Mediterranean diet cuts risk of heart disease!

This story is making headline news yet again, including at the New York Times, where Gina Kolata describes a new study done in Spain of the effects of adding extra virgin olive oil or nuts, fish, fruits, legumes and red wine to the diet, and reducing the consumption of red meat and processed baked goods.  (Note that the pyramid below was adapted from Consumer Reports Nov 1994.)

From www.womensheart.org

The total study group of about 7000 men (aged 55-80) and women (aged 60-80) was divided into 3 subgroups, and asked to follow the Mediterranean diet (MD) with extra virgin olive oil, the MD with nuts, or a low-fat diet.  They weren't asked to lose weight or to exercise.  The MD groups were given olive oil or nuts every week, and counseling as to how to follow the diet, while the low-fat diet group (the control group) was given a pamphlet on how to follow the diet when enrolled in the study, and then annually until 2006, when it was recognized that their adherence to the diet was poor, at which point researchers added further intervention. This group still never could consistently follow the low-fat diet, and instead were essentially eating their usual diet. 

Recruits were people without heart disease but with type 2 diabetes, or at least three major risk factors ("smoking, hypertension, elevated low-density lipoprotein cholesterol levels, low high-density lipoprotein cholesterol levels, overweight or obesity, or a family history of premature coronary heart disease").  They were followed up until either they dropped out or until the year 2010. The total person-years in the study was about 12,000, 11,000 and 10,000, by group. 

The paper is published in the New England Journal of Medicine.  The researchers assessed primary and secondary outcomes, with the former being heart attack, stroke or death from cardiovascular disease, and the latter being heart attack, stroke, death from cardiovascular disease or death from any other cause. 

The median follow-up period was 4.8 years. A total of 288 primary-outcome events occurred: 96 in the group assigned to a Mediterranean diet with extra-virgin olive oil (3.8%), 83 in the group assigned to a Mediterranean diet with nuts (3.4%), and 109 in the control group (4.4%). Taking into account the small differences in the accrual of person-years among the three groups, the respective rates of the primary end point were 8.1, 8.0, and 11.2 per 1000 person-years.  Outcomes According to Study Group.). The unadjusted hazard ratios were 0.70 (95% confidence interval [CI], 0.53 to 0.91) for a Mediterranean diet with extra-virgin olive oil and 0.70 (95% CI, 0.53 to 0.94) for a Mediterranean diet with nuts.

And,

In this trial, an energy-unrestricted Mediterranean diet supplemented with either extra-virgin olive oil or nuts resulted in an absolute risk reduction of approximately 3 major cardiovascular events per 1000 person-years, for a relative risk reduction of approximately 30%, among high-risk persons who were initially free of cardiovascular disease.

There was no effect of diet on mortality from all causes.  That is, the difference in total number of deaths between groups was not statistically significant.  The effect of diet on cardiovascular disease was apparently through stroke, not heart attack.  People following the Mediterranean diet did not lose weight, nor reduce the amount of fat in their diet, so the effect, the researchers say, was of dietary components alone. 

Interestingly, when researchers compared the merged MD groups with controls recruited before and after 2006, they found that "adjusted hazard ratios were 0.77 (95% CI, 0.59 to 1.00) for participants recruited before October 2006 and 0.49 (95% CI, 0.26 to 0.92) for those recruited thereafter (P=0.21 for interaction)."  Remember that the researchers decided to do more intervention with the control group after 2006. 

But how is that to be interpreted?  Were controls adhering better to the low-fat diet after 2006, and it turns out to be significantly worse than the MD?  Or vice versa?  Did the act of intervention itself made a difference somehow, or were the people recruited after 2006 metabolically different, older, sicker or something else from those recruited before?  Whatever the reason for the difference, it does suggest that comparison between the three groups is not a simple comparison of three different diets.

If this study is as significant as the write-ups are saying, it means that the effect of changing diet can be significant enough to be detectable within a relatively short time span (this study spanned 2003-2010, with subjects apparently included for varying lengths of time).  This suggests that the effect of a non Mediterranean diet over a lifetime is reversible, and thus that risk isn't as genetic as some think (not a surprise!), risk factors like blocked arteries may be reversible by diet, cholesterol can be changed by diet (this is also well-known), being overweight is not a significant risk factor (results on this issue go back and forth), and low-fat diets aren't protective (this, too, has been shown before, though doesn't seem to have caught on with the public). 

And, while this study does confirm things that have already been known, actually rather well and for quite a long time, it also means that people on the Mediterranean diet still die of cardiovascular disease, particularly heart attacks.  Rather than 11 CVD deaths per 1000 person-years, there were 8.  So, the difference may be statistically significant, but it's not qualitatively huge, like 25 vs 2, or even 11 vs 2.  Though, of course if you're one of those three, that's an incalculable difference.  Further, we don't know whether it's eliminating red meat and baked goods rather than adding olive oil and wine and nuts that makes the difference.

One can ask to what extent we should even be doing more and more studies of the same basic idea, once we have systematic data in its favor (dramatic disease benefits were found in a major North Karelia Finland project a long time ago, for example, in a huge dietary intervention study in a place that had, at the time, the highest CVD rates in the world).  There are various ways to measure effects and benefits, and to define outcomes, and these are relevant to evaluating any studies of diet and health.

If nothing else, this study is a reminder that if you reduce deaths from one cause, deaths from other causes go up.  People do still die of something.