Arch Gynecol Obstet (****) ***:*** ***

DOI **.***7/s00404-009-0988-1

O R I G I N A L A R T I CL E

Homeopathic treatment of patients with dysmenorrhea:

a prospective observational study with 2 years follow-up

Claudia M. Witt Rainer L dtke Stefan N. Willich

Received: 24 November 2008 / Accepted: 2 February 2009 / Published online: 20 February 2009

Springer-Verlag 2009

Abstract Abbreviations

Purpose Evaluating homeopathic treatment for dysmen- WHO World Health Organization

International ClassiWcation of Diseases

orrhea. ICD

Methods Prospective multicenter observational study in NRS Numerical rating scale

primary care, using standardized questionnaires to record QoL Health-related quality-of-life

for 2 years diseases, quality of life, medical history, consul- MOS SF-36 Medical outcomes trust 36-item short form

tations, all treatments, other health services use. survey instrument

Results Fifty-seven physicians treated 128 women (age KINDL KINDer Lebensqualit tsfragebogen

32.4 7.5 years, mean SD) and 11 girls (13.7 4.0). Cn nth centesimal potency

Women had dysmenorrhea for 11.6 9.0 (girls 3.1 1.5) Qn nth quinquagintamillesimal potency

years. Patients received 7.5 6.5 (5.9 3.7) homeopathic GP General practitioner

prescriptions. Diagnoses and complaints severity improved RCT Randomized controlled trial

markedly [at 24 months, dysmenorrhea relieved by > 50%

of baseline rating in 46.1% (59) of the women and 45.5%

(5) of the girls] with large eVect sizes (24 months: Cohen s Introduction

d from 1.18 to 2.93). In addition, QoL improved

(24 months: SF-36 physical component score: 0.25, mental Dysmenorrhea, either primary without associated organic

component score 0.25, KINDL sum score 0.27). Conven- disease or secondary, is the most frequent gynecological

tional medication changed little and use of other health ser- problem. With a high prevalence (18 81%, depending on

deWnition and the used survey method) [1] it causes consid-

vices decreased.

Conclusions Patients with dysmenorrhea improved under erable activity limitations and absenteeism from school or

homeopathic treatment. Controlled studies should investi- work [2]. The production of uterine prostaglandins that

gate eYcacy and eVectiveness. stimulate the contraction of the myometrium and cause

ischemia receive increasing attention for their role in its

Keywords Dysmenorrhea Homeopathy in usual care pathogenesis in primary and secondary dysmenorrhea, but

Prospective observational study other mechanisms can also be causative for the latter [2].

Treatment options depend on causes; they include simple

analgesics, NSAIDs, COX2 inhibitors (withdrawn in many

C. M. Witt S. N. Willich

countries), contraceptive hormones, levonorgestrel, and

Institute for Social Medicine,

surgery. Varying degrees of therapeutic eVectiveness as

Epidemiology and Health Economics,

well as side eVects cause patients to stop seeking medical

Charit University Medical Center, 10098 Berlin, Germany

e-mail: abpqa7@r.postjobfree.com help. Generally, dysmenorrhea has been seen as underdi-

agnosed and undertreated. [1, 2] The available comple-

R. L dtke

mentary therapies are often insuYciently researched [2],

Karl and Veronica Carstens-Foundation,

however, acupuncture is often successfully used in usual

Am Deimelsberg 36, 45276 Essen, Germany

123

604 Arch Gynecol Obstet (2009) 280:603 611

care and was found to be clinically eVective and cost-eVec- [16] (women, 17 years), and the KINDL [17, 18] (girls

tive [3]. Patients also use homeopathy, but no research has aged 7 16 years) questionnaires.

focused yet on its eVectiveness. The Wrst questionnaires were handed out by study physi-

Homeopathy is practiced in many regions of the world cians and completed before treatment. Patients sent them in

sealed envelopes directly to the study oYce, from where

[4], especially in high-income countries where it ranks the

most popular among traditional, complementary, or alterna- they received follow-up questionnaires at 3, 12, and

tive medicines [4 6]. In Germany, it is used in 83% of hos- 24 months, with every complaint being transferred to the

pital-associated clinics for gynecology or obstetrics [7]. A follow-up questionnaires to ensure continuous assessment.

diagnosis can be treated with diVerent remedies in diVerent At the same times (0, 3, 12 and 24 months), the participat-

patients ( individualization ), depending on varying side ing physicians recorded up to four diagnoses per patient and

symptoms. Homeopathic drugs ( remedies ) are produced assessed their severity on identical NRS. On a continuous

by alternating steps of diluting and agitating a starting sub- basis, they recorded the homeopathic treatment (which was

stance; the resulting potencies may reach dilutions mostly following the classical homeopathy style, see dis-

beyond Avogadro s number where the probability of even cussion), use of any conventional therapy, and all referrals.

