Angelina Jolie’s 87% cancer risk

Summary

---

More or Less – "Angelina Jolie’s 87% cancer risk"

BBC Radio 4, first broadcast May 17, 2013
Host: Tim Harford

---

Episode Overview

This episode of More or Less delves into the numbers and statistics that regularly surface in public debate and the news. The show particularly focuses on the accuracy and interpretation of Angelina Jolie's widely reported "87% cancer risk," statistical misunderstandings about Romanians and crime in the UK, the manipulation of survey data in public discourse, how official jobs and unemployment statistics can mislead, and the often-neglected issue of publication bias in reporting.

---

Main Segments & Key Insights

1. Angelina Jolie's 87% Cancer Risk (00:34–04:47)

Key Discussion Points

• Angelina Jolie’s Story: Jolie underwent a preventative double mastectomy after her doctors estimated she had an 87% lifetime risk of developing breast cancer.
• How the Figure Was Calculated:
  • The risk number likely comes from a computer algorithm factoring in genetic information, family history, and other health data.
    • Dr. Kat Arney (Cancer Research UK): "There are various computer programs... you put genetic information, family history, all sorts of stuff in, and it spits a number out..." [01:06]
  • After surgery, her risk dropped to about 5%, lower than the average woman.
• BRCA1 Gene:
  • Everyone has the BRCA1 gene; a faulty (mutated) version increases cancer risk significantly.
  • If a parent has the mutation (as Jolie’s mother did), their children have a 50% chance of inheriting it.
  • Risk among BRCA1 carriers: “...your risk of developing breast cancer goes from about 12% on average to between 50 to 80% or even more...” – Dr. Kat Arney [02:15]
• Contextual Risk:
  • Average UK woman lifetime risk: 12% (1 in 8).
  • Age-based incidence:
    • Before 30: 1 in 2000.
    • Before 50: 1 in 50.
    • Before 70: 1 in 13. [02:49]
  • Global variation: Highest in Western countries, lowest in Middle/Eastern Africa, attributed to differing childbirth and breastfeeding patterns.
• General Cancer Risk:
  • UK-wide: 1 in 3 men/women diagnosed in their lifetime.
  • Future projections: For men born in the 2020s, this may rise to 1 in 2.
• Modifiable Risk Factors: Obesity, smoking, excessive alcohol, inactivity, poor diet all increase risk, but can be addressed.

Notable Quote

• Dr. Kat Arney: “...between men and women, there's roughly a 1 in 3 lifetime chance of being diagnosed with cancer. That's getting close to one in two for men...” [04:04]

---

2. Romanians and Crime in UK – Debunking a Statistic (05:12–10:25)

Key Discussion Points

• Claim: Christine Hamilton stated that of 68,000 Romanians in the UK, 28,000 had been arrested for serious offenses since 2008—implying a rate of over 40%.
• Breaking Down the Data:
  • Population Underestimation:
    • ONS figures show c.93,000 Romanians in the UK, making claim percentages off.
  • Arrest Numbers:
    • 28,000 arrests refers to tally in London only, and multiple arrests of the same individuals are counted, not unique persons.
    • Tim Harford: "It could be, in theory, one totally badass Romanian who's been arrested 28,000 times..." [06:35]
• Relative Context:
  • Romanians: 2nd most arrested group in London (after Polish nationals), but only 9th most populous immigrant group.
  • Arrest ≠ conviction, and not all are for serious crimes; most are for petty offenses (begging, shoplifting, minor thefts). [08:07]
  • Only 624 Romanians were in prison at end of 2012; this is comparatively high relative to their population.
  • The quality of such statistics is problematic due to poor data and reporting inconsistencies.
• Alternative Explanations:
  • Romanian ambassador suggests close police cooperation may result in higher arrest stats, not necessarily higher crime.
  • Importance of learning host country rules cited to reduce future issues.

