42

It's not hard to find claims like these in the media or on the web from time to time.

Is there any scientifically verifiable evidence that prayer (in addition to appropriate medical/surgical treatment) has a benefit for a patient? Please keep answers to demonstrable evidence, such as:

  • Do prayed-for patients heal faster after surgery?

  • Do they have lower rates of post-operative infection?

  • Do they statistically demonstrate better outcomes?

Sklivvz
  • 78,578
  • 29
  • 321
  • 428
Monkey Tuesday
  • 24,456
  • 19
  • 109
  • 136
  • 2
    There's a fairly readable discussion of evidence on [Peter Norvig's site](http://www.norvig.com/prayer.html). He goes through a long list of papers, and finds that the ones that suggest prayer works had serious methodological flaws. (Arguably, this could have been due to looking harder for flaws in the papers with the conclusion Norvig disagreed with, but the flaws he found do seem major.) – David Thornley Apr 06 '11 at 02:21

2 Answers2

81

The American Heart Journal has published a three-year study of the therapeutic effects of intercessory prayer of 1800 patients undergoing heart bypass surgery.

The study divided subjects into three groups: those receiving no intercessory prayer, those that did, but didn’t know about it, and those that did and did know about it.

Its conclusion is very clear indeed:

Intercessory prayer itself had no effect on complication-free recovery from CABG, but certainty of receiving intercessory prayer was associated with a higher incidence of complications.

In other words: prayer doesn’t help. And if you know that you are prayed for, you do worse. Only slightly, but statistically significantly.

Scientific American has published a break-down of the study and its results.

