Can prayer assist healing?

This page last updated February 17th, 2017

This page in brief

Believers the world over pray to their God, or gods, when they are sick, in the hope that God will heal them, or at least ease the pain. But is there any evidence that this prayer achieves anything?

Several dozen scientific medical studies have investigated the matter, without a definitive answer, but you may be surprised at the conclusions so far.

Faith and wellbeing

It is already established that religious faith is commonly associated with better recovery from sickness and higher levels of physical and mental health. This may be explained as the activity of God, or as the result of people with faith having a greater sense of peace, which assists their recovery, but the results are well established. But the possible impacts of prayer, sometimes offered at a distance and anonymously so there can be no ‘placebo’ effect, are much more contentious.

Two kinds of prayer studies

Two quite different kinds of prayer studies have been attempted.

Randomised controlled trials utilise the common scientific method of dividing the subjects (in this case patients recovering from illness, accident or operations) into two groups, one of which is prayed for and the other isn’t. Results are measured and compared statistically.

This methodology is easiest to test statistically, but faces the problem that it doesn’t measure how christians normally pray. Some studies utilise people to pray who aren’t normally involved in prayer for healing. So this approach doesn’t really measure christian prayer. It is also open to many other criticisms.

Within-subjects trials more directly and accurately address christian prayer by testing or obtaining medical documentation of sick or impaired people before they are prayed for by christians, and then again afterwards. Any improvement is an indication of possible healing.

While this method gets closer to the issue being investigated, its experimental design is less rigorous and it requires good statistical testing to draw a conclusion. Nevertheless it is a useful way to develop an understanding of an issue.

This page

This page addresses randomised controls. I have looked at within-subject studies and some individual examples in Healing miracles and God and Healing of sight and hearing.

Believers and unbelievers

There is a lot at stake here. Believers want to be assured that their prayers ‘work’, that God is really there, listening and responding – perhaps to assist their own faith, or perhaps to help convince unbelievers. Many unbelievers, on the other hand, see prayer as an unscientific idea and would find their unbelief under attack if the effectiveness of prayer was proven scientifically. So we can expect some bias in how some people interpret the results, some strong claims and counter claims. Fortunately a strong medical ethic of taking every possible opportunity to improve health has led to medical scientists of many different beliefs being willing to follow where the evidence leads.

Obviously I have my point of view too. But I have gathered together information from every reputable medical study I could find and summarised the results in Intercessary prayer and healing. In these notes, I have tried to summarise the evidence fairly.

A summary of the evidence

I have found 21 individual medical studies of prayer and healing, and 5 reviews of the evidence from available studies. (Note that some studies examine ‘distant therapies’, i.e. any attempt to influence the health of a patient from a distance, not using medical means. Prayer is by far the most common of these therapies, and most studies focus on prayer alone.) These studies yield inconsistent results, summarised as follows:

  • The studies examined a wide range of medical conditions (including coronary care, AIDS, mental health, alcoholism) and a wide range of methodologies. The reviews also used different methods to draw conclusions from the many separate studies considered. Varying results can perhaps be expected.
  • 14 of the studies found prayer had a statistically significant and positive effect on patient health and recovery, whereas 7 found no effect, or even a small negative effect. 3 of the reviews found the overall impact of prayer was positive, whereas 2 found no overall effect.
  • The largest study (in terms of the number of participants) found a positive effect, whereas the most expensive, and possibly most rigorous, study found no significant improvement overall, and people who knew they were being prayed for actually had worse results.
  • The difficulty of conducting such studies is illustrated by methodological problems found in reviews. Nevertheless, the largest review assessed the internal validity of 90 studies to be 75%, which is apparently quite good.
  • Commentators are divided on the value of these studies, though the most common view seems to be that they are worth pursuing further.

What can we learn from the evidence?

With about two thirds of the studies and reviews indicating that prayer assists healing, it is hard to deny that something is happening.

The sceptical view

A person holding to the view that prayer has no effect would expect all the studies to produce negative results. But the majority of studies and reviews say otherwise, so the sceptic has to argue that the data are in error, whether by bias or poor experimental design.

It is natural for most modern western people to be somewhat sceptical of claims that prayer might actually ‘work’. There are no known natural mechanisms for prayer to work through, and science only really works on natural processes. Many sceptics oppose the expenditure on these studies because they think they are unscientific.

Thus the sceptic, bound to accept the scientific method which includes acceptance of statistically significant results, faces a dilemma in this case, because statistically prayer “works” more often than it doesn’t. I have found many resolve the dilemma by focusing on the large Benson et al study which had negative results and ignoring or dismissing the rest.

The accepting view

A person holding to the view that prayer is helpful, whether via natural or supernatural causes, might expect more positive results than have been found so far. But perhaps some studies have met the (as yet unknown) requirements for a natural or supernatural benefit resulting from prayer, and some did not.

Christians have also (along with sceptics) expressed some misgivings about these studies, believing that God isn’t predictable and that the studies de-personalise prayer. Theologian NT Wright said: “Prayer is not a penny in the slot machine. You can’t just put in a coin and get out a chocolate bar.” Nevertheless many christians have participated in the studies.

Overall, christians will have mixed, but likely positive, feelings about these results.


The studies may be trying to measure something that is not predictable by science, and for which there is no clear methodology. Nevertheless it seems likely that medical treatment could benefit from such studies. All viewpoints might be disappointed by the results so far, but the sceptic seems to have the harder position to justify from the evidence.

Photo: MorgueFile.

Feedback on this page

Was this page helpful to you? little

Comment on this topic or leave a note on the Guest book to let me know you’ve visited.