Notable Quotes

• Tim Harford: "Therefore you can't say that it's a percentage of the number of people who've been arrested. It could be, in theory, one totally badass Romanian who's been arrested 28,000 times..." [06:35]
• Dr. John Ginger (Romanian Ambassador): “64% of all crimes committed by Romanians in the United Kingdom are the so-called petty crimes...” [07:51]

---

3. The Reliability of PR Surveys in Public Debate (10:25–13:57)

Key Discussion Points

• Teenagers and History Survey: Michael Gove cited surveys showing shocking ignorance among teenagers, e.g., that 58% believe Sherlock Holmes was real.
• Source Quality:
  • Underlying surveys often conducted by PR firms with small, possibly unrepresentative online panels.
  • Many supposed “child” respondents may actually be adults incentivized by small payments.
• Press Coverage Hypocrisy:
  • Newspapers criticize PR-based survey stats, yet run similar stories themselves.
  • Even the BBC falls into this trap with trivial surveys (e.g., "lucky underwear" during exams [13:25]).
• Prevalence: 30–40 such PR-generated stories appear in the UK media each week.

Notable Quotes

• Michael Marshall (Merseyside Sceptic Society): "There are limits, the number of people who can take each survey. So there's a massive incentive on people to take as many surveys as they can and not necessarily care an awful lot about the results..." [12:00]
• Tim Harford (dryly): "At least the BBC would never stoop so low." [13:23]

---

4. The Curious Case of Private Sector Job Creation Statistics (14:40–17:36)

Key Discussion Points

• Government Claims: Over 1 million new private sector jobs since current government took office—more net jobs in 2 years than Labour managed in 10.
• Statistical Manipulation:
  • Many “new” jobs were previously public sector positions (e.g., college reclassification in education) simply moved by bureaucratic changes.
  • Similarly, during Labour's tenure, bank nationalisation shifted jobs into the public category.
• Workfare and Employment Status:
  • Some people on government training schemes or unpaid work placements are classified as "employed" in official statistics, regardless of pay.
  • Nick Palmer (ONS): "It's not a key part of the definition of people on schemes of whether they're receiving income for the work or not." [17:19]

Notable Quotes

• Tim Harford: “So yes, Suzanne, there are people out there who are defined as being employed despite the fact that they're not earning a wage.” [17:36]

---

5. The UK’s “Productivity Puzzle” – Why Isn’t Unemployment Higher? (18:33–19:46)

Key Discussion Points

• Despite deep recession and slow economic growth, unemployment hasn’t risen as much as economists would expect.
• Possible Explanations (per Prof. John Van Reenen, LSE):
  • Statistics are likely accurate; GDP underestimation doesn't explain it.
  • Increase in part-time/self-employment and decreased hours partially explain resilience.
  • Main factor: Real wages have dropped dramatically, making labor cheaper, supporting higher employment but lowering productivity (less capital per worker).

Notable Quotes

• Prof. John Van Reenen: "...people's real wages have fallen very dramatically in this recession... So what that means is that because people's wages have fallen, it's cheaper for firms to employ them. So because of that, you see more employment." [19:29]

---

6. Publication Bias: Reporting Only the "Interesting" Stats (20:51–23:47)

Key Discussion Points

• Case Study: Falklands War veterans' suicide rates—persistent claim that more veterans killed themselves post-war than died in the conflict.
• Findings: Only 95 suicides in 30 years, compared to 237 war deaths—giving a suicide rate below the general population.
• Media Response: Major interviews, including with Prof. Simon Wessely, were cancelled because the numbers were deemed "not interesting", illustrating publication bias—important statistics not reported when they contradict prevailing narratives.

Notable Quotes

• Prof. Simon Wessely: “What is a good piece of research but doesn't reveal something terribly wrong with the mental health of those who served in our forces isn't a story.” [22:17]
• Tim Harford: "...you can mislead with statistics simply by not bothering to report that they exist." [22:31]

---

Conclusion

This episode of More or Less exposes the often hidden workings and misuses of statistics: from health risks and crime numbers to job statistics and survey data. The team effectively debunks misleading claims (both in media and government) and reveals the story-telling power and pitfalls of numbers:

• Jolie's 87% cancer risk is specific to personal genetic and family factors, not universally applicable.
• Crime figures cited for Romanians in the UK are misleading due to statistical misinterpretations.
• PR surveys often produce dubious data cited in serious debates.
• Official claims about jobs and employment are frequently massaged via classification tricks.
• Important but unexciting facts (e.g., about veteran suicide rates) may be ignored by newsrooms, an example of publication bias.

