This blogpost is about this site:

NHS Choices

Ten days ago I wrote to NHS Choices, at their request for feedback.   I was looking forward to reading their outcomes data,  however the more I read it, the more confused I became.  In fact, by the end I had many more questions than answers, and far from providing ‘transparency’  things ended more blurred and confused.

They did say it was a work in progress and they wanted feedback.  So I wrote to them. My email is at the bottom of this post, and you are welcome to read and hopefully comment.

Generally :

  • There is no information on what the data is being compared to.
  • There are no time frames
  • There is no information on trends – getting better or getting worse?

I find their rating system particularly baffling

  • 60% of Bolton’s population are overweight or obese, their rating is ..OK, while Trafford also has a 60% population who are overweight or obese  get a … green tick.
The inconsistency is one thing, the symbols suggesting that it’s ok or even commendable to have 60% of your population of an unhealthy weight is surely damaging?
  • Cancer Screening programmes are marked out of two programmes.
I’m pretty sure there are more screening programmes than that – does anyone know which two? why they were chosen? and most importantly if you do know – how did you find out?

 

Making data accessible, should not mean making it so simple that it fails to convey an accurate message.

Outcomes and Public Health are of huge importance. This data is badged as being open data for transparency – I would really love to know what others think.

Oh and if you ask for feedback, it’s good to respond to those who give it.

Email to Feedback to NHS Choices dated 27th September 2014
Thank you for the opportunity to feedback on the NHS service
performance site
http://www.nhs.uk/service-search/performance/search#performance.
I work with organisations to help them evidence their outcomes, mainly in the area of public health and health inequalities.  I have spent some time looking at your site and attempting to make sense of the data.  I think that overall the data presented gives a confusing and, at times misleading picture. I would like to give you some examples.
I have particularly looked at public health. West Berkshire, Wokingham and Bracknell Forest  breastfeeding prevalence rate at 6 – 8 weeks is 56% and is given an OK rating, whereas Luton, Windsor and Maidenhead, Buckingham, Slough and Reading also have a prevalence of 56% but are given a green tick.  The information says compared to a similar area, but there is no further information. I would like to know if it is against a regional rate, a Core Cities rate  or a national rate then the figures may start to make sense. Although in my experience most areas are benchmarked against a national figure. 

When  outcomes are presented in this way (with a figure and a symbol) it is very confusing if the symbols and the figures are not consistent, even more so when there is no explanation of how the symbols were arrived at. This comment applies to a significant amount of the symbol rated data.

In terms of the three columns on the right areas are scored out of 4, 14 and 2 – I have no idea of what they are: When hovering over the question mark you either get a repeat of what is above,  or a lengthy explanation about the benefits of screening and immunisation.  eg:I know there are more than two types of cancer screening programmes, however, I have no idea which two you are talking about. 

I move on to the outcomes, although understanding why breastfeeding prevalence is a ‘service’ and prevalence of smoking is an ‘outcome’ is unclear.

The main confusion here is along the lines of the issues in the other data set.  Hampshire has 149 preventable deaths per 100,000 that is OK but, Cambridge also has 149 preventable deaths per 100,000 that gets a green tick.  I am unable to decipher your ratings as I don’t know what they are being measured against. 

Generally  there is a lack of information about:

  * Trends – are the areas improving? 
  * What date this data refers to? – there is no year
  * Time frames – What period of time does it cover?
  * Links to more detailed data for those that want it.  

The symbols used give a misleading picture to the public – ie: a green tick for Dorset with a 62% overweight or obese population suggests that’s this is acceptable even good.  70% of Bolton’s population are overweight or obese – the symbol says OK This may give a very misleading picture. 

Making data accessible, should not mean making it so simple that it fails to convey an accurate message. 

I hope that this feedback is constructive. 

Vicki Fitzgerald

 I am at risk of looking particularly stupid, I may have missed a BIG BUTTON, that says PRESS HERE and all this information will become clear, but it is a chance I am willing to take.