Gathering the evidence
Monitoring the effectiveness of a homeopathic practice
through an outcomes-based data collection system
by Barbara J Braun BSc(Hons) DO Dip Ac DHoM MIRCH
and Ruth M Odendaal LCPH
large-scale funding for the project,
It is common knowledge that homeopathy as and also to develop some means
a therapy is currently undergoing a difficult period, of monitoring other factors in the
and there is a need to provide good and unbiased day-to-day running of a large-scale
evidence to demonstrate exactly how effective it is. homeopathy programme. These
It was in this context that the Swaziland Homeo- factors include: remedy use for
pathy Project, a charity registered in the southern keeping track of the pharmacy;
African country of Swaziland, decided to set up details of individual clinics; fre-
Barbara Braun was Ruth Odendaal was
a system for gathering evidence of patient outcomes quency of symptoms; and statistics
born in Swaziland, born in Zimbabwe
as a result of homeopathic treatment. Initiated on general patient population
and completed and trained at the
in June 2008, the Swaziland Homeopathy Project demographics.
a BSc (Hons) in College of Practical
Zoology at Reading Homeopathy, grad-
(SHP) currently conducts on average 19 clinics
University in the uating in 2004. per month. Most of these are free, mobile outreach Collecting the information
UK. She then She has been clinics into the impoverished rural areas of northern Since the majority of the project’s
qualified as an a director and Swaziland. clinics are held outdoors in rural
osteopath in the UK practising homeo- areas, often in awkward condi-
and later also as path for the Swazi- tions, cases are written, not com-
an acupuncturist land Homeopathy In the process of justifying the need puterised; and so a cover sheet
and homeopath. Project for the past for further homeopathy clinics, was devised to include all the
She has been in three years; prior some of the questions that needed details on each patient that need
private practice to that she prac- to be addressed were: to be entered on the database. The
in Swaziland since tised at an HIV clin-
• Is homeopathy actually working? cover sheet has been designed in
1986. In 2008 she
ic in Kings Cross,
London and at • Are the patients actually show-
a user-friendly way so that the
information is easily transferred.
Neals Yard Reme-
dies. She was also • Are the improvements noted
just a result of taking the case
However, in a different clinical
setting, data could be captured
with two British a manager for the
and having someone listen? directly as the case is taken.
homeopaths, Richmond Shop
Julie Hodgson and Therapy • Are enough of the patients
For each patient, the following
overall details are recorded: name,
and Shirley Reece. rooms. She prac-
The methodology tises a number of • Which remedies are being pre-
scribed most frequently?
age, sex, HIV status and clinic
location. Then for each individual
was based on the healing modalities
Maun Homeopathy such as Pleiadian • How many of the patients are
testing HIV positive?
consultation, the following are
noted: the date, the prescription,
Project initiated by light work,
• Can the individual symptoms
be tracked to see if they have
the remedy justification (symptoms
or rubrics for the remedy), a symp-
improved, changed or aggravated? tom score (VAS – see info box
• Is it worth all the effort?
A straightforward data collecting
on page 32), an evaluation of the
treatment, and finally a patient and
system was set up to try to provide practitioner overall health score
clear answers to some of these out of 10 (based on the Karnofsky
questions. The motivation for this scale in which 10 = completely
was principally to see whether healthy – see info box on page 38).
it would be worth applying for These two scoring systems are the
30 Homeopathy in practice Autumn 2011
basis for the analysis, provid-
ing both subjective and objective
The details of each consultation
are then entered on a weekly basis,
and the patients can then be con-
stantly and easily monitored and
When deciding the evaluation
scores, it was thought best to stay
with existing scientific systems.
The VAS Symptom score is on
a scale of 0-5, where 5 = highest
severity, 0 = symptom is cleared.
This is evaluated by the practi-
tioner, although it would also be
possible to have both patient and
practitioner score this.
Some case examples
The following cases illustrate how
the patient scores change over
time and indicate improvement or
otherwise. Swaziland Homeopathy
Project adopts the Layers method
(Eizayaga model) for prescriptions
considering miasmatic, constitu-
tional, fundamental and lesion lay-
ers. This is because many of our
cases are of advanced pathology
and often complicated with drugs
and are complex with multiple
aetiologies. yellow phlegm. She has financial Barbara with Nat mur 200: thrice weekly: cough.