A recurring message: Always question the context and provenance of the numbers you hear.

---

Transcript

[A]

Thank you for downloading More or Less from the BBC. This is the version of the programme first broadcast on BBC Radio 4 on the 17th of May 2013. Here's Tim Harford.

[B]

Hello and welcome to More or Less Melting the Statistical chocolate chips in the great news cookie of life. This week we try to unravel the mystery of the UK unemployment figures. And is Christine Hamilton right about Romanians?

[A]

68,000 of them are already here, of whom 28,000 have been arrested for serious offences since 2008.

[C]

Come on.

[B]

But first, earlier this week, Hollywood actor Angelina Jolie wrote in the New York Times that she'd undergone a preventative double mastectomy. Jolie's doctors had estimated that she had an 87% risk of developing breast cancer. That's a very specific figure, so we've been having a closer look at it. And Ruth Alexander's here. Hello, Ruth.

[A]

Hello. It is very specific and high. So how would this figure have been calculated? Well, we asked Dr. Katani, a senior science manager at the charity Cancer Research uk.

[D]

There are various computer programs where you put all sorts of things in. You put genetic information, family history, all sorts of stuff in, and it spits a number out, which is why she's probably got this very precise sounding number. By having her breasts removed, you remove most of the tissue that could become cancerous. She still says that she has around a 5% chance of developing breast cancer, which is about half of that of an average woman in the general population. But by having this operation, she's got rid of most significant amount of the kind of tissue that could become cancerous and cause her a problem.

[B]

So we think that effectively a computer algorithm has created this figure of 87% based on Angelina's genetic makeup and her family history.

[A]

That's right. But it was the presence of a certain mutated gene that stacked the odds against Angelina and made her take this drastic step. The BRCA1 gene is present in everyone, but mutates in roughly one in every 1,000 women and then can be a trigger for both breast and ovarian cancer. However, if a parent has the mutated gene and Angelina's mother did, she died of ovarian cancer at the age of 56, then their children have a 1 in 2 chance of inheriting it. Kids, Katani explains.

[D]

We know it is actually more common in certain groups, but certainly if you carry this faulty gene, you do have a greatly increased risk of developing the disease. So for breast cancer, your risk of developing breast cancer goes from about 12% on average to between 50 to 80% or even more, depending on exactly the kind of fault you have and a number of other factors. For example, the other genetic variations that you have on top of that, too.

[B]

Now, Kathani said there that the risk of getting breast cancer is on average 12%. That's much lower than Angelina's estimate of an 87% risk, but it's still very high.

[E]

Ruth?

[A]

Yes, a woman in the UK has a 12% chance of getting breast cancer during her lifetime. So it's a lifetime risk. And a third of all incidences of breast cancer occur in women aged over 70. The chance of a woman developing breast cancer before the age of 30 is very slim. One in 2000, that goes up to one in 50 before the age of 50 and one in 13 before the age of 70.

[B]

So it's quite common in later life. But these are statistics from the uk. I guess the story changes depending on where in the world you are.

[A]

It does. The UK statistics, though, are similar to those in the United States and in other Western European countries. The lowest breast cancer rates are in middle and eastern Africa, where women are four times less likely to be diagnosed with the disease than women in the uk. Here's Qatarnia Gun.

[D]

In a lot of African countries where women have children younger, they have more children and they breastfeed them for longer, that does decrease breast cancer risk. So rates of breast cancer are relatively low. But unfortunately, there are many other types of cancer that tend to affect people in those countries. More things like cervical cancer and cancers linked to infectious diseases.

[B]

So what are people's chances of developing any form of cancer during their lifetime?

[A]

Well, frankly, the likelihood is high. Here's the bad news and a little about what to do about it.

[D]

Overall, between men and women, there's roughly a 1 in 3 lifetime chance of being diagnosed with cancer. That's getting close to one in two for men. And probably in the next couple of decades, a man born in, say, 2025, 2027 will have a 1 in 2 lifetime chance of being diagnosed with cancer. And that's for a number of reasons. We're all living longer. And as you get older, your risk of cancer does go up significantly. We know there are lifestyle factors involved. Obesity is linked to many thousands of cases of cancer in the uk. There's also the risk factors that we know about. Obviously, smoking is the big one. Drinking excessive alcohol, not exercising enough, not eating a healthy diet. All of these things can increase the risk of cancer. And these are things people can do things about.

