Biofield healers have worked with virtually every organ system of the body, and with virtually every pathology. Outcome studies, however, are important. In the literature, one can mainly find research results for biofield healing of pain, cancer, anxiety, cardio-pulmonary issues, enhancement of the immune system, dementia, and laboratory value changes (e.g., blood count).
When I first decided to undertake this inquiry, I assumed that there were very few outcome studies of healing in the human biofield, so that it would be easy to detail them. But as I did a literature search, I was surprised at the amount of data actually in print. Though I was quite selective about what I gathered, I wound up with a very large box containing stacks of studies.
Up-front problems adhere to many studies in the field. Study populations are small, target systems or tissues are diverse, techniques vary widely, practitioner training and expertise vary and are often not reported, frequency and length of healing sessions vary, and study data are often incomplete.
To simplify things, I am going to refer primarily at this point to a 2009 study by Drs. Shamini Jain and Paul J. Mills writing in the International Journal of Behavioral Medicine: “Biofield Therapies: Helpful or Full of Hype? A Best Evidence Synthesis.” Drs. Jain and Mills looked closely at 66 clinical studies to try to determine their quality and then assessed outcomes using a set of criteria that I believe stand up well. Patient populations varied widely, and a number of different kinds of therapies were represented. The descriptive criteria they assessed enlarge upon the issues listed here.
Here are their criteria. The studies had to be in English in peer-reviewed journals, healing had to be “local” (but included a variety of techniques), session-length had to be indicated (which varied from 3 to 90 minutes, with an average of 23 minutes and a median of 17.5 minutes), the number of treatment sessions had to be indicated (varying from 1-15, with an average of 4 and a median of 3), and some kind of quantitative outcome measures had to be established. 85% of the outcome measures were self-report or psychological, 54% were “objective” functional or biological, and 9% were qualitative. 54% of the studies had two or more endpoints (generally biological plus psychological).
Drs. Jain and Mills excluded studies in which patients were in groups but not randomized to groups, studies with mixed interventions so that the modalities could not be differentiated (e.g., a group receiving hands-on-healing and also practicing meditation), and descriptive studies where qualitative analysis was lacking. Their exclusion criteria led to 22 studies being eliminated from the original group of 88, for a final tally of 66 studies they reviewed carefully.
They looked at approximately 27 different descriptive criteria as part of their assessment, including information about age, gender, number of participants, biofield modality, session duration, session frequency and total number, number of practitioners and their experience/training, study design (between- or within-subjects), drop-out rates, and a variety of outcome measures and outcome ratios (biological, psychological and qualitative).
Taking a combination of criteria relating to methodology and design, statistical methods applied, and outcome determinations, Drs. Jain and Mills assigned numerical values to each study. A study could earn from 0-8 points for methodology; statistical methods earned from -3 to 6 points (“minus” points were given for suboptimal data analysis and for failure to report statistics in such a way that inferences could be supported). Outcome reporting allowed -1 to 2 points on the basis of the validity and reliability of assessment measures. In total, a study could earn between -3 and 16 points overall. Once they had established numerical values for the quality of each study, Drs. Jain and Mills separated the studies into population groups by diagnosis, then by outcomes evaluated. Having done that, they determined “levels of evidence” for each group of studies.
To demonstrate overall quality review ratings and constituent ratings for each study, Drs. Jain and Mills fashioned a chart which can be seen in their original publication. The median numerical value of quality score ratings within a diagnostic group defined those studies as being of higher or lower quality. “Effect” was determined by p<0.05 null hypothesis rejection of the studied factor(s). For a group of studies to qualify as having having level 1 efficacy (“strong evidence”), two or more randomized and controlled studies within the group had to be of high quality, with other studies within the group showing “generally consistent evidence.” The “moderate evidence” group, or level 2, had to have two randomized controlled studies, one of which had to be of higher quality and the other of lower quality, or the second one a non-randomized and controlled study but quasi-experimental and of higher quality. A level 3 group (“limited evidence”) had to have a randomized and controlled study of lower quality and/or a quasi-experimental study. And level 4 (“conflicting evidence”) encompassed studies where the outcomes were a balance of positive versus null results.
All of the study characteristics and results were summarized in chart form by Drs. Jain and Mills. Of the 66 studies included in Drs. Jain and Mills’ assessment, ten were excluded from analysis due poor statistical techniques. The remaining 56 studies had an overall quality range of from 1 to 12 (out of 16 possible points), with a mean of 6.4 and a median of 6. Drs. Jain and Mills therefore assessed the overall quality of the studies to be medium.
Zero to 8 points were possible for methodology, with an actual range of from 1 to 6, a mean of 2.0 and a median of 3. Statistical quality (-3 to 6 points possible) fell between -2 and 5, with a mean of 2.9 and a median of 3. Most studies lost a point for small sample size (i.e., <31 participants). Overall outcome (-1 to 2 points possible) ranged from 0 to 2 (mean=1.4, median=1.5). The validity and reliability measures (-1 to 1 possible) averaged .89, indicating that studies as a whole used appropriate measures of outcome, but all possible outcomes were frequently not measured. Outcomes in the ten studies that were excluded due to poor statistical techniques were nonetheless subjected to linear regression analysis, and were found to be similar to those of the included studies.
Only some outcomes could be assessed
A limited number of healing intervention outcomes were included in Drs. Jain and Mills’ study, due largely to limited numbers of well-designed studies and to the fact that some topics tend to be researched more than others. But what Drs. Jain and Mills did find can be summarized. With regard to pain, there was strong evidence for a reduction of pain intensity in non-hospitalized patients. For hospitalized pain patients and cancer patients with acute pain, there was moderate evidence for pain reduction. Regarding “comprehensive” or chronic pain (mixed hospitalized versus non-hospitalized individuals?), biofield healing interventions were evaluated equivocally. (Drs. Jain and Mills point out that for “comprehensive” pain, an affective component plays a considerable role.) With cancer patients, studies of biofield work showed equivocal outcomes for fatigue and quality of life but, as just stated, moderate pain-reduction outcomes. Dementia patients treated with biofield techniques showed behavioral symptom improvement in studies determined to have “moderate” outcomes. Anxiety reduction was a focus in some studies: moderate evidence for a decrease in anxiety in hospitalized patients was found, but there was equivocal evidence (limited data) for a decrease in anxiety in cardiovascular patients.
In Conclusion. . . .
Biofield healers go wherever they are called, so to speak. Thus an individual healer may work with individuals whose complaints cross a very broad spectrum. And each individual healer may have particular skills with particular disorders, depending on his or her native gifts, training, and experience. The study just discussed summarizes reliable data for a limited number of healing foci. What is significant is that under rigorous review, positive biofield treatment outcomes are clearly present for work with pain and anxiety, and also for behavioral problems exhibited by dementia patients. Where outcomes from studies were equivocal, more study participants and better and more focused studies would give greater outcome clarity. That outcomes were not reported for most of the myriad of problems that biofield healers actually treat does not mean that the healers are ineffective with those problems (or conversely, that they are effective). It means merely that much more investigation within the realm of biofield healing is called for. We have a promising start!