This involves giving different stresses as husband not working. a patient at an Patient score: 5/10.
remedies in sequence over a period Burning itching eyes. Can’t focus outreach clinic
of a week. The remedies are given her vision. Has a stye in her right NB: Patient didn’t return until ten
to last for one month. We aim to eye. Deaf in right ear. Heartburn in months later. Says all cough etc
see patients on a monthly basis stomach when eating bitter things. was cleared.
and they will follow up until they She is breastfeeding, therefore no Patient score: 8/10 on original
feel well. menstruation. Itching skin with symptoms.
sores on the thighs and buttocks.
Case 1: Female, 27yrs, HIV negative Watery discharge, worse in rainy Follow-up 9.3.10
weather. Now HIV positive. Waiting to
1st consultation 24.2.09 start on ART (Anti-retroviral
Presenting complaint: Backache, Prescription therapy). CD4: 253 (see CD4
headache, dizziness, cough. Has Tuberculinum bovinum 200: twice Count info box on page 39)
a history of TB. December 08: weekly: history of TB with recur- Presenting new symptoms: cough
Completed TB treatment with local rent symptoms of cough, dizziness, has developed. Chest pain with
hospital. Has four children. Sole headache, sweating. cough burning ‘like a sore inside’,
wage earner, weaving baskets for Natrum muriaticum 30: thrice yellow sputum. Loss of appetite.
Tintsaba crafts. Breastfeeding. Has weekly: dizziness with dry cough. ++ thirsty, worse with cloudy
a throbbing pain in backbone from Itching skin with watery blisters. weather. Has had this for one
top of spine to bottom which is Chronic headaches throbbing from week now. Headache with cough
worse at night. Suffers from sleep- morning till evening. Itching, burn- in forehead and occiput, worse in
lessness due to this. Has headaches ing eyes. heat and light.
all the time which make her cry. It Patient score: 3/10. Has new mental stresses; husband
is a throbbing pain, mostly in fore- is in love with another woman. He
head, morning till evening. Has Follow-up 12.5.09 now has a job but is supporting his
had this for the past year. These Dizziness has now gone. Cough lover and her three children, not
headaches started with a cough now expectorating, whitish watery. her. She has four children to sup-
accompanied by dizziness, heat, No smell. Pain on right side with port. He has abandoned his family
sweating and palpitations. The cough. with no support. Feels very sad
cough is a dry cough worse in cold about this. Cries by herself, feels
weather with a pricking or sharp Prescription: repeat remedies in better for it. Is still breastfeeding.
pain. No sputum. higher potencies.
Her child is ill and this worries Tub bov 1M: twice weekly: Prescription
her. He has a constant cough with history TB. Arsenicum album 200: thrice
Homeopathy in practice Autumn 2011 31
weekly: burning pains with Liver drainage support (Carduus
cough. Yellow sputum. ++ thirst. marianus + Chelidonium + Visual Analogue Scale (VAS):
Financial worries. Poverty. Hydrastis) 30x: four times weekly.
Headache with cough. Loss of ART side effects: burning, numb- This is commonly used in the medical field, especially
appetite. ness, loss of appetite. oncology, to assess change in symptomology.
Tub bov 30: twice weekly: repeat Patient score: 4/10.
history of TB. Symptom evaluation:
Patient score: 5/10. Case 2: Female, 43yrs, HIV negative 0: Completely better, no symptoms any more
1: Noticeable, mild, irritating, occasional, part time,
Follow-up 13.4.10 1st consultation 24.2.09 an inconvenience, ‘It hurts a bit.’
Started ART (Anti-retroviral Presenting complaint: Headaches. 2: Most of the time, can’t easily ignore, bit restric-
therapy); CD4 count 253. Backache. Ovarian tubes tied two tive, ‘It bothers me a lot.’