[B]

Kat Arnie from Cancer Research UK and Ruth Alexander. Thank you. Our esteemed colleagues at Radio 4's Any Questions? Got in touch with us to say that one of their panelists, Christine Hamilton, had come out with a baffling figure that got quite a rise from the audience. She began by saying that a campaign group had estimated that 50,000 people will come to Britain every year from Romania and Bulgaria when entry restrictions are lifted.

[A]

68,000 of them are already here, of whom 28,000 have been arrested for serious offences since 2008.

[F]

Come on.

[B]

40% have been arrested. Well, Christine Hamilton appears to have raised a few doubts there. Can she be right? Charlotte MacDonald's here. Now, Charlotte, I have a theory that Christine Hamilton was confused and was actually talking about Romulans, the highly aggressive alien race from Star Trek, and not Romanians at all. What do you think?

[G]

I'm fairly sure she did mean Romanians, Tim, on any questions, she actually applied these numbers to both Romanians and Bulgarians. But the figures she quotes are in fact just related to Romanians. Now, first, let's look at the population figures. 68,000 is actually a very low estimate for the number of Romanians who are in the UK already. It is just an estimate, but the Office for National Statistics annual figures suggest there are around 93,000.

[B]

OK, so there are probably more Romanians here than Christine Hamilton thinks, and that means the percentage of Romanians who've been arrested is a little bit lower, but still it's 30%, three out of every 10 Romanians, if the figure of 28,000 people being arrested is correct.

[G]

Well, actually, that figure of 28,000 only refers to London, and the number of Romanians in the capital is about 48,000. So that would make the arrest rate higher than she thinks. Except this is a tally of arrests,

[B]

not people, so the same person could be arrested several times. Therefore you can't say that it's a percentage of the number of people who've been arrested. It could be, in theory, one totally badass Romanian who's been arrested 28,000 times or more realistically, that there's a relatively small number of repeat offenders and they are making up the majority of these 28,000 arrests. But broadly speaking, Christine Hamilton, she's got the details wrong, but she seems to have the big picture right. She's saying there's a problem and she has a point.

[G]

Well, I took a look at the data for other foreign nationals who've been arrested, and only Polish nationals have been arrested more than Romanians over the four year period that these arrest figures span, and there are around three times as many Polish nationals in London. The Romanians are the second most arrested immigrant group, but they're the ninth most populous immigrant community.

[B]

Do we have any idea of whether

[G]

these are serious offences or in fact offences at all? Because being arrested isn't the same as being found guilty of a crime. But we do know that the arrests were for all crimes, not just serious crimes, as Christine Hamilton suggested.

[B]

Well, this still looks rather worrying. Somebody should talk to the Romanian ambassador.

[G]

Good point. I took the liberty of arranging an interview, Tim. His name is Dr. John Ginger and this is what he has to say.

[H]

I can tell you that 64% of all crimes committed by Romanians in the United Kingdom are the so called petty crimes. So low risk criminality.

[G]

What kind of low level crimes?

[H]

The so called petty crime, which means begging, thefts, shopliftings, but nothing to do with violence, driving crimes, for instance.

[G]

Now, certainly we shouldn't imagine that every single one of the 28,000 arrests led to a successful prosecution for armed robbery or murder. There were only 624 Romanians in prisons in England and Wales at the end of 2012. But note that's actually several times the incarceration rate of the UK population as a whole and the fourth highest prison population of foreign nationals behind Jamaica, Poland and Ireland.

[B]

I feel like we're dancing around this somehow. Can't we get some kind of league table of different nationalities and their propensity to get arrested or charged with serious crimes or imprisoned?

[G]

Well, I'm sure many media outlets would love to produce such a league table, Tim, but it would only be the start of a serious analysis. It would tell you nothing about a nation, only about the demographics of the people who come to a particular country.

[B]

So you mean the French think that British immigrants are a collection of sweet middle class pensioners who've retired to Provence, while the Cypriots think the average Brit is 18 years old and too drunk to walk?

[G]

Something like that. And the other problem with producing this fancy league table you have in mind is that the data are terrible. It's hard to say where people are from. Is it place of birth, is it nationality plus the ons stress? It's always hard to estimate foreign national populations in the uk and the figures we've been looking at are for arrests, not convictions.

[B]

Well, I take your point. Even so, these figures do seem to suggest that there is some kind of problem with Romanian nationals and crime. Even if it isn't necessarily serious or violent crime.

[G]

Yes, maybe. Although the Romanian ambassador has another theory. He thinks that the embassy's good relationship with the police has something to do with the statistics.

[H]

Romanian authorities have developed an excellent cooperation with the Metropolitan Police and the British Transport Police. Because this cooperation is so good, we got results. So if you get results, you arrest people. It is about understanding the rules of the host country. Driving in Britain is a bit different from driving in Europe, for instance, but once the rules are learned, the level of crime will certainly diminish.

[B]

The Romanian ambassador who was speaking to more or less's Charlotte MacDonald. Thank you, Charlotte. The Education Secretary, Michael Gove, has been a friend of More Or Less ever since he appeared on the programme to talk about his love of the game to Dungeons and Dragons. If you want to hear that interview, please visit BBC.co.uk more or less. But we cannot let our shared nerd heritage get in the way of the truth. And Mr. Gove has received unwelcome attention of late. He justified his reforms to the history curriculum by pointing out that survey after survey shows that teenagers are ignorant of key historical events. No, Sherlock Holmes was not a real historical figure, as 58% of teenagers apparently believe. And yes, Winston Churchill was real. I have nothing to offer but blood,

[E]

toil, tears and sweat, contrary to the

[B]

views of one in five young people. But where was Mr. Gove getting these shocking statistics? A swift Freedom of Information request later, and it turns out that the source was the likes of Premier Inn or UK Gold tv, who'd commissioned the surveys and turned them into press release releases designed to push their products. Still, at least fine, upstanding hotel chains can be relied upon to commission their surveys in a thoroughly rigorous fashion.

[I]

This is the real problem. They're absolutely not rigorous at all.

[B]

Ah, that's Michael Marshall of the Merseyside Sceptic Society, who also writes a blog called Bad pr. He can probably be described as the country's top slayer of lightweight surveys.

[I]

They're actually almost always done by a very small number of polling companies who tend to be affiliated with the PR company themselves. And these polling companies have a user base of maybe 100,000 people. And these people will sign in every day, and they're almost always done online, which means you're not actually monitoring who the person is who's filling it in. So when the survey says, have you got a child who's 11 to 16 and is that child with you? You can't tell when the person said yes, whether the child's actually there. So I actually suspect there's a very good possibility that many of the surveys that were cited by Gov weren't actually taken by just children at all. I imagine there's some adults who've crept in there just to earn a little bit of money, because people get paid a very small amount of money for every survey that they take. And there are limits, the number of people who can take each survey. So there's a massive incentive on people to take as many surveys as they can and not necessarily care an awful lot about the results, because as far as they're concerned, it's just a little bit of fun. And these stories are absolutely phenomenally ubiquitous. I think that something around the level of 30 to 40 of these stories appear in the British press every week.

[B]

It sounds to me that it's a great thing that the newspapers have picked up on this and the Daily Telegraph, I noticed, has criticised Michael Gove. So well done, the Daily Telegraph.

[I]

Well, not so fast on well done, the Daily Telegraph. Because at the same time as complaining that Michael Gove uses PR surveys to back up his points, the Daily Telegraph also a week earlier published a PR survey of the Raw. That was about the worst time for your neighbours to be mowing their lawn. And this was a survey put out by B and Q.

[B]

At least the BBC would never stoop so low.

[I]

Well, you may say that, but unfortunately, just last week the BBC ran a story about students and exams using lucky underwear. And this was a survey put out by BIC which helpfully said, don't use your lucky underwear. What you should do is revise thoroughly and and bring a spare pen.

[B]

And that ran on the BBC website.

[I]

It absolutely did.

[B]

Well, at least it can't get any worse.

[I]

I wish that were the case, but unfortunately the Independent who did criticise Michael Gove actually published the very survey that he ended up citing. In fact, the Independent essentially said, this MP is wrong for believing what we've written.

