PhD thesis: Evaluating
Homeopathic Medicine: Clinical and Social Research, withPre-Menstrual Syndrome
(PMS) as a case Study
By Michal Yakir (with Prf BenTwvitch as the PhD mentor. The the Jerusalem university, Israel)
medicine (CAM), including homeopathy, the focus of this study, is utilized
increasingly by patients and by medical institutes (Ernst and White 2000,
Sharma 1992, Borkan et al. 1994, Eisenberg et al. 1993, and Shoval and Anson
2000). Despite the growing public use of homeopathy, controlled and rigorous
clinical trials for determining the efficacy of homeopathic treatments are
scarce. Moreover, methodological difficulties inherent in such trials present a
major obstacle (Effective Health Care 2002).
evaluation and research are needed to obtain more sound evidence of
homeopathy’s effectiveness and the scope of its influence (Kleijnen 1991,
Cucherat et al. 2000). Furthermore, in order to understand the social context
in which it operates, investigating the opinions, experiences and expectations
of health consumers and health providers is of great importance. This study
strives to put homeopathy to the test of scientific research procedures, using
quantitative and qualitative interdisciplinary methods. Pre-Menstrual Syndrome
(PMS) serves as a case study, in terms of both social science and clinical
varied for different parts of the study.
the purpose of understanding the perceptions, attitudes, expectations,
experiences and levels of satisfaction among health consumers and providers who
use or practice homeopathy or other CAM, an explorative qualitative study of
limited size was performed using Grounded Theory (Pope 1995). Interviews were
conducted with three populations: the women participating in the clinical
study, gynecologists, and homeopaths. We inquired about their experiences in
the treatment of PMS with conventional therapies or CAM, the factors associated
with the choice of treatment and their reflections upon it. With doctors and
homeopaths, we used the Half Open interview method, placing more emphasis on
their dealings with PMS patients and their willingness to incorporate the other
medicine into the context of the PMS treatment. The findings provide background
to the present status of homeopathy in society and medicine.
determine the efficacy of homeopathic treatment over placebo, and the
reproducibility of homeopathic research in the treatment of PMS, we ran
clinical studies that evaluated the efficacy of two different homeopathic
methods: treatment tailored to the individual and treatment with identical
medication for each woman. A novel methodological approach for homeopathic
clinical trials was incorporated into the studies: a simple yet repeatable
protocol for diagnosing the needed homeopathic remedy (medication). Evaluation
of another aspect of the placebo effect in homeopathic treatment was enabled by
tests for traits such as suggestibility (Kreitler et al. 1990) and
repressibility (Weinberger et al. 1995).
The study, initiated in 1992 as part of PhD thesis in the Haddasa Jerusalem univesity, Israel, was conducted as two
repetitions of a randomized, controlled double-blind clinical trial. The first
stage was a pilot study; the second, a trial with a larger number of cases. In
every trial, two parallel methods were investigated:
intervention was chosen individually for each patient according to an
especially developed model of symptom clusters.
homeopathic intervention – Folliculinum – was given to every participant of the
Design: A three-year
randomized controlled double-blind clinical study.
Recruitment: Women volunteers,
suffering from confirmed PMS for at least one year previously, were recruited.
Age limitation: between 20 and 50 years. Exclusion: chronic illnesses, use of
hormonal or psychopharmacological drugs.
Location: The Hadassa hospital, Jerusalem, Israel.
The homeopathic protocol model: In the individual
treatment study, the medication was prescribed on the basis of a specially
designed protocol of symptom clusters. Patients were given a questionnaire,
comprising 70 questions grouped in clusters, and scored with reference to the
key symptoms that indicated one of 14 remedies used for individualized
treatment. Patients with unclear scoring results were automatically excluded
and entered into the Folliculinum group – that is, the second study. Each
patient was treated randomly with a single oral dose of the chosen homeopathic
remedy or with its parallel masked placebo.
Procedure: Each participant
underwent a two-month baseline daily assessment, was assigned to one of the
studies, was given the homeopathic medication once, and underwent a
post-intervention daily follow-up for three more months.
measures of change:(a) Changes in scores of a daily menstrual distress questionnaire (MDQ, see
Moos 1989) before and after treatment (picture 3); (b) reduction in medication
consumption and sick-days; (c) self-assessment; and (d) psychological tests for
suggestibility. These tests were used to examine the possible effects of
suggestion on the placebo effect. Other traits, such as anxiety, social
desirability, and repressibility, were screened for through questionnaires,
before and at the end of treatment.
Qualitative study results
Interviews with women before the clinical study demonstrated some reasons why
women turn to CAM: Unease with medications, hormones and medical treatment,
discouraging answers from doctors, the feeling that "I will not be
understood anyway" and sometimes bad experiences with medical
treatment turned to feelings of
disappointment, alienation and distrust with CM. This led women to find
therapeutic alternatives, mainly "natural" products bought off the shelf,
without guidance, supervision or regulation. The desire for a personal
connection during treatment, beliefs in a natural life style and the
association of homeopathy with this were also factors that led women to seek
out CAM treatments – or to apply to the study. The findings that discontent and
disappointment were greater factors in not utilizing CM, rather than a
deep-seated belief in the greater effectiveness of CAM or homeopathy,
underscore the conclusions of McGregor and Peay (1996) and Dunfield (1996) and
run contrary to those of Furnham and Kirkcaldy (1996), Eisenberg et al. (2001)
Interviews with doctors and homeopaths. Homeopaths’ and doctors’ patterns of
perceptions and reference to PMS and to the "other" system of
treatment (doctors to homeopathy and vice-versa) was parallel: each displayed
little knowledge of the other system or interest in its treatment options for
PMS. Both admitted to rarely referring women to the other system or otherwise
cooperating with it. The lack of referrals bears more weight in conventional
medicine, since most women first turn to conventional medicine before trying
Clinical study results
Setting: Hadassah hospital
outpatient gynecology clinic, Jerusalem, Israel.
study(1992-94). Individual treatment: Nineteen women, aged 20 to 48, suffering from
PMS for ten years on average.
significant decrease in baseline MDQ scores (i.e., improvement), from 0.44 to
0.13 (P<0.05) was observed in the treatment group (verum), but not in
the control group (from 0.38 to 0.34) (see also figure 1 in the Hebrew version),
thus leading to a higher rate of improvement (figure 2,3) and lower
self-assessment of PMS (figure 4).
Sick-days before menses were reduced from 0.75 to 0 in the verum group and not changed in the
control group (0.56 to 0.53; table 4.). Medication consumption was also reduced
more in the verum group (table 5).
(For detailed information, see paper or Yakir et al. 2001.)
treatment:105 women; 96 of whom completed the study (53 in control and 43 in verum (homeopathic treatment). Mean age
was 37.2 years; all participants had high school or university education. The
drop-out rate during the study was 8.6 % (n = 9), caused either by pregnancy or
non-compliance taking medication or filling out study forms. The most
frequently administered study medications were Nux-v, Nat-m, Sep, Lach, Ars,
Calc-c and Phos.
decrease in mean PMS symptoms (monthly mean; see figures 12 and 14)) was
significantly more pronounced in the verumgroup compared to controls: in verum: 0.443 before versus 0.287 after
treatment; in the control group: 0.426 before versus 0.34 after (ANOVA, F=3.01,
p=0.04, 1-tailed). The rate of
improvement was significantly higher in the verumgroup than among the controls (0.35 vs. 0.2 respectively, t=-1.62, p=0.05). The
improvements in the placebo group were mainly limited to those patients who
displayed low initial MDQ-scores at the outset of the study, whereas in the verum group even severely affected women
improved considerably (see table 14 and figure 16). Overall, women in the verum group assessed themselves with greater
improvement (figure 20), and perceived themselves as less in need of treatment
at the end of the study (1.8 vs. 2.37 in control, p=0.037 1-tailed).
The consumption of other drugs for PMS, such as
pain killers or sedatives, was reduced after treatment by 75% in the verum group and 36% in the control group
(p<0.05). The highest rate of reduction was found in the verum group in women who suffered most
from PMS, whereas in the placebo group such women increased consumption of
concomitant drugs (table 17, figure 23). Another striking finding was the
difference in sick-days before menses. Absence from work due to PMS decreased
in the verum group by 91% and did not
change in the control group (ANOVA, F=3.409, p=0.03, 1-tailed) (table 16 and
figure 21). Confirming that, general improvement in health during the rest of
the month (non PMS period) was higher in the verum group (0.21 vs.
0.08, p=0.05). For full details, see paper in appendix 2.
treatment in the pilot study: Folliculinum 9c. 30 women completed
the study. No significant difference was found between verum and control groups. In verum,
mean PMS symptoms decreased from 0.66 to 0.53. Among controls it decreased from
0.38 to 0.34 (table 9 and figure 7). Improvement rates were higher in verum but
not significantly different (verum: 0.66, control 0.38).
study: Folliculinum 30c. 35 women completed the study. again,
no significant difference in improvement rates between verum and control
groups was found (table 48, figure 37). In the verum group, symptoms decreased from 0.37 to 0.22. Among controls
symptoms decreased from 0.31 to 0.24. In the verum group medication
reduction was only 25% (compare to individual treatment). Among controls
medication consumption actually increased (table 52). Overall, more women
described themselves as either highly improved or as worse in the verum treatment
group (figure 41) and more assessed themselves as needing treatment (1.4 versus
1.9 in the control).
evidence was found in either study, of suggestibility or any other
psychological trait to influence the (figure 26) effect of homeopathic
treatment. Furthermore, with the individualized treatment, some psychological
traits – such as social desirability – were found to be somewhat improved
homeopathic treatment (classical homeopathy) was significantly and reproducibly
associated with amelioration of PMS symptoms and improved well-being., Pain,
sick-days and Consumption of medication were greatly reduced. Homeopathic
treatment was effective in improving both physical symptoms and emotional
issues, and the success of treatment could not be accounted for by placebo
effects or such personality traits as suggestibility. These results lend
support to using homeopathy as an established treatment for women suffering
from PMS. Uniform homeopathic treatment was not found to be this effective over
placebo. These results, comparing two treatments at the same time in a single
study setting may imply invalid results in meta-analysis that encompasses both
methods. The symptom cluster approach offers a new promising tool that can be applied
to other clinical homeopathic trials and facilitate research in homeopathy by
enabling reproducible protocol and efficient masking of medication, thus
overcoming the obstacle of methodological difficulties often encountered in
homeopathic research. To solidify the conclusions, further and independent
replication of the study in other centers is called for.
literature survey demonstrates that turning to homeopathy is becoming more and
more common, which can account for the higher results observed in the main
study control group compare to the pilot study controls. During the years past
between the studies, increase in awareness and social discussion occurred. These
factors as well as greater utilization of homeopathic treatment may have
enhanced placebo response in the controls.
interviews with health providers and consumers elucidate the background of the
growing distance of patients from CM and the reasons for higher inclination
towards CAM and homeopathy. As opposed to some studies, we found
dissatisfaction with conventional medical care to be a more significant factor
than belief in the appropriateness or efficacy of homeopathy. That can be
attributed to the fact that there is no satisfactory answer to the problem of
PMS in medicine. Nevertheless, our subjects’ perceptions of medical services
should sound an alarm and not be ignored. The results emphasize the importance
of creating a dialogue between alternative and conventional medicine – for the
sake of the patient.