Does coffee make you live longer?

Behind the Headlines

Wednesday July 12 2017

Pregnant women shouldn't drink more than two mugs of coffee a day

Coffee may contain useful antioxidants

"Drinking three cups of coffee a day could add years to your life, suggest studies," reports the Metro.

It follows the results of European and US studies that looked at the relationship between how much coffee people drink and death.

The European study included more than 450,000 people. Researchers found men who drank the highest amounts of coffee had a 12% overall reduced risk of death at follow-up from causes including cancer and cardiovascular, digestive and respiratory conditions.

Women had a 7% reduced risk overall, but a greater risk of dying of cancer the more coffee they drank.

These findings need to be interpreted with caution – the research doesn't prove coffee reduces the risk of death. Many other factors that might have played a role weren't taken into account.

Drinking coffee can be part of a healthy, balanced diet. Current guidelines recommend drinking no more than around four cups a day. 

Pregnant women are advised to consume no more than 200mg of caffeine a day, the equivalent of two mugs of instant coffee.

There are no magic shortcuts (or magic coffee beans) for achieving good health and living a longer life.

Leading a healthy lifestyle by having a varied, healthy diet and getting regular exercise is the best way to achieve this. 

Where did the story come from?

The study was carried out by researchers from a range of more than 20 academic and health institutions across Europe, including Imperial College London and the International Agency for Research on Cancer in France.

The European research was funded by a number of institutions, including the European Commission Directorate General for Health and the Consumers and International Agency for Research on Cancer.

Two authors declared potential conflicts of interest, naming grants from the pharmaceutical companies Biogen, Merck and Pfizer, although the companies weren't involved in this study.

Another author declared receiving grants from Unilever and FrieslandCampina, two consumer goods companies also not involved in the research.

The study was published in the peer-reviewed journal Annals of Internal Medicine.

A second study from the US looking at the same topic was published in the journal at the same time, and reported similar positive results.

The UK media's coverage of this research was generally accurate, with The Guardian rightly highlighting the fact "scientists say that the link might just be down to coffee drinkers having healthier behaviours". 

What kind of research was this?

This cohort study looked at data from people enrolled in the European Prospective Investigation into Cancer and Nutrition (EPIC) to see if there were links between coffee consumption and overall deaths, as well as deaths from specific diseases.

This type of study is good for looking at this kind of issue, as it involves people already participating in other research and allows data on a large number of people to be examined.

But cohort studies can't show cause and effect, so aren't able to prove that drinking coffee decreases or increases likelihood of death.

randomised controlled trial where people are put into groups to either drink coffee or not drink coffee until they died would be needed to prove this, something that wouldn't be feasible.

What did the research involve?

The researchers took data from 451,743 participants, mostly over the age of 35, from the EPIC study and looked at their coffee consumption and death from all causes and specific causes.

Participants were recruited between 1992 and 2000, mostly from the general population of 10 European countries: Denmark, France, Germany, Greece, Italy, the Netherlands, Norway, Spain, Sweden, and the UK.

People who reported having cancer, heart disease, diabetes or a history of stroke at the start weren't included in the study.

Similarly, people who reported an extremely high or extremely low calorie consumption weren't included, as these people would not be representative of the population at large.

People were also excluded when follow-up information and information on coffee consumption was missing.

Participants recorded the number of cups of coffee they drank each month, week or day through self-reported questionnaires or interviews.

Coffee consumption (in ml a day) was calculated using the typical cup sizes for each institution involved per country.

The amount consumed was split into four quartiles:

  • non-consumers
  • quartile 1 (low consumption) – up to 83ml a day for UK data
  • quartile 2 (low to medium consumption) – up to 380ml for UK data
  • quartile 3 (medium to high consumption) – up to 488ml for UK data
  • quartile 4 (high consumption) – above 488ml

The quartiles were country specific, with the average daily amount ranging from 93ml a day in Italy to 900ml a day in Denmark.

Data on cause and date of death was collected from cancer registries, local health organisations and death records, as well as through active follow-up from other sources.

Specific causes of death included digestive, respiratory, circulatory and cerebrovascular causes, as well as ischaemic heart disease, cancer, suicide, and external causes.

Information on the following potential confounding factors was recorded and taken into account during the analysis:

  • education
  • smoking
  • alcohol consumption
  • physical activity
  • diet
  • body mass index
  • use of oral contraceptives and menopausal hormone therapy, as well as menopausal status

What were the basic results?

