News/Death Ratios

Back in May 2009 Hans Rosling introduced the world to a new indicator for media hype: the News/Death ratio. His example compares the frenzied media reporting of swine flu to deaths from tuberculosis over the same thirteen day period; the point being that the former was attracting extensive news coverage despite a trivial impact on global mortality, while the latter was responsible for numerous under-reported deaths. So why not extend his idea and look at some other stuff?

It’s not clear why Rosling picked tuberculosis for the comparator, as the same point could have been made with HIV, malaria or indeed many other significant diseases of the developing world. It’s important to say that there’s nothing very quantitative or systematic about this methodology- the number of news reports are estimated from the google news archives search facility, and are dependent on the search terms used and whatever mysterious algorithms google uses to archive and filter its approximate (‘about’) search results. And the crude mortality count is just that; while it’s important, it can’t possibly be considered a complete or sole indicator of significance. With that disclaimer out of the way, let’s look at some more diseases/events/stories:

Health data from WHO
*search terms in google news archive as indicated [except childhood diarrhoea = children diarrhoea + children diarrhea]
**pale=neglect dark=hype

News collection periods were selected either according to the ease of finding relevant data on the WHO website— usually specific-cause mortality for a specific year (with the google news archive search carried out for the same period); or for discrete events, the neat histogram feature on the google news archive allows you to pick an interval which captures most news reports for that event. Often a year suffices, but the still-frequent reports referencing the September 11 attacks means that in this case, the news collection period was extended right to the present day.

I’ll say a quick word or two about some of these selections, and what’s missing. I’ll warn you again- there’s nothing very systematic about any of this.

Cardiovascular disease, cancer and road traffic accidents are major causes of global mortality missing from the table. You could make a good case for including any of them, but when it came down to it, doing the searches was just too complicated. I considered including the 2003 Iraq war… but which numbers to use? Do we believe the controversial Lancet statistics?

As for what was included: diarrhoea in the under 5s, HIV and malaria are significant causes of death in less developed countries, but with the (internet) media being concentrated in the developed world, it seemed likely these were being neglected in the news. So it proved. Dengue fever and yellow fever are both tropical viral diseases carried by mosquito vectors. Dengue has been increasing in the last 25 years, lacks a vaccine, and has been designated as a ‘neglected tropical disease’ by the WHO. Yellow fever has a very effective vaccine with improving global coverage in the last twenty years, but in terms of ‘media relevance’ it appears to be marginally more neglected.

The natural disasters are another arbitrary selection. I remember hearing about the March 2005 Sumatran earthquake by chance several months after it happened, and guessing that the ongoing tsunami clean-up stories had pushed it out of the news; but the ratio here suggests that the reports were there, and at an appropriate level— indeed, slightly more hyped than the tsunami itself according to the N/D measure.

I doubt the comparisons are particularly meaningful at this level, but the hype/relevance/neglect orders-of-magnitude feel about right for most of these examples. More effective media hype-ometers might be possible with different indicators— incorporating different measures of significance— and more rigorous search methodologies.

While I’m not sure whether this a useful exercise, at least it’s an excuse to link to gapminder. It almost makes health/social statistics, well, sexy. Well worth a look (and play).

OK, the table should be right now and match up with Hans Rosling’s News/Death ratios (though his NDR scale looks like it goes higher— perhaps because he used the google news search, rather than the google news archive). Sorry if any confusion from earlier overhasty correction.


3 Responses to “News/Death Ratios”

  1. thinkingdan Says:

    Nice post. Its worth noting that death itself is a bad choice of newsworthiness – future threat of death is more important from a practical point of view. For example, coverage of a single bad event that is not likely to repeat should be lower than for a similar scale disaster that requires action to prevent a recurrence. This would justify an increased level of coverage for infectious diseases (and maybe terrorism… though maybe not), when compared to natural disasters. Infectious diseases, in particular, have the potential to kill far more “readers” than anything else on the list above (even if they usually don’t).

    Of course your point stands regardless.

  2. Katherine Says:

    I’d love to see this done on a larger scale, and with some way of factoring in potential deaths that were prevented by action following media coverage (very few I’d think, as with the case of infectious disease most of the prevention methods should be being worked on before the media/public are made aware of the threat, rather than the media making the relevant authorities aware of the issue).

  3. Media Chaos and measles outbreaks | Quiet Peregrine Says:

    […] I’ve highlighted Hans Rosling’s death/news ratios as a way of highlighting how poorly the impact of a health problem or disaster, in terms of actual […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: