“Eating just two small bits of chocolate a week can cut the risk of heart failure by up to a third,” according to an article in the Daily Express .
The news story is based on a study of whether chocolate affects the risk of heart failure in elderly and middle-aged women. Women who ate moderate amounts of chocolate (one-to-two servings a week or one-to-three servings a month) were found to be at lower risk of heart failure. This association was not found in women who ate three or more servings a week.
This was a large and well-conducted study, but it is not strong evidence that chocolate reduces the risk of heart failure. Problematically, chocolate intake was only assessed once, at the start of this nine-year study, and so any subsequent changes in diet or lifestyle have not been taken into account.
The study also relied on women recalling their precise intake of chocolate and other foods over a year, which is likely to introduce the risk of error, particularly when a “serving” can mean different things to different people. Many studies have looked at the potential health benefits of chocolate, but these are far from conclusive and further research is needed.
Where did the story come from?
The study was carried out by researchers from Harvard Medical School and Harvard School of Public Health in Boston, and from the Karolinska Institute in Sweden. It was funded by the Swedish Research Council and published in the peer-reviewed medical journal, Circulation: Heart Failure.
The study was widely reported, with most papers correctly pointing out that the association between chocolate and a lower risk of heart failure only applied to eating chocolate in moderation. The Daily Express and the BBC included warnings from health experts about the fat and calories found in chocolate.
The BBC and the Daily Mirror both reported that dark chocolate can be good for the heart, when the study made no distinction between types of chocolate. The BBC’s report goes on to say that while most of the chocolate consumed was milk, the concentration of cocoa solids in Swedish milk chocolate was equivalent to dark chocolate by UK standards. However, this is not correct; the study reports that approximately 90% of chocolate consumption in Sweden is milk chocolate, which contains approximately 30% cocoa solids; dark chocolate in the UK usually contains 70% cocoa solids.
The Express said “two small bits” of chocolate a week would cut the risk, but it is unclear what size the portions consumed in the study were.
What kind of research was this?
This was a prospective cohort study of nearly 32,000 women, investigating whether chocolate has any effect on the risk of heart failure. The researchers point out that clinical trials have shown that chocolate can reduce blood pressure, which is a particularly strong risk factor for heart failure. In addition, observational studies have also found an inverse association between chocolate and cardiovascular disease. This is the first study to examine whether chocolate intake is associated with the risk of heart failure.
This is a cohort study, a type of study design that can be used to investigate whether certain factors (in this case, chocolate intake) are associated with health outcomes (in this case, heart failure incidence). However, on its own, a cohort study cannot be certain about cause and effect. A randomised controlled trial would provide firmer evidence of effect, but this study design is not always feasible.
What did the research involve?
The study included 39,227 women from central Sweden, aged between 48 and 83. The women were all participants in a larger, ongoing study assessing the relationships between various lifestyle factors and the risk of certain chronic (long-term) diseases. The women were asked to complete a questionnaire on their health and lifestyle, including detailed questions about diet and chocolate intake. Women who failed to complete the questionnaire correctly or who had a history of heart failure, heart attack or diabetes or a previous diagnosis of cancer were excluded from the study, leaving data from 31,823 women available to use.
The questionnaire included a validated food frequency questionnaire that asked participants how frequently they had consumed 96 different foods and beverages over the previous year. This included whether they ate chocolate, with eight predefined responses ranging from never to three or more servings a day. The study did not seem to ask about the size of the servings, but the researchers say that diet records indicate that for Swedish women, the average portion of chocolate would be 19-30 grams depending on age. Chocolate bars vary in size, usually from 25 to 50 grams.
The study makes no distinction between mild and dark chocolate, but points out that in Sweden in the 1990s, most chocolate consumed was milk chocolate and that this contains about 30% cocoa solids.
The women were followed up from 1998 through to the end of 2006. Hospitalisation or death from heart failure was recorded using national inpatient and cause-of-death registers.
The researchers then used standard statistical methods to examine whether chocolate intake had any effect on the incidence of heart failure. They took account of other factors that might affect this in their analysis, including the women’s age, education, physical activity, smoking habits and alcohol intake, family history and self-reported history of high blood pressure and high cholesterol. They also examined the associations between chocolate intake, heart failure and milk consumption, as they claim that milk can inhibit the absorption of flavenoids, the substances in chocolate thought to have health benefits. They also analysed the women’s intake of other snack foods.
What were the basic results?
Over the nine years they were followed up, 419 women were hospitalised for the first time or died of heart failure, which corresponds to a rate of about 15 cases per 10,000 person years (the accumulated amount of time that all the women were being followed up).
The researchers found that compared to women who did not eat chocolate regularly, heart failure rates were:
- 26% lower among those eating one-to-three servings a month (95% CI 0.58 to 0.95)
- 32% lower in those eating one-to-two servings a week (95% CI 0.50 to 0.93)
- not significantly affected by consuming three-to-six servings a week (HR 1.09, 95% CI 0.74 to 1.62) or one or more serving a day (HR 1.23, 95% CI 0.73 to 2.08)
The association was similar in both high and low dairy groups. The consumption of other high-fat snacks such as cakes and biscuits was not associated with heart failure.
How did the researchers interpret the results?
The researchers say that moderate, regular chocolate intake was associated with a lower rate of heart failure. They suggest that the flavenoids in chocolate may have a beneficial effect on cardiovascular risk factors.
This study has strengths in its large size and that it had a relatively long follow-up period. However, it has several limitations, some of which are noted by the researchers:
- Although they tried to take account of other lifestyle and medical factors that can affect heart failure risk, residual or unmeasured “confounding” cannot be ruled out. The researchers also relied on the participants reporting whether or not they had high blood pressure and high cholesterol. This introduces the possibility of error, as some women may have not answered this question correctly.
- Chocolate consumption, diet and other lifestyle factors were only measured once at the start of the study, so there is no information about how any changes in these factors might have affected heart failure risk.
- The questionnaire relied on the women recalling what they had eaten over the past year. This is a relatively long time and there is a good chance that some women misremembered what they had eaten.
- It is not clear how big the servings of chocolate were, so it is difficult to say exactly how much chocolate was consumed. A “serving” is likely to mean different things to different people.
- Only heart failure cases that resulted in hospitalisation or death were included in the analysis.
- Other cardiovascular outcomes such as heart attack and stroke were not assessed.
- A longer follow-up period may have produced different results since more women may have developed heart failure after the nine years of the study.
As nutritional experts have pointed out, the flavenoids in chocolate can also be found in fruit and vegetables, but without the associated fat and calories.
Overall, this study cannot tell us for sure if chocolate reduces the risk of heart failure. Chocolate is high in fat, sugar and calories and if consumed excessively is a risk factor for obesity, which increases the risk of heart disease and diabetes. The current advice is to eat chocolate as an occasional treat, rather than a regular part of the diet.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
BBC News, 17 August 2010
Daily Express, 17 August 2010
Daily Mirror, 17 August 2010
The Daily Telegraph, 17 August 2010
Links to the science
Circulation: Heart failure 2010; Published online before print August 16
Cochrane Database of Systematic Reviews 2007, Issue 4