Intended for healthcare professionals

Rapid response to:

Papers

Modified Mediterranean diet and survival: EPIC-elderly prospective cohort study

BMJ 2005; 330 doi: https://doi.org/10.1136/bmj.38415.644155.8F (Published 28 April 2005) Cite this as: BMJ 2005;330:991

Rapid Response:

"The two unit problem"

Editor- When Trichopoulou et al collapsed all the participants in
their multicentre prospective cohort study together, they found that a two
unit increment in a Mediterranean diet score was associated with a risk
reduction of 7% for overall mortality (1).

If this is true the more one adheres to a Mediterranean diet, the
lower the risk of death, but it is evident that the shapes of the
prediction lines were not especially straight in most of the cohorts in
this study. In my opinion it is therefore problematic to use statistics
in such a way.

Let us take a look at Table 6 in the original paper. We here find
that among the nine cohorts, only in United Kingdom, Sweden and Denmark
there were a linear risk reduction across the diet score groups. In
France, Spain and Greece a moderate diet score (4-5) were associated with
increased risk and good diet score (6-9) with a reduced risk, compared to
a poor diet score (0-3). In Italy and Netherlands a moderate diet score
were associated with lower risk, and a good diet score with higher risk.
In Germany, higher diet scores (moderate and good) were actually
associated with higher risk compared to the poor diet score group.

In conclusion a two unit increment on this Mediterranean diet score
will often be a poor predicator on mortality.

1. Trichopoulou A, Orfanos P, Norat T, Bueno-de-Mesquita B, Ocke MC,
Peeters PH et al. Modified Mediterranean diet and survival: EPIC-elderly
prospective cohort study. BMJ. 2005 Apr 30;330(7498):991. Epub 2005 Apr 8.

Competing interests:
None declared

Competing interests: No competing interests

13 May 2005
John Roger Andersen
Assistant professor
Sogn and Fjordane University College. Faculty of Health Studies. 6800 Førde. Norway