Tuesday, June 16, 2015

Chocolate and heart disease


When I was at the barber's earlier today, I picked up The Daily Mail (an entertainingly but highly untrustworthy British newspaper) to find a massive headline emblazoned across the front page, pronouncing that two chocolate bars a day are good for our hearts. The headline of web-site article reads: "How eating two chocolate bars A DAY 'cuts your risk of heart attack and stroke by up to 25%' - and milk is just as good as dark".

I have taken a look at the study, which was published in the reputable Journal 'Heart' (open access). It is interesting but we've got some way to go before the government adds two chocolate bars to it's '5-a-day' recommendation. The finding is one of association so any talk of cauality remains s pure speculation.

On the positive side, the sample size in the study is comprehensive (21,000 people), and the authors include a meta-analysis of nine past studies. However, on the downside:
- The study was unavoidably survey based and subjects are known to mis-state when it comes to self-reporting (re-call bias).
- While the authors tried to account for other influencing factors, they point out that residual effects of confounding factors may still be present. Indeed, when you think of all the other factors that may influence the results, such as smoking, weight, diabetes, physical activity, you can fairly imagine that it's nigh on impossible to properly disentangle these factors. The authors do try but how accurate is this statistical jiggery-pokery, I wonder.
- The authors note that reverse causality cannot be ruled out.
- Looking at the table of results, it is a little odd that highest chocolate consuming quintile has a much higher energy intake (around 400 calories than the other groups) and yet the BMI and hip-to-waist ratios are similar to the other groups, while physical activity isn't that much different.
- The highest chocolate consuming quintile group ranges from 15.6-98.8g/day i.e. the range is from a couple of piece of dairy milk up to almost two bars a day. How many people were actually on two bars a day?
- The findings are statistically significant but could still be down to chance. The authors state that 'The calculations showed that compared with those who ate no chocolate, higher intake was linked to an 11% lower risk of cardiovascular disease.' This sounds impressive but needs to be looked at in the context of the overall population's risk of suffering cardiac disease, which was 14%. An 11% variation on 14% is something, but it could juts as easily be nothing i.e. noise.

All these issues aside, I do like the take home message of the study:

"There does not appear to be any evidence to say that chocolate should be avoided in those who are concerned about cardiovascular risk."

Even if it's not a prescription to eat two chocolate bars a day, it is still helpful to know that chocolate bars needn't be feared as death sticks.

No comments: