Does coffee cause or prevent heart failure?

“Two cups of coffee a day can reduce the risk of heart failure (but five can be bad for you),” the Daily Mail reported today. This seemingly contradictory headline is actually a fair but slightly overenthusiastic reflection of a review of research that looked at the relationship between regular coffee drinking and risk of heart failure.

Heart failure is a condition in which the heart cannot pump blood to meet the body’s demands, this can have many causes, the most common being a heart attack. This review of five large studies questioned people about how much coffee they drank and then examined whether they went on to develop heart failure. The combined results suggested that, compared to no coffee consumption, four to five servings of coffee a day was associated with an 11% lower risk. Drinking excessive amounts of coffee had no benefit – and is likely to give you the jitters.

Although a systematic review can provide good evidence, the results of this one should be viewed with some caution. All of the included studies relied on self-reported coffee drinking and this can affect the reliability. Coffee consumption was only measured once and did not take into account any changes in consumption over the years. It is also unclear how the studies defined the outcome of heart failure. Because some of the studies included people who had already had a heart attack, it is not clear whether they were currently in some stage of heart failure when their coffee drinking was assessed. The studies were inconsistent in how they adjusted for other factors that can be related to heart failure risk, such as high blood pressure and obesity.

The type, strength and volume of coffee consumed by people in this study will also have varied widely, making it hard to get any clear message about the benefits or risks of coffee consumption.

Overall, this review does not prove that coffee is good for the heart - only a possible association between coffee drinking and the risk of heart failure. The relationship between coffee consumption and heart health remains confusing, with previous studies showing inconsistent results. As in most things, moderation is key.

Where did the story come from?

The study was carried out by researchers from Harvard Medical School, Harvard School of Public Health, Brigham and Women’s Hospital and the University of Alabama. It was funded by the US National Institute of Health. It was published in the peer-reviewed medical journal Circulation Heart Failure.

Coverage of the study in the press is confusing. Both the Daily Express and the Mail said that two cups of coffee can reduce the risk of heart failure, but this number appears to be a conversion of US serving sizes and servings in coffee chains.

The Express claim that coffee ‘can dramatically reduce the risk of heart failure’ was not borne out by this research, which found a relatively small decrease in risk associated with moderate coffee drinking. Both papers reported that drinking excessive amount of coffee is dangerous, but this is also not borne out by the research, which found a slight increase in risk associated with 10-11 cups a day. The newspapers also did not identify the important limitations of the studies included in this review.

What kind of research was this?

This was a systematic review and meta-analysis of studies that looked at the association between risk of heart failure and the number of cups of coffee people reported drinking. The authors point out that the results of previous studies looking at this relationship have been inconsistent, while US guidelines on heart failure say that coffee may increase the risk.

Prospective cohort studies can help assess the relationship between lifestyle factors, such as diet and people’s health, because they can follow large groups of people for a number of years. However, they cannot prove cause and effect.

A systematic review and meta-analysis of prospective cohort studies is the best way of identifying and combining the results of all relevant research that has addressed the question of whether a coffee drinking is associated with the risk of heart failure.

However, though it is a high standard of evidence, the review will often have limitations due to the different methods of the studies it included. These studies may have:

included different populations
followed them for different periods of time
assessed exposures and outcomes differently
variably adjusted for different confounding factors that could also affect the relationship

For more on why this matters, see How to read health news.

What did the research involve?

The authors searched several electronic research databases to find coffee and heart failure data published between 1966 and 2011, following an established meta-analysis protocol.

They included in their review prospective cohort studies that examined the association between coffee consumption and heart failure. They also included studies that had calculated risk figures for new cases of heart failure or mortality. For each study included in the review, they extracted information about:

characteristics of the participants (including health at the start of the study start)
level of coffee intake
number of people at each different coffee consumption level
estimates of risk at different levels of consumption.

They used the study results that had been adjusted for the greatest number of potential confounders.

What were the basic results?

The researchers included five studies looking at coffee consumption and the risk of heart failure, four carried out in Sweden and one in Finland. In total, they included 140,220 participants and 6,522 cases of heart failure. Three of the studies consisted of participants with no history of heart attacks; one consisted of participants with a history of heart attacks; and one included separate analyses for people with and without a history of diabetes or heart attack.

