Jeremy types “systematic review randomized trial fasting” into a search engine. He finds a systematic review supporting the benefits of fasting.
An ounce of practice is generally worth more than a ton of theory.
—Ernst F. Schumacher
How empty is theory in the presence of fact!
—Mark Twain
Ignaz Semmelweis was the Hungarian director of the maternity ward at the Vienna General Hospital in the 1840s. The maternity ward was a big room divided into two halves called Ward 1 and Ward 2. In Ward 1, doctors and medical students—mostly male—delivered the babies. In Ward 2, midwives—all female—delivered the babies. Semmelweis noticed that 10 percent in Ward 1 were dying and fewer than 4 percent were dying in Ward 2.
He tried to figure out why so many mothers and babies were dying in Ward 1. At the time, people believed the “miasma theory,” an idea that diseases were caused by “bad air.” This was not as crazy as it sounds, because people were often ill in crowded urban areas where the air was in fact very smelly. In fact, the word malaria just means “bad air” in Italian. But Semmelweis realized the difference in maternal deaths could not be bad air, because both wards were in the same room with the same air.
The Viennese professors suggested it was foreign student doctors who were causing the deaths because they were rough with patients. Semmelweis agreed that the foreign doctors were associated with more deaths, but it was because they worked harder and treated more patients, not because they were less well qualified or more rough. Desperate, Semmelweis even persuaded the priest, who normally walked through the first ward on the way to the second ward, to take a roundabout route and enter via the second ward. He thought the priest might frighten the patients into being more likely to die. The priest reluctantly agreed to change his point of entry, but the difference in deaths remained the same.
Then something very tragic happened to Semmelweis’s colleague Jacob Kolletschka. While performing an autopsy, Kolletschka was accidentally stabbed in a finger by a student with a knife that had been used on the corpse. He soon became ill with the same symptoms as the mothers who were dying. Since Kolletschka’s death and the mothers’ deaths appeared to happen in a similar way, Semmelweis had a lightbulb moment. As it seemed that the doctor died because he was stabbed with a knife that had touched the dead body, maybe some particles from the dead body were reaching the mothers. But how? The autopsy room was in the basement, whereas the maternity wards were upstairs.
The link was simple: the doctors moved freely between the autopsy room and the maternity ward. They washed their hands with soap and water, but they did not disinfect their hands. Semmelweis thought that although the doctors washed their hands, some small particles from the dead bodies must remain under the doctors’ fingernails. It seems obvious to us today that doctors should disinfect their hands after an autopsy before treating patients. But the reason it is evident now is because of what people like Semmelweis discovered.
To be sure, Semmelweis couldn’t see the particles—he didn’t have a powerful enough microscope. However, he instituted a policy of doctors and students washing hands with a solution of chlorinated lime, what we now call chlorine bleach, before entering the maternity ward.
Soon after the students started disinfecting their hands, the death rates in Ward 1 reduced to below the death rates in Ward 2. Very happy, Semmelweis wrote about his discovery in a book, and in medical publications across the world. His work eventually became very famous, with his name on coins, stamps, and even a university named after him in his native Budapest.
But all his fame came after he died. At the time, the reaction to Semmelweis was hostile. The other famous doctors in Europe rejected Semmelweis’s findings. A British doctor named James Young Simpson said that the Brits already knew about Semmelweis’s discovery. This was untrue, because 60 percent of people who had surgery in London hospitals died of postoperative infections. Semmelweis’s Viennese colleagues also rejected his policy of disinfecting hands. He was demoted and forbidden from treating patients. More women in Ward 1 quickly started dying again.
Disgusted, Semmelweis went back to Budapest, where he began drinking and slowly going insane. He was forced into an asylum, where he was severely beaten by guards, secured in a straightjacket, and confined to a dark cell. He died two weeks later from a gangrenous wound, possibly caused by the beating.
