| |
|
The Research Corner
Complementary and Alternative Medicine Research Review:
Reiki and Functional Recovery for Patients Poststroke
Reviewed by Pamela Potter, MA, MSN, APRN, DNSc (c)
New Haven, CT
pamela.potter@mindspring.com
Designing, conducting,
and reporting research with complementary and alternative medicine (CAM)
therapies calls to mind the dual meaning of the Chinese character for
“crisis,” which signifies both “danger” and “opportunity.”
Similarly, reviewing a CAM research study suggests danger as well as opportunity.
As a doctoral student, I appreciate the possibility that almost any critique
can feel harsh and judgmental at first glance. However, the opportunity
lies in generating insights for improving research and contributing to
knowledge. With this in mind, the purpose of this article is to review
a CAM intervention study, identify design strengths and limitations, suggest
possibilities for building on the research, and provide a forum for discussion.
A recently published study of a Reiki intervention given to patients in
poststroke rehabilitation concluded that Reiki did not demonstrate any
clinically useful effect on the speed or level of functional recovery
in hospitalized patients with subacute stroke (Shiflett, Nayak, Bid, Miles,
& Agostinelli, 2002). Although the study did not draw from a population
with cancer, it addressed an intervention, Reiki, which has reported use
among people living with cancer.
The design was described as a placebo-controlled, modified double-blinded
clinical trial with an additional historical control condition. The sample
consisted of 50 patients with subacute ischemic stroke who were recruited
from the stroke unit of a major rehabilitation hospital. Thirty-eight
patients were assigned randomly to one of three treatment conditions:
Reiki master, Reiki practitioner, and sham Reiki practitioner. However,
8 of the 30 patients were lost by attrition. Additionally, 20 patients
represented historical controls and were selected randomly from charts
of comparable patients who were treated at the study institution within
six months prior to and six months after the study. In addition to receiving
standard rehabilitation treatment, subjects in the three treatment conditions
received up to ten 30-minute Reiki treatments over 2.5 weeks. Two instruments
measured study outcomes: the Functional Independence Measure (FIM) (Deutsch,
Braun, & Granger, 1996) and the Center for Epidemiologic Studies-Depression
(CES-D) scale (Shinar et al., 1986). Both instruments have demonstrated
validity and reliability with the population of interest.
Shiflett et al. (2002) made a valiant attempt to meet the rigorous requirements
of experimental research by designing a double-blinded study with placebo
control. After a brief review of the limited Reiki research and anecdotal
reports of its effectiveness for people recovering from stroke, the case
was made that Reiki intervention warranted further investigation.
The three study objectives serve as a structure for this review. These
are
1) to evaluate the effectiveness of Reiki as an adjunctive treatment for
patients with subacute stroke who were receiving standard rehabilitation
as inpatients; 2) to evaluate a double-blinded procedure for training
Reiki practitioners; and 3) to determine whether or not double-blinded
Reiki and sham practitioners could determine which category they were
in (Shiflett et al., 2002, p. 755).
Evaluating the Effectiveness of Reiki
The three treatment arms compared Reiki treatments given by an experienced
practitioner with treatments given by someone newly attuned to Reiki and
those given by a sham Reiki practitioner. The groups then were compared
with usual care historical controls. No significant effects from the intervention
were demonstrated by the outcome measures.
The researchers identified three possible reasons for the results. First,
outcomes from the FIM were limited to functional outcomes as a result
of missing cognitive data that had not been routinely entered in the charts
as anticipated. Second, the measures, although commonly used to measure
outcomes for this population, may not have been adequate to detect the
subtle effects of the intervention. Third, the result that Reiki has a
smaller effect size than anticipated may have resulted from an insufficient
sample size and the possibility of Type II error.
Evaluating the Double-Blind Procedure for Training Reiki Practitioners
The 14 practitioners in the study, both sham and Reiki, received identical
instruction in the principles and practice of Reiki. However, half the
practitioners did not receive the initiation procedure considered essential
for becoming a Reiki practitioner. This was accomplished by adapting the
initiation procedure to the second-degree Reiki distance healing technique.
To blind the healers to whether they were among those attuned, the potential
practitioners sat in a row of chairs while the master attuned half of
them at a distance instead of using the traditional laying-on-of-hands
initiation. Those practitioners who had not received the initiation were
given it at the completion of the study.
By blinding the practitioners to their status as initiated or uninitiated,
the researchers said that beliefs about practitioner status and unconscious
intentionality were randomized across the new practitioner groups. Therefore,
because intentionality was controlled for, it was not a factor in study
results.
