In medicine, a lot of what we know about how to cut the risk of an early death is rooted in high quality evidence from clinical trials or health policies. This is particularly true when it comes to the influence of smoking, obesity, diabetes, very high cholesterol and blood pressure.
However, what is less certain is the impact of specific foods or drinks. Coffee is a prime example. On the one hand, multiple studies have suggested that those who drink it tend to develop heart disease, diabetes, cancer and other conditions less often, and live longer than those who do not. It also appears that the more cups of coffee you drink, the lower the risks.
On the other hand, there is concern that too much coffee may cause some harms due to caffeine content. Caffeine stimulates the brain and generally causes alertness, but too much can lead to mild dependence and withdrawal symptoms, such as poor sleep, irritability and headaches.
Advertisement
So, is coffee good or bad for us? The hunt for the answer goes on. The latest contribution is a , which looked at around half a million people in the UK.
Eight cups a day
It reported that coffee drinking was linked to lower risk of early death, including in those drinking eight or more cups per day. It also showed that this finding was consistent irrespective of whether individuals were genetically fast or slow at metabolising caffeine, or whether the coffee was caffeinated or decaffeinated.
Based on these results, the study concluded that coffee drinking could be part of a healthy diet and that non-caffeine constituents might be responsible for proposed health benefits. This prompted headlines declaring that .
As a lifelong tea drinker, might I be persuaded by such evidence to move to coffee to lessen my risks of a range of illnesses? No. Here鈥檚 why. With all observational studies 鈥� those that link one factor to another but cannot show cause and effect 鈥� there is a risk of confounding. This means that some other thing about people who drink coffee, and not the coffee itself, may be the reason for lower risk of early death.
That coffee seems to 鈥減rotect鈥� against a broad range of diseases, ranging from cancers to diabetes and neurological conditions, lends some support for such confounding. Few proven health interventions influence such a diverse range of conditions, with perhaps smoking being the exception.
Reverse causality
Furthermore, 鈥渞everse causality鈥�, whereby people who develop an illness or start to feel unwell lessen their intake of coffee, could be a factor behind the apparent health benefits for coffee. This to my mind has not been examined as a potential confounder. It ; many people stop drinking alcohol when ill, which may be why we often hear about studies that suggest non-drinkers have worse health than moderate drinkers.
Finally, observational studies, even if they seem consistent, can give incorrect messages. A good example was hormone replacement therapy (HRT). For many years, numerous studies reported women taking HRT had lower risks of heart disease, a finding completely overturned by the first randomised trial.
I am not saying don鈥檛 drink coffee 鈥� quite the opposite. If you enjoy it and don鈥檛 experience any symptoms, such as headaches, continue to do so. But I would counsel against hoping for some magical effects of coffee to lessen other health risks you may have. Rather, the usual lifestyle advice and risk checks should apply.
If a future randomised trial, one that can show cause and effect, confirmed coffee drinking was beneficial to health, then we would have a more solid footing to recommend it. Until that time, I will continue to enjoy my tea.
Read more: Coffee: The demon drink?; Good hydrations: Is tea or coffee better for you?