Presenting: Cough still remains. yrs ago. Terrible headaches with 3: Constant, all of the time, can’t ignore, restricts
Burning pains in chest. Cough throbbing in the temples for past activity at times, ‘My pain is …’
now dry. No sputum. Worse in six months. Itching eyes. With 4: Severe, all the time, intense, strong, incapacitat-
evenings. The cough comes with lachrymation. Basket weaver for ing, overwhelming, ‘I’m in pain.’
a stomach ache with a desire Tintsaba crafts. The weaving 5: Critical, intense, overwhelming, all consuming,
to vomit; she ends up vomiting. makes it worse. She says the ‘Help me,’ desperate.
Also a desire to sit up with cough. headaches and eye problems are (http://en.wikipedia.org/wiki/Visual_analogue_scale)
Still ++ thirst and loss of appetite. connected.
Throbbing headache on right side Lower backache started two backache with ‘lightning pains’.
with cough. Worse in evening, months ago. The pain is like a Worse with motion.
worse in cold. ‘lightning pain’, especially bad Patient score: 4/10.
Still sadness over family prob- when getting up and with any
lems. Can’t see a change. ‘Hopeless movement that is quick, also with Follow-up 23.6.09
of future.’ Won’t confront her hus- side-to-side movement. Has an irri- Presenting complaint: headaches
band about the financial and other tating cough but no pain. have gone completely >>>. They
problems. She has no food. Is expe- Has huge financial stress. Five had been there for six months.
riencing dizziness and numbness on children in school. Sole wage earn- Cough now has an expectorant
right side of body with the ART. er. Husband is ill. Feels very angry which is yellow in colour. Not
when can’t provide for children. painful. Has gone in two-week
Prescription: NB: case compro- Her head hurts with this. cycles since Oct last year. It started
mised by ART. Has a slight cough with clear when caring for sick mother in
Phosphorus 200: thrice weekly: phlegm. No pain. Nothing serious. hospital; she had a cough (pneu-
HIV. Cough with vomiting. Can’t sleep due to her problems of monia / TB). Had to sleep outside.
Burning pains in chest with cough. finances. Thinks a lot at night. It was cold. All else OK.
Grief. ++ thirst. Worse when cold.
Carcinosin 30: twice weekly: HIV. Prescription Prescription
Immune deficiency. Grief. Over- Sulphur 30: twice weekly: terrible Hepar sulph 200: thrice weekly:
responsibility. Drug suppression. headaches with throbbing in tem- constant cough with yellow spu-
Tissue salts 6x (all tissue salts in ples and itching eyes. Fear of tum. Worse with cold.
combination): four times weekly: poverty. Constitutional. Tub 200: twice weekly: consistent
immune boost. Hypericum 1M: twice weekly: Mpofu clinic cough. Past eight months. Family
Patient score: 8/10.
NB: New problem. All previous
Presenting complaint: Vision is
bad, gets fogginess in front of eyes.
They are itching. Worse strain with
handicraft. All else clear.
Nat mur 30: thrice weekly: eyes,
vision, bad, worse from eyestrain.
Eyes itching in morning.
Patient score: 8/10.
NB: New problem. All previous
Presenting complaint: right ankle
pain. Worse after exertion i.e. long
walking. Worse on further exertion
i.e. any movement after resting.
32 Homeopathy in practice Autumn 2011
Rhus toxicodendron 200: thrice
weekly: ankle strain. Worse after
Patient score: 8/10.
Symptoms got better, then returned
again after three months.
Presenting complaint: pain under
heel. Throbbing. Worse in cold
water, walking long distance. Now
is starting to extend up the ankle.
Rhus tox 30: thrice weekly: old
sprain injury. Painful throbbing in
Patient score 8/10.
Ankle now all >>>.
Presenting complaint: cough with
headache, i.e. common cold.
Cough is dry with pain ‘like a tear-
ing apart’ in middle of chest. She
has thirst and is worse in evening.
Headache is worse in heat and
heavy work. Also with thinking.
She has financial stress with school
fees. Young daughter also preg-
nant. She has throbbing on side of
the head. Gets angry. Cries a lot
with her troubles in evening, will
then go to bed.
Bryonia 200: thrice weekly: dry
cough with sharp pains. Thirst
and financial worries. Common
cold with headache. Throbbing in
Patient score: 8/10.
Case 3: Female, 55yrs. Not tested
First consultation 27.11.09 She has constipation and burning Patients at Mgululu Hormonal balance.
