Is coffee bad for you?

Summary

---

More or Less – "Is coffee bad for you?"

Host: Ruth Alexander
Guests: Dr. Vivek Muthu (Economist Intelligence Unit, Bayesian), Emily Oster (University of Chicago)
Date: August 24, 2013

---

Episode Overview

This episode of More or Less explores the question: Is coffee bad for you? Host Ruth Alexander examines recent media headlines about coffee's impact on mortality, breaks down the numbers behind the research, and investigates advice for pregnant women. Guests Dr. Vivek Muthu and economist Emily Oster bring statistical clarity and personal insight to the debate about coffee, pregnancy, and risk.

---

Key Discussion Points & Insights

Coffee and Mortality: Understanding the Research

• Media Reporting and Listener Confusion
  The episode begins with Ruth Alexander discussing a BBC News item claiming "People who drink more than four cups of coffee a day could be shortening their lives... by more than 50%."
  Listeners question how the risk of dying, already at 100%, can somehow increase "by a further 50%." ([00:47])

• Interpreting "Increased Mortality Risk"
  Dr. Vivek Muthu clarifies the study’s findings:

  "At any given moment in time, people who drink more than 28 cups per week, they had a 50% greater chance of dying than those who didn't drink." ([01:24])
  "At any given moment in time they'd be one and a half times the risk of dying in the coffee drinking group compared with the non-coffee drinking group." ([01:24])

  He emphasizes this is a relative figure, so the absolute risk might still be low:

  "If your risk of dying at any given moment is very low, then elevating it to one and a half times that is still a low number." ([01:47])

• Caveats and Limitations
  Dr. Muthu highlights several important caveats:

  • Correlation does not equal causation: heavy coffee drinkers may also smoke, drink alcohol, or have other habits.
  • Statistically adjusting for confounders has limits; residual confounding may persist.
  • The danger of over-interpreting a single study:

    "It's a single study... Even if there's a mythical study that's perfectly conducted, you can have random variation in the results." ([02:06])

  • Reliability comes from accumulating evidence across studies:

    "What you need to do is repeat the experiment several times.... That gives you a statistically more reliable result." ([02:06])

• Personal Choices
  Dr. Muthu:

  "I don't plan to give up his two or three cappuccinos a day anytime soon." ([03:13])

---

Coffee and Pregnancy: Navigating the Advice

• Confusing Guidelines
  Economist Emily Oster recounts frustration over contradictory advice about coffee in pregnancy:

  "I found people disagree. But then even the books disagreed and my doctor didn't always agree with the books." ([03:42])

  Recommendations range from "none" to up to six cups a day. ([04:18])

• Delving into the Evidence
  Emily Oster researched the medical literature herself:

  "Certainly it is fine to have two cups... I think much of the evidence also suggests that [three to four cups] is fine." ([04:33])

  But there is some evidence that very high intake (six to eight cups) may be risky.

• Why the Evidence is Unclear
  Ruth Alexander explains why studies in this area fall short:

  • No randomized controlled trials (RCTs) for ethical reasons; instead, researchers rely on observational studies.
  • Habits and wellbeing confound the data (e.g., sicker pregnant women may naturally drink less coffee, and nausea is associated with healthier pregnancies). ([05:29])

  Emily Oster:

  "It's very hard to know whether the differences you see in what happens to them are really about the coffee, or they're really about something else that's different." ([05:29])

  "Being sick is a sign of a healthy pregnancy. And so when we see that women who drink less coffee also have more successful pregnancies, we don't really know if that's just about coffee or whether it's really a confounding factor from this nausea." ([05:29])

---

Risk, Food, and Personal Agency in Pregnancy

• Alcohol
  Oster examined data suggesting that while "excessive drinking, heavy drinking in pregnancy is very dangerous," having an occasional drink is not linked with adverse outcomes ([06:15]):

  "...children of women who drink occasionally have very similar outcomes to the children of women who abstain." ([06:15])

  Personal conclusion:

  "I had maybe three glasses of wine total in the first trimester, and then in later trimesters, maybe half a glass, three or four times a week of wine." ([06:46])

• Food Restrictions
  Oster laments the overwhelming list of foods to avoid:

  "I was carrying around this list, trying to sneak it out during lunch so people wouldn't know that I was pregnant." ([07:02])

  By examining Centers for Disease Control data on listeria outbreaks, she found some foods (unpasteurized soft cheese, deli turkey) are more risky, while others (ham, cantaloupe, celery) are less predictable:

  "There was no particular reason to avoid ham." ([07:51])

• Simplification versus Empowerment
  Oster advocates empowering pregnant women with nuanced data:

  "Much of the information that we get is a little oversimplified... it's time for women to sort of think through these decisions and understand these decisions, really the details of them, not just some broad outlines." ([08:03])

  "...that will ultimately really improve the quality of medical care. Because then you can come into your doctor and say, look, now I understand what's going on. Let's talk about how this applies to my particular situation." ([08:03])

