A disservice to public health
I write to respond to the publication last week of the Hooper et al
(1) review of risks and benefits of omega-3 fats.
As scientific adviser to a consortium of omega-3 interests I last
year coordinated an application to the Joint Health Claims Initiative
(JHCI)for a label claim supporting the role of long chain omega-3
polyunsaturates in respect of cardiovascular health. The process involved
a literature search and review. Our conclusions were quite different. The
conclusions were endorsed by the independent Expert panel of assessors
appointed by JHCI. The Cochrane review process has previously been
questioned (2) over it’s suitability for use in nutrition studies of this
sort. The fact that previous Cochrane reviews by Hooper have concluded
that there was no evidence to support the hypothesis that reducing dietary
salt intake decreased blood pressure, or that reducing saturated fat
intake decreased heart disease risks seems to me to make the point very
The Hooper review is flawed by the indiscriminate combination of
primary and secondary prevention studies, and the consequent failure to
compare like with like. They also dismiss as unreliable, cohort studies
involving many thousands of subject-years, on the grounds that there is a
risk of confounding. But then they include studies such as the Burr 2003
(3) study, the authors of which accept there may be confounding issues.
They conclude that with a RR of 0.86 and a 95% CI of 0.70 to 1.04 in
the 12 RCT studies involving fish or fish oil in CV mortality, that there
is no clear evidence of benefit. With a RR of 0.65 and 95% CI of 0.48 to
0.88 from like-with-like cohort studies, I submit that to put out the
message that there is no clear evidence to support a reduction in risk of
CV death from long chain omega-3 usage is highly irresponsible.
When Bucher et al (4) reviewed this subject in 2002, they surveyed
the literature, drew up their criteria and carried out a meta analysis of
the 11 RCT studies published at the time, and concluded that eating more
omega-3 significantly reduced risk of death by 25-30%. The Burr 2003
study was included in the Hooper review, but not Bucher, primarily because
it was not published at the time. In conducting the search and review for
the JHCI application process, I carried out an update of the Bucher
process, with a closing date of September 2003. The search revealed a
small number of relevant papers, which met the Bucher inclusion criteria.
Interestingly, the Burr 2003 study did not meet the Bucher criteria, as I
confirmed with the lead author, Professor Burr in a personal
communication. I submitted the additional data to a statistician, and
included the results in the JHCI submission (5).
The Burr 2003 study is anomalous for a number of reasons, and is at
odds with the very extensive published literature in this field. There are
significant uncertainties in this trial, which the author and others have
pointed out. To use it uncritically as Hooper et al did, is not science.
To claim “long chain omega-3 does not offer any protection from heart
disease” based on an uncritical interpretation of selected papers is a
gross disservice to the communication of science, and to the health of UK
citizens and others. To publish this flawed study, as the BMJ chose to do,
is a similar disservice.
I urge the authors and the BMJ to correct this situation by
1.Hooper L,Thompson RL,Harrison RA,Summerbell CD,Ness AR, Moore HJ, et al.
Risks and benefits of omega3 fats for mortality, cardiovascular disease
and cancer: a systematic review. BMJ 2006.
[Epub ahead of print; doi = 10.1136/bmj.38755.366331. 2F].
2.Truswell A.S. Some problems with Cochrane reviews of diet and chronic
disease. Eur.J.Clin.Nutr. 2005,59; Suppl 1:S150-154.
3. Burr ML, Ashfield-Watt PA, Dunstan FD, Fehily AM, Breay P, Ashton T,
Zotos PC, Haboubi NA, and Elwood PC. Lack of benefit of dietary advice to
men with angina: results of a controlled trial. Eur J Clin Nutr. 2003; 57:
4. Bucher HC, Hengstler P, Schindler C, & Meier| G. N-3
Polyunsaturated Fatty Acids in Coronary Disease: A Meta-Analysis of
Randomized Controlled Trials. Am.J.Med., 2002, 112:298-304.
Scientific Consultant to the Omega-3 Group (A consortium of organisations involved in the marketing of fish)
Competing interests: No competing interests