Presenting complaint: fibroids for and throbbing in her feet. Worse in clinic Patient score: 5/10.
the past 20yrs, untreated. Peri- hot weather. Her arms ache with
menopausal. Headaches with the handicraft work. She earns a Follow-up 12.2.10
menses. Painful knees. She has living from handicraft. This is how Presenting complaint: no more pain
‘lumps in the womb’, experiences she supports her family. Her knees with menses >>>. Can feel the
painful menstruation with heavy ache from long sitting when doing lumps reducing. Feet still burning.
bleeding. Like ‘labour pains’, starts handicraft. Better after stretching. Knees >>>. Tiredness better but not
with a ‘kicking sensation’. Loses gone. She has developed an itching
energy a lot with this. She also has Prescription rash with whitish spots that darken
a headache with this menstruation. Medorrhinum 200: twice weekly: after scratching.
This has all been since the birth fibroids, burning feet and rotten
of her last child. discharge. Prescription
Husband died two yrs ago, after Sepia 30: thrice weekly: fibroids Medorrhinum 30: twice weekly.
the birth of her last child. She wor- with ‘labour pain’ with bleeding. Sepia 30: thrice weekly.
ries a lot about looking after her NBWS birth of last child. Head- Thyroidinum 30x: twice weekly.
children. aches with bleeding. Menopause Patient score: 7/10.
She has itching in her vagina starting.
with yellow discharge which smells Thyroidinum 30x: thrice weekly: Follow-up 12.3.10
rotten. organ support. Fibroid drainage. Presenting complaint: can still feel
Homeopathy in practice Autumn 2011 33
fibroid reducing. Menstruation treatment may be effectively
flow is much lighter. No pain evaluated. Table 3: Example of individual clinic statistics
either. Only has tiredness left in Swaziland Homeopathy Project – July 2011
shoulders. Has started skipping Compiling the information
menses i.e. menopause. Skin has Two databases were set up: one for Summary by location
now cleared >>>. all the clinics and patients overall, Change between consultations, for first four consultations
and a separate database for a study Endzingeni
Prescription of the HIV positive patients as part Number of patients: 83 HIV status:
Sepia 30: thrice weekly. of ongoing research into the effect Number of patients Not tested 17
Medorrhinum 30: twice weekly. of an integrated approach in the with follow-ups: 45 Tested negative 28
Thyroidinum 30x: twice weekly. treatment of HIV / AIDS. Tested positive 21
Patient score: 8/10. Homeopaths are often criticised Progress at each follow up: Unknown 17
for not doing sufficient research
From these case studies, it can be and so-called clinical trials. How- Patient Practitioner
seen that a simple score can easily ever, it is possible to provide solid Aggravated 12% Aggravated 9%
demonstrate the changes in the information from pragmatic obser- No change 24% No change 24%
case, and in this way homeopathic vational studies that are conducted Improved 64% Improved 67%
Table 1: Patient status, gender and age groups Number of patients 48 HIV status:
Number of patients Not tested 6
Swaziland Homeopathy Project – July 2011
with follow-ups 30 Tested negative 25
Total 1799 Tested positive 9
Gender Progress at each follow up: Unknown 8
Aggravated 16% Aggravated 15%
HIV status Total Male Female Age distribution No change: 17% No change 19%
Tested positive 320 33 287 Under 10 109 Improved 67% Improved 66%
Tested negative 662 53 609 10 – 19 83
Not tested 392 68 324 20 – 29 239
Suspected positive 11 6 5 30 – 39 315
Table 4: Example from HIV database
Unknown 413 92 321 40 – 49 358
Patients on ART 155 19 136 50 – 59 292 Swaziland Homeopathy Project – July 2011
60 – 69 216
70 – 79 101 Most frequently used remedies
Over 79 41 (Remedies prescribed at least 10 times)
Remedy No. of times
Table 2: Overall changes in evaluations Carcinosin 30
Swaziland Homeopathy Project – July 2011 Natrum muriaticum 22
Change between consultations, for first four consultations Sulphur 19
Arsenicum album 18
Patient Practitioner Phosphorus 15
All follow-ups: 1514 All follow-ups: 1514 Medhorrhinum 15
Improved 940 62% Improved 962 64% Rhus toxicodendron 13
No change 373 25% No change 368 24% Pulsatilla 11
Aggravated * 200 13% Aggravated 184 12% Bryonia 10
Follow-ups – HIV+ 319 Follow-ups – HIV+ 319
Improved 195 61% Improved 208 65% on an ongoing basis. This would
No change 75 24% No change 68 21% serve as concrete evidence for
Aggravated 49 15% Aggravated 43 13% the efficacy of homeopathy. Indi-
vidual cases as well as large groups
Follow-ups – patients on ART 182 Follow-ups – patients on ART 182 of patients may be practically
Improved 111 61% Improved 118 65% analysed.
No change 41 23% No change 38 21%
Aggravated 30 16% Aggravated 26 14% Monthly reports and analyses
Follow-ups – patients not on ARVS 145 Follow-ups – patients not on ARVS 145 1. Patient statistics and demographics
Improved 88 61% Improved: 94 65% This information is important for
No change 36 25% No change 32 22% general statistical purposes, and
Aggravated 21 14% Aggravated 19 13% gives an idea of the overall status
of the patients seen at clinics (see
* ‘Aggravated’ does not mean a remedy aggravation; it reflects the patient’s current overall Table 1). This is very useful data in
health. It includes new symptoms as well as changes in previously observed symptoms. general. Due to stigma over HIV /
AIDS in Swaziland, many people
34 Homeopathy in practice Autumn 2011
are not tested and are unwilling to
Table 5: Example from overall database do so. The last census conducted
Swaziland Homeopathy Project – July 2011 in this country was in 2007 and
most of the statistics are projec-
Analysis of changes in symptoms severity tions made on the results of this
census. Any accurate information is
Total symptom follow-ups 3419
therefore very useful in this regard.
Improved 214 63%
No change 799 23%
2. Overall patient evaluations
Aggravated 477 14%
This report gives an overview of
Symptom follow-ups – all the patients who have returned
HIV+ 837 for follow-up treatments along with
Improved 507 61% the patient and practitioner scores
No change 218 26% (see Table 2). From this, a steady
Aggravated 112 13% average overall improvement of
60-65% in patient scores can be
Symptom follow-ups – observed. This applies to both the
patients on ART 508 patient and practitioner evaluations.
Improved 303 60% These scores are for the previous
No change 133 26% six-month period of follow-ups.
Aggravated 72 14%
Symptom follow-ups – 3. Individual clinic statistics
patients not on ART 350 This report allows each clinic to be
Improved 215 61% individually tracked and checked
No change 92 26% to see how many patients are fol-
Aggravated 43 12% lowing up and how this can be
improved if necessary (see Table 3).
Note: If symptoms have not been scored within a six-month period, they are not included in This is particularly useful for pre-
this analysis. paring feedback to the women’s
handicraft groups that partner the
project. It also enables monitoring
Table 6: Example: HIV database (300 patients) of the success of the individual
clinics. Each of these clinic loca-
Swaziland Homeopathy Project – July 2011 tions is visited once a month.
Comparison of symptom changes by practitioner
4. Analysis of remedy use
Practitioner Improved No change Aggravated Total The remedy / prescription analysis
Practitioner A 206 65% 69 22% 44 14% 319 provides interesting information
Practitioner B 184 63% 75 26% 34 12% 293 regarding the nature of the treatment,
Practitioner C 11 50% 6 27% 5 23% 22 and in the case of the HIV database,
Practitioner D 11 46% 13 54% 0 0% 24 some insight into the ‘genus epidemi-
Practitioner E 3 38% 3 38% 2 25% 8 cus’ of the HIV / AIDS pandemic
Practitioner F 3 100% 0 0% 0 0% 3 in Swaziland (see Table 4).
Practitioner G 3 75% 1 25% 0 0% 4 As can be seen, the most com-
Practitioner H 2 100% 0 0% 0 0% 2 monly used prescriptions are tuber-
Practitioner I 1 100% 0 0% 0 0% 1 cular remedies with the exception
Practitioner J 1 25% 3 75% 0 0% 4 of Rhus tox. The presence of Rhus
tox is probably due to the high
prevalence of herpes zoster in HIV
Swaziland Homeopathy Project – July 2011 positive patients.
Overall database (1,800 patients)
5. Summary of symptom changes
Comparison of symptom changes by practitioner This report gives an overview
of the total symptoms that were
Practitioner Improved No change Aggravated Total scored at the follow-up treatments,
Practitioner A 1033 62% 374 23% 247 15% 1654 and gives an idea of whether the
Practitioner B 824 62% 314 24% 185 14% 1323 symptoms are improving as a result
Practitioner C 97 59% 51 31% 16 10% 164 of treatment within a period of six
Practitioner D 63 64% 22 22% 13 13% 98 months (see Table 5). There is also
Practitioner E 31 84% 6 16% 0 0% 37 an additional symptom report which
Practitioner F 24 62% 11 28% 4 10% 39 scores the frequency of symptoms;
Practitioner G 23 61% 8 21% 7 18% 38 for example, it can be seen from this
Practitioner H 15 88% 2 12% 0 0% 17 report that grief, cough and headache
Practitioner I 12 75% 1 6% 3 19% 16 are the most common symptoms
Practitioner J 11 61% 6 33% 1 6% 18 treated. This ties in with the tuber-
Practitioner K 10 77% 2 15% 1 8% 13 cular nature of the chronic disease
most commonly encountered.
Homeopathy in practice Autumn 2011 35
6. Practitioner symptom analysis
When the symptoms are scored, Table 7: Analysis of patient and practitioner overall evaluations – July 2011
they are linked to the practitioner Swaziland Homeopathy Project ART Study
who prescribed at that consulta- No treatment figures based on all first consultation scores i.e. pre-treatment
tion. This enables the database Treatment figures based on last consultation for each patient receiving treatment where they have followed up
Comparison of patient and practitioner evaluations for patients receiving homeopathic treatment compared with
to track the number of symptoms no treatment
improved, aggravated or unchanged
as per prescription. In this way the Patient scores
effectiveness of the practitioner can HIV+ ART
be measured (see Table 6). No. of patients Average score std dev
No treatment 146 5.58 0.11
The scores only really start to Homeopathic treatment 71 7.28 0.18
have significance at a higher num- Difference 1.70 1.70
ber of symptoms, but it can be seen significant at 99.9% confidence levels
that practitioners A and B (in both t* 71.89
approx df * 48
tables) have on average a score of highly significant: t for 50 df
about 60-65% improved symptoms 0.001 probability = 3.496
on the VAS score analysis. This is 99.9% confidence interval 1.70
similar to the overall patient and +/- 0.09
range from 1.61
practitioner Karnofsky score analy- to 1.80
sis, which is also in this range. So
with these two scores from differ- Practitioner scores
ent scales one may fairly accurately HIV+ ART
No. of patients Average score std dev
say that homeopathy has brought No treatment 146 5.38 0.10
about a 60-65% improvement in Homeopathic treatment 71 7.18 0.19
the health of this sample of 1,800 Difference 1.80 1.80
patients. This is evident in both significant at 99.9% confidence levels
HIV and overall databases (see t 76.94
approx df 45
Table 6). highly significant: t for 40 df
The above reports are compiled 0.001 probability = 3.551
on a monthly basis. In addition 99.9% confidence interval 1.80
to this, it is possible to carry out +/- 0.09
range from 1.71
further statistical analysis of the to 1.90
database results. The patients
on the HIV database were statisti- Patient scores
cally analysed using the Student’s HIV+ No treatment
No. of patients Average score std dev
t-test (see info box on page 39). ART 146 5.58 0.11
No ART 154 6.17 0.43
ART study background Difference 0.59 0.59
In Swaziland, patients are only eli- significant at 99.9% confidence levels
gible for free anti-retroviral treat- approx df 88
ment (ART) if their CD4 count highly significant: t for 80 df
is below 350. Many patients are 0.001 probability = 3.416
therefore not yet eligible for ART. 99.9% confidence interval 0.59
Those that are on ART and suffer- +/- 0.12
range from 0.47
ing from side effects such as peri- to 0.71
pheral neuropathy, nausea, vertigo
and headache, can get a great deal Practitioner scores
of relief from homeopathic HIV+ No treatment
No. of patients Average Score std dev
treatment. Homeopathy does not ART 146 5.38 0.10
interfere with ART and the No ART 154 5.58 0.12
patients are more able to adhere to Difference 0.21 0.17
their treatment regime. significant at 99.9% confidence levels
It should be noted that the project approx df 218
is not claiming to cure HIV / AIDS highly significant: t for 120 df
and recommends that patients 0.001 probability = 3.373
test to determine their status. The 99.9% confidence interval 0.21
patients who are not yet on ART range from 0.17
are even more in need of assistance, to 0.25
and homeopathy is able to help
in maintaining the health of these * t = t distribution df = degrees of freedom
patients by boosting their immunity All comparisons showed highly significant differences, using the Student's t test for statistical analysis.
as well as facilitating the recovery
from opportunistic infections. Interpretation: Comparisons between no treatment and treatment showed a significantly better evaluation of health
From the database information, irrespective of whether the patients were on ART and whether the scoring was by the patient or the practitioner.
a statistical analysis of practitioner Based on both patient and practitioner evaluations, prior to homeopathic treatment, patients not on ART scored a
and patient scores shows the higher overall evaluation of health.
following (see Table 7).
36 Homeopathy in practice Autumn 2011
This study provides
that as a result of
No. of patients Average score std dev
154 6.17 0.43
78 7.10 0.20
significant at 99.9% confidence levels
there is a significant
positive measure of
highly significant: t for 120 df
0.001 probability = 0 3.373
99.9% confidence interval 0.93
1.09 change in overall
No. of patients Average score std dev health scores
154 5.58 0.12
78 8.19 1.15
significant at 99.9% confidence levels
approx df 39
highly significant: t for 40 df
0.001 probability = 3.551 What can be inferred from these
99.9% confidence interval 2.61 statistics?
+/- 0.51 It is probable that patients on ART
range from 2.09
to 3.12 were experiencing some drug side
effects and therefore scored lower
at the start, prior to the homeo-
Treatment pathic treatment. However, once
No. of patients Average Score std dev
71 7.28 0.18 these patients had received homeo-
78 7.10 0.20 pathy, those on ART scored higher
0.18 0.00 overall – higher even than those
significant at 99.9% confidence levels not yet on ART. It may be that
approx df 147
after the homeopathic treatment,
highly significant: t for 120 df the side effects were alleviated and
0.001 probability = 3.373 the patients generally felt better.
99.9% confidence interval 0.18 This analysis was from a sample
+/- 0.08 of about 300 patients and gives
range from 0.10
to 0.26 a good indication that there is
a significant difference in patients
getting homeopathic treatment as
Treatment well as the standard ART.
No. of patients Average Score std dev
71 7.18 0.19 This study provides empirical
78 8.19 1.15 evidence that as a result of homeo-
1.01 0.65 pathic treatment there is a signifi-
significant at 99.9% confidence levels cant positive measure of change in
approx df 40 overall health scores.
highly significant: t for 40 df
0.001 probability = 3.551 Conclusions
99.9% confidence interval 1.01 As a result of maintaining simple
range from 0.65
and consistent records, the data-
to 1.37 base can be used to provide
answers at some level to all the
Comparison between patients on ART and not ART showed an interesting difference between original questions that were posed.
patient and practitioner scoring, where the patient scores show that patients on ART evaluated
their health at a higher level than the evaluations by the practitioners, to the extent that they Is homeopathy actually working?
scored higher than patients not on ART when looking at patient scores in contrast to the Yes: 60-65% of patients reported
reverse when looking at practitioner scores. improved overall health scores.
