DR. SUMIT GOEL M.D. (Hom)
In order to be sure of the integrity of our work, we
must demand three essential things:
1.
The quality
of the drug must be pure; it must be free from all mixture with other drugs,
and it must possess all its active properties.
2.
The prover
must possess the proper balance in functions and be in a normal, healthy state,
so that we can estimate and weigh the amount of the disturbance caused when we
deliberately upset the balance of health.
3.
The
circumstances surrounding the prover must be those of his normal surroundings,
so that the drug can express its action under conditions and circumstances
normal to the prover, that any deviation from normal in the prover's condition
cannot be attributed to different circumstances and conditions of his life, but
directly to the action of the drug.
Problems arise because of placebo symptoms and the
drug specifity of individual symptoms can only be expressed in probabilities.
Statistical analysis of homoeopathic drug tests has to based on comparision
between the totality of symptoms elicited during exhibition and the totality of
symptoms recorded on placebo. Statistical confirmation of drug specificity for
a single symptom in a test is a methodological impossibility as individual
subjects differ in reactivity. Statistical analysis and placebo studies do not
serve to demonstrate drug specificity for a particular symptom in a test, nor
exclude it.
Many current techniques were developed at a time when
precise methods of evaluation were unknown. New techniques should not be
generally applied without proof of effectiveness.
The basic components of proving trials are
1.
The procedure
to be evaluated
2.
The
comparision procedure (the control)
3.
The measures
of effectiveness
The comparision procedure or the control is defined
with the same care as the study procedure and is an alternative activity that
reasonably might accomplish the same goal.
Prior designation of measures of effectiveness is
absolutely fundamental. If not designated before undertaking the proving trial,
the investigator can find in almost any study a post hoc characteristic in
which the study program seems to surpass the control program.
The measures of effectiveness in program reviews
involve more immediately evident observations and are related to the
characteristics of organizational structure and the process.
MEASURES OF
EFFECTIVENESS
The main measures of effectiveness include the study
of structure, process and outcome.
STRUCTURE – Structure includes the settings available and
utilized
Personnel – number, type,
training, motivation
Facilities – number, type, accreditation, standards
Organization – coordination, continuity,
accessibility, appropriateness
Information systems and Records
Financing
Process refers to the activities of the personnel
involved in the proving project.
Factors weighing against outcome measures are their
cost, their limitation in identification of the mechanism whereby a program
succeeds or fails.
The structural variables include the qualification
and training. Process variables include the appropriateness of the probes used
in eliciting medical history, the relevance of the physical examination
performed, the choice of the clinical laboratory tests.
The patient’s history and medical record is taken as
the basic source of information.
STUDY DESIGN
OF PROVING TRIALS
Different types of population must be considered in
planning an experimental trial.
Reference Population
à Experimental
Population à
Sampling Frame à
Study and Control Group
The reference population is the population to which
the results of the proving will be generalized.
Since the Proving Master cannot anticipate the
characteristics of all groups to which generalization will be made, his
obligation is to make his methods and findings so clear so as to be universally
applicable.
The experimental population is the population within
which the experimental proving trial is actually done. This group must be
clearly defined. Those commonly drawn include
It is suggested that experimental populations should
be a representative of the total population of the geographic area including
the religion and cultural variance.
ASSIGNMENT OF
STUDY AND CONTROL POPULATION
The internal validity of the proving study is
determined by the similarity of study and control groups. The basic principle
of validity requires that study and control groups must be comparable with one
another prior to introduction of the programs to be evaluated. The defining
feature of the experimental study is that comparability is obtained by
selecting each group in such a way that both are representative of the same
sampling frame.
The most common and most satisfactory method of
choosing comparable groups from a sampling frame is through the use of random
numbers.
It is not useful to make any useful estimate of the
number of observations necessary to test an hypotheses.
Blind procedures are techniques employed to overcome
potential bias in evaluation.
Blind assignment is the use of random assignment or
other assignment technique that selects study and control groups without
reference to characteristics of the individuals. The purpose is to avoid
selection bias.
Blind assessment is the use of a procedure for
categorizing outcome without knowing which program the subject belongs to; or
in other words, assessment without knowing whether the prover has the test dose
or the placebo dose.
Double blind procedure is the practice of concealing
the nature and identity of test dose from each member of the project. The
purpose of double blind procedure is to avoid observation bias in assessing
outcome.
In reprovings, the measure of effectiveness is
comparable with the results of previous proving trials and comparing the data
of the reproving with the known data of the test substance.
To enhance the acceptance and publication of provings
in homoeopathy, the use of standardized methodology, presentation checklists
pre-registration and a central pooling authority is recommended.
Typical features of exploratory research are lack of
randomisation, multiple outcomes and investigation of diverse variables.
Even the best investigation will have little impact
if presented in an inadequate manner.
Checklists using predefined criteria is recommended
as a standard method in systematic reviews for assessing the methodological
quality of publications.
It is recommended that proving groups should seek
advice from more experienced provers and study closely the few good examples
that have already been published.
DR. SUMIT GOEL M.D. (Hom)