Get documents about "Homeopathy in childhood asthma
Description
Homeopathy in childhood asthma
Document Sample
Downloaded from thorax.bmj.com on 7 March 2007
Homeopathy in childhood asthma
A M Li, A Bush, N M Wilson, F Dantas, S B Brien, G Lewith, P Fisher, K Chatfield, R
Mathie, R Leckridge and A White
Thorax 2003;58;826-828
doi:10.1136/thorax.58.9.826
Updated information and services can be found at:
http://thorax.bmj.com/cgi/content/full/58/9/826
These include:
References This article cites 6 articles, 2 of which can be accessed free at:
http://thorax.bmj.com/cgi/content/full/58/9/826#BIBL
2 online articles that cite this article can be accessed at:
http://thorax.bmj.com/cgi/content/full/58/9/826#otherarticles
Email alerting Receive free email alerts when new articles cite this article - sign up in the box at the
service top right corner of the article
Topic collections Articles on similar topics can be found in the following collections
Complementary Medicine (265 articles)
Asthma (1209 articles)
Children (1806 articles)
Notes
To order reprints of this article go to:
http://www.bmjjournals.com/cgi/reprintform
To subscribe to Thorax go to:
http://www.bmjjournals.com/subscriptions/
Downloaded from thorax.bmj.com on 7 March 2007
Thorax 2003;58:825–828 825
PostScript ..............................................................................................
data, and chest radiographic findings are rou- as to the further diagnosis of the unclassified
LETTERS TO THE EDITOR tinely documented in a database. The inclu- subgroup (fig 1). Our results also show that
sion criteria were symptoms of COPD (chronic stage I disease (FEV1/FVC <70% and FEV1
cough with chronic sputum production for >80% predicted) was very rare, constituting
more than 2 years) and radiographic findings only 4–5% of the patients. This indicates that
If you have a burning desire to respond to of COPD (hyperinflation, diaphragmatic flat- the distribution of the stages, especially stage
a paper published in Thorax, why not tening). Patients with a history of asthma I, is inhomogeneous.
make use of our “rapid response” option? (variability of spirometric parameters, im- Despite its retrospective design, this study
provement in forced expiratory volume in 1 was strengthened by the fact that lung
Log on to our website
second (FEV1) of >20% after inhalation of β2 function data, chest radiographic findings,
(www.thoraxjnl.com), find the paper that
agonists, symptoms predominantly at night, and the results of a standard clinical examina-
interests you, and send your response via tion were available for all patients. It therefore
seasonal allergies, allergic rhinitis, or eczema)
email by clicking on the “eLetters” option were excluded from the study, as were those offers the chance to investigate the clinical
in the box at the top right hand corner. in whom FEV1 and forced vital capacity (FVC) impact of the GOLD classification, especially
Providing it isn’t libellous or obscene, it differed by more than 5% according to the in patients with mild COPD.
will be posted within seven days. You can American Thoracic Society (ATS) guidelines2 Our study therefore suggests that GOLD
retrieve it by clicking on “read eLetters” and patients with an abnormal chest radio- criteria miss an important subgroup of
on our homepage. graph or chronic cough caused by a disease patients with clinically diagnosed COPD,
The editors will decide as before other than COPD. which reduces its usefulness as a clinical tool.
whether to also publish it in a future FEV1 and FEV1/FVC were determined three D Köhler
paper issue. times. The predicted values for FEV1 were Krankenhaus Kloster Grafschaft, D-57392
taken from the European Respiratory Society Schmallenberg, Germany
(ERS) guidelines.3 The individual values of
FEV1 and FEV1/FVC for all patients are shown J Fischer, F Raschke
in fig 1. Almost 14% of patients clinically Klinik Norderney, D-26548 Norderney, Germany
Usefulness of GOLD diagnosed as having COPD could not be clas- B Schönhofer
classification of COPD severity sified because they had an FEV1/FVC ratio of Abteilung für Pneumologie und Intensivmedizin,
>70%, despite having a reduced FEV1 (<80% Klinikum Hannover, Podbielskistrasse 380, 30659
In 2001 the US National Heart, Lung and predicted). This combination is not repre- Hannover, Germany
Blood Institute (NHLBI) and the World sented in the GOLD classification. Less than
Health Organization announced guidelines 5% of all patients were classified as GOLD Correspondence to: Dr D Köhler, Krankenhaus
for the diagnosis, management, and treat- stage I. Kloster Grafschaft, D-57392 Schmallenberg,
ment of COPD (Global Initiative for Chronic The finding that the GOLD classification Germany; D.Koehler@fkkg.de
Obstructive Lung Disease, GOLD).1 One key missed an important subgroup of patients References
aspect of these guidelines is that COPD is with mild COPD challenges any proposed 1 Pauwels RA, Buist AS, Calverley PM, et al.
classified by severity into five stages which advantage of this classification scheme over Global strategy for the diagnosis,
constitute the basis of treatment recommen- existing guidelines from the ATS4 and ERS.5 management, and prevention of chronic
dations. However, to date there has been little Only six patients not classified as having obstructive pulmonary disease. NHLBI/WHO
evidence for the usefulness of these severity COPD by GOLD were missed using the ATS Global Initiative for Chronic Obstructive Lung
stages. criteria (stage I: FEV1 >50%) and ERS criteria Disease (GOLD) Workshop summary. Am J
Respir Crit Care Med 2001;163:1256–76.
We retrospectively reviewed 1000 patients (mild: FEV1 <70% and FEV1/VC >88% for 2 American Thoracic Society. Standardization
with COPD diagnosed clinically in 2001; 500 men and >89% for women). Obviously, any of spirometry, 1994 update. Am J Respir Crit
patients originated from a pulmonary reha- arbitrary classification of a continuous vari- Care Med 1995;152:1107–36.
bilitation hospital. Patients’ symptoms (based able such as FEV1 and FEV1/FVC results in a 3 Quanjer PH, Dalhuijsen A, Van Zomeren BC.
on a standardised interview), findings of a borderline group of patients. The GOLD Summary equations of reference values. Bull
standardised lung examination, lung function classification, however, provides no guidance Eur Physiopathol Respir 1983;19:45–51.
4 American Thoracic Society. Standards for
the diagnosis and care of patients with
chronic obstructive pulmonary disease. Am J
100 Respir Crit Care Med 1995;152:S77–121.
Not classified 0 5 Siafakas NM, Vermeire P, Pride NB, et al.
Optimal assessment and management of
90 chronic obstructive pulmonary disease
(COPD). The European Respiratory Society
Task Force. Eur Respir J 1995;8:1398–420.
80
Sahaja yoga in asthma
70 Since the publication of our paper on Sahaja
FEV1 /FVC (%)
yoga in the management of moderate to
severe asthma1 we have received a large
60 number of enquiries. One issue that has been
raised about the technique used in the study
warrants clarification and further acknowl-
50 edgement.
The Sahaja yoga meditation technique used
in the study was not developed by the authors.