As outcome measures, we deWned: the severity of the

one molecule of the starting substance being present

approaches zero. Such high potencies are often used, pain due to dysmenorrhea, mean severity of all baseline

however, their eVects constitute a subject of scientiWc con- diagnoses (pooled physician assessment), mean severity of

troversy [8]. Meta-analyses of placebo controlled trials all complaints (pooled patient assessment), and QoL scores.

Statistical analysis (using SAS/STAT v8.2 software) fol-

(pooling a great variety of diseases and ailments) have

shown inconsistent results [9, 10]. lowed the intention-to-treat approach: every included

We did a Wrst step and globally evaluated use and eVects patient entered Wnal analyses. We replaced missing values

of homeopathy under the conditions of usual care. For this, as follows: Cured complaints: severity = 0 in subsequent

we followed 3,981 patients over 2 years in a prospective records; deceased patients: severity = 10. The remaining

observational study [11 13]. This paper presents the sub- missing values were multiply imputed according to Rubin

[19]. Each was given Wve distinct, but plausible values,

group of 139 patients consulting a homeopathic physician

based on correlations with non-missing values and reXect-

because of dysmenorrhea of any etiology.

ing the overall variability of data. This generated a total of

Wve distinct complete data tables, each without any missing

Methods value. These were analyzed separately (see below), and the

results pooled to calculate treatment eVects and P values.

For each imputed data set, treatment eVects were esti-

In this prospective multi-center observational study,

patients were included consecutively upon their Wrst mated on the basis of a generalized multiple linear regres-

consultation with a participating physician (almost all sion model: In complete analogy to the recommendations

located in Germany and some in Switzerland, about 1% by Diggle et al. [20] we assumed the treatment course to be

of all certi Wed homeopathic physicians in Germany), mixed of a piecewise linear part (0 3 months and 3

regardless of prior homeopathic treatment elsewhere, or 24 months) and a quadratic term (starting at month 3). The

of any other criteria than age (>1 year) and informed serial correlation was assumed to be exponential with time.

EVect sizes were calculated by dividing treatment eVects as

consent. They were followed up for 24 months using

standardized questionnaires. This paper analyses the estimated above by baseline standard deviations (Cohen s

patients su Vering from dysmenorrhea (ICD-9: 625.3, d). Their absolute values were classiWed: as d > 0.8, large;

ICD-10: N94.6, painful menstruation [14]). Study 0.8 d > 0.5, medium; 0.5 d > 0.2, small; d 0.2

physicians were required to have passed certi Wed train- clinically not relevant. To test whether the QoL changes are

regression to the mean eVects, we applied Mee and Chua s

ing in classical homeopathy and 3 years of experience

in its practice (details of recruitment [12 ]). Written test [21] under the assumption that the patients had the

informed consent and approval by ethics review boards same QoL as the general German population [16].

were obtained.

Before treatment (at baseline), patients independently

from their physicians recorded the complaints that insti- Results

gated homeopathic treatment, and rated their severity on a

numeric rating scale (NRS, 0, cured; 10, maximum sever- We included 139 patients in the present analysis (Table 1),

ity) [15]. For girls ( 16 years), the parents provided medi- who were treated by 57 physicians (all in Germany except 5

cal information and severity ratings. The health-related patients of 3 Swiss physicians). Among them were 11 girls

quality-of-life (QoL) was recorded with the MOS SF-36 (7 16 years) whose QoL was evaluated with the KINDL

123

Arch Gynecol Obstet (2009) 280:603 611 605

lasted for at least 4.5 years in women and 3.1 years in girls

Table 1 Demographics and baseline status

(Table 2).

Women Girls

The consultations consisted of an extensive initial con-

sultation (Table 3), followed by the analysis of the case.

Sample

Almost all patients received the Wrst remedy on the same

Patients 128 11

Age (years)a day; administration to one woman each was delayed

32.4 7.5 13.7 4.0

1 week and 1 month, to 1 girl > 1 month. The subse-

10 years School 69.5% (89) /

quent consultations (8.4 8.2 by women, 6.9 5.8 by

Patient expected: homeopathy (%, n)

girls), about half of them telephone calls (3.9 5.4 and

Will help 71.1% (91) 72.7% (8)

2.9 2.9), were much shorter (Table 3). The last homeo-

Will maybe help 28.1% (36) 27.3% (3)

pathic medication was recorded for women after on average

Will not help 0.8% (1) 0% (0)

13.7 10.5 and for girls after 14.5 10.5 months. The

Baseline diagnoses

majority of patients (54% of the women, 63.7% of the girls)

Total, numbera 3.3 0.8 3.1 0.9

continued homeopathic care at study end (Table 3) or had

Severity (NRS)a 6.2 1.5 5.4 1.3

suspended it temporarily.

a

Chronic, number 3.2 0.8 3.1 0.9

Over the course of the study, the women had received on

Any baseline diagnosis pretreated (%, n)

average 7.5 6.5 homeopathic prescriptions, the girls

Any treatmentb 94.5% (121) 90.9% (10)

5.9 3.7.

Medication 78.9% (101) 72.7% (8)

The most frequent prescribed remedies in all patients

Surgery 21.9% (28) 9.1% (1)

were (identiWed with traditional abbreviations): sep 14.1%;

Other 55.5% (71) 54.5% (6)

nat-m 6.7%; puls 6.6%; phos 5.3%; calc 5.1%; sulph 5.1%;

NRS numerical rating scale: 10 maximum, 0 cured nux-v 4.5%; lyc3.9%; sil 2.8%; carc 2.7%. This means that

a

Mean SD more than half of all prescriptions were covered by 10

b

homeopathic remedies, but in total, 108 diVerent remedies

Excluding homeopathy

were applied in women, 26 remedies in girls, which sup-

questionnaire. All patients suVered from dysmenorrhea that ports the claim of individualized prescriptions in homeopa-

had lasted for 11.6 9.0 (mean SD) years (women) and thy. The most used potencies in all patients were: c200

3.1 1.5 years (girls) (Table 1). Almost all accompanying (35.0%), c1000 (25.8%), c30 (12.4%), c10000 (8.7%), d12

diagnoses assessed at baseline were chronic diseases that (3.0%), q6 (2.5%), q1 (2.1%), and c12 (1.9%). In total,

usually had been under mostly conventional treatment 89.4% of the used potencies implied a dilution above

before (Tables 1, 2), the most frequently recorded had Avogadro s number.

Table 2 Baseline diagnoses

ICD-10 code Patients (%, n) Severity (NRS) Duration (years)

Women

Dysmenorrhea N94.6 100.0% (128) 6.5 1.8 11.6 9.0

Headache R51 14.1% (18) 5.6 1.5 7.8 6.5

Frequent infections R68.8 9.4% (12) 6.1 1.7 7.6 8.2

Chronic sinusitis J32.9 7.8% (10) 7.0 1.4 12.6 11.5

Premenstrual tension N94.3 7.8% (10) 7.1 1.6 12.4 8.1

Dermatitis L30.9 7.8% (10) 4.9 1.9 5.9 7.6

Migraine G43.9 7.8% (10) 6.7 2.0 14.6 9.9

Allergy T78.4 7.0% (9) 6.9 2.3 10.5 9.3

Sleep disturbance G47.9 6.3% (8) 7.0 2.1 5.4 5.4

Chronic rhinitis J31.0 5.5% (7) 6.0 2.2 4.5 2.5

Girls

Dysmenorrhea N94.6 100.0% (11) 6.0 1.8 3.1 1.5

NRS numerical rating scale: 10

maximum, 0 cured. Only diag- Headache R51 27.3% (3) 5.0 3.0 3.0 1.7

noses seen in 5% of the women Frequent infections R68.8 18.2% (2) 4.0 1.4 5.5 4.9

or 2 girls

123

606 Arch Gynecol Obstet (2009) 280:603 611

Discussion

Table 3 Consultations and continuance of homeopathic treatment at

study end

This prospective multicenter observational study was

Women Girls

aimed to give an unbiased representation of contemporary

Consultations (mean SD) homeopathic health care and its outcome in 128 dysmenor-

1st consultation (min) 116-**-***-** rhea patients. Assessments of disease severity and health-

Case analysis (min) 43 41 31 27 related quality-of-life (QoL) consistently showed substantial

Follow-ups number, all 8.4 8.2 6.9 5.8 improvements, although the disease was long-standing,

Telephone 3.9 5.4 2.9 2.9 chronic, and conventionally treated. Similarly, all accompa-

Practice 4.1 5.2 4.0 3.9 nying diseases (almost all chronic) (Table 2) were mark-

edly ameliorated. The major improvements were seen

FUs duration (min), all 20.1 15.0 17.2 6.0

within the Wrst 3 months of homeopathic treatment. QoL

Telephone 7.5 4.8 6.6 3.4

increased with the severity improvements, conventional

Practice 29.5 14.7 24.8 7.6

medication changed little, and the use of health-care ser-

FUs cumulated (min), all 190.6 176.4 121.3 82.5

vices decreased.