How could Semmelweis’s lifesaving discovery have been rejected? Some believe it was because he was a Hungarian in Vienna, and the Austrians might have been prejudiced against him. This prejudice may have mattered in Vienna itself, but not elsewhere in Europe where his work was published. The latest research suggests the reason for his failure was that he didn’t have a theory to explain how disinfecting hands could work. His colleagues didn’t know about germs and didn’t understand how a tiny amount of material from the corpses could cause the mothers to die. Their refusal to believe Semmelweis’s observations was due to their obsession with theories, and Semmelweis didn’t have one because he couldn’t see the small particles under the fingernails of the student doctors.
A few years after Semmelweis died, Joseph Lister did some experiments proving that germs caused disease. Supported by the germ theory of disease, disinfecting hands became (and still is) one of the most important things that all doctors learn to do. Did you know that the mouthwash Listerine comes from the name Lister?
Too much focus on theory over observations can be dangerous for introducing new treatments, too. Indeed, before modern medicine most treatments, such as bloodletting and leeching, fell into this category. Bloodletting was based on the theory of the four humors. According to this theory, our body contains four liquids, or “humors”: blood, phlegm, yellow bile, and black bile. Diseases (again according to this theory) were caused when these four humors were out of balance.
Blood was thought to be the dominant humor, so when people were ill, it was often believed to be because there was too much blood. The cure, therefore, was to cut the person to release some blood, usually about two pints. (The most prominent medical journal in the United Kingdom, The Lancet, derives its name from the knife used in bloodletting.) Ironically, bloodletting probably had a huge placebo effect, so it might have made people feel better, at least temporarily. At the same time it was almost certainly harmful for many of the diseases it was intended to treat. In fact George Washington was probably bled to death by his four doctors.
Here is a more recent example of a bad theory leading to a harmful treatment. Irregular heartbeats called arrhythmias (irregular heartbeats) are known to increase the chances of dying from a heart attack. So when drugs that reduced irregular heartbeats were discovered, doctors gave them to people having arrhythmias. The theory was that reducing irregular heartbeats would also reduce death after heart attacks. On this basis, hundreds of thousands of people were given the drugs. However, some American cardiologists insisted on doing a trial comparing the drugs against placebos. Proponents of the drugs accused them of being unethical, and they thought those taking the placebo would die. The trial began in 1987, and 1,727 patients were randomized to receive the drugs or a placebo.
The results? Researchers analyzed the outcomes halfway through the trial and found that eight out of ten people taking the drug died, but only three out of ten people in the placebo group died. This showed that the drugs were killing people. Because of this, they stopped the trial early and the drugs were withdrawn as a treatment for irregular heartbeats. More than 50,000 people who took the drugs died each year, which is more than the total number of Americans who died in the Vietnam War.
Theories are important for telling us where to look for new treatments. For example, Louis Pasteur developed the Germ Theory of Disease, and this led him to see that rabies might be cured by a vaccine. Stories are useful for teaching and remembering. And some theories or stories about treatments will turn out to be supported by evidence, but many will not, and we cannot be sure until we actually test them in good trials. A problem is that experts often use theories and stories without checking the evidence. We have to remember that just because someone is an expert, it doesn’t make them right, and just because a theory about a treatment makes sense, it doesn’t mean the treatment works.
I remember the first time I went to a medical conference as an invited speaker. I had just finished my PhD in philosophy. At philosophy conferences, they used to put us up in student dorms and give us sandwiches for lunch. So I was surprised when I was put up in a fantastic five-star hotel and given lavish meals at the medical conference. (It was an academic conference, not an industry conference, by the way.)
I was so surprised that I told a senior medical colleague how great it was compared with philosophy conferences. His response was (I am withholding his name so I don’t slander him, he was a nice guy): “Jeremy, this is nothing. You should have seen in the 1990s before strict rules about taking money from pharma came into effect. They used to fly me and my family first class to the South of France, we’d have a meeting in the morning, go to the beach in the afternoon, and all we’d have to do is sign up to a consensus statement at the end.”
The consensus statement was then used as “proof” that the new treatment worked. Until the emergence of the evidence-based-medicine movement, consensus statements were the main way of deciding whether treatments worked. That is why Trish Greenhalgh called the consensus method the GOBSAT method, which stands for “Good Old Boys Sat Around a Table.” The conflict of interest and industry bias has not disappeared. However, at least now the experts must produce or refer to evidence to support their views.