40 The technique was taught to subjects in the
intervention group by experienced Sahaja
30 III IIb IIa I
yoga practitioners free of charge. The tech-
n = 1000 nique itself was developed by yoga expert H H
Shri Mataji Nirmala Devi and she permitted
20 the investigators to conduct the study on the
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 following reasonable conditions: (1) that no
FEV1 (% predicted) part of the technique be misrepresented, mis-
appropriated or commercialised by the inves-
Figure 1 Plot of % predicted forced expiratory volume in 1 second (FEV1) against the ratio of tigators; (2) that the founder and practition-
FEV1 to forced vital capacity (FVC) (%) of the total population (n=1000). ers of the process be appropriately
www.thoraxjnl.com
Downloaded from thorax.bmj.com on 7 March 2007
826 PostScript
acknowledged as the true source and custodi- on the homeopathic remedy, and to return treatments on quality of life are likely to
ans of the technique and its associated again 4 weeks later (visit 2) to assess the involve relatively small effect sizes even when
knowledge; and (3) that it be made clear that response after stopping the remedy. The one treatment is clearly superior.4
the Sahaja yoga technique is, as a matter of spirometric test results of one patient from My paper on the safety of homeopathy5 is
policy and philosophical conviction, always the first and second visits were missing. misquoted; it does not in any way imply that
made available free of charge. No side effects were reported and all the rate of exacerbations is a “hallmark of suc-
The authors sincerely regret any misunder- subjects were compliant with the homeo- cessful treatment”. Instead, I stated that “one
standing that may have led readers or pathic remedy. Using the Wilcoxon signed needs to consider the way practitioners are
members of the public to believe otherwise. ranks test, there was no significant difference informing patients of the possibility of such
They sincerely and gratefully acknowledge at baseline and at visits 1 and 2 in FEV1 (86% aggravations after using homeopathic medi-
the important and crucial role played by HH (interquartile range (IQR) 81.1–93.3) v 89% cines, thus creating some expectations that
Shri Mataji Nirmala Devi and the Sahaja yoga (85.0–100.0) v 85% (74.0–89.0), respectively) will fulfil what was said in the consultation”.
practitioners of Australia in the execution of and eNO (54 ppb (IQR 36.2–99.6) v 68 ppb Finally, I cannot agree with the statement
this study, and sincerely regret not having (37.0–87.0) v 76 ppb (43.6–131.4), respec- that the trial was designed with the input of
made more appropriate acknowledgements in tively). This could be because of the small experienced homeopathic practitioners for
the original article. sample size or because the homeopathic optimal conditions: individualised prescrip-
remedy genuinely did not have any anti- tion of homeopathic medicines needs a good
R Manocha inflammatory effect. medical understanding of asthma to discrimi-
Natural Therapies Unit, Royal Hospital for Women, This study provides important baseline data nate between disease-specific and patient-
Locked Bag 2000, Randwick 2031, Australia; for the calculation of the sample size needed specific or peculiar symptoms. Treatment was
r.manocha@unsw.edu.au by non-medically trained homeopaths with-
to carry out a randomised, placebo controlled,
Reference double blind study. A sample size of 65 out proper medical supervision, and this has
subjects per treatment arm would have 80% implications on the selection of medicines.
1 Manocha R, Marks GB, Kenchington P, et al.
Sahaja yoga in the management of moderate power to detect a difference of 10% in mean Medical doctors prescribing homeopathic
to severe asthma: a randomised controlled FEV1, assuming a standard deviation of medicines know what the patient has in terms
trial. Thorax 2002;57:110–5. difference of 28.86, using a paired t test with a of conventional diagnosis and can distinguish
two sided significance level of 0.05. features typical of the disease from those spe-
cific to the individual patient. This was not
Homeopathy in childhood A M Li, adequately considered by the authors in plan-
asthma Department of Paediatrics, Prince of Wales ning the study.
Hospital, Shatin, NT, Hong Kong Taken together, these biases seriously un-
We read with interest the article by White et al dermine the validity of the claimed results.
on the use of homeopathy as an adjunct in the A M Li, A Bush, N M Wilson
Such shortcomings should be eliminated
treatment of childhood asthma.1 We also Department of Paediatric Respiratory Medicine,
Royal Brompton Hospital, London, UK from future trials of homeopathy for asthma
obtained negative findings in an open study in published by respected journals such as
which we assessed the effects of homeopathy Correspondence to: Dr A M Li, Department of Thorax.
on spirometry and exhaled nitric oxide (eNO) Paediatrics, Prince of Wales Hospital, Shatin, NT,
in children with stable asthma. Hong Kong; albertm68mcli@yahoo.com F Dantas
Twelve asthmatic children (4 boys, median Professor of Homeopathy, Federal University of
age 13.5 years, range 7–18) who satisfied the References Uberlândia, Visiting Professor, Federal University of
following inclusion criteria were recruited: (1) 1 White A, Slade P, Hunt C, et al. ˜
Sao Paulo, Rua Pedro de Toledo 920, Sao ˜
Individualised homeopathy as an adjunct in Paulo-SP, 04039–020, Brazil;
stable asthma with no clinical indication for
the treatment of childhood asthma: a fdantas@climed.epm.br
change in treatment, on any dose of inhaled
corticosteroid and any other asthma medica- randomised placebo controlled trial. Thorax References
tions; (2) raised eNO level at the start of the 2003;58:317–21.
1 White A, Slade P, Hunt C, et al.
study despite clinical stability; (3) identifiable Individualised homeopathy as an adjunct in
sensitivity to house dust mite (HDM, n=3) or Homeopathy deserves to be scientifically ap- the treatment of childhood asthma: a
cat and HDM (n=9) by history and skin prick praised by good quality studies and the results randomised placebo controlled trial. Thorax
test (SPT); (4) no hospital admission or emer- published without bias which could distort 2003;58:317–21.
2 National Asthma Education Program.
gency department attendance for asthma in future meta-analyses. The study on childhood Guidelines for the diagnosis and management
the previous 3 months; (5) no history of con- asthma by White et al1 published recently in of asthma. Publication No 91-3042.
sumption of oral corticosteroid in the previous Thorax has critical flaws which seriously Bethesda, MD: National Heart, Lung and
3 months; (6) no homeopathic treatment undermine its conclusion. The main weak- Blood Institute, National Institutes of Health,
within the previous 6 months, allergen desen- nesses of the study, which were mentioned by 1991.
sitisation within the previous year, or HDM the authors but not given due attention, were 3 Enright PL, Lebowitz MD, Cockroft DW.
avoidance measures or removal of household the limitations of the primary outcome meas- Physiologic measures: pulmonary function
pet to which the subject had a positive SPT in ure and the mildness of the children’s asthma. tests. Am J Respir Crit Care Med 1994;149:
S9–18.
the previous 3 months. However, there is also concealed selection and 4 Richards JM, Hemstreet MP. Measures of life
At baseline all recruited patients under- measurement bias which could have been pre- quality, role performance and functional status
went SPT if this had not been done within the vented when planning the trial. in asthma research. Am J Respir Crit Care
previous 2 years, eNO measurement (NIOX, Available guidelines for the diagnosis of Med 1994;149: S31–9.
Aerocrine, Sweden), and spirometric testing asthma were not properly used for inclusion 5 Dantas F, Rampes H. Do homeopathic
(Vitalograph, Buckingham, UK) measuring of patients, leaving room for doubt as to medicines provoke adverse effects? A
forced expiratory volume in 1 second (FEV1). whether or not the included patients had systematic review. Br Homeopath J 2000;89
The mean of three best efforts was recorded asthma. Classification of asthma severity (Suppl 1):35–8.
and the result was expressed as percentage could be established at entry by using
predicted. The homeopathic remedy was published international paediatric asthma The efficacy and clinical effectiveness of
prescribed according to the child’s SPT result. consensus or guidelines.2 Better physiological homeopathy engenders considerable debate;
This was a preparation of HDM or cat dander measures could have been used—peak expira- it is therefore essential that clinical trials are
(or both, if appropriate) in the form of two tory flow results are less reliable than forced accurately interpreted and reported. The
lactose globules. The preparation was made up expiratory volume in 1 second, which is the recent publication by White et al1 has high-
according to the principles laid out in the most reproducible pulmonary function lighted this issue.
