JRSM >
Volume 96, Number 3 >
Pp. 156-157
Complementary Medicine, Peninsula Medical School, Universities of Exeter & Plymouth, 25 Victoria Park Road, Exeter, Devon EX2 4BN, UK
Dr Linde and Dr Willich (January 2003 JRSM1) have done a great service to the medical community by showing us how systematic reviews (SRs) can differ confusingly regarding a range of methodological aspects. As they point out, SRs are designed to avoid biases and make conclusions as objective as possible. Given the diversity of the SRs analysed, I was impressed with the level of agreement many of them reached in their conclusions. Table 1 summarizes five subject areas where our own SRs were cited. The concordance of the conclusions is stunning.
|
There are exceptions, of course. The prime example is acupuncture for back pain, where we have published a ‘positive’2 and others a largely ‘negative’ SR3. The difference is, however, readily explicable: we pooled the data from all trials where this was possible2 while van Tulder et al.2 basically relied on a vote count. If studies tend to be small and produce non-significant positive trends, pooling will generate a positive overall result, while vote-counts will yield a negative one.
And then there is homeopathy, which is perhaps the most controversial subject of all. Linde et al. have published their landmark SR concluding that ‘clinical effects of homeopathy are not completely due to placebo’4. Two years later they cast considerable doubt on this by reanalysing the data and concluding that ‘there was clear evidence that studies with better methodological quality tended to yield less positive results’5. I have recently reviewed all eleven SRs of homeopathy published since Lind's meta-analysis4. This collective evidence led me to conclude that ‘the best clinical evidence for homeopathy available to date does not warrant positive recommendations for its use in clinical practice’6.
REFERENCES
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
