Thus it leaves open the possibility that screening might save people from dying of colorectal cancer, only for them to die of something else at the same age. What I want to know is whether screening is likely to extend my life, not just prevent me from dying of colorectal cancer. This is an important consideration if a disease is not especially common by comparison with other possibly fatal conditions.

I would also want to know the absolute risk of dying from colorectal cancer, not merely the reduction in relative risk. This is because a large reduction in relative risk may translate into a trivial reduction in absolute risk if the condition is rare to begin with, while a smaller reduction in the relative risk of a common condition may be much greater in absolute terms.⁴ I am not saying that this is the case here, but the paper gives no means of working out the absolute risk. I have long made it a principle to mistrust papers where only relative and not absolute risks are given, or that make it very difficult to calculate the reduction in the absolute risk.

By its omissions, the paper gave me just what I wanted, then: a reason not to be screened.

Since writing this, I have had the opportunity (March 2019) to read Professor Renaud Piarroux’s book, Choléra: Haiti 2010–2018 histoire d’un désastre (CNRS Éditions). From this it appears that, having followed the NEJM, I have considerably underestimated the cholera epidemic’s lethality, perhaps as much as by eight times. ▣

1 It is necessary to be a little cautious in tracing the death rates from cholera because of different definitions of cases. Most people who are infected with the causative organism, Vibrio cholerae, do not suffer from the disease. They are cases of cholera infection, but not of cholera. The death rate therefore depends very considerably on case definition.

2 Nepali troops have been prominent in United Nations peacekeeping missions since 1959. The tradition continued even during the civil war in Nepal that lasted from 1996 to 2008. Thanks to Nepal’s poverty, the troops are cheap and there has always been a large discrepancy between what the troops receive in pay and what the Nepali army receives from the United Nations. For Nepal, therefore, peacekeeping is a valuable source of income.

3 For a brilliant and riveting account of the affair, see Ralph R. Frerichs, Deadly River: Cholera and Cover-Up in Post-Earthquake Haiti (Ithaca, N.Y.: Cornell University Press, 2016).

4 Suppose that I have a one in a million chance of dying of disease X, and that by eating broccoli every day I can reduce the risk to one in three million. That is a reduction by two-thirds, but not very significant in absolute terms. It would hardly be worth the effort.

January 19, 2017

“Behind mountains, more mountains,” say the peasants of Haiti. The more things are transparent, the more they are opaque; or perhaps they simply raise further difficult questions.

An article titled “Transparency and Trust—Online Patient Reviews of Physicians” raised more questions than it answered. The author, in impeccably bureaucratic prose, extols the potential value of reviews of doctors by patients, published on the internet:

Patient reviews offer clinicians valuable performance feedback for learning and improving, both individually and across a system. Receptivity to performance feedback, which depends heavily on physicians’ acceptance of the data’s validity, facilitates a culture of continuous learning and patient-centeredness.

Note that what is yet to be proved is here taken to be established fact: what is surmised to be possibly or desirably the case actually is the case. I have sat through many meetings, and read many documents, in which bureaucrats have made precisely this assumption. The world, I have found, is often refractory to the best of intentions (the worst being easier to fulfill).

“Whenever I hear the word culture,” Goering is reputed to have said, “I reach for my Browning [pistol].” Whenever I read bureaucratic prose, if any meaning at all can be discerned in it, I reach for my objections. Sometimes a still small voice whispers somewhere in the back of my head (it really feels located there) that my objections to whatever is proposed are really a manifestation of fear and dislike of change. The particular change discussed in this article hardly matters to me now that I am retired, but I wouldn’t have wanted to know what my patients thought of me.

What are my objections to such internet reviews, which treat doctors as if they were restaurants? Among them are the possibility that patients may value the wrong things in their doctors. A patient is not a customer, and a doctor is not merely a provider of a service, like a shopkeeper who sells whatever his customers wish to purchase from his stock; he is very importantly an adviser also. A doctor may decline to do what his patient wants of him because he considers it not in the patient’s interests. This might not win him a good review from the patient; but on the other hand, Dr. Harold Shipman was generally well regarded by his patients, and he turned out to be the most prolific serial killer in British history.

My wariness of online reviews was not entirely allayed by the three examples evaluating a surgeon named Courtney L. Scaife that were cited in the article. These reviews read:

1. Was thoroughly impressed with the physician.

2. Dr. Courtney is an excellent surgeon, she explains things very clearly, is very detailed and just amazing. I would recommend Dr. Scaife very highly.

3. It did feel like the provider is not as concerned for my condition as I am.

There is very little of any substance in these comments. One can easily be impressed by charlatans; indeed, whole books, often very amusing, have been written about the gullibility of humans, sometimes many thousands of people at once. Then too, only the survivors of surgery can give a review of their surgeon, so a doctor could decimate a countryside and receive glowing reports.

As for the critical remark among the three, would you trust the judgment of someone who wanted or expected his doctor to be “as concerned for my condition as I am”? A doctor who was concerned for his patient in precisely the same way and to the same degree as the patient was concerned for himself would be useless or worse, possibly dangerous. The doctor should be sympathetic, of course, and empathetic as well; but his sympathy and empathy for his patient are not those of the close friend or relative. The doctor is concerned for his patient but also detached from him, otherwise he would soon find himself paralyzed by emotion. A doctor cannot and should not grieve over the death of his patient in the way that a spouse, a child or a parent grieves. Not all is callous that is unemotional.

The comment that Dr. Scaife was not as con...