Meta-Analysis Indicates Prayer Effective

[Solitaire]

Quote:

David R. Hodge, an Assistant Professor of Social Work in the College of Human Services at Arizona State University, conducted a comprehensive analysis of 17 major studies on the effects of intercessory prayer – or prayer that is offered for the benefit of another person – among people with psychological or medical problems. He found a positive effect.

Do Prayers For The Sick Make A Difference?

[Dancing David]

Quote:

Perhaps surprisingly, many social workers appear to use intercessory prayer in direct practice settings.

Unethical crap, I hate that!

Rant, foam, fall over backwards.

Prayer has no place in the clinical relationship.

If you want to be spiritual counselor then call yourself that.

Arrgh, aggle-aggle, booya!

[paiute]

So prayers are effective, eh? I will make my usual observation.

Now that we know God is listening and can be tested, we can find out which kind of prayer he (sorry, big guy, I mean He) listens to. The protocol uses prayers of the Baptist, Catholic, Muslim (both Sunni and Shia), etc. Every one of the hundred thousand sects will be sent up.

The sect with the most positive response will be the One True Faith. The rest will be slain as heretics.

Amen.

[strathmeyer]

Quote:

Do intercessory prayers (those said on behalf of another person and no, I'm not talking about having your friends quickly pray that the approaching police officer doesn't give you a ticket) have an effect on the recovery from illness above and beyond what medical treatment can provide?

Uh... yeah. It's important to put something at the top of you fraudulent articles in order to prevent intelligent people from reading it.

[casebro]

Oookaay, so including psychologocal illnesses means what? That prayer has a placebo effect on disorders of the mind ? Whereas the studies about physical diseases show no advantage. Who would have suspected that?

Reminiscent of the old saw about "How come there's all those crutches at Lourdes, but NO wooden legs?"

[Geckko]

Read your entire link. It goes on to cite a Harvard study, with this conclusion:

Quote:

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

Sounds like prayer is downright dangerous.

Any conventional medical treatment that should evidence of being associated with complications would be a candidate for banning.

[bpesta22]

odd, I keep trying to pull this article, but can't. I wonder if the editor yanked it...lol

Never heard of the journal, but browsing its TOC, it publishes a lot of stuff on spirituality.

I think we should read this article. If done right (well controlled, double blind and all that) any effect size greater than zero would present a significant problem for materialism.

Note the key words there are "if done right"...

[Ersby]

Hmmm, person wants to prove power of prayer with statistics.

Gets a result of p=0.015

But statistics would set the significance level at p=0.01, perhaps more, at p=0.001. Certainly not at p=0.05 (ie, 1 in 20)

Person doesn't care about that bit of statistics

Person declares significant result anyway

[bpesta22]

Originally Posted by Ersby

Hmmm, person wants to prove power of prayer with statistics.

Gets a result of p=0.15

But statistics would set the significance level at p=0.01, perhaps more, at p=0.001. Certainly not at p=0.05 (ie, 1 in 20)

Person doesn't care about that bit of statistics

Person declares significant result anyway

I got the article now, the p value is .015, not .15. So, the effect is significant.

[Ersby]

Sorry, that was a typo - my point still stands.

Thanks for the correction.

[EternalSceptic]

Originally Posted by Solitaire

Do Prayers For The Sick Make A Difference?

From the comments:

"I think it's very significant that Jesus was always more concerned with spiritual health than with physical healing, although he performed more than a few of those as well.

We will all die eventually. Prayer may not have any impact on the method or speed of that death, but I believe that the impact of prayer on spiritual healing is beyond measure. "

"is beyond measure" - how true. LMAO

[Dr Richard]

The problem with this paper is that it looks like a very bad meta-analysis.

The author is aware of some of the problems:

1. Publication bias - positive studies get published more often than negative ones. To control for this, authors of a meta analysis should make an active search for studies that were done but not published. The author admits this is a possibility but makes no such effort.
2. Presentation of the data - no funnel plot, no relative risk, no 95% confidence intervals, no explicit statements on population size/weighting
3. Heterogeneity of the sudies combined in the analysis: including the effect on pregnancy rates on in vitro fertilization, outcomes after surgery and, my favorite study of all: a RETROSPECTIVE randomized controlled trial on the efficacy of prayer.

I cannot believe it is valid to combine such widely divergent outcome measures.

But if you are bored, check out the link above, and then the rapid response comments following it. This is truly a mind-bogglingly stupid paper and I was amazed to see it included in this "meta analysis".

