|Clarification: Influence of Prayer on IVF-ET|
|Authors:||Kwang Y. Cha, M.D.|
To the Editors:
Since the publication of our article "Does Prayer Influence the Success of in Vitro Fertilization-Embryo Transfer: Report of a Masked, Randomized Trial," by Cha and Wirth (2001;46:781-787), a substantial amount of discussion has been generated about it both through scientific correspondence and in the lay press. At the request of the editor-in-chief of the Journal of Reproductive Medicine(r), the major issues of concern surrounding the study are addressed below.
This blinded study did not require written informed consent. The study was approved by our local institutional review board with full knowledge that patients would not be signing informed consent forms. Not signing such a form is permitted in clinical trials. We thought that the requirement of written informed consent would introduce a bias or variable in the study by encouraging patients to pray for themselves or not, depending on their own religious persuasion. This might have interfered with the potential effects, if any, of intercessory prayer as an independent variable in our study.
It is regrettable that coauthor Daniel P. Wirth has been accused of fraud in an unrelated matter. This alleged crime was entirely unrelated to this study. The study was completely blinded, and it is impossible for Mr. Wirth to have influenced the outcome, whether or not he was motivated to do so. While some have accused Mr. Wirth of having participated in "questionable" studies involving faith healing, he was introduced to this author as someone who was knowledgeable about intercessory prayer and could help organize prayer groups for the study. Patients from Korea were randomized in the U.S. into 2 groups, 1 that received intercessory prayer and 1 that did not. This randomization was carried out without the knowledge of the clinicians in Korea. Further, Mr. Wirth had no knowledge of clinical proceedings or outcomes experienced by the patients in Korea during the conduct of the study. The patients who were prayed for were presented to this author for the first time when the code was broken, only after all clinical outcome data were available. Once the code was broken, an independently contracted statistician carried out the statistical analyses and adjusted for confounding variables. This deliberate design constraint made it impossible for Mr. Wirth to have played any role in manipulating or altering the data.
The prayer groups have been criticized because the design was perceived to be complicated. This complexity is considered by some to be a "flaw" in the study. The design was chosen by Mr. Wirth, who thought that in this area of inquiry, this was the best and most rigorous design to use. The study was not intended to assess the quantity or quality of the intervention, intercessory prayer. This is not a flaw in the design of the study, for the study was designed to answer the question (yes/no) as to whether intercessory prayer would have an effect on IVF-ET outcomes.
It has also been suggested that Mr. Wirth might not have organized the prayer groups and that patients might not have actually been prayed for. While this author did not have information about the composition and conduct of the prayer groups during the study (a design consideration to avoid potential investigator bias, or Hawthorne effect), there is no reason to think that Mr. Wirth would have been motivated not to organize prayer groups when such groups are his area of interest. At the outset of the study, patients were randomized into 2 groups; at the conclusion, analysis of the data revealed a significant difference in clinical outcomes.
A valid question that has been raised concerns the observed pregnancy rates. In looking at the average pregnancy rates at our center during the study period (32.8%), we would not have expected the rates in the prayer group to differ. However, it is well known that over a period of time, variations in pregnancy rates in IVF-ET programs may occur. It is only in a randomized, controlled trial that one can assess the effect of a treatment on outcome. Indeed, averaging the rates of the 2 groups in the study equates them with the overall pregnancy rates in our center during the study period.
It has also been suggested that the dropout rates and patient cancellations in the study are another flaw. In the conduct of virtually all randomized clinical trials, there is an anticipated dropout rate (usually 20-30%) that is unavoidable and is used as a basis of calculating the sample size and power of the study. In this study the dropouts were expected and distributed equally between the 2 groups.
The purpose of conducting a randomized, double-blind, prospective trial is to eliminate bias. Whether an investigator does or does not believe in the possible outcome of a study will not influence the results in a blinded, randomized trial, prospective.
The authors were very careful not to overinterpret the findings of the study. Our paper clearly states several times that the finding of enhanced pregnancy rates might not have been a direct result of intercessory prayer and would need to be tested in
future research. In undertaking to study a controversial area with a scientific approach, the best way to determine whether such interventions as intercessory prayer affect clinical outcomes is to design and conduct further prospective, randomized trials.
|Keywords:||fertilization in vitro, embryo transfer, prayer, complementary medicine, alternative medicine|
|Acrobat Reader 7.0 is recommended to properly view and print the article.
Reader can be downloaded from