Practitioner Category Recognition
From a 41-item questionnaire about practitioner experience, only one item
yielded a significant difference at the 0.05 level. Sham Reiki practitioners
reported a greater frequency of feeling heat in their hands (t = 2.44,
p < 0.03). Although not significant, the Reiki practitioners were less
confident that they actually had been initiated than sham practitioners
(t = -2.12, p < 0.06). The researchers suggested that these findings
may be explained by the presence in both groups of those with natural
healing abilities to sense subtle energies. The researchers concluded
that initiated practitioners and sham practitioners truly were blinded
because reported experience and sensation did not differ significantly
between groups.
Study Constraints
This study has been presented along with the researchers’ conclusions
about research limitations and the meaning of the findings. Four constraints
stand out as warranting further discussion: predictive power, unplanned
institutional issues, complexity of the research question, and the nature
of assumptions about what is being studied.
Power to observe significant differences, if they truly exist, is key
to any interpretation of study findings. Based on a power analysis for
the FIM, to achieve a statistical power of 0.80 with a moderate effect
size (an 8–10 point change), the researchers concluded that a total
of 50 subjects would be needed for the study. It is unclear from the article
how the power analysis was formulated and whether the number of groups
was considered in that formulation. As the number of groups in a study
increases, the power to detect mean differences decreases; therefore,
to retain power, the sample size must be increased accordingly. Using
Cohen’s conventional value for medium effect (0.06), a study with
a power of 0.80 at an alpha of 0.05 suggests a sample size of 45 subjects
per group in a four-group study (n = 180) (Polit & Hungler, 1999).
Whether the possibility of Type II error resulted from assuming too high
an effect size when calculating the study sample or from a failure to
consider the number of groups when calculating the sample size, the study
appears underpowered to conclude that a Reiki intervention for poststroke
patients has no impact on functional status.
Unplanned for institutional issues, like poor record keeping on measures
essential to evaluating study outcomes, can limit research findings. The
unavailability of cognitive scores on the post FIM measure also may have
contributed to a loss of predictive power. This is a hindsight lesson
from a pilot study that can be applied to designing future studies either
by taking steps to ensure the desired data will be available or identifying
other means for evaluating the outcome variable.
Trying to answer too many questions in one study may detract from finding
evidence of intervention efficacy. Assuming the study did have enough
power to detect outcome differences among practitioner types, the idea
of “distance attunement” added to the complexity of the study.
Whether or not the distance attunement actually is comparable to the traditional
initiation attunement is unknown. Alternatively, the possibility of those
not intentionally initiated by the master practitioner receiving attunement
by association with those actually attuned also is beyond apprehension.
Making assumptions about the nature of what is being studied may have
further contributed to insignificant outcomes. Assumptions were made that
laying-on-of-hands touch by a non-initiated practitioner is quantitatively
different from an initiated practitioner when the difference may be more
qualitative. It cannot be said that uninitiated touch is inert. Further,
the attempt to control for intentionality through a blinded initiation
procedure might have had a negative placebo, or nocebo, effect on the
initiated practitioners wherein they doubted the verity of initiation.
Implications for Research
Because this study was labeled as a “pilot,” results must
be interpreted in terms of a pilot rather than a full clinical trial.
The study appeared to lack the power to conclude that Reiki is not an
effective intervention for hospitalized patients with subacute stroke.
Rather, lessons learned from this pilot can be applied to future research.
Because little is known about responses to Reiki, researchers might want
to conduct a Level I trial sufficiently powered to detect differences
in cognitive and functional improvement between those with post subacute
stroke who receive a series of Reiki treatments by experienced practitioners
and a usual care control. Triangulation with qualitative data may yield
further insights into the effect of the intervention with this population.
Placebo attunement and sham practitioner treatment appear to confound
because their active ingredients are unknown. This is another level of
research requiring possible biological and electromagnetic measures to
determine differences between practitioners.
References
Deutsch, A., Braun, S., & Granger, C. (1996). The functional independence
measure and the functional independence measure for children. Ten years
of development. Critical Review of Physical Medicine Rehabilitation,
8, 267–281.
Polit, D.F., & Hungler, B.P. (1999). Nursing research: Principles
and methods (6th ed.). Philadelphia: J.B. Lippincott.
Shiflett, S.C., Nayak, S., Bid, C., Miles, P., & Agostinelli, S. (2002).
Effect of Reiki treatments on functional recovery in patients in poststroke
rehabilitation: A pilot study. Journal of Alternative & Complementary
Medicine, 8, 755–763.
Shinar, D., Gross, C.R., Price, T.R., Banko, M., Bolduc, P.L., & Robinson,
R.G. (1986). Screening for depression in stroke patients: The reliability
and validity of the center for epidemiologic studies depression scale.
Stroke, 17, 241–245.
|
|
Jo
Ann (Jody) Owen, BSN, RN
Gwen
Wyatt, RN, MSN, PhD
Pat
Snowden Dittig, RN, MSN, CRNP