But the

[fls]

Originally Posted by Dr Richard

my favorite study of all: a RETROSPECTIVE randomized controlled trial on the efficacy of prayer.

I cannot believe it is valid to combine such widely divergent outcome measures.

But if you are bored, check out the link above, and then the rapid response comments following it. This is truly a mind-bogglingly stupid paper and I was amazed to see it included in this "meta analysis".

This appears to be a paper published in the Christmas BMJ issue. I think the BMJ, like several other medical journals, publishes silly, off-beat, satirical, etc. articles as a special present at that time of year.

I don't have access to the meta-analysis, but if that study was included, then the meta-analysis can be safely ignored as invalid.

ETA: Are you also saying that this possibly fraudulent study was included in the meta-analysis?

Linda

[TruthSeeker]

I'm pretty sure the retrospective RCT is a joke - they prayed for the patients up to 10 years after their illness!

[JoeTheJuggler]

I'm trying to think of what it would mean anyway if the effect is only proven in a meta-analysis--that prayer works a tiny amount that's only detectable over huge numbers?

So it's not like every hair is numbered, not a sparrow falls, whatso'er you ask it shall be granted kind of thing I guess.

I'm sure this is the argument the faithful are looking for.

[patnray]

Ah, the paradox. No study of christian prayers could possibly show a positive effect without contradicting the bible.

If a study shows no effect, christian apologists are quick to point out that the bible says god can not be tested. Therefore, the negative result validates their belief!

But if a study shows a positive effect, then the claim that god can not be tested is proven false. This inconvenient truth will be ignored, of course, and the positive result will validate their beliefs....

[Skeptic Ginger]

I can get access to this paper but I have to physically go to the campus library. Apparently we lowly alumni can't access journals online.

Short of that, could one of you with access list all of the 17 actual studies analyzed such as the one listed above, it would be worth looking for ourselves.

I'll be true to science and say, "show me the evidence", before calling BS on this. However, I have looked at all prayer studies that I have become aware of over the years and haven't found a single one with valid positive results. Regardless, claims the results are "significant" always seem to be made about the results.

Re the claim god cannot be tested for, that comes from both sides, faithers and sciencers alike but usually in different contexts. Sciencers talk about the esoteric god outside of the natural world despite the fact few if any religions describe such a god (that would be the god that never answers prayers and/or always covers its tracks), and faithers use the, "you can't test for god" to excuse the lack of positive findings in research such as intercessory prayer.

Considering the conclusions in this meta-analysis claim a "small but significant effect" one has to wonder why 'God' would be so miserly in granting favors. After all, are those Christians forgiven or not?

[patnray]

Originally Posted by skeptigirl

Re the claim god cannot be tested for, that comes from both sides, faithers and sciencers alike but usually in different contexts. Sciencers talk about the esoteric god outside of the natural world despite the fact few if any religions describe such a god (that would be the god that never answers prayers and/or always covers its tracks), and faithers use the, "you can't test for god" to excuse the lack of positive findings in research such as intercessory prayer.

It is not just an idle speculation among believers. The bible specificaly states that god can not and will not be tested. Further, the bible commands that believers are to believe on faith alone and are not to test god or ask for proof. Therefore, anyone who undertakes such a test must not be a true believer...

And science says any claims that god meddles in the real world are subject to test and verification.

[TruthSeeker]

I have the pdf and am happy to email it to anyone who sends an address via PM.

Yes, this is just a ploy to fill my PM Inbox. Popularity is important to me.

[Skeptic Ginger]

Originally Posted by Dr Richard

The problem with this paper is that it looks like a very bad meta-analysis.

The author is aware of some of the problems:

1. Publication bias - positive studies get published more often than negative ones. To control for this, authors of a meta analysis should make an active search for studies that were done but not published. The author admits this is a possibility but makes no such effort.
2. Presentation of the data - no funnel plot, no relative risk, no 95% confidence intervals, no explicit statements on population size/weighting
3. Heterogeneity of the sudies combined in the analysis: including the effect on pregnancy rates on in vitro fertilization, outcomes after surgery and, my favorite study of all: a RETROSPECTIVE randomized controlled trial on the efficacy of prayer.

I cannot believe it is valid to combine such widely divergent outcome measures.

But if you are bored, check out the link above, and then the rapid response comments following it. This is truly a mind-bogglingly stupid paper and I was amazed to see it included in this "meta analysis".