Sir Iain Chalmers, who cofounded the Cochrane Collaboration (now known simply as Cochrane), a group that specializes in producing systematic reviews, tells another interesting story about why we can’t trust everything “experts” say, or trust stories about how things work. He was working at a refugee camp in the Gaza Strip and saw many children with measles. Measles is a viral disease, and antibiotics don’t work against those. He’d been told again and again by his expert teachers at medical school not to use antibiotics to treat viral infections such as measles. So he conserved his limited supply of antibiotics and did not give them to children with measles. The children were often malnourished or in poor health and had other complications. Sadly, some of them died a few days after Iain saw them.
Iain’s Palestinian colleague was seeing similar children with measles, but the children he treated seemed rarely to die. After a year or so, one of the Palestinian doctors gently pointed out to Iain that giving antibiotics to the children with measles helped, because these vulnerable children often developed bacterial infections in addition to measles. Iain changed his practice and started prescribing antibiotics to children with measles and he later found out that this was consistent with the evidence from controlled trials. This experience taught him to be very skeptical of what he was told by “experts,” and he began a lifelong crusade to insist on good evidence.
Iain’s experience is not isolated. Textbook recommendations (written by experts) for treatments intended for heart attack often don’t mention the latest advances. In some cases they keep recommending treatments long after they have proved to be harmful, or they fail to recommend treatments that have proved to be effective! If we add that many experts have a conflict of interest because they are paid by the industry, there is even less reason to trust experts.
Unfortunately, the problem with researchers who produce the evidence having financial conflicts of interest is a very common one. A recent study of the most influential experts within the American Psychiatric Association (APA) showed that many of its members had declared being paid by the pharmaceutical industry. Given that many of these experts are often involved in deciding what is an “official” psychiatric disorder and what is not, this raises questions about some of the more controversial psychiatric classifications.
Until 1973 the APA classified homosexuality as a disease, and in their most recent report published in 2013 they added to their list of “diseases” a bunch of things that most of us would consider normal differences between people. Besides the adult attention deficit disorder already mentioned in the introduction to this book, they have named restless leg syndrome (a desire to move one’s legs) and caffeine intoxication disorder (a “disease” you have if you get excited after having three or four cups of coffee). Most people I know get excited after drinking three straight cups of coffee, and many people I know sometimes move their legs.
Now the odd person may have such a severe case that they need serious treatment, but the vast majority are simply human. How did these normal things get classified as diseases? Did the conflicts of interest play a part? It is a fair question, given that most of the experts who have promoted these normal behaviors as diseases have declared ties to the same companies that make the medications used to treat the disorders.
Psychiatry is not the only industry with rampant conflicts of interest. As I described in the section on hidden bias (see Chapter 2), these financial conflicts of interest have been shown time and time again to influence results, often in ways that are difficult to detect.
Okay, I can hear you think, in Chapter 2 you said systematic reviews of randomized trials are great, but hidden biases and other stuff make them hard to trust. Now you are saying that people in studies are different from us and that we cannot trust theories or experts either. So what can we trust?
The answer to this question is that we have to keep asking questions, but it’s not all doom and gloom either. We are different from the average person in a trial because we all are unique, yet human bodies are also similar in a lot of ways, and many treatments work for most people. Even though stories and experts are problematic, they can still be useful in many cases—especially when supported by evidence. And while there are problems with systematic reviews of randomized trials, they are better than the alternatives.
If we remain skeptical and do a bit of work to find good evidence, we can often be quite confident that a treatment is better than a placebo. This brings us to the question, How powerful are placebos?
About once a week someone I know tells me about a new diet or exercise routine or food supplement they swear has changed their life. These treatments are often based on best-selling books or statements from beautiful experts who are producing lots of YouTube videos. If I am totally honest, I have tried some of these things myself. Most of these treatments are unlikely to hurt you, and a lot of them work because of the placebo effect. It is still important to check the evidence for two reasons:
So let’s go over a conversation between me and someone who recently told me about a diet where they were supposed to drink just pomegranate juice (and eat nothing else) for ten days. It is a true story. I will call the person who told me about this diet “Sabrina.”