After an average follow-up of 16.4 years, there were 41,693 deaths. Among these, 18,003 were from cancer, 9,106 from circulatory diseases, 2,380 from cerebrovascular diseases, and 3,536 from ischaemic heart diseases.

For all causes of death:

  • Men who drank the highest amount of coffee had a 12% lower risk of death than non-consumers (adjusted hazard ratio [aHR] 0.88, 95% confidence interval [CI] 0.82 to 0.95).
  • Women who drank the highest amount of coffee also had a 7% lower risk of death than non-consumers (aHR 0.93, 95% CI 0.87 to 0.98).

For specific causes of death:

  • Men who drank the highest amount of coffee versus non-consumers and low consumers had a 59% lower risk of death from digestive disease (aHR 0.41, 95% CI 0.32 to 0.54).
  • Women who drank the highest amount of coffee versus non-consumers and low consumers had a 40% lower risk of death from digestive disease (aHR 0.60, 95% CI 0.46 to 0.78).
  • Women who drank the highest amount of coffee versus non-consumers had a 22% lower risk of death from circulatory disease (aHR 0.78, 95% CI 0.68 to 0.90).
  • Women who drank the highest amount of coffee versus non-consumers had a 30% lower risk of death from cerebrovascular disease (aHR 0.70, 95% CI 0.55 to 0.90).

One negative finding was that women who drank the highest amount of coffee had a 12% higher risk of death from cancer (aHR 1.12, 95% CI 1.02 to 1.23). No other associations were seen between coffee consumption and the other causes of death studied.

The US study showed similar findings in that higher coffee consumption was linked to a lower risk of death.

How did the researchers interpret the results?

The researchers concluded that, "Our results suggest that higher levels of coffee drinking are associated with lower risk for death from various causes, specifically digestive and circulatory diseases."

They added: "Because coffee consumption is so widespread and intakes are modifiable, its potentially beneficial clinical implications should be carefully considered." 

Conclusion

This study, conducted on a large number of people across Europe, was backed up by similar findings in the US. It appears to show some association between people who drink higher amounts of coffee and a reduced risk of death.

But the "potentially beneficial clinical implications" need to be considered carefully for a number of reasons:

  • Although the analyses were adjusted for some confounding variables, there may be a number of other factors that differ between the groups that account for the differences in death, such as socioeconomic status, family history, other medical conditions, and use of medication to name a few.
  • Participants with a range of illnesses, including cancer, heart disease, stroke or diabetes, were excluded from the study. These people may have different coffee habits from those included in the study, biasing the results.
  • Coffee consumption was self-reported and might have been over or underestimated, leading to inaccuracies in the results.
  • Coffee consumption was only assessed at one point in time – people's habits might vary greatly over days, months and years, so one snapshot might not give an accurate picture of lifelong coffee drinking habits.
  • Combining different cut-off levels of coffee per country may lead to inaccurate results.
  • Lots of analyses were carried out on a range of diseases, most of which weren't significant, and the likelihood of finding some significant results by chance would be fairly likely. Those significant results reported therefore need to be treated with caution.
  • Not all outcomes were positive: women had a greater risk of death from cancer if they drank higher amounts of coffee.

The media like to run stories on one single drink or "superfood" that will "guarantee" good health. This, of course, is nonsense: the only way to increase your chances of leading a longer, healthier life is to have a healthy, balanced diet and exercise regularly.

Analysis by Bazian

Edited by NHS Choices

Links to the headlines

Drinking three cups of coffee a day could add years to your life, suggest studies. Metro, July 11 2017

Coffee cuts risk of dying from stroke and heart disease, study suggests. The Guardian, July 10 2017

Coffee drinkers live longer – perhaps. BBC News, July 11 2017

Drinking coffee could reduce your chance of death, scientists say. The Independent, July 10 2017

Links to the science

Gunter MJ, Murphy N, Cross AJ, et al. Coffee Drinking and Mortality in 10 European Countries: A Multinational Cohort Study. Annals of Internal Medicine. Published online July 11 2017

Ratings

How helpful is this page?

Average rating

Based on 7 ratings

All ratings

6  ratings
1  ratings
0  ratings
0  ratings
0  ratings

Add your rating