But the

Can someone explain to me what "retroactive intercessory prayer" is in the above study?

Quote:

All adult patients whose bloodstream infection was detected at a university hospital (Rabin Medical Center, Beilinson Campus) in Israel during 1990-6 were included in the study ...In July 2000 a random number generator (Proc Uniform, SAS, Cary, NC, USA) was used to randomise the patients into two groups..

No mechanism known today can account for the effects of remote, retroactive intercessory prayer said for a group of patients with a bloodstream infection. However, the significant results and the flawless design prove that an effect was achieved.

Let me get this straight, are these researchers actually claiming the prayer changed the past? Or are they claiming God influenced the random generator used to split the groups into controls and prayed for? What did they do, pray over the medical records?

Seriously. This is one bizarre paper. And the authors claim it's akin to knowing something works but not why it works. I'll be happy to explain to them why it worked. In this study they had a 50:50 chance of the prayer intervention being assigned to the half of the randomly generated group that randomly did better and they ran the test once.

We have two groups, they differ in outcome by x. Now we toss a coin and the group that did better randomly gets assigned to the prayer intervention group. Now we declare prayer made the difference.

For something like the above to be significant, it would have to be repeated many times. If it happened over and over, you'd need to explain why luck always fell with the prayed for group.

I'd say this is evidence the p# needs to be reconsidered as to where significance is claimed. Was the expected difference by chance between these 2 randomized groups properly determined?

[my_wan]

From the physorg.com link given on that article David R. Hodge is quoted to say;

Quote:

Overall, the meta-analysis indicates that prayer is effective. Is it effective enough to meet the standards of the American Psychological Association’s Division 12 for empirically validated interventions? No.

So is it effective overall... No.

[Skeptic Ginger]

Originally Posted by patnray

It is not just an idle speculation among believers. The bible specificaly states that god can not and will not be tested. Further, the bible commands that believers are to believe on faith alone and are not to test god or ask for proof. Therefore, anyone who undertakes such a test must not be a true believer...

And science says any claims that god meddles in the real world are subject to test and verification.

Can you cite the Biblical passages so I can interpret them directly?

You are talking about "specific tests" and I can't imagine that is the language used in the Bible.

[Skeptic Ginger]

Originally Posted by TruthSeeker

I'm pretty sure the retrospective RCT is a joke - they prayed for the patients up to 10 years after their illness!

You mean I wrote that whole post and it was something like an Onion story?

[Skeptic Ginger]

Originally Posted by my_wan

From the physorg.com link given on that article David R. Hodge is quoted to say;

So is it effective overall... No.

One of the comments from the physorg site from anonymous:

Quote:

This is totally idiotic. Imagine undertaking a study where I analyze 10 studies. I shall call me study a "meta" study:

2 studies show a slight positive effect.
8 studies show no effect.

I average them and guess what? A net positive effect. Why? Because there is no way to get a negative effect to balance out the positive (only a zero effect). This idiot has no idea what he is doing. My guess is the positive outcome tests show some sort of error (non-double blind, etc) which was meticulously avoided in the most recent study, which is why it is considered the final word. Averaging out 1000 well conducted studies which sum to zero with 3 retarded studies which sum to 8% will yield a net positive result. Way to throw in the buzz word "meta."

I think that goes to what Dr Richard was saying as well about the positive bias in published papers. If the effect detected is small, it is a stretch to be calling it significant because of the expected positive bias.

[Skeptic Ginger]

Well, just got Truthseeker's pdf. Thank you very much.

The above Onionesque study is indeed included. Can we say that might be a good reason to suspect the knowledge of the scientific process the Hodge study researchers possess?

I'll see what else is in here.

[Skeptic Ginger]

Five studies, "not significant, trend favors prayer group"; five with "not significant results".

Can you find the glaring bias here?

The five not noted to have the favorable trend - does that mean they had an unfavorable trend the author doesn't want to mention? Why mention trend at all if it isn't significant, it isn't significant.

[Skeptic Ginger]

Here's another study included I won't bother looking at,

Quote:

Quasi-experimental using pre and post tests; 26 primarily White, elderly, born-again females with arthritis; results significant [when prayed for] in person, not significant [when prayed for] distantly.

Let's review, born-again patients with arthritis report feeling better when prayed for but when they are prayed for and don't know it, they don't report feeling better. Right.