Konrad Rudolph
  • 12,427
  • 3
  • 56
  • 84
  • That's for intercessory prayer. Praying for yourself has been correlated to increased healing, but no more than meditation or yoga. – Konerak Apr 04 '11 at 14:10
  • 4
    @Konerak Well the question was explicitly about “prayed-for patients”, i.e. intercessory prayer. – Konrad Rudolph Apr 04 '11 at 14:43
  • 11
    Excellent answer, I was about to cite the study myself.And before someone calls "BIAS!" on the study: It was funded by the (in)famous Templeton Foundation that doles out lots of money to scientists willing to say nice things about god. – Lagerbaer Apr 04 '11 at 14:56
  • Excellent answer. I found the published study on medscape. And even though it's only one study, it is sound in its methodology and conclusions. It seems likely that its results will be reproducible. In short, you make a very convincing argument, which will be difficult to counter. – Monkey Tuesday Apr 04 '11 at 21:50
  • @Konrad: I really can't find that in the question or the title. It says "can prayer help to heal" or "does prayer hold any benefit to a patient", but nothing about whp's praying. Am I reading it wrong? – Konerak Apr 04 '11 at 22:05
  • @Konerak The title is a *summary* of the question. The question I quoted in my above comment is in the question’s main body of text. – Konrad Rudolph Apr 05 '11 at 07:16
  • 2
    I see what you mean. Sorry, I'm not a native English speaker - from a logical point of view, a "prayed-for" patient does not include the prayer being the patient itself... after all, if you pray for yourself, are you not prayed-for? :) – Konerak Apr 05 '11 at 10:16
  • 1
    @Konerak Logically you have a point there. But I think that the usual usage dictates that a prayed-for person doesn’t pray for itself. – Konrad Rudolph Apr 05 '11 at 11:03
  • 1
    How did they prevent anonymous prayers? In many churches, there is anonymous praying going on "for the poor, for the sick, for the lonely...". Is there a proven shielding technique? :) – user unknown Apr 08 '11 at 14:37
  • @userunknown The study’s design doesn’t really suffer from anonymous prayer: that’s just background noise that can be eliminated statistically (it’s a systematic bias). – Konrad Rudolph Apr 09 '11 at 08:47
  • Doesn't 'background noise' imply, that the expected result of prayers is linear to the amount of prayer? If you have the assumption, that it is a digital question, of whether there is some prayer or not, I would follow that you can't exclude anybody, because you cannot shield anybody from the prayers. I fail to see how this can be 'background noise'. The problem is, that the idea of praying for healing is too vague, to be provable, and I guess it is that vague, because it isn't provable. The vagueness is a reflex to empirical observations. – user unknown Apr 10 '11 at 12:49
  • @userunknown “Doesn't 'background noise' imply, that the expected result of prayers is linear to the amount of prayer?” No. We expect some kind of connection but not necessarily a linear one. The underlying assumption is “more prayer => more chance of healing”. And that hypothesis was formulated by people who *wanted* it to be true. Of course you can always find a formulation that isn’t, in principle, empirically verifiable but unless you can show that this proposition has some merit, why should it be considered? If all people are prayed for and nothing more matters, specific prayer is useless – Konrad Rudolph Apr 10 '11 at 16:34
  • Well, you know, I don't use the arguments myself, but I knew a lot of religious people, and they all have different arguments, and combine them in an eclectic way. They are - maybe not a complete list - as follows: `a`: the person praying is a bad sinner. `b`: the prayed for person is a bad sinner `c`: it is a challenge (for one of them) to their faith `d`: Thou shalt not challenge god, the lord! (because he doesn't like science and prof, but he likes to challenge you, to prove you, which is a double moral standard, but hey, that's how god is). 'e': God is mystery, so nobody is able to ... – user unknown Apr 10 '11 at 17:05
  • ... (able to) understand HIM. Which is a contradiction to religion as such, which always claims to know something, but intellectual honesty has never been a religious virtue. For centuries, theologists tried to prove the existence of god (the whole idea of wonders is, to prove it), but nowadays, you shall believe from pure faith. The religious system is a system - most religious leaders avoid statements, which can be disproven, and when their knowledge about science increases, and they realize, they are disproven, they don' t stop their claims, but change them. So we may debunk some statements – user unknown Apr 10 '11 at 17:13
  • ... (statements), but run from excuse to excuse. If you debunk one argumentation, there is another excuse. Science can't disprove every combination of believes and excuses why praying for healing doesn't work. – user unknown Apr 10 '11 at 17:26
  • 1
    @userunknown “Science can't disprove every combination of believes and excuses why praying for healing doesn't work” – so what? Science *doesn’t have to* disprove every combination of anything. Proof, or indeed science, doesn’t work that way. If you claim something, you prove it. If you are unwilling to do so you are acting intellectually dishonest and will be ignored. This is known as [Russell’s teapot](http://en.wikipedia.org/wiki/Russell%27s_teapot). Anything that falls into this category is strictly off-topic on skeptics.SE, and for good reason. – Konrad Rudolph Apr 10 '11 at 17:44
  • So what? I say, in absence of a clear claim, when and how prayer heals, one shouldn't answer with a study which demonstrates, that `some prayers` don't work. – user unknown Apr 10 '11 at 17:57
  • 3
    @userunknown The study demonstrates that prayer, as implicitly defined by a lot of people (I’d claim that it’s the clear majority), and furthermore the only *meaningful* notion of prayer within the realm of empiricism, doesn’t work. Furthermore, as I’ve already said, I’ve answered the kind of question that’s on topic here. Any other kind of prayer is off-topic on skeptics.SE. I **still** don’t fully comprehend what exactly you don’t like about the answer. – Konrad Rudolph Apr 10 '11 at 18:09
  • No, I think you got it. No more questions on my side. – user unknown Apr 10 '11 at 19:46
  • Hmm. Interesting discussion going on here. However, @userunknown: if you agree a study can demonstrate "some prayers don't work", what empirical basis would there be to think that any amount of other prayers would? – Monkey Tuesday Apr 19 '11 at 08:15
  • You get me wrong if you think, that I believe prayers could be beneficial, or if you think I think they could be proven to be beneficial. I say, that the whole idea of prayers and religion is to fishy, to grasp it, you can only show the fishyness. People, who aren't healed are known by the churches for centuries and millenniums. They are well prepared for such arguments. – user unknown Apr 20 '11 at 19:14
  • Please move this discussion to chat, thanks :-) – Sklivvz Apr 20 '11 at 21:28
  • 1
    @Sklivvz FWIW I think the discussion *does* add to the answer since it shows the limitations of the research that can be done, and why this doesn’t matter. IMHO, a perfect example of good comment use. – Konrad Rudolph Apr 21 '11 at 08:35
  • 1
    `And if you know that you are prayed for, you do worse. Only slightly, but statistically significantly` - or could it be that patients with poorest prognosis are most likely to be prayed for? – Konrad Morawski Aug 29 '12 at 16:23
  • 3
    @KonradMorawski No, since the study was blinded. They praying people didn’t select who to pray for, and didn’t know their prognosis. – Konrad Rudolph Aug 29 '12 at 18:25
  • My main concern with this study is the same that was expressed as the first point in the letter to the editor named [A STEP in the right direction](http://www.sciencedirect.com/science/article/pii/S0002870306006351#) – mnicky May 08 '14 at 13:11
  • 2
    The point goes: `One flaw with the STEP trial and others like it has been the lack of a true placebo group (...) In the STEP trial, group 2 received no prayer (...), but it was reported that 96.8% of the 597 patients in this group “believed that friends, relatives, and/or members of the religious institution would be praying for them.”` – mnicky May 08 '14 at 13:12
  • The point by @mnicky goes well with what userunknown was originally saying. You cannot just dismiss unaccounted-for prayer as noise. If the study has 3 groups (1: no prayer, 2: don't know they get prayers, 3: know they get prayers), then much of group 1 could still be equivalently group 2 or 3 without the study knowing. Some of group 1 might believe they got prayers when they didn't really, or they could have prayers that they (and the study) didn't know. It is very common even for non-religious people to be getting lots of prayers without knowing because friend or family asks from church. – Aaron Nov 27 '18 at 22:17
0

I'm struggling to find links to actual scientific papers, but I am aware they exist and tend to show that if someone is prayed for and doesn't know it then there is no demonstrable increase in healing, but if someone is told that a group is praying for them they tend to heal quicker.

Mark Rogers
  • 11,324
  • 11
  • 55
  • 71
Mark Pim
  • 117
  • 3
  • 7
    " but if someone is told that a group is praying for them they tend to heal quicker." I remember seeing a study which concluded exactly the opposite: when the "target" did not know on the prayer, it had no effect, when he knew about it, the healing was slower and there were more complications. – Suma Apr 04 '11 at 08:42
  • 2
    Your argument as it stands is quite thin, wikipedia is not a great reference, and the godisimaginary link actually links to sites that better serve your argument (like the wired article). You'll have a lot more going for you if you can link to the specific study cited in the times online article, the one you mentioned (or anything published and peer-reviewed since Galton, and I know it's out there) or at least add where it can be found. – Monkey Tuesday Apr 04 '11 at 08:45
  • @Suma Exactly right. I’m searching for the studies … – Konrad Rudolph Apr 04 '11 at 09:30
  • 1
    The fact that the patient has to be informed someone is praying for them in order for it to work *strongly* suggests they're looking at placebo effect at best. I'd also theorize it depends heavily on the patient and their state of mind when they're informed of said prayer; not everyone would simply accept prayers in the spirit they're undoubtedly intended in... – Shadur May 26 '16 at 07:07