Sabrina: Have you heard of this new fasting diet called the pomegranate diet? You just eat pomegranate and nothing else for three days. It is supposed to clean your system, boost your immune defenses, and make your skin shine.
Jeremy: I haven’t heard of it, but it sounds interesting. Do the people promoting it provide you with any evidence?
Sabrina: Yes, it has been tested on people at a research university. And they explain how it works. Basically, pomegranate is a superfood. It has a lot of vitamin C, which can boost your immune system. And fasting cleans out stuff that gets stuck. This all makes your body cleaner and healthier.
Jeremy: Well, there are three things you are asking about. First, is fasting good for health? Second, are pomegranates good for health? And third, is fasting on pomegranate good for health?
Sabrina: I suppose, yes.
Jeremy: Most cultures and religions have some kind of fasting and have been doing it for thousands of years. So I doubt very much that fasting in general is deadly (unless you do something really dumb like stop drinking anything for days, or stop eating altogether for months). But let’s check the evidence.
Sabrina: Okay.
Jeremy types “systematic review randomized trial fasting” into a search engine. He finds a systematic review supporting the benefits of fasting.
Jeremy: Obviously a complete investigation would take years. However, a quick search shows there are a few systematic reviews of what is called intermittent fasting, which means on-and-off fasting. It seems that there is likely to be a health benefit, but more evidence is required.
Sabrina: Good!
Jeremy: Well, hold on, just because fasting in general is good, it doesn’t mean that pomegranate fasting is good.
Sabrina: True.
Jeremy: The next thing I’m going to check is whether pomegranates are good for our health . . . Pauses to search . . . Again, this is a quick search so we can’t be a hundred percent sure. However, one systematic review shows that pomegranates can help lower blood pressure, and weaker evidence shows that it can slow the progress of prostate cancer. I didn’t find any evidence that pomegranates cause any damage.
Sabrina: See, I told you!
Jeremy: Well, hold on again, just because fasting is good and pomegranates are good, it doesn’t mean that fasting on pomegranates is good.
Sabrina: Okay. You academic types can be really annoying sometimes.
Jeremy: Pauses to search . . . I couldn’t find any randomized trials or systematic reviews of pomegranate fasting.
Sabrina: Hmm.
Jeremy: That doesn’t mean it is bad, it just means there is no evidence.
Sabrina: So what is the bottom line?
Jeremy: Well, based on what we know, it probably won’t harm you. And eating pomegranates as part of a regular diet seems good for you. However, to fast on pomegranates alone means you would be experimenting on yourself like a guinea pig.
Sabrina: What should I do?
Jeremy: Well, I can’t tell you what to do, I can just share two things that might help you make a decision. First, you can go with the intermittent fasting method that is recommended by the systematic review. Or, if you want to try something not supported by a randomized trial, at least choose something that has stood the test of time rather than something promoted by a small group of people who haven’t used it for long enough even to use their experience to measure the long-term effects. I can tell you what I do.
I fast from time to time, for example at the beginning of Lent, and as part of my yoga discipline. However, when I do this I am using methods that have existed for thousands of years, which means that while there are no randomized trials, there is a wealth of casual observational data and experience. I also surround myself with experts who have done the type of fasting I’m doing. So if I feel funny, or if something goes wrong, I can ask an expert.
Many Christians fast before Easter. Many Jews fast on Yom Kippur (and some on other days described in the Ta’anit), and many Muslims fast during Ramadan. Indian yogis have been doing various types of fasting for thousands of years. All of these traditions have guidelines regarding the best things to eat before, during, and after a fast. Why choose something new that might be great, but could also be harmful, and might not be as good as what’s been around for centuries where you can get proper guidance?
Sabrina: That makes sense. I might try a more traditional method of fasting, and add pomegranates into my regular diet. Thanks for your advice!
Jeremy: You are welcome! And thank you for asking the question, because I’ve learned something, too. I’ve learned that pomegranates probably have a health benefit, at least for lowering blood pressure. I happen to like how they taste, too, so I’m going to eat more of them.