---

Notable Quotes

• Dr. Vivek Muthu:

  "It's a single study... What you need to do is repeat the experiment several times or repeat the observation several times, and then that gives you a statistically more reliable result." ([02:06])

• Emily Oster:

  "I found people disagree. But then even the books disagreed and my doctor didn't always agree with the books." ([03:42]) "Much of the information that we get is a little oversimplified... it's time for women to sort of think through these decisions and understand these decisions." ([08:03])

---

Timestamps for Important Segments

• 00:13 – Ruth Alexander introduces the coffee study & listener confusion
• 01:24 – Dr. Muthu explains what "increased mortality risk" means
• 02:00 – Caveats: confounders, single-study limitations
• 03:13 – Dr. Muthu's personal coffee habits
• 03:42 – Emily Oster's quest for caffeine clarity during pregnancy
• 04:33 – Oster's analysis and recommendations for coffee in pregnancy
• 05:29 – Discussion on limitations of observational studies
• 06:15 – Oster on alcohol risk during pregnancy
• 07:02 – Oster on food restriction and listeria data
• 08:03 – Oster's concluding thoughts on empowering decision-making in pregnancy

---

Summary

More or Less carefully debunks sensational headlines, stresses statistical nuance, and encourages listeners—especially pregnant women—to make more informed, individualized choices based on a realistic understanding of scientific evidence. High coffee consumption may be associated with increased mortality risk, but absolute risks are typically low, underlying studies are limited, and personal circumstances matter greatly. When it comes to pregnancy, individual agency, nuanced data, and consultation with healthcare providers are more valuable than rigid, oversimplified rules.

---

Transcript

[BBC Announcer]

Thank you for downloading from the BBC. For details of our complete range of podcasts and our terms of use, go to bbcworldserveys.com podcasts

[Ruth Alexander]

this is the short edition of More or Less, first broadcast on the BBC World Service. Hello and welcome to More or Less on the BBC World Service. I'm Ruth Alexander. The BBC News announced this recently. People who drink more than four cups of coffee a day could be shortening their lives. A study of more than 40,000 people suggests that at that level, coffee can increase the risk of dying by more than 50%. Eagle Eared listeners Stuart and Andrew were quick to email us at more or lessbc.co.uk how does coffee raise a likelihood

[BBC Announcer]

that I understand to be already 100% by a further 50%.

[Ruth Alexander]

Does this mean their chances of dying

[BBC Announcer]

go from 100% to 150%?

[Ruth Alexander]

I think we should be told now what the research which was carried out on people in the United States actually said was. Men and women aged under 55 who drank 28 cups of coffee a week had a greater than 50% increased mortality risk. Now what that means is explained by Dr. Vivek Muthu, Director for Healthcare at the Economist Intelligence Unit and chief Executive of the healthcare evidence consultancy Bayesian.

[BBC Announcer]

What they mean is not that you have 150% chance of dying, it's that at any given moment in time, people who drink more than 28 cups per week, they had a 50% greater chance of dying than those who didn't drink. So in other words, any given moment in time they'd be one and a half times the risk of dying in the coffee drinking group compared with a non coffee drinking group.

[Ruth Alexander]

That does sound quite alarming.

[BBC Announcer]

It does sound quite alarming, but this is a relative figure. So if your risk of dying at any given moment is very low, then elevating it to one and a half times that is still a low number.

[Ruth Alexander]

So should we reduce the amount of coffee we're drinking if we're drinking more than four cups a day?

[BBC Announcer]

Well, I mean, I think what makes the headlines is this big figure. What is less sexy to put on the news is all the caveats behind the study, which, to be fair to the authors, they were very explicit about in the actual research. When you look at it, there could be lots of other factors other than the coffee that could be contributing to this effect. People who drink coffee more, maybe smoke more or drink more alcohol, and that may be having a contributory effect. What they've done is try and statistically strip that out using certain methods. But even though you can try to strip it out statistically, these factors can still have an effect. The other thing that's really important about this, it's a single study. Even if there's a mythical study that's perfectly conducted, you can have random variation in the results. So if I toss a coin 10 times, an unbiased coin, quite easily you can imagine I might end up with eight heads. That's just chance. What you need to do is repeat the experiment several times or repeat the observation several times, and then that gives you a statistically more reliable result. So it's really important to have a body of evidence of similarly designed studies before you come to a very definitive conclusion.

[Ruth Alexander]

Dr. Vivek Muthu, who doesn't plan to give up his two or three cappuccinos a day anytime soon. The one group of people for whom coffee forms a particular question mark for their health is pregnant women. But when Emily Oster, an associate professor of economics at the University of Chicago, became pregnant, she found the advice on coffee to be frustratingly conflicting. But she figures she had the statistical training to get to the facts of the matter. She's written up her findings in a book, Expecting Better.