Telephone 45.3 73.0 24.1 17.3

The methodological strengths of our study include the

Practice 162.9 160.9 102.0 81.6

consecutive patient enrollment and the use of standardized

Last consultation (month) 15.6 9.5 17.2 9.8

outcome instruments. The participation of about 1% of all

Homeopathy at study end

certiWed homeopathic physicians in Germany (=14% of the

Treatment ongoing 35.2% (45) 36.4% (4)

members of an association for physicians practicing classi-

Changed homeopath 0.8% (1) 0% (0)

cal homeopathy, the Hahnemann Association) in the main

Currently not treated 18.0% (23) 27.3% (3)

study makes the study and the subgroup presented in this

Treatment ended because of

paper a representative sample for contemporary homeo-

Cure or amelioration 3.1% (4) 0.0% (0)

pathic practice. For quality assurance purposes, we avoided

Reason outcome-unrelated 3.9% (5) 0.0% (0)

selecting a random sample of homeopathic physicians,

No eVect or aggravation 21.1% (27) 18.2% (2)

choosing instead to recruit physicians trained and certiWed

Not stated reason 1.6% (2) 0.0% (0)

in classical homeopathy. Our results are, therefore, repre-

No answer to treatment status 16.4% (21) 18.2% (2)

sentative only for the classical type of homeopathy, which

is the type of homeopathy that is accepted and certiWed by

the German Medical Association. In contrast to randomized

trials, our study describes patients from everyday practice

The strongest improvement of the severity of diagnoses with multiple morbidities and a large variety of life styles.

and medical complaints was seen in the Wrst 3 months, it This ensures a high degree of external validity that allows

generally continued during the full observation period extrapolation to usual medical care. Our study was

(Tables 4, 5), only dysmenorrhea in girls temporarily designed to evaluate homeopathic treatment in patients

suVering from various diagnoses. This disallowed the use

relapsed during months 4 12. Physicians severity assess-

of a more complex disease-speciWc measurement instru-

ments tended to be more positive than patients assess-

ments. All severity changes since baseline were of large ment. We used a numeric rating scale which is validated,

eVect size (at 24 months 1.18 2.93), whereas improve- often used [15] and accepted to measure pain. In addition,

ments in health-related QoL were small or medium (at we used generic QoL questionnaires.

24 months SF-36 physical component score 0.25, mental As a general observation, especially for industrialized

component score 0.52, KINDL 0.27). Mee Chua tests for countries, homeopathic patients tend to be younger and bet-

the SF-36 conWrmed a treatment eVect for physical compo- ter educated than conventional patients, of higher socioeco-

nent score after 12 and 24 months (0.0088, and 0.0017, at nomic status, and more often female [22]. These factors

3 months P = 0.0635), and for the mental component score could be indicative for a health-awareness above average

after 3 months (P = 0.0417, after 12 and 24 months 0.5650 and an inclination to self-treatment for lesser ailments [23].

and 0.4061). As a result, accompanying chronic diseases were strongly

After 24 months, the dysmenorrhea was relieved predominant in our study, as was seen in other observations

by > 50% of the baseline rating in 46.1% (59) of the [23 27]. Additionally, waiting lists of sometimes several

women and 45.5% (5) of the girls and the other baseline months would preclude the shorter periods of acute ill-

diagnoses were considerably relieved (Table 6). Conven- nesses, and the reputation of homeopathy as a medicine for

the whole person (reXected in the extensive initial case

tional medication changed little, and the use of health-care

services decreased (Table 7). taking) may cause a self-selection of patients seeking more

123

Table 4 Diagnoses, complaints, quality of life (estimated means and 95% conWdence intervals from the statistical model)

Status (95% CI) Change (95% CI)

Month 0 Month 3 Month 12 Month 24 Months 0 3 Months 0 12 Months 0 24

Disease severity

(NRS)

Dysmenorrheaa Women 6.51 3.64 2.58 1.89 2.86*** 3.92*** 4.62***

Arch Gynecol Obstet (2009) 280:603 611

(6.12; 6.89) (3.26; 4.03) (2.20; 2.97) (1.50; 2.27) ( 3.23; 2.50) ( 4.38; 3.47) ( 5.12; 4.12)

Girls 6.00 3.64 4.60 1.72 2.36* 1.40 4.28**

(3.89; 8.11) (1.70; 5.57) (2.66; 6.53) ( 0.22; 3.65) ( 4.55; 0.17) ( 3.96; 1.15) ( 6.96; 1.61)

All diagnoses Women 6.19 3.74 2.50 1.73 2.45*** 3.69*** 4.46***

(mean)a (5.90; 6.48) (3.45; 4.03) (2.21; 2.79) (1.44; 2.02) ( 2.72; 2.19) ( 4.02; 3.35) ( 4.82; 4.09

Girls 5.37 3.09 3.24 1.34 2.28*** 2.13** 4.03***

(4.47; 6.28) (2.26; 3.92) (2.41; 4.07) (0.51; 2.17) ( 3.19; 1.37) ( 3.20; 1.06) ( 5.16; 2.90)

All complaints Women 6.39 3.71 3.48 2.99 2.67*** 2.91*** 3.40***

(mean)b (5.98; 6.80) (3.27; 4.16) (3.10; 3.87) (2.62; 3.36) ( 3.11; 2.24) ( 3.43; 2.38) ( 3.95; 2.85)

Girls 6.97 2.46 4.07 4.88 4.51* 2.89 2.09

(4.56; 9.37) ( 0.34; 5.25) (2.48; 5.67) (2.84; 6.92) ( 8.26; 0.76) ( 5.69; 0.09) ( 5.04; 0.87)

Quality of lifeb

SF-36 Physical Women 47.48 48.97 50.29 50.60 1.48* 2.80** 3.11*

component score (45.29; 49.68) (46.72; 51.21) (48.13; 52.44) (48.48; 52.71) (0.27; 2.69) (0.86; 4.74) (0.68; 5.54)

Mental component Women 37.41 43.89 43.78 44.06 6.48*** 6.37*** 6.64***

score (35.06; 39.76) (41.39; 46.38) (41.53; 46.04) (41.88; 46.23) (4.43; 8.53) (3.57; 9.17) (3.53; 9.76)

KINDL sum score Girls 63.14 65.70 65.44 65.68 2.56 2.30 2.54

(56.00; 70.28) (56.67; 74.72) (60.05; 70.83) (58.69; 72.66) ( 7.69; 12.80) ( 5.79; 10.40) ( 7.22; 12.30)

NRS numerical rating scale: 10 maximum, 0 cured. Quality of life: higher values = better; CI 95% = 95% conWdence interval

a

Physicians answers

b

Patients answers

* P

** P 0.8,

Girls 2.56* (4.68; 0.43) 1.64 (3.23; 0.05) 1.18 (2.86; 0.49)

large; 0.8 d > 0.5, medium;

Quality of lifeb

0.5 d > 0.2, small; d 0.2

clinically not relevant. CI SF-36 physical component score

95% = 95% conWdence interval

Women 0.12* (0.02; 0.22) 0.23** (0.07; 0.39) 0.25* (0.06; 0.45)

a

Physicians answers

Mental component score

b

Patients answers

Women 0.50*** (0.34; 0.66) 0.50*** (0.28; 0.71) 0.52*** (0.27; 0.76)

* P 10% 1.6% (2) 0.0% (0)

[32], but might be compensated by their low frequency. On

Responders, all diagnoses (diagnoses, percent, n)

average, conventional consultations take place about 24

Total 298 23

times in 24 months per patient with a resulting doctor

Fully cured 33.9% (101) 43.5% (10)

workload of about 190 min in 2 years [33].

Better by 50% baseline 27.5% (82) 26.1% (6)

Our study focused on the widespread individualizing

Better by 10% 10% 1.7% (5) 0.0% (0)

lars, remedies were selected for symptoms both typical of

the diagnoses and outside the predominating pathologies

than a quick Wx for a single issue. That other diagnoses ( constitutional ). The broad variety of chosen remedies,

besides dysmenorrhea were ameliorated as well in our and the similar frequencies of the leading remedies in dys-

patients supports the whole person approach of the menorrhea treatment and the overall observational study

observed treatment. Besides dysmenorrhea, the patients [11] support this impression. The predominant use of high

suVered from headache/migraine, allergies, and eczema. potencies is also typical for this line of homeopathy.

Those diagnoses were also observed among the most fre- According to the predominant opinion in the homeo-

pathic community, a longer period until a clear eVect would

quent in other homeopathic observational studies [26, 28].