[Skeptic Ginger]

To recap, we are left with 5 studies claimed as prospective, double blind, randomized controlled studies.

There were 10 studies with no significant effect, 5 supposedly fell on the significant side of not significant and 5 didn't. It isn't worth looking into what the authors claim indicates a favorable trend but if someone else wants to, be my guest.

We have 1 studies that is absurd. Any legit researcher wouldn't have included that study. Claiming it is legit despite not knowing how it worked in that case is mind bogglingly stupid. Anyone with common sense can tell you why "it worked" and it wasn't because praying had anything to do with it.

I would conclude the 26 born-agains that felt better when they knew they were prayed for and not when they were unaware of the prayer shows pretty clear placebo effect. By all means, send those born-agains with arthritis to church. Why not?

So you end up with 11 studies against effect and 5 maybes, (I'll have to look), and at least one that shouldn't have been included. But so far, the meta-analysis is not looking too good.

I might also mention the only study with a large sample was the one which prayed for the dead people. (I assume some of them were indeed dead 4 years after they were in the hospital with sepsis.)

The total sample size left between the remaining 5 studies is 787.

[Skeptic Ginger]

Study #1 of 5, Byrd, 1988

Quote:

Positive therapeutic effects of intercessory prayer in a coronary care unit population; Author: Randolph C. Byrd, M.D.; Institution: Medical Service, San Francisco General Medical Center, CA.; Source: Southern Medical Journal 1988 Jul; 81(7): 826-9

Abstract: The therapeutic effects of intercessory prayer (IP) to the Judeo-Christian God, one of the oldest forms of therapy, has had little attention in the medical literature. To evaluate the effects of IP in a coronary care unit (CCU) population, a prospective randomized double-blind protocol was followed. Over ten months, 393 patients admitted to the CCU were randomized, after signing informed consent, to an intercessory prayer group (192 patients) or to a control group (201 patients). While hospitalized, the first group received IP by participating Christians praying outside the hospital; the control group did not. At entry, chi-square and stepwise logistic analysis revealed no statistical difference between the groups. After entry, all patients had follow-up for the remainder of the admission. The IP group subsequently had a significantly lower severity score based on the hospital course after entry (P less than .01). Multivariant analysis separated the groups on the basis of the outcome variables (P less than .0001). The control patients required ventilatory assistance, antibiotics, and diuretics more frequently than patients in the IP group. These data suggest that intercessory prayer to the Judeo-Christian God has a beneficial therapeutic effect in patients admitted to a CCU.

The critique:

Quote:

Critical readers of reports relating prayer to illness recovery have reason to question reported outcomes. In 1988, the Southern Medical Journal published Randolph Byrd, M.D.'s study on intercessory prayer (prayer by outsiders, at a distance, as opposed to personal prayer) in treating cardiac care unit patients.1 Among the prayer group, as compared to the controls, the article reported lesser incidence (5% to 7% reductions) in a number of complications, such as congestive heart failure, pneumonia, need for diuretic therapy, and respiratory intubation and ventilation. But Byrd found no significant difference in variables such as death rate, length of CCU stay, and total days hospitalized. Byrd also included a summary of several previous attempts to demonstrate the healing power of prayer, with results ranging from mixed to negative.

Critical examination of this study revealed flaws that appear to render the effort no more useful than those that preceded it. For example, the complications for which he reported significant reductions were not independent variables -- the development of one (e.g., congestive heart failure) may automatically have lead to a cascade involving numerous others such as diuretic therapy, respiratory intubation/ventilation, and pneumonia).2

Larry Dossey, M.D., a retired internist affiliated with the NIH Office of Alternative Medicine, has authored several books about the power of prayer in the practice of medicine. The Introduction to Healing Words 3 cites Byrd's study as the seminal event that led Dossey on his own quest to document prayer's healing powers. Yet later in the book (p. 185), he cites my critique4 of Byrd's study, and acknowledges that the study had actually "missed the mark . . . [S]tatistically significant life and death effects . . . simply did not occur."

Re the not independent variables, that would be like counting the parts of your body and claiming you had more people because the number was higher.

[Skeptic Ginger]

And, Csicop has weighed in on the Harris study (#2 of the 5 remaining studies with significant results).
Also, here.

If you leave those 2 studies out, now you have 129 subjects left in the not so 'meta'-analysis.

[Skeptic Ginger]

Skepdic weighs in on #3 of the remaining 5.

This is sure saving time.

