Abstract
In recent times there has been increasing evidence for the healing effects of prayer. Double-blind studies have been conducted in various medical settings confirming the positive effects of prayer in cardiac patients and those suffering from AIDS.[1],[2] There is also evidence that alternative treatments such as prayer, meditation and herbal remedies can be of benefit for psychiatric conditions.
In 1998 Dr Elisabeth Targ and her colleagues at California Pacific Medical Center in San Francisco, conducted a controlled, double-blind study of the effects of ‘distant healing’, or prayer, on patients with advanced AIDS.[2] In this study, those patients receiving prayer survived in greater numbers, got sick less often and recovered faster than those not receiving prayer. Prayer in this study looked like a medical breakthrough.
In 1988 Dr Randolph Byrd conducted a study at San Francisco General Hospital involving patients with heart attack or severe chest pain.[1] He found that patients receiving prayer did much better clinically than those who did not.
Dr Mitchell Krucoff, at Duke University Medical Center in Durham, North Carolina, studied the effects of prayer on patients undergoing cardiac procedures such as catheterization and angioplasty.[3] Patients receiving prayer had up to 100% fewer side-effects from these procedures than people not prayed for.
In addition, there are also impressive triple-blind studies, meaning that it is not known, to the researchers or the patients, who is receiving prayer and who is not.[1],[4] The people doing the praying are also blind to the identity and condition of the patient. This eliminates, or at least reduces, the placebo effect, which is the power of suggestion or positive thinking.
The studies that are most impressive are not those done on humans.[5] For example, when bacteria are prayed for, they tend to grow faster; when seeds are prayed for, they tend to germinate quicker; when wounded mice are prayed for, they tend to heal faster. These studies can be done with great precision, and they eliminate all effects of suggestion and positive thinking because we can be sure the effects are not due to the placebo effect. Mice, seeds, and microbes presumably do not think positively.
A survey on the East Coast of America found that 75% of hospitalized patients believed that their doctor should be concerned about their spiritual welfare, and 50% wanted their physician to pray not only for them, but with them.[6] In response to the evidence showing a positive role of prayer and religious practices in health, nearly 50 medical schools in America currently offer courses in this area.
In a survey of parents of child psychiatric patients, we found that more than half had interest about spiritual issues and a similar proportion wanted their treating clinician to be aware of this in the assessment and treatment of their children.[7]
Prayer and positive intentions are non-local, that is, they work at a distance.[8] Evidence suggests that they are as effective from the other side of the earth as from the bedside. This means that lay groups, off site, can successfully employ these methods.
A Cochrane review on intercessory prayer for the alleviation of ill health[9] concluded that there are very few completed trials of the value of intercessory prayer.[6],[10] The evidence presented so far is interesting enough to support further study. A Cochrane review on distant healing found that 57% of studies showed a positive effect, which merited further research.[11]
The aim of the present study was to investigate whether intercessory prayer had an effect on the outcomes of a group of children with psychiatric conditions.
METHOD
Forty consecutive patients from the list of children attending a child and adolescent mental health service (CAMHS) were selected for the study. Half of the patients (n = 20) were randomly assigned to the intervention prayer group and the other half (n = 20) to the control group. A committed group of six people selected by the chief investigator prayed once a week for the patients in the intervention group; they were blind to the identity and the condition of the child, who was denoted by a numerical code. Because this was a pilot study, none of the patients or their families was aware of the study, that is, the clinician and patient and child's family were blind to the fact that prayer was taking place. This cancelled any positive effects, relating to positive expectations, that may have confounded the findings. Ethical issues were considered; the study was approved as an audit project because negative effects were considered negligible and no one associated with the patients knew that prayer was happening. The outcome evaluation was the same as for normal clinical practice. Both groups were given the outcome questionnaires at 3 months after first entry into the service, whether they were still in treatment or otherwise. Two standard routine outcome measures were used to measure change.
The Strengths and Difficulties Questionnaire (SDQ), a brief behavioural questionnaire completed by parents of 4–16-year-olds, was used as a preand post-intervention measure.[12] This instrument is routinely used by CAMHS, and has been used in a previous study by our research group.[13],[14] The second instrument was the Health of the Nation Outcome Scales for Children and Adolescents (HoNOSCA),[15] another outcome measure used by CAMHS.
RESULTS
Forty patients were entered into the study but four had to be excluded because of incomplete data. Thirty-six patients were recruited; three-quarters (n = 27) of this final sample was male. The mean age was 9 years (range 4–14 years).
Of the 36 participants, 17 were in the intervention prayer group and 19 in the control group. At 3 month follow up, 33 of the children had HoNOSCAs completed by their clinicians (92%) and 22 parent-completed SDQs were returned (61%).
Examining the sample overall, there was no difference in SDQ scores from time 1 to time 2. There was, however, a significant reduction in HoNOSCA scores across time (F =36.546, p < 0.001). These means are depicted in Table 1.
Means at times 1 and 2 for HoNOSCA and SDQ
Effect of the intervention
The prayer and intervention groups were compared at time 1 to ensure that there was no bias in the assignment to the groups. The prayer and control groups did not differ statistically at time 1 on their mean SDQ total difficulties scores, indicating that the participants did not differ significantly in their SDQ scores upon allocation to the two groups. To examine the differences between groups, a repeated measures ANOVA was conducted with ‘time’ as the within-subjects factor. The means are presented in Table 2.
Comparison of the means
To examine the difference between the means in the repeated measures ANOVA, only the data of the participants who had both time 1 and time 2 questionnaires were used (nine participants in the prayer group and 13 participants in the control group). That is, the children who did not have time 2 questionnaires were excluded. This difference between the control and intervention groups on their mean SDQ scores was not statistically different. The difference at follow up between the control and intervention groups on the clinician-rated mean HoNOSCA scores was also not statistically different.
DISCUSSION
The results were disappointing in light of other studies that showed a more positive outcome with prayer. However, there are a few aspects that need to be considered. The follow-up data at 3 months was based on only half the initial sample. Because the study was blinded to the patients, it was not ethical to keep reminding parents to return the follow-up questionnaires. Nearly 50% of patients were discharged or failed to return at 3 months. This also made it difficult for us to do the follow up. Of the follow-up questionnaires that were sent, only two were returned. The data that were obtained were therefore not representative of the sample that was studied. However, it is interesting to note that in this sample the SDQ did not show a significant change at 3 months. The HoNOSCA did show a significant change but there was no difference between the control and intervention groups.
This may be due to the perception by clinicians that such a study was being undertaken, although they were blind to who was in the experimental group. The parents were not aware of this. This raises questions about double-blind studies where patients are aware that they are in a trial and the possible effect of this on their perception of change. Triple-blind studies are difficult to undertake in other trials for ethical reasons; we were able to do this in the present study because the treatment was non-interventionist.
Three months may be too short a time to elicit any changes in this group, based on SDQ responses. A longer study with a larger sample may show a more definite change as revealed at 6 month follow up of patients receiving ‘treatment as usual’ in our previous study. Knowledge that they were being prayed for may also make a difference in outcomes, although this could also be a placebo effect.
In summary, the present study was unable to show any additional benefits to patients who received intercessory prayer compared to those who received treatment as usual.
Footnotes
Acknowledgements
We thank clinicians of the Community Child and Adolescent Mental Health Team, Wyndham, for their support and cooperation, and prayer partners at Waverley Christian Fellowship, Melbourne.