Praying to heal from illness is possibly the oldest form of alternative medicine. When that prayer is made to a divine being on behalf of someone else鈥攍ike asking a god to heal an ailing relative鈥攊t is known as an intercessory prayer, from the word that means 鈥渢o intercede.鈥�
Prayers may seem to lay outside of the scope of science. If the being at the other end eludes the laws of our universe, how can science test the value of a prayer?
The classic counterargument is to say that, since the prayer asks for a measurable improvement in someone鈥檚 health, its effectiveness can be studied. We may not be able to explain how, scientifically, prayers work, but we should be able to detect that people prayed for survive an illness longer, don鈥檛 require as many medical interventions, and are cured much more frequently that those not prayed for.
This has been the justification behind a trickle of studies of intercessory prayers for health outcomes鈥攚e can鈥檛 test for a god but we can test for health improvements. Many of these studies show that intercessory prayers do not work, but some studies report a slight benefit. One even shows that prayers uttered now can travel back in time and improve the health of a person years into their past.
Are you scratching your head yet? Because studying how effective a prayer is at improving someone鈥檚 health is like trying to nail Jell-O to a wall.
Pray for pregnancy
There is no shortage of anecdotes on the power of prayer, and some have even been inducted into the medical literature. A woman聽聽from Parkinson鈥檚 disease鈥攁 diagnosis which had been backed up by a dopamine scan of her brain鈥攁fter being prayed for during a Christian conference. Her neurologist was stunned; the paper which relates this story notes, however, that nine years later, the disease relapsed. There鈥檚 also the historical case of聽, was blind for 12 years, then regained her vision in an instant when her husband prayed with her.
Short of a thorough investigation, it鈥檚 hard to parse these miraculous tales; it鈥檚 also worth wondering if prayer had not been involved, would these cases have received the attention they got. Spontaneous remission is surprising, but a spontaneous remission聽following prayer聽certainly has that extra mass appeal.
Can our sight really be improved by someone praying for us? A single case won鈥檛 be enough, but what about 11? A聽聽was conducted in rural Mozambique where a dozen patients reported having an impaired vision. They were given an eye chart test, then were hugged and prayed for by an evangelizing member of a Pentecostal ministry and took the test again. Using the same eye chart. Vision improved thanks to prayer... or maybe the patients simply habituated to the chart and were able to read an extra line the second time around. Without a control group, we can鈥檛 know.
Every study on praying is flawed in some way, which means you can cherry-pick the ones you like. Do you believe in the healing powers of Christ? You can cite聽聽reporting that six hours of in-person intercessory prayer and six months of distant prayer significantly improved the lives of people living with rheumatoid arthritis. Are you an atheist? Then you can cite a聽聽that came out the year after that shows that praying for people with cardiovascular disease after they leave the hospital makes no difference in their outcome.
But these experiments are small potatoes. What about the keystone studies, the ones that get cited over and over again? Get ready to parse through a long list of criticisms and allegations of fraud. There鈥檚 the infamous聽, linked to Columbia University, which stunned the world by showing that women undergoing聽in vitro聽fertilization were聽twice as likely聽to get pregnant if they were prayed for at a distance. Cue the Facebook ads telling you that fertility experts hate this one simple trick.
Unfortunately for anyone looking for an easy way out of a difficult conception, the study聽. Of its three authors, the last one, a department head, later claimed he was told of the study鈥檚 existence at least six months聽after聽it had been completed. And聽聽has a history of false identities, fraudulent charges, and an interest in proving that the paranormal exists. I don鈥檛 think would-be parents should put any stock in this study.
Then there鈥檚 the聽聽by Professor Leonard Leibovici where he showed that praying聽now聽can improve the health of people聽in the past. After all, we should not assume that time鈥檚 arrow only points in one direction or that a god would be unable to go back into the past. This study of 鈥渞etroactive prayer鈥� was, believe it or not, taken so seriously it was included in some meta-analyses of studies looking into the power of prayers. But the point Leibovici was making鈥攊n the Christmas edition of the聽British Medical Journal, no less, a special issue dedicated to聽real science of zany topics鈥攚as that you could do a study that looked rigorous but that tested something completely bonkers, and some people would take you at your word.
That strict adherence to methodology plagued the聽聽of prayer for ill health. Cochrane reviews are widely regarded as one of the best summaries of what works and what doesn鈥檛 in healthcare, but they are not flawless. An early version of this particular review聽聽for adding the Leibovici study of praying backwards in time to the pile, as if it had been a serious experiment. Its authors also threw prior plausibility out the window when they called for聽尘辞谤别听studies of prayer to figure out if it really worked. Ask any atheist if prayer works when you鈥檙e in the hospital and they will remind you that we have yet to see an amputee whose limb miraculously grew back through divine intervention. Now, that鈥檚 a study I鈥檇 like to see.
