Mantra Study - BBC TV

[Jaggy Bunnet]

There was a program shown last night in the BBC Everyman series (general religious / current affairs) on the Mantra study into the effect of prayer on patient outcomes.

Basically, they followed 750 patients, half of whom were randomised to receive prayer from a variety of different religious groups around the world and followed to see if it had any effect on their outcome.

The surprising result.... no statistically significant effect. The program ended with the researchers flicking through the statistical data looking for blips. The main medic involved got excited about the impact of tier 2 prayer (a prayer group praying for the prayer group praying for the patient) as the results for that were "up near statistical significance".

Did anyone else see it? Any views on the study?

http://www.bbc.co.uk/religion/programmes/everyman/index.shtml

 

[hammegk]

I saw some of it.

My first thought was 'how in the world are they measuring "effects"'? Lots of imo subjective patient questionnaires seemed to be much of it.

The chief medic -- rational throughout the proceedings -- ended with the comment (paraphrased) "If any major drug company saw the 'near-statistical-significance' result in any clinical trial, they would be pouring big bucks & efforts into further research to pin down the effect".

 

[Jaggy Bunnet]

I saw some of it.

My first thought was 'how in the world are they measuring "effects"'? Lots of imo subjective patient questionnaires seemed to be much of it.

The chief medic -- rational throughout the proceedings -- ended with the comment (paraphrased) "If any major drug company saw the 'near-statistical-significance' result in any clinical trial, they would be pouring big bucks & efforts into further research to pin down the effect".

There was unfortunately little info on how they were measuring it. I did have two favourite bits though:

Firstly, the amazing discovery that playing people relaxing music in a quiet environment made them less stressed! This seemed to be in relation to a test of the effectiveness of "music/image/touch therapy" but it wasn't covered in any depth.

Secondly, the various groups explanations for why the study was not evidence that prayer didn't work. I'm sure they would have made similar arguments if it had produced results more to their liking!

What I found deeply sad was the attitude of certain of the groups (particularly the Christian ones) when asked about the effect of the prayers by followers of other religions. Seems a Christian God will not intercede on behalf of a Christian patient if asked by a non-Christian as these people are "wrong" and their prayers will be "ignored".

 

[Lothian]

A few things struck me.

Two thirds of the way through the trail they struggled to get volunteers because of 9/11. No-one wanted Muslims and Bhuddists praying for them. They therefore decided to test the touch/music therapy. And had a second tier of prayers.

The problem with previous studies was that the sample size was too small. 750 was the minimum. These late changes would therefore only have a max of 250 in this study so the results can not be taken that seriously.

Two conclusions.
1. Overall prayer has no effect
2. However if the people praying are themselves prayed for then there is a significant positive effect.

These two taken together suggest that when prayers are made (with no secondary praying) the effect is detrimental. I presume this as the latter addition of (significantly improved effect) secondary praying brought the overall effect up to no difference between those prayed for and those not.

I was bothered by the independence of the testers and potential flaws in the testing process.

Many prayer groups were in the immediate vicinity of the hospital. Those groups knew the names of those being prayed for. It is therefore possible for the researched to find out who is in the prayed for group.

The touch / music therapy experiment was did not appear to be blind never mind double blind. Both victim and tester were aware who was in which group. (I use the term victim as although these people felt better about the operation the exposure to music and healing touch killed them quicker )

 

[Jaggy Bunnet]

Two conclusions.
1. Overall prayer has no effect
2. However if the people praying are themselves prayed for then there is a significant positive effect.

These two taken together suggest that when prayers are made (with no secondary praying) the effect is detrimental. I presume this as the latter addition of (significantly improved effect) secondary praying brought the overall effect up to no difference between those prayed for and those not.

Point 2 I think is wrong. I think this is what they were talking about when they said it was "getting up there towards statistically significant". Effectively claiming a positive outcome even though it was not statistically significant - great research!!

I had the same thought about the effect of unreinforced prayer - but then I thought it might be possible for the overall to be within the bounds of statistical significance even if part of it (secondary praying) was significant if looked at on its own.

Then I remembered that I know very little about statistics and was probably wrong, and anyway the idea that praying for people makes them worse suggests a really sick God, so I accepted it.

 

[Lothian]

Then I remembered that I know very little about statistics and was probably wrong, and anyway the idea that praying for people makes them worse suggests a really sick God, so I accepted it.

Yes I realised that the smaller sample size in the latter would not mean that the effect was opposite on the single group but.....
I just loved the idea of someone being ill. Someone offering to pray for them. And the recipient pleading with them not to until they can find someone to pray for them praying.

I wondered if it like multiplying by a negative

If someone prays for you it has a small negative effect.(-1.1) However if someone also prays for those praying then you have a positive effect (-1.1 x -1.1 = +1.21).
But if someone prays for the people praying for the people praying you go back to a larger negative effect and so on...

 

[Jaggy Bunnet]

Yes I realised that the smaller sample size in the latter would not mean that the effect was opposite on the single group but.....
I just loved the idea of someone being ill. Someone offering to pray for them. And the recipient pleading with them not to until they can find someone to pray for them praying.

I wondered if it like multiplying by a negative

If someone prays for you it has a small negative effect.(-1.1) However if someone also prays for those praying then you have a positive effect (-1.1 x -1.1 = +1.21).
But if someone prays for the people praying for the people praying you go back to a larger negative effect and so on...

So if you don't feel like you are getting better you should pray for anybody praying for you in case that is what is stopping the recovery?

Or do you need to know who is praying for you and specifically reinforce them rather than just a general reinforcement?

But if nobody is praying for you, would praying for them make you worse?

This religion lark is pretty complex.