SHOCKER – CANCER UP 30% (Or just bad reporting?)

April 2, 2013 — 4 Comments

If you’re a man and watch or read the news today you may, understandably, be worried.

Bowel cancer rates in men soar by a quarter“…

…proclaims the Daily Mail.

Bowel cancer rates for men rise by 29%” writes The Independent, and The Guardian reports a rise of “nearly 30% in 35 years”. Stick ‘bowel cancer’ into Google News and you’ll see similar reporting across the board.

These headlines are actually relatively accurate. However, as much as headlines need to sell a story, they shouldn’t scaremonger or mislead; which I believe these do. I would think it reasonable for a person reading these headlines to, at a glance, believe their likelihood of getting bowel cancer to be up by nearly a third, or that 30% more people they know will be getting it. But that’s not the case at all.

Let me argue for a moment that today’s headlines could have, in fact, been more like:

“Bowel cancer survival up, as rate of new patients with the disease remains unchanged”

The numbers

I’m not a statistician, but broadly speaking…

The key word in all of this is rate. Cancer rates have risen 29% in 35 years (for bowel cancer in men, according to the stats from Cancer Research UK). The rate refers to the number of people getting the disease per 100,000, which is a world away from simply ‘the number of people getting the disease’. You might reasonably assume the phrase ‘rate of cancer’ refers to the latter if you didn’t know better.

In 1975-77, the rate of bowel cancer for men was 45 in 100,000. In 2008-10, it’s 58. So after 35 years, 13 more men per 100,000 are likely to get bowel cancer. That’s 1.3 per 10,000. That’s 0.13 per thousand. These numbers suddenly seem less scary than something involving THIRTY PERCENT.

35 years later (assuming 50/50 men/women for the following examples), you could say that the number of men who might get bowel cancer in the City of Ely, Cambridgeshire, has gone up by one person. At a full capacity concert, Wembley Stadium would hold maybe seven more men who might get it – but only if almost a quarter of the crowd were 60 years old or above (more on that in a minute…).

Don’t get me wrong – cancer’s on the increase and it’s not good. But it’s as true to say “SOME CANCER RATES UP 30%!” as it is to say “the number of men per thousand getting bowel cancer in the last 35 years has risen by 0.13”.

An aging population

A key part of this story is to be found in the news copy beneath the terrifying headlines.

“The age group with the biggest rise [in bowel cancer] is those in their 60s and 70s, who experience 23,000 new cases a year.”

Ah, ok. So how many new cases are there each year, in ‘real’ numbers? The total number of new cases in the UK for 2010 was 40,695 (this is no longer gender specific).

So over 50% of new bowel cancer cases are now in the 60+ age group. We know there’s an ageing population, and a greater proportion of the UK is over 60 compared to 35 years ago. If bowel cancer is far more common in this age group, we would be expecting the rate to rise as the population ages.

But why let that dent a strong headline?

Create your own headlines!

Today’s bold page-headers refer to changes in bowel cancer rates over 35 years. But with the same data today’s news came from, let’s try out a headline for, say, the change over ten years – again with male bowel cancer rates.

In 1998-2000 the rate was 57 per 100,000. As we know the rate for 2008-10 is 58 (I’m looking at Figure 1.2 here).

So how’s this for a headline:

Bowel cancer rates remain largely unchanged over the past ten years

Doesn’t quite have the same ring to it… There’s another key bit of info in today’s news too – that cancer survival rates have doubled in the last 40 years.

So why don’t we go for:

Bowel cancer survival up, as rate of new patients with the disease remains unchanged

That’s not right either

Cancer is a huge problem. Trends over time show that rates are increasing. Anecdotally most people know this – sadly – from personal experience.

I think headlines like the rather boring one I wrote just above, are just as irresponsible as ones that blindly report 30% increases.

My main argument here is that unless you think the majority of people reading your headline know that ‘cancer rates’ refers to ‘the number of people per 100,000’, you should perhaps think again about how you portray the story.

If you compare today’s news articles with the original press release you’ll see the majority are largely cut-and-paste jobs. When it comes to something like cancer, just regurgitating the stats from a press release isn’t always good enough.

 

An incredible book for any journalist is ‘The Tiger That Isn’t’ by Michael Blastland and Andrew Dilnot. It will make you think twice before reporting any statistic – from cancer rates to unemployment figures and beyond. I can’t recommend it enough.

4 responses to SHOCKER – CANCER UP 30% (Or just bad reporting?)

  1. Well said, and please keep saying it. There’s far too much bad reporting of this type of story and it’s not good for my blood pressure.

    David Spiegelhalter, Professor for the Public Understanding of Risk, talks brilliantly about this. I can’t recommend him highly enough as a speaker, and if you could get him to educate a room full of journalists in how to better interpret this type of press release, you’d be doing a service to mankind.

  2. This is something that drives me mad when watching/reading/listening to the news. It happens all too often and for journalists to read ‘The Tiger That Isn’t’ is a brilliant suggestion!

  3. Excellent, what a web site it is! This web site gives useful information to us, keep it
    up.

  4. My partner and I absolutely love your blog and find the majority of your post’s to be precisely what I’m looking for.
    Does one offer guest writers to write content in your case?

    I wouldn’t mind composing a post or elaborating on most of the subjects you write regarding here.
    Again, awesome blog!

Leave a Reply to Fiona Kate Cancel reply

Text formatting is available via select HTML. <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

*