The rare studies of intercessory prayers with large effects have come under suspicion鈥攖oo good to be true鈥攚hile the ones with much smaller effects can be explained quite simply. It鈥檚 not that prayer works; it鈥檚 that when you are testing small groups of people and comparing them in multiple ways鈥攈ow long they stayed in the intensive care unit, how long they stayed in the hospital, how long they had this symptom or that one鈥攜ou鈥檙e bound to find differences by chance alone. This is the pitfall of聽补苍测听scientific study, and it should be kept top of mind, especially when extraordinary claims are tested in a group of only 11 people.
And the scientists behind these experiments should not escape our scrutiny, either. One of the first major trials of intercessory prayer (this one for patients in a coronary care unit) was performed by Dr. Randolph Byrd, who ended聽聽by thanking 鈥淕od for responding to the many prayers made on behalf of the patients.鈥� Because religion is such a hot-button issue, believers are highly incentivized to prove prayer works and atheists, to show that it doesn鈥檛. There鈥檚 also the funding issue. Many of these trials derive their money from the John Templeton Foundation鈥攁 philanthropic endeavour accused of using science to validate religious beliefs鈥攐r the Global Medical Research Institute鈥攁 nonprofit funding research into Christian Spiritual Healing. Somehow, I don鈥檛 think these benefactors are interested in studies showing that being prayed for does bupkis after a heart attack.
When we leave the experimental details behind, much more important questions come into focus and we realize how absurd it is to ask scientists to test prayers the way they test aspirin.
A moratorium on prayers for the control group
In the simplest form of a human trial, a control group does not receive the intervention. It is given something that聽濒辞辞办蝉听like it but that is inert.
Now imagine what a control group for intercessory prayer looks like. Can you prove that no one in that control group was prayed for? No, and that鈥檚 one of the inherent problems with these studies. People are prayed for by hospital chaplains, by their relatives, their friends, by their congregation, by well-meaning people all over the world, and no scientist can put a stop to that. It鈥檚 impossible to have a control group.
And it gets worse from here. The people praying in these studies are rarely given the patients鈥� full names because of medical data privacy. They may be given a surname, a pair of initials, a number code. That might explain why the prayer didn鈥檛 work, then: the deity called upon simply did not know whose help was requested. But isn鈥檛 the Christian god鈥攁nd almost all of these studies are fundamentally anchored in Christianity鈥攕aid to be all-knowing? And also all-powerful and all-good? Will this being let someone die in a hospital bed unless they are prayed for?
There鈥檚 also the problem that if the patient dies after receiving a prayer, it could be seen as their god answering said prayer and delivering them from their suffering. Therefore, heads, I win; tails, you lose. Whatever the outcome, prayer comes out on top.
Ethicists have also given this issue some thought, and while their questions聽, they鈥檙e not wrong to ask them. In some trials, the patients receiving a prayer are never informed of it: consent is waived. Why? Because the risk is seen as nonexistent. An ethicist would say, not so fast. There may be risks in attempting to communicate with an all-powerful divine entity and begging for a favour; since this entity can鈥檛 be known scientifically, how can you explain the risks of the procedure to a patient?
At the end of this long line of interrogations, I fear there鈥檚 only one solution left: science needs to butt out.
鈥淒o not put the Lord your God to the test as you did at Massah鈥�
One of the most illuminating articles I read on this topic was written by聽鈥擨 swear, this is not the start of a joke. They tried to design an experiment to test the efficacy of intercessory prayer on people with major depression and they became, well, depressed.
Prayer is not a drug, and yet if you are studying it in the context of a human trial, you must think of it like one. How much prayer would work? What kind of prayer, anyway? What are the magic words? How long should the prayer last? How many prayers in a week? How fervent should the people praying be? Is it important that they belong to a faith? If so, which faith? How many of them should be praying for a single patient? What if you pit Christian prayers against Jewish ones and the former shows efficacy but not the latter? Have you proven the existence of Jesus Christ? And what about that line in the Bible where the Judeo-Christian god proclaims that聽? Aren鈥檛 you testing a deity by relying on him for a clinical trial?
The Protestant, Catholic, and Jewish academics never ran their trial. Maybe they drowned their sorrows at the local bar.
It does not matter how many researchers defend their study by claiming that it didn鈥檛 actually test the existence of a god; prayer could work through some unknown physical mechanism, some sort of ectoplasm, perhaps, that moves invisibly from the intercessor and into the body of the sufferer. That argument does not pass the sniff test. Given that deities, if they exist, work in mysterious ways, as we keep being told, to constrain them to a scientific study is not going to work.
Moreover, believers will not stop praying if enough studies show that it doesn鈥檛 work. Popping an analgesic when you have a headache is based on science and on past uses, and if new evidence convincingly shows it doesn鈥檛 work, many will stop taking it鈥攂ut praying? That won鈥檛 be swayed by a Cochrane review.
We have better things to research with the limited funds at our disposal. Let鈥檚 pray future scientists stop testing fickle divinities to see if they might deliver us from the intensive care unit one day early.聽
Take-home message:
- Scientific studies of intercessory prayer for better health usually show no effect.
- Studying the efficacy of prayer arguably should not be done by scientists as it is impossible to prove that the control group was not prayed for by anyone, and it tests an alleged divinity that will not behave the way a medication does.