The study found that compared to no coffee consumption, the lowest risk of heart failure was associated with four or five servings of coffee daily. Above four to five cups, the size of the risk decrease (that is, the benefit) tended to get smaller. At around 10 cups of coffee a day, people had about an equivalent risk to those who drank none. Above 10 cups there was a trend for the risk to increase compared to those who drank none. For example:

those who reported one or two servings a day had a 4% lower risk of heart failure than those who drank no coffee (relative risk [RR] 0.96, 95% confidence interval [CI] 0.90 to 0.99)
those who drank three or four servings a day had a 10% lower risk (RR 0.90, 95% CI, 0.82 to 0.99).
those who drank four or five servings a day had 11% lower risk (RR 0.89, 95% CI 0.81 to 0.99

For levels of coffee consumption higher than this, the apparent benefit gradually disappeared. At 11 servings a day, there was a slight increase in risk (RR 1.03 95% CI 0.89 to 1.19). However, it should be noted, that although there was a trend for the risk decrease to become smaller above four or five servings, and then begin to increase above 10 servings, none of the risk calculations for servings of between 5 and 11 cups a day were actually statistically significant. This means that there is no real evidence of a difference in risk between no coffee consumption and consumption of between 5 and 11 cups, only a trend.

This type of result, in which the associated benefits and risks are dose related, is called a J-shaped curve (because of the shape the data makes when plotted on a graph). It suggested that while moderate coffee consumption is associated with benefits for the heart, the benefits begin to fall and the risk begins to rise at a certain level of consumption. The 11% reduction in risk of heart failure associated with four to five servings of coffee a day was at the bottom of the J-shaped curve.

How did the researchers interpret the results?

The researchers conclude that moderate coffee consumption is associated with a reduced risk of heart failure, with the most reduced risk associated with four servings a day. At higher levels of consumption, they suggest there is a potentially higher risk of heart failure. They speculate that this apparent benefit may be linked to the possible effect of coffee on two important risk factors: diabetes and high blood pressure.

Conclusion

Although a systematic review is considered a high level of evidence, its results should be viewed with some caution.

Limited number of studies included

hough the study included a relatively large population size, it only included five studies. A larger number of studies may give a more reliable indication of the whether a clear relationship between coffee drinking and heart failure risk exists.

Unclear definition of heart failure

It is unclear how the studies defined the outcome of heart failure. Some studies included people with prior heart attack and some did not. It is therefore unknown whether some people already had heart failure at the start of the study.

Differences between what the studies accounted for

The studies adjusted for factors that may also have been related to risk of heart failure. Four out of five studies adjusted for important confounders of age, body mass index, alcohol consumption and smoking status. Other relevant factors such as diabetes and cholesterol were inconsistently accounted for, while other relevant factors such as high blood pressure do not appear to have been accounted for in any study.

One of the five studies made no adjustment for any confounders. All of these things can affect the reliability of the pooled risk estimates.

Lack of statistical significance

The review authors observed a J-shaped trend in risk but not all of the risk calculations were statistically significant. For consumption between 1-2 and 4-5 servings a day there was a small but significant decrease in risk compared to no consumption (with the greatest decrease in risk at 4-5 servings). For consumption between 5 and 11 cups a day, none of the risk calculations were actually statistically significant (that is, there is no real evidence of a difference in risk between no coffee consumption and consumption of between 5 and 11 cups, only a trend).

Self-reporting at a single time point

The studies relied on self-reported coffee drinking, which may mean the results are less reliable. The studies also only measured coffee consumption at a single time point so took no account of changes in coffee consumption over the years.

Differing strengths and sizes of coffee

The study did not take into account the strength of coffee that people reported drinking. Nor did it examine:

whether caffeinated or decaf was drunk
the method of preparation used (filtered, boiled, espresso)
the volume of coffee drunk
whether the coffee was drunk with milk or sugar

There is some confusion in the newspaper reports about the size of servings used in the review. The authors say that four cups of coffee a day is a northern European serving (125ml-150ml), which equates to slightly more than two cups in US coffee chains, where a standard serving varies from 295-590ml. In the UK, servings in coffee shops vary, with one popular chain serving a range of sizes from 89ml for an espresso to 470ml for a grande.

This review cannot prove that coffee is good for the heart and can only show a possible association between coffee drinking and risk of heart failure. While the study is not particularly helpful with regards to making lifestyle choices, drinking 11 cups of coffee a day is probably not a good idea for other reasons.