[B]

Blogger and skeptic Michael Marshall, you're listening to More or Less with me, Tim Harford. If you have an hilariously bad piece of PR based journalism or any other statistical story you'd like us to take a look at, please write to more or lessbc.co.uk. we haven't heard from the Chancellor of the Exchequer yet this week. Let's put that right. When I was in the West Midlands a couple of weeks ago, there had been 67,000 new private sector jobs in that one region alone. George Osborne isn't the only senior member of the Government to tell us that the private sector is creating jobs by the dozen. The Prime Minister, David Cameron, made this claim at last Autumn's Conservative Party conference.

[E]

Since this government took office, over 1 million new jobs have been created in the private sector. That is more net in the last two years than Labour managed in 10 years.

[B]

Well, this sounds surprising. Mr. Cameron took office in the wake of a deep recession and the economy has been pretty flat since then. Loyal listener Suzanne Jarrett has been wondering whether there isn't some kind of statistical tomfoolery going on.

[D]

Are outsourced public sector jobs counted as

[B]

jobs created by the private sector? She wondered. We've passed her ponderings on to the man who should know, Nick Palmer, senior labour market statistician at the Office for National Statistics. Yes, he told me if a public sector organization outsources some of its work to the private sector, then that work shows up in the official statistics as new private sector jobs. And sometimes there are major reclassifications, as there was last year in education.

[C]

Further education colleges and sixth form colleges in England were reclassified from the public to the private sector as a result of the way they were being funded. So that effect kind of moved people from the public to the private sector.

[B]

So Mr. Cameron was chalking up a million private sector jobs, but that includes nearly 200,000 jobs in education moved from the public to the private sector by a reclassification exercise. And this works both ways. The failure of the Labour government to preside over many new private sector jobs in the previous 10 years is clearly partly because of the deep recession of 2008. But it can't have helped that over 200,000 private sector banking jobs suddenly became public sector jobs, thanks to the effective nationalisation of some major banks. A good statistical spot from loyal listener Suzanne Jarrett then. And she had another suspicion.

[D]

Are unemployed people on compulsory workfare jobs counted as employed?

[B]

Nick Palmer from the ons the key

[C]

to it really is asking about their activity whilst they're on that program. So again, we asked them directly, well, what were you doing last week when you're on this program? So anyone who says they're doing some kind of work experience or paid work or doing some work based training or activity within the workplace, their class is in employment. But other people on schemes who are doing some kind of job search activity or help with writing their CV or something, they're usually classed as unemployed. And even if they're not doing either of those things, if they're doing some kind of training or studying at a college and those people are classed as neither employed or unemployed, they're classified as economically inactive. So they're outside the workforce.

[B]

So 300,000 people on some form of government work placement or training scheme, and about 2/3 of them, 200,000, end up being classified as employed. Do we know how many of those are actually drawing down a wage and how many of them are doing it because that's the way that they keep their benefits?

[C]

No, we don't actually ask for that information. We don't collect information on whether they're being paid by an employer or whether they're in receipt of Job Seekers Allowance or something. The reason we don't collect that information because it's not actually used in the definition of whether employed or not. It's not a key part of the definition of people on schemes of whether they're receiving income for the work or not.

[B]

Nick Palmer from the ons and so yes, Suzanne, there are people out there who are defined as being employed despite the fact that they're not earning a wage. Now, you might want to subtract them from the ranks of the employed or add them to the ranks of the unemployed, but you can't because the ONS can't tell us how many people there are in that position. Staying with the subject of the labour market, unemployment figures out this week show 15,000 more people are out of work than in January. The total number of unemployed people stands at 2.52 million. That sounds bad and it is bad, but it's not nearly as bad as we might expect. Given that the recession was very deep and the slump has been unusually long lasting, most economists are wondering why the UK's poor economic growth hasn't thus far led to far higher levels of unemployment. I've been speaking to John Van Reenen, professor of Economics at the London School of Economics.