British Homeopathic Pharmacopoeia. The parameter.3 All patients were using β adrener- The study—which assessed classical home-
patients were told to take the globules daily gic inhalers and more than two thirds had opathy as an adjunctive treatment for child-
for the next 4 weeks while continuing with been prescribed inhaled steroids at baseline hood asthma—concluded that, based on the
the same conventional asthma treatment. A and were well controlled; at least 50% of primary outcome (the active quality of living
diary was given to each child to encourage patients in the homeopathic group had had subscale of the Childhood Asthma Question-
compliance and to document any break- no asthma event in the previous 12 months, naire), classical homeopathy was not superior
through symptoms or side effects from the suggesting a design bias against homeopathy to placebo. We disagree with this conclusion.
remedy during the study period. The subjects (ceiling effect). Sample size was calculated The scale used to assess the primary outcome
were told to return for eNO measurement and without a pilot study and did not allow for the was inappropriate; it does not distinguish
spirometric assessment after 4 weeks (visit 1) fact that comparisons of the impact of asthma between asthmatics and non-asthmatics2 and
www.thoraxjnl.com
Downloaded from thorax.bmj.com on 7 March 2007
PostScript 827
is more suitable as a cross sectional measure scale validation study cited in the paper by Methods such as that developed by Jadad et
than as a longitudinal outcome, and the abil- White et al.2 Furthermore, the CAQ QoL scale al4 would assess this as a high quality study,
ity to identify any therapeutic improvement does not discriminate between non-asthmatic and the primary outcome appears to be nega-
was severely reduced due to ceiling/flooring and mildly asthmatic children. tive. As we have shown, this interpretation is
effects in both the primary and some second- Similar floor and ceiling effects are seen in fundamentally flawed. We believe a correction
ary outcome data. For example, baseline many of the secondary outcome measures— should be published which should focus on
scores identified that the study population for example, in the homeopathy group at (1) the inappropriate scope of the original
had good quality of life, and that two of the entry the peak expiratory flow rate was conclusions, and (2) clarification of the
three age groups studied had mild asthma. 100.4% of expected and the median number secondary outcomes and the conclusions
Any therapeutic improvement would there- of asthma episodes in the preceding year was drawn from them.
fore be hard to identify, let alone quantify. zero. These are all “hard” floor/ceiling effects;
Other design issues were apparent—for no improvement at all could have been P Fisher
example, no data were reported on homeo- expected. There is also a strong suggestion of Director of Research, Royal London Homoeopathic
pathic exacerbations (an indicator of the floor/ceiling effects in other outcomes such as Hospital, London W1W 5BP, UK;
healing response), and the security of blind- days lost from school, but we cannot be peter.fisher@uclh.org
ing was not assessed. Yet, despite these certain from the published data. Other sec-
limitations, some encouraging therapeutic ondary outcomes show relative floor/ceiling K Chatfield
effects were apparent. For example, a clini- effects—for instance, the mean final value in Director of Research, Society of Homeopaths
cally relevant improvement in asthma severity the CAQA parental severity score was 5.5 on a Faculty of Health, University of Central Lancashire,
scale of 5–19. Since this was an intention to Preston PR1 2HE, UK
(unadjusted scores) was seen in two of the
three groups, and a favourable pattern in the treat analysis, 20% of the values were simply
pretreatment values carried forward. R Mathie
days off school/days attended was seen in the
Other CAQ subscales analysed as secondary Research Development Adviser,
homeopathic treated children (although no Faculty of Homeopathy, Hahnemann House,
data were presented). outcomes consistently favour homeopathy.
29 Park Street West, Luton LU1 3BE, UK
We suggest that a balanced and accurate For the severity subscales the improvement
conclusion to these data would be that no was statistically highly significant (p=0.01)
definitive conclusions could be drawn but that with 95% confidence intervals not including References
further investigation is needed. We therefore zero. This again was an intention to treat 1 White A, Slade P, Hunt C, et al.
hope that the authors’ inaccurate conclusions analysis, and while there are good reasons for Individualised homeopathy as an adjunct in
performing such analyses, effect size esti- the treatment of childhood asthma: a
neither dampen future research nor bias randomised placebo controlled trial. Thorax
future systematic reviews.3 mates should be based on data for subjects
who have actually taken the treatment and 2003;58:317–21.
2 French DJ, Christie MJ, Sowden AJ. The
S B Brien, G Lewith had its impact evaluated; in this case, 20% did
reproducibility of the childhood asthma
Complementary Medicine Research Unit, Royal not. In addition, there was a floor effect (see questionnaires: measures of quality of life for
South Hants Hospital, Southampton SO14 0YG, above) in one of the severity scales. A similar children with asthma aged 4–16 years. Qual
Hampshire, UK; s.brien@soton.ac.uk pattern is seen for other subscales of the CAQ, Life Res 1994;3:215–24.
but no statistical analysis is presented. 3 Linde K, Clausius N, Ramirez G, et al. Are
References The question most frequently posed about the clinical effects of homeopathy placebo
1 White A, Slade P, Hunt C, et al. homeopathy is “is it all placebo effect?” Most effects? A meta-analysis of placebo-controlled
Individualised homeopathy as an adjunct in meta-analyses have concluded that it is not. If trials. Lancet 1997;350:834–43.
the treatment of childhood asthma: a 4 Jadad AR, Moore RA, Carrol D, et al.
the outcome measures which could not have
randomised placebo controlled trial. Thorax Assessing the quality of reports of randomized
2003;58:317–21. improved are excluded, the results of this trial
accord with those of the largest meta-analysis clinical trials: is blinding necessary? Control
2 French DJ, Christie MJ, Snowden AJ. The Clin Trials 1996;17:1–12.
reproducibility of the childhood asthma of homeopathy; they “are not compatible with
questionnaires: measures of quality of life for the hypothesis that the clinical effects of
children with asthma aged 4–16 years. Qual homeopathy are completely due to placebo”.3 The study by White et al1 of quality of life in
Life Res 1994;3:215–24. The treatment effect size was relatively small, children with asthma treated with homeo-
3 White P, Lewith G, Berman B, et al. Reviews but classical homeopathy is a complex and pathy is fatally flawed. The Childhood Asthma
of acupuncture for chronic neck pain: pitfalls non-standardised intervention. The practi- Quality of Life instrument used was validated
in conduting systematic reviews.
Rheumatology 2002;41:1224–31.
tioners involved had no particular experience in a study by French et al.2 The children
of asthma. There is thus considerable scope entered into the study by White et al had
for refinement. scores consistent with those of normal chil-
The paper by White et al1 in the April issue of Regrettably, the conclusions do not ad- dren who don’t have asthma. For a statisti-
Thorax purports to show that adjunctive equately reflect the shortcomings of the trial. cally significant improvement to occur in this
homeopathic treatment has no effect on the The authors state that “there was no evidence score, the treated group would have to develop
quality of life of asthmatic children. The of a clinically relevant change in quality of life scores of around 100%—that is, better than
primary outcome measure was the active score”, but omit to mention that none was normal non-asthmatic children. This is clearly
Quality of Life (QoL) scale of the Childhood expected since the QoL scores were normal at highly unlikely. In addition, a similar “ceiling
Asthma Questionnaire (CAQ). This measure entry. There is no reference to the many floor/ effect” applies to the peak expiratory flow
was inappropriate, and no such conclusion is ceiling effects. readings which at entry were 100.4% and
justified. As table 1 shows, the children Our greatest concern is that the bias in the 96.9% of expected for the verum and placebo
enrolled had QoL scores virtually identical to interpretation of the results will carry groups, respectively.
those of non-asthmatic children in a large through to future meta-analyses and reviews. This is a very poor quality trial which does
absolutely nothing to further our understand-
ing of the potential value of homeopathic
treatment in children with asthma. In fact,
Table 1 Comparison of Childhood Asthma Questionnaire (CAQ) active
the press release from the journal has been
quality of life scores in in paper by White et al1 and CAQ validation study by picked up by the media and used to support
French et al2 the headline “Homeopathy of no use in
asthma”.
