Creating heat-maps using Many Eyes

I came across an article on the Guardian’s datablog called ‘Antidepressants in England and Wales: a map of GP prescriptions’.

The article includes a map of England and Wales – the Primary Care Trusts (PCTs) which prescribe the most anti-depressant drugs are represented by darker colours, like Blackpool PCT. This type of visualisation is called a heat-map (or an intensity map on Google Fusion Tables).

The article also includes a small heat-map of London but it only uses two colours.

Below, it says, ‘depression rates are known to be higher in deprived areas, where people struggle with unemployment and poverty.’

The subject immediately interested me so I set myself three tasks:

1/to create a more detailed heat-map showing how much different London PCT spend anti-depressant prescriptions

2/to make an unemployment map of London boroughs

3/to compare the two: is the amount spent by London PCTs on anti-depressant prescriptions linked to unemployment levels?

Step 1: the PCT heat-map

The Guardian datastore does provide a dataset but I was keen to find the data myself by exploring

I found the data I wanted pretty easily by typing ‘mental health London’ into the search bar. It appeared in a ZOHO spreadsheet, which listed the number and cost of prescriptions in different London PCTs. I chose to use the cost column in my visualisation so copied and pasted it to Many Eyes. Here’s the result:

Click here to see the interactive map.

Many Eyes is created by the American technology giant IBM. This could explain why it’s not as clued-up on UK maps as it could be. It didn’t recognise the names and boundaries of some of the PCTs so I’m unsure about the accuracy of the final result.

Step 2: The Unemployment heat-map

The Guardian offers a dataset of unemployment benefit claimants in all constituencies. I considered using this data to create my second heat-map but Many Eyes didn’t recognise the constituency boundaries at all. Plus, in order to compare the two maps effectively I thought it would be better to find unemployment statistics following borough borders. So I went on another data hunt. The PCT figures were from 2009-2010 and eventually I found another Guardian dataset from 2009 that showed unemployment rates in London boroughs.

Again, using Many Eyes, I came up with this result:

Click here to view the interactive map.

Step 3: Comparing the two heat-maps

The results were interesting. In some boroughs, the unemployment rate seemed to mirror the amount spent by PCTs on anti-depressant prescriptions.

However, Bromley PCT spent the most in anti-depressant prescriptions in 2009-2010, yet had one of the lowest unemployment rates on the map. In Hackney, the unemployment rate was high at 20.6% but the amount spent on anti-depressant prescriptions was relatively low at £586,726.

The initial Guardian article provided a possible explanation for these discrepancies:

‘While the startling variations in prescription rates reflect a complex mixture of circumstances, two major underlying issues emerged during inquiries by the Guardian – the availability of talking therapies and the readiness of GPs to offer them to their patients instead of pills.’

In population sampling there are so many trends evolving that it is difficult to draw objective conclusions from borough-wide data. The best tests ring-fence a specific population set, for instance unemployed people in a specific age group against antidepressants in the same age group.

All in all, I think heat-maps are amazing visualisations. Once you have sorted out area codes, they produce immediate and clear results by replacing numbers with varying shades of colour.

Having said this, they remain fairly inaccurate because they cannot  illustrate the represented area’s characteristics and, with Many Eyes, the precision of some of the boundaries were compromised.

Claire Gilmore (@ClaireEGilmore)

This entry was posted in Data Journalism Experiment, Uncategorized and tagged , , , , , , , , , , , . Bookmark the permalink.

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 )

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s