[Emily Oster]

When I first found out I was pregnant, I really wanted to have a cup of coffee. It was like first thing in the morning. And then I thought all of a sudden, oh my gosh, am I even allowed to have one cup of coffee? And so I sort of started to look into that on the Internet and you know, I found people disagree. But then even the books disagreed and my doctor didn't always agree with the books. And I ended up actually trying to go into the medical literature and go into the studies and data on this and use basically my training and statistics to try to figure out what's the real evidence on this.

[Ruth Alexander]

Say, what are people told about coffee and pregnancy?

[Emily Oster]

Some people will say don't have any. Some people and organizations will say it's fine to have two cups, don't have more than that. Some will say three. I read books that said six.

[Ruth Alexander]

So what did you find out?

[Emily Oster]

Certainly it is fine to have two cups. When I started to think about, I'm more of like a three to four cup a day coffee lady. And when I started to look into that, I think much of the evidence also suggests that that is fine. When you look at more like six or eight cups of coffee, there is some more evidence that that might be risky.

[Ruth Alexander]

Emily found it impossible to come to a definitive conclusion because most of the Studies are not what are called randomized controlled trials. Randomized controlled trials, as loyal listeners will know, are the gold standard. They are experiments where the subjects under study have been split into two groups at random. One group would drink coffee, the other group would not, and the scientists could easily see any effects coffee had, but it wouldn't to divide pregnant women into two random groups and experiment on them like this. So scientists have to study people's normal habits, and that makes the task of isolating the effect of coffee difficult.

[Emily Oster]

There are a lot of differences between people in the studies who drink coffee and people who don't. And it's very hard to know whether the differences you see in what happens to them are really about the coffee or they're really about something else that's different. So in the case of pregnancy, the biggest issue is that caffeine consumption correlates very strongly with how nauseous you are. So a lot of pregnant women are very sick, especially early on. The women who are sicker tend to drink less coffee. But we know that being sick is a sign of a healthy pregnancy. And so when we see that women who drink less coffee also have more successful pregnancies, we don't really know if that's just about coffee or whether it's really a confounding factor from this nausea.

[Ruth Alexander]

So once you'd sort of looked at coffee, you there were other things that were worrying you.

[Emily Oster]

I looked into alcohol, excessive drinking, heavy drinking in pregnancy is very dangerous. That's something you see very clearly in the data. But when I looked at having an occasional drink, the evidence that we have on that suggests that that is safe. So we don't have large randomized trials, but we do have a lot of high quality studies which show that the children of women who drink occasionally have very similar outcomes to the children of women who abstain.

[Ruth Alexander]

What did you conclude about what was safe for you to drink?

[Emily Oster]

So I had maybe three glasses of wine total in the first trimester, and then in later trimesters, maybe half a glass, three or four times a week of wine.

[Ruth Alexander]

And then there are foods. I'm sort of struck if I'm at a lunch with a pregnant woman that they have to be very careful about what they eat. Do you agree?

[Emily Oster]

Sort of. One of the things I found overwhelming about the food list that I got was it was just so long, there were so many things to not eat. And so I was like carrying around this list, trying to sneak it out during lunch so people wouldn't know that I was pregnant. And so What I came to think is, look, I need to understand why these foods are restricted. So I just went and looked at the Centers for Disease Control in the US data on what are the sources of listeria infection in the last 15 years. And I found, you know, about 20% of those outbreaks can be linked to unpasteurized soft cheese, about 10% to Deli Turkey. But there was like one outbreak linked to ham, there's one outbreak linked to cantaloupe, there's one outbreak linked to celery. And I came to think for a lot of these things, there's kind of no way to predict this. And so there was no particular reason to avoid ham.

[Ruth Alexander]

And so overall, through this quite personal journey of research, what did you conclude about what pregnant women are told and what information they're given?

[Emily Oster]

I think that much of the information that we get is a little oversimplified. So there just isn't time when we're going to the doctor for four minutes a month for the doctor to explain everything that is going on and also help us think through the decisions. And so I kind of, a lot of what I came to think is that it's time for women to sort of think through these decisions and understand these decisions, really the details of them, not just some broad outlines, and that that will ultimately really improve the quality of medical care. Because then you can come into your doctor and say, look, now I understand what's going on. Let's talk about how this applies to my particular situation. And that's where really productive conversations can happen.

[Ruth Alexander]

Emily Oster, the proud mother of a healthy two year old girl. It's important to note, though, that Emily's conclusions differ from the advice given by organisations. The National Health Service in England, for example, says that the best thing to do in pregnancy is to avoid alcohol altogether because the potential damage to your unborn baby can be serious. If there are any numbers you'd like us to take a closer look at, do email us. Our address is more or lessbc.co.uk and you can also sign up for a free download of our program at our bbcworldservice.com/more or less.

[BBC Announcer]

There are dozens of different podcasts now available from the BBC, including news, documentaries, science, business, arts and sports. The details of them all go to bbcworldservice.com podcasts.