Quote:

The 1998 Sicher-Targ study appears to be an extremely well-designed double-blind, controlled study. Even so, it is very unclear exactly what causal processes were to be measured. Worse, the researchers—at Targ’s urging—changed the goal of the study and had a statistician mine the data after it had been completed. As noted above, the original goal was to see if DH could lower the death rate for AIDS patients. Since only one of the patients in the study died, there was nothing to report on that count. The published study, however, claimed that the aim was to measure DH against a long list of AIDS-related symptoms. A biased researcher (Sicher) went through all the data after the study was completed to determine which patients had which symptoms. It was a bit of a fluke that Sicher and Targ’s deception was exposed by Po Bronson in Wired magazine four years after the study was published. By that time, Targ had bankrolled it into another big grant from NIH.

Both the Csicop and the Skepdic articles address more than one of these 5 remaining studies with "significant results" in the Hodge paper.

Csicop on the Sicher, Targ study

[Skeptic Ginger]

No easy access to #4, Furlow and O'Quinn, published in the Spring 2002 issue of The Jnl of Christian Nursing. One needs a subscription. The abstract gives one a hint at how unscientific this journal is and the study involved 21 Christian cardiac patients.

Quote:

Research: Does Prayer Really Help? by Leslie Furlow and Josie Lu O'Quinn
Building on previous studies on the influence of intercessory prayer on the recovery of patients hospitalized with a cardiac diagnosis, Furlow and O'Quinn based their research on the question: "Does intercessory prayer to the Judeo-Christian God have an effect on the patient's medical condition and recovery while in the hospital?" Thirty-nine patients were in the study; twenty-two were in the experimental group and seventeen in the control group. Findings were compared with previous studies; the conclusions were significant, although further research with a larger group and rigorous controls needs to be done.

Somehow I don't think this one is going to prove the meta-analysis conclusion case.

[Skeptic Ginger]

Last one in the meta-analysis. Maybe praying for the sick is the wrong thing. God might just not like changing his mind after making people fall ill.

Perhaps he/she responds better to requests for babies. I have posted the abstract and a few snips, but the whole article is at the link.


Does Prayer Influence the Success of in Vitro Fertilization–Embryo Transfer?
Report of a Masked, Randomized Trial; Kwang Y. Cha, M.D., Daniel P. Wirth, J.D., M.S., and Rogerio A. Lobo, M.D.

Quote:

OBJECTIVE: To assess the potential effect of intercessory prayer (IP) on pregnancy rates in women being treated with in vitro fertilization–embryo transfer (IVF-ET).
STUDY DESIGN: Prospective, double-blind, randomized clinical trial in which patients and providers were not informed about the intervention. Statisticians and investigators were masked until all the data had been collected and clinical outcomes were known. The setting was an IVF-ET program at Cha Hospital, Seoul, Korea. IP was carried out by prayer groups in the United States, Canada and Australia. The investigators were at a tertiary medical center in the United States. The patients were 219 women aged 26–46 years who were consecutively treated with IVF-ET over a four-month period. Randomization was performed after stratification of variables in two groups: distant IP vs. no IP. The clinical pregnancy rates in the two groups were the main outcome measure.
RESULTS: After clinical pregnancies were known, the data were unmasked to assess the effects of IP after assessment of multiple comparisons in a log-linear model. The IP group had a higher pregnancy rate as compared to the no-IP rate (50% vs. 26%, P=.0013). The IP group showed a higher implantation rate (16.3% vs. 8%, P=.0005). Observed effects were independent of clinical or laboratory providers and clinical variables.
CONCLUSION: A statistically significant difference was observed for the effect of IP on the outcome of IVF-ET, though the data should be interpreted as preliminary. (J Reprod Med 2001;46:781–787)

...There was a consistent statistically higher pregnancy rate for IP in the 30–39-year group and in the >39-year group but not in the <30-year group. IP vs. NIP in the 30–39-year group was 51% (29/57) vs. 23% (14/62) (P=.0013), 42% (5/12) vs. 23% (3/13) in the >39-year group and 53% (10/19) vs. 67% (4/6) in the <30-year group. We could not identify a difference in pregnancy rates in women <30, in whom the pregnancy rates were extremely high....

But it seems God might have robbed Peter to pay Paul.

Quote:

The overall pregnancy rate for IVF-ET during the study (December 1998–March 1999) was 38.5% when all pregnancies (both groups) were taken into account. This rate was similar to the historical rate for the center’s program; the rate during the preceding months, January–November 1998, was 32.8%.