Are the patients actually showing
Homeopathy in practice Autumn 2011 37
Karnofsky performance status scale score: Our modified version, ‘Patient score’ in the cases, is
The Karnofsky score runs from 100 to 0, where 100 is ‘perfect’ Modified Karnofsky:
health and 0 is death. Although the score generally uses inter- 10 Normal, no complaints or evidence of disease (I am well)
vals of 10, a practitioner may choose decimals if he or she feels 9 Able to perform normal activity; minor signs and symp-
a patient’s situation holds somewhere between two marks. It toms of disease (I feel the problem only a little)
is named after Dr David A Karnofsky, who described the scale 8 Some signs and symptoms of disease (My problem is still
with Dr Joseph H Burchenal in 1949. there sometimes but not very bad)
Karnofsky score is a subjective measure of how well the 7 Able to perform normal activity with effort; signs and
patient is doing. symptoms of disease (I can work but my problem is
100 Normal, no complaints or evidence of disease always there)
90 Able to perform normal activity; minor signs and symptoms 6 Cares for self but difficulty performing normal activity or
of disease doing active work (The problem makes it difficult for me
80 Able to perform normal activity with effort; some signs to work)
and symptoms of disease 5 Requires occasional assistance but is able to care for
70 Cares for self, unable to perform normal activity or to do most of own needs (I can manage but the problem makes
active work me weak and prevents me from doing the things I need
60 Requires occasional assistance but is able to care for most to do)
of own needs 4 Partially disabled; able to walk only with assistance (I am
50 Requires considerable assistance and frequent medical care weak and need help to walk)
40 Requires special care and assistance; disabled 3 Severely disabled although death not imminent (I am
30 Hospitalisation indicated, although death not imminent; very, very weak)
severely disabled 2 Hospitalisation necessary; active supportive treatment
20 Hospitalisation necessary; active supportive treatment required, very sick (I cannot get up or walk)
required, very sick 1 Fatal processes progressing rapidly; dying state (I am very
10 Fatal processes progressing rapidly; dying state sick and weak, cannot eat or get up)
0 Dead. 0 Death.
38 Homeopathy in practice Autumn 2011
The project is not
claiming to cure
HIV / AIDS
Yes: 60-65% of symptoms showed How many of the patients have
improvement. tested HIV positive? Student’s t test
18% of patients have been tested
Are the improvements noted just HIV positive; however, 45% of The Student’s t test is a standard
a result of taking the case and hav- patients have not been tested at all. statistical analysis:
ing someone listen?
This is difficult to determine from Can the individual symptoms be T = t distribution
the data. From the case studies, it tracked to see if they have im- df = degrees of freedom
can be seen that many patients suf- proved, changed or aggravated?
fer from poverty and social prob- Yes, the database makes this easy. A t test is any statistical hypoth-
lems, and have no access to coun- esis test in which the test statis-
selling or other social services. Is it worth all the effort? tic follows a Student’s t distribu-
Spending time listening to the The practitioners have found that tion, if the null hypothesis is
patient is an important part of the the process of collecting data has supported. It is most commonly
healing process. For example, in improved their overall practice, applied when the test statistic
one instance, on being asked to because it provides a clear over- would follow a normal distribu-
give a score a patient commented: view of the day-to-day work of tion if the value of a scaling
When I came in I was definitely 5, but the project. term in the test statistic were
now that I have told you my story I am known. When the scaling term
10 out of 10(!), when the remedies Barbara Braun can be contacted at is unknown and is replaced by
had not even been prescribed yet. email@example.com. an estimate based on the data,
However, infants and small children the test statistic (under certain
who were not directly questioned conditions) follows a Student’s
showed similar improvements in t distribution.
symptoms and overall health.
Are enough of the patients report- Explicit expressions that can be
ing improvement? used to carry out various t tests
The improvement rates are high are given below. In each case,
enough to justify the continuation the formula for a test statistic
of free clinics. that either exactly follows or
closely approximates a t distri-
Which remedies are being bution under the null hypothesis
prescribed most frequently? is given. Also, the appropriate
Mainly tubercular remedies. degrees of freedom are given
in each case. Each of these
statistics can be used to carry
The improvement out either a one-tailed test or
a two-tailed test.
Once a t value is determined,
People infected with HIV lose
helper T lymphocytes (CD4+
rates are high a p value can be found using
a table of values from Student’s
t distribution. If the calculated
cells) in three phases over
months or years. A healthy
person has a CD4+ lymphocyte
enough to p value is below the threshold
chosen for statistical signifi-
cance (usually the 0.10, the
count of roughly 800 to 1300
cells per microlitre of blood. justify the 0.05, or 0.01 level), then the
null hypothesis is rejected
in favour of the alternative
(The Merck Manual of Medical
Information (1999, Home continuation of hypothesis.
Edition) 187, 1012 – 1013)
free clinics Student’s_t-test)
Homeopathy in practice Autumn 2011 39