The long duration of the diseases was also observed in ear- become noticeable was to be expected. The 3-months

lier studies [24, 28, 29]. improvements might include several aspects and could be

inXuenced by changed life style, reduced conventional

The latter, and the high rate of previously treated

medications, or reXect context eVects induced by expecta-

patients might indicate that patients turn to homeopathy

after Wnding conventional care not satisfactory for their tions from the waiting list time, or the long and detailed

123

Arch Gynecol Obstet (2009) 280:603 611 609

Table 7 Use of other treatment

Baseline (%, n) 3 months (%, n) 12 months (%, n) 24 months (%, n)

and health care services

Patients using conventional drugsb

ATC-class G (genito-urinary system and sex hormones)

Women 11.7 (15) 5.5 (7) 10.2 (13) 5.5% (7)

Girls 9.1 (1) 9.1 (1) 0.0 (0) 18.2% (2)

ATC-class H (systemic hormones)

Women 12.5 (16) 12.5 (16) 14.1 (18) 11.7% (15)

Girls 0.0 (0) 0.0 (0) 0.0 (0) 0.0% (0)

Analgetics

Women 8.6 (11) 7.0 (9) 7.8 (10) 8.6 (11)

Girls 9.1 (1) 9.1 (1) 9.1 (1) 9.1 (1)

Baseline 0 3 months >3 12 months >12 24 months

Patients consulting other health-care providersb

Any physiciana

Women 97.7% (125) 39.8% (51) 71.9% (92) 79.7% (102)

Girls 100.0% (11) 54.5% (6) 72.7% (8) 81.8% (9)

Data related to dysmenorrhea

Gynecologist

only

Women 62.5% (80) 19.5% (25) 43.8% (56) 51.6% (66)

a

Data related to treatments for

Girls 36.4% (4) 9.1% (1) 27.3% (3) 45.5% (5)

all diagnoses as well as routine

checks (e.g., dentist, gynecolo- CAM treatmentsa

gist)

Women 21.1% (27) 3.1% (4) 6.3% (8) 10.2% (13)

b

Patients answers

Girls 45.5% (5) 0.0% (0) 9.1% (1) 18.2% (2)

Multiple answers possible

eVects requires the mean of the target population to be

initial case taking. Future research should look into these

(and other) contexts of the treatment, e.g., searching for known or plausibly assumed. The observed QoL improve-

correlations between time spent in consultations and the ments are unlikely to have been caused by regression

toward the mean. They were signiWcantly greater than

subsequently observable improvement, or vary consultation

patterns systematically. could be expected and assuming chronically ill patients

The eVect size of the severity ratings after 12 and with often several severe diseases to have the same QoL as

24 months was large. This may be partly explained by pla- the general German population was itself an extremely con-

cebo and/or regression to the mean eVects that our study servative approach. Moreover, patients received homeo-

was not designed to control (eVect sizes in between-group pathic treatment after years of other treatment and a waiting

comparisons are usually smaller). We also cannot rule out period regression toward the mean would long have faded

overestimation of the treatment eVect. The QoL improve- out by then. The reduction in conventional or alternative

ments, on the other hand, may have been greater than medication and treatments also may not be due to the

recorded: The SF-36 is unlikely to overestimate changes; homeopathic remedies alone. Homeopathic physicians are

its mental scales have even been found to be less sensitive known to use conventional means with a certain hesitation,

than the mental und social scales of other instruments such thus functioning as a kind of gatekeeper .

as the Duke Health ProWle. [24] The version of KINDL that Primary dysmenorrhea under homeopathy has not been

was available at study time has been updated by its authors researched so far. Studies on endometriosis mention its allevi-

since to correct its insensitivity to change [34]. We can only ation [36 38], but this area presents special challenges [39,

speculate that the new version would have reported much 40]. A study on premenstrual tension syndrome [41] with

greater eVects, as would a diagnosis-speciWc tool to mea- 54.5% of the patients being permanently relieved by >70% of

sure the psycho-social part in dysmenorrhea suVering. baseline severity and 36.4% by 30 70% (placebo 12.5 and

Usually, patients seek treatment when their health is out 25%) does not mention dysmenorrhea explicitly.

Our study does not support conclusions as to the eVec-

of average (such as severe pain, low QoL, and so on). A

natural alleviation of their diseases (regression to the mean) tiveness of the homeopathic remedies, because no method-

can be mistaken for an eVect of the beginning treatment ology for this purpose (control group, randomization,

[35]. Separating regression to the mean from treatment blinding) was built into its design and patients could use

123

610 Arch Gynecol Obstet (2009) 280:603 611

additional conventional therapies. Further research is clinics practicing gynecology and obstetrics in Germany. Results

clearly warranted to explore the eVects of homeopathic care of a questionnaire survey. Gynecol Obstet Invest 55:73 81

8. Ernst E (1998) Are highly dilute homoeopathic remedies place-

on dysmenorrhea of speciWc etiologies. It should include bos? Perfusion 11:291 292

objective data, diagnosis-speciWc instruments, and special- 9. Linde K, Clausius N, Ramirez G, Melchart D, Eitel F, Hedges L,

Jonas W (1997) Are the eVects of homoeopathy placebo eVects? A

ized physicians should be involved. The aim of our study

meta-analysis of randomized, placebo controlled trials. Lancet

was to provide for the Wrst time systematic and detailed

350:834 843

information about status and eVects of homeopathic medi- 10. Shang A, Huwiler-M ntener K, Nartey L, J ni P, D rig S, Sterne

cal care in routine medical practice. These data should JAC, Pewsner D, Egger M (2005) Are the clinical eVects of

homoeopathy placebo eVects? Comparative study of placebo-con-

build a good basis for the planning of further research pro-

jects on homeopathy, which could include speciWc instru- trolled trials of homoeopathy and allopathy. Lancet 366:726 732

11. Becker-Witt C, L dtke R, Wei huhn TER, Willich SN (2004)

ments and control groups. Diagnoses and treatment in homeopathic medical practice. Forsch

Komplement rmed Klass Naturheilkd 11:98 103

12. Witt CM, L dtke R, Baur R, Willich S (2005) Homeopathic

medical practice: long-term results of a cohort study with 3981

Conclusions

patients. BMC Public Health 5:115

Witt C, L dtke R, Willich SN (2005) EVect size in patients treated

13.

The patients in our study suVered from long-standing dys- by homeopathy diVers according to diagnosis results of an

menorrhea and other chronic diseases. Under homeopathic observational study. Perfusion 18:356 360

14. World Health Organization (WHO) (2008) 2008 ICD-9-CM Diag-

treatment the severity of the diseases and the QoL

nosis 625.3 [Dysmenorrhea]. Alkaline Software Inc, Fort Erie,

improved substantially, which supports the whole person Canada. http://www.icd9data.com/2007/Volume1/580-629/617-

approach prevailing in contemporary homeopathy, and 629/625/625.3.htm

mostly in the Wrst 3 months. The data represent a good basis 15. Huskisson E, Scott J (1993) VAS Visuelle Analog-Skalen; VAPS

Visual Analogue Pain Scales; NRS Numerische Rating-Skalen;

for the planning of further research projects on homeopa-

Mod. Kategorialskalen. In: WesthoV G (ed) Handbuch psychoso-

thy. zialer Me instrumente ein Kompendium f r epidemiologische

und klinische Forschung zu chronischer Krankheit. Hogrefe

Acknowledgments We want to thank the participating physicians G ttingen, pp 881-885

for their work and the patients for their cooperation. We thank Elvira 16. Bullinger M, Kirchberger I (1998) SF-36 Fragebogen zum

Kr ger for data acquisition and to Karin Weber and Katja Wruck for Gesundheitszustand Handanweisung. Hogrefe, G ttingen

data management. This work was supported with a grant by the Karl 17. Bullinger M, von Mackensen S, Kirchberger I (1994) KINDL

und Veronica Carstens-Foundation, D-Essen, for SNW and CMW. ein Fragebogen zur Erfassung der gesundheitsbezogenen

Lebensqualit t von Kindern. Zeitschrift f r Gesundheitspsycholo-

ConXict of interest statement None. gie:64 70

18. Ravens-Sieberer U, Bullinger M (1998) Assessing health-related

quality of life in chronically ill children with the German KINDL:

Wrst psychometric and content analytical results. Qual Life Res

References 7:399 407

19. Rubin DB (1987) Multiple imputations for nonresponse in sur-

1. Latthe PML, Say L, Gulmezoglu M, Khan KS (2006) WHO sys- veys. Wiley, New York

tematic review of prevalence of chronic pelvic pain: a neglected 20. Diggle P, Liang K, Zeger S (1994) Analysis of longitudinal data.

reproductive health morbidity. BMC Public Health 6:177 Clarendon Press, Oxford

2. Proctor M, Farquhar C (2006) Diagnosis and management of 21. Mee R, Chua T (1991) Regression toward the mean and the paired

dysmenorrhoea. Br Med J 332:1134 1138 sample t test. Am Stat 45:39 42

3. Witt CM, Reinhold T, Brinkhaus B, Roll S, Jena S, Willich SN 22. Marstedt G, Moebus S (2002) Gesundheitsberichtserstattung des

(2008) Acupuncture in patients with dysmenorrhea: a randomized Bundes Inanspruchnahme alternativer Methoden in der Medizin.

study on clinical eVectiveness and cost-eVectiveness in usual care. Robert Koch Institut, Berlin

HoVmann S (2007) Diagnosespektrum in hom opathischen

Am J Obstet Gynecol 166:e161 e168 23.