That leaves one to question why the resulting babies were unevenly distributed between the prayed for and not prayed for groups, while statistically the praying didn't actually change the number of resulting pregnancies for the group as a whole.

To be fair, this is the only study I found with valid results. But just as many studies which have gone before, there is a strong suggestion these results will not be consistently repeatable. Unless one hypothesizes a god doling out a fixed number of babies favoring those who are prayed for, then the fact there was on average the same fertility rate once a larger sample was compared to a control sample, even given only half the larger sample was prayed for, it still suggests random variation in the prayed for group has not been ruled out.

[Skeptic Ginger]

So much for the meta-analysis.

The usual wishful view through the tunnel.

[Kopji]

And how come all these wacky professors need to be from Arizona? Isn't it embarrassing enough to live here? Have they no pity that my tax dollars help support this kind of research? Surely there is a small school somewhere in Alaska or Vermont in need of instructors.

[Jon.]

Originally Posted by skeptigirl

Last one in the meta-analysis. Maybe praying for the sick is the wrong thing. God might just not like changing his mind after making people fall ill.

Perhaps he/she responds better to requests for babies. I have posted the abstract and a few snips, but the whole article is at the link.


Does Prayer Influence the Success of in Vitro Fertilization–Embryo Transfer?
Report of a Masked, Randomized Trial; Kwang Y. Cha, M.D., Daniel P. Wirth, J.D., M.S., and Rogerio A. Lobo, M.D.But it seems God might have robbed Peter to pay Paul.That leaves one to question why the resulting babies were unevenly distributed between the prayed for and not prayed for groups, while statistically the praying didn't actually change the number of resulting pregnancies for the group as a whole.

To be fair, this is the only study I found with valid results. But just as many studies which have gone before, there is a strong suggestion these results will not be consistently repeatable. Unless one hypothesizes a god doling out a fixed number of babies favoring those who are prayed for, then the fact there was on average the same fertility rate once a larger sample was compared to a control sample, even given only half the larger sample was prayed for, it still suggests random variation in the prayed for group has not been ruled out.

I believe this one has been fairly thoroughly discredited, although not withdrawn by the journal in which it was published. I think Randi has addressed it a few times in his commentary.

[Skeptic Ginger]

Originally Posted by Kopji

And how come all these wacky professors need to be from Arizona? Isn't it embarrassing enough to live here? Have they no pity that my tax dollars help support this kind of research? Surely there is a small school somewhere in Alaska or Vermont in need of instructors.

Gosh, speaking of tax dollars, that one researcher who data-mined for positive results after the study failed to find any, and didn't bother to mention that in the study, got 1.5 million grant from the NIH to do further research.

It pays to deceive.

[Skeptic Ginger]

Originally Posted by Jon.

I believe this one has been fairly thoroughly discredited, although not withdrawn by the journal in which it was published. I think Randi has addressed it a few times in his commentary.

If you find the links, let me know. I ran out of steam at the end of my rant. I can't tell you how many prayer studies I have looked at after seeing the claims. None of them hold muster, none! So doing a meta-analysis when the critiques of these studies were easily found shows you how deluded believers are. We have another thread going about being about being Christian and a skeptic. This thread on prayer studies supports the conclusion people claiming both have a skeptic blind spot.

[my_wan]

Originally Posted by skeptigirl

One of the comments from the physorg site from anonymous:

Quote:

This is totally idiotic. Imagine undertaking a study where I analyze 10 studies. I shall call me study a "meta" study:

2 studies show a slight positive effect.
8 studies show no effect.

I average them and guess what? A net positive effect. Why? Because there is no way to get a negative effect to balance out the positive (only a zero effect). This idiot has no idea what he is doing. My guess is the positive outcome tests show some sort of error (non-double blind, etc) which was meticulously avoided in the most recent study, which is why it is considered the final word. Averaging out 1000 well conducted studies which sum to zero with 3 retarded studies which sum to 8% will yield a net positive result. Way to throw in the buzz word "meta."

I think that goes to what Dr Richard was saying as well about the positive bias in published papers. If the effect detected is small, it is a stretch to be calling it significant because of the expected positive bias.

Skeptigirl trumped me here . I should have read the comments

[Skeptic Ginger]

Hey, we're all in this together. I might not have even thought of positive bias if it weren't for what everyone else wrote.