CAQ validation study (French et al2), n=535
(median) Publishing this quality of research at best
White et al1, n=93 does not improve our necessary evidence base
(mean at entry) Asthmatic Non-asthmatic and, at worst, contributes to the denial of
services which may indeed be of value to
CAQA: range 10–40 35.2 34 34 patients. A close analysis of the study shows
(4–7 years) that the treatment group had a trend to better
CAQB: range 7–35 28.1 28 29
outcomes than the placebo group. If this were
(8–11 years)
a pilot study, it would be indicating that there
CAQC: range 8–36 29.4 No data No data
(12–16 years) is indeed a potential benefit to asthmatic
children from homeopathy which should be
Higher values indicate better quality of life. investigated with a proper trial of good meth-
odological quality.
www.thoraxjnl.com
Downloaded from thorax.bmj.com on 7 March 2007
828 PostScript
R Leckridge scale measures symptoms only, and active understand the current literature. Professor
Glasgow Homeopathic Hospital, Glasgow quality of life is much more appropriate for Coggon moves rapidly through types of data
G12 0XQ, UK; bob.leckridge@virgin.net the holistic approach of homeopathy. (continuous, ordinal or univariate and multi-
The claim by Fisher et al and Dantas that the variate) to methods of summarising data on
References homeopaths were inadequate to the task is which a fair amount of time is spent. Tabular
1 White A, Slade P, Hunt C, et al. speculative and one we reject. Dantas would and graphical (dot, line, bar and pie chart)
Individualised homeopathy as an adjunct in presentations are discussed with numerous
the treatment of childhood asthma: a
prefer us to have used more rigorous criteria
for inclusion and assessment, but the study illustrations from everyday clinical practice.
randomised placebo controlled trial. Thorax
was especially designed to reflect “real life” The interpretation of graphical data and its
2003;58:317–21.
2 French DJ, Christie MJ, Sowden AJ. The pragmatically by rigorously applying the limitations—a very important part of under-
reproducibility of the childhood asthma criteria used by the GPs, the children, and standing current medical research—are thor-
questionnaires: measures of quality of life for their families. We cited his paper as the only oughly discussed. The concept of probability is
children with asthma aged 4–16 years. Qual systematic and objective report on homeo- introduced and combining probabilities is
Life Res 1994;3:215–24. pathic aggravations that we were aware of, explained. Sensitivity and specificity are de-
and if we gave the impression that he stated fined here but could more appropriately be
Author’s reply placed later as they are, in fact, properties of
that aggravations are a hallmark of success,
These authors are to be thanked for their con- then we regret it. statistical tests. Hypothesis testing, confi-
tribution to the debate about the interpret- dence intervals, and the basis of sample size
ation of the results of our trial, and it is A White calculations (though not how to calculate the
conceded that ceiling effects may exist which Complementary Medicine, Peninsula Medical size of a sample) are also discussed. The
limit the interpretation of our results. If the School, Universities of Exeter and Plymouth, Exeter author explains the two most common meth-
children were already effectively medicated, it EX2 4NT, UK; adrian.white@pms.ac.uk ods of statistical modelling—linear regression
may not have been possible to show any ben- and survival analysis—and concludes with a
efit from homeopathic treatment in quality of section on meta-analyses and the importance
life. It would require a much larger study to BOOK REVIEW of involving statisticians very early in the
show any differential change in conventional planning stage of a study.
This is an excellent introduction to practis-
medication or global indicators, which were Statistics in Clinical Practice ing statistics in medicine and will be ex-
absent from the results. Because of ethical
tremely useful for medical students and clini-
issues, it may be difficult to conduct a defini- David Coggon. 2nd edition. London: BMJ cians alike. Medical researchers will, however,
tive trial in children with severe asthma. Books, 2002. £14.95, 120 pp. ISBN need to follow this text with a more advanced
Leckridge, Fisher et al, and Brien and 0727916092 one.
Lewith suggest that we should have concen-
trated instead on the small changes in the This is a very clearly written introduction to T Seemungal
severity subscales, the estimate of which is statistics, suitable for medical students and Barts and The London Medical School, London, UK;
not clinically relevant. However, the severity doctors who need a quick update in order to tseemungal@aol.com
www.thoraxjnl.com