4. Ong CK, Bodeker G, Grundy C, Burford G, Shein K (2005) WHO Arztpraxen eine prospektive Beobachtungsstudie und ein Vergl-

Global Atlas of Traditional, Complementary and Alternative eich zu konventionellen Arztpraxen. Medizinische Fakult t,

Medicine. World Health Organization. Centre for Health Develop- vol Dr med. Charit, Universit tsmedizin Berlin, Berlin

ment, Kobe 24. Anelli M, Scheepers L, Sermeus G, Van Wassenhoven MH (2002)

5. Eisenberg D, Davis R, Ettner S, Appel S, Wilkey S, Van Rompay Homeopathy and health related quality of life: a survey in six

M, Kessler R (1998) Trends in alternative medicine use in the European countries. Homeopathy 91:18 21

United States, 1990 1997: results of a follow-up national survey. 25. Fortes L, Fraiz IC (2002) Homoeopathy from the patient s stand-

J Am Med Assoc 280:1569 1575 point: an empirical study in the City of Curitiba (Brazil), 1998

6. H rtel U, Volger E (2003) Inanspruchnahme und Akzeptanz von 1999. In: Dinges M (ed) Patients in the history of homoeopathy.

klassischen Naturheilverfahren und alternativen Heilmethoden in European Association for the History of Medicine and Health Pub-

lications, SheYeld, pp 301 316

Deutschland: Ergebnisse einer repr sentativen Bev lkerungsstu-

die. Das Gesundheitswesen 65:A35 26. Jacobs J, Chapman E, Crothers D (1998) Patient characteristics

7. Beer A-M, Ostermann T (2003) On the use of classical naturop- and practice patterns of physicians using homeopathy. Arch Fam

athy and complementary medicine procedures in hospitals and Med 7:537 540

123

Arch Gynecol Obstet (2009) 280:603 611 611

34. Ravens-Sieberer U, Bullinger M (2007) KINDLR Homepage :

27. Kerek-Bodden H, Koch H, Brenner G, Flatten G (2000) Diagnos-

espektrum und Behandlungsaufwand des allgemein rztlichen Pat- Revised Children Quality of Life Questionnaire. vol 2007. http://

ientenklientels. Ergebnisse des ADT-Panels des Zentralinstituts www.kindl.org/indexE.html

f r die kassen rztliche Versorgung. Z rztl Fortbild Qualit tssich 35. Barnett AG, Van der Pols JC, Dobson AJ (2005) Regression to the

94:21 30 mean: what it is and how to deal with it. Int J Epidemiol 34:215 220

28. Spence DS, Thompson EA, Barron SJ (2005) Homeopathic treat- 36. Schantz M (2005) Klassisch-Hom opathische Therapie bei Endo-

ment for chronic disease: a 6-year, University Hospital outpatient metriose. AHZ 250:117 126

observational study. J Altern Complement Med 11:793 798 37. Hunton M (1993) Endometriosis and homoeopathy. An audit

29. Sharples F, van Haselen R (1998) Patients perspective on using a study of 8 consecutive patients. Br Hom J 82:92 96

complementary medicine approach to their health: a survey at the 38. Gerhard I, Schantz M, Sillem M, Runnebaum B (1998) Hom o-

Royal London Homoeopathic Hospital. NHS Trust pathische Behandlung der Endometriose. Arch Gynecol Obstet

30. Jain A (2003) Does homeopathy reduce the cost of conventional 261(suppl 1):181

drug prescribing? A study of comparative prescribing costs in 39. Schm ck M, Gerhard I, L dtke R, Seidenabel P (1995) Behand-

General Practice. Homeopathy 92:71 76 lung von Zyklusst rungen mit Hom opathie und die Problematik

31. Van Wassenhoven M, Ives G (2004) An observational study of des Wirksamkeitsnachweises mittels einer randomisierten place-

patients receiving homeopathic treatment. Homeopathy 93:3 11 bokontrollierten Doppelblindstudie. EHK 44:556 562

32. Deveugele M, Derese A, van den Brink-Muinen A, Bensingand J, 40. L dtke R, Schm ck M, Gerhard I (1997) Methodische berlegungen

De Maeseneer J (2002) Consultation length in general practice: zum Wirksamkeitsnachweis der hom opathischen Einzelmittelbe-

cross sectional study in six European countries. Br Med J handlung von Zyklusst rungen. Forsch Komplement rmed 4:28 32

325:472 477 41. Yakir M, Kreitler S, Brzezinski A, Vithoulkas G, Oberbaum M,

Bentwich Z (2001) EVects of homeopathic treatment in women

33. Bahrs O (2003) Mein Hausarzt hat Zeit f r mich Wunsch und

Wirklichkeit. GGW 1:17 23 with premenstrual syndrome: a pilot study. Br Hom J 90:148 153

123

Contact this candidate