[E]

One explanation is that numbers are wrong, that simply that GDP is much higher than we think it is. I don't think that's very plausible because the revisions historically that we see for GDP are relatively small. So it's unlikely that GDP will be revised so much. It's going to resolve the productivity puzzle. The second explanation is that what's happened is that although the total number of employees has increased or recovered to what it was before the recession, the amount of hours hasn't self employed people, part time employed people. So when you take that into account, productivity hasn't fallen so badly. However, even when you look at GDP per hour, that has also performed poorly. So I don't think that's the full explanation either. The most plausible explanation for what's happened is people's real wages have fallen very dramatically in this recession, much more so than they did in previous recessions. So what that means is that because people's wages have fallen, it's cheaper for firms to employ them. So because of that, you see more employment. You also see a fall of capital relative to workers. And this means that productivity falls because there's less capital intensity and there's lower wages.

[B]

Capital intensity is the economist's attempt to measure the tools and equipment available to each worker. If I had one fewer microphone, for example, my ability to present more or less might be diminished. Thanks to John Van Reenen, director of the LSE center for Economic Performance. You're listening to More or Less, and I'm Tim. Two microphones Harford. Last week we discussed a claim that surfaced over 13 years ago that the number of Falklands war veterans that have killed themselves since the conflict exceeds the number of servicemen killed in the campaign. This was in anticipation of a report published on Tuesday this week that would tell us definitively if the claim were true or not. And we invited Simon Wesley, professor of Military Health at King's College London, onto the programme to talk about it. The main point we gleaned was that the absolute figures are not the most telling. It's the rate of deaths which is more likely to tell you whether there's a particular problem of suicides among war veterans. Now, when the day of the big report came, 5 live breakfast did an early morning interview with Simon Wesley at his home in anticipation of the report's publication later that morning.

[G]

This is BBC Radio 5 live.

[I]

The true figures on this will be released for the first time by the Ministry of Defence. Professor Simon Wesley is from the Institute of Psychiatry at King's College in London. Good morning, Professor.

[C]

Good morning.

[B]

Now, by half past nine when the report came out, Simon Wesley was in a car booked by the BBC, being whisked into the central London studios to talk to other BBC outlets about these numbers. And then his phone rang. The report was saying that 95 veterans had killed themselves over the 30 years since the conflict ended, far fewer than the 237 who died in the war. And in fact, that gives a suicide rate lower than that of the general population. That's positive news, right? Worth talking about? Apparently not. The BBC called Simon in his cab to stand him down for the interviews they'd booked. The number was just too small to be interesting. Well, Charlotte MacDonald and I called Simon at home to find out what had happened.

[F]

I think the word was uninteresting was used. I mean, I think there's a serious point. One is, you know, this had been running for over 10 years and how no one had ever pointed out that actually it's a silly figure. Anyway, I thought that was a story. But there's a wider thing which is actually more of a concern that what is a good piece of research but doesn't reveal something terribly wrong with the mental health of those who served in our forces isn't a story.

[B]

The interviews were pre booked, you were in the car, you were ready to go, the numbers came out and the numbers weren't what the news desks had been expecting and everything was cancelled. So you can mislead with statistics simply by not bothering to report that they exist.

[F]

You know, one should pay tribute to the people in the defence statistics who did put a lot of investment into this and I think they'll probably be a little disappointed to find out that other than me and you, no one else seems to be taking much interest.

[G]

It was news gathering that contacted you, wasn't it?

[F]

Sorry, can I just. It was you. Scan. Just one second. This might be them again. Just one second. Hello? It is, yes, who's this? Angus. Oh, it's Angus from the BBC. Just one second, Angus. You just stay on the line. Just one second. Well, one of your colleagues, Angus Crawford, has just called me, so.

[B]

Excellent. Has he called to cancel or has he called to arrange you?

[F]

Ok, I'll talk to Angus. Alright, bye bye, Angus.

[B]

It turns out that Angus, a BBC news journalist, had phoned Simon, thankfully not to cancel another interview in, but to have what we in the business call a background chat. We often discuss publication bias here on More or Less. It's good to be reminded it doesn't just apply in academia, it applies to the media as well. And that's all we have time for this week. But you can download more editions of the programme via our website, BBC.co.ukmoreorless the website is also the place to read more or to send us your comments and your questions. Until next week, goodbye.

[A]

More or Less was presented by Tim Harford, the Financial Times undercover economist. The producer was Ruth Alexander, the editor, Richard Varden. You can Download many more BBC Radio 4 programmes for free. Find these at BBC.co.uk radio4.