Home   About Kurtosis   Courses   Course Calendar   Booking Information   Ideas   Technique   Clients   Contact

The Opposite of Data

Do we have to choose between patient stories and data? Or can we have both?


Yesterday Alex Neil, Scotland’s Health Secretary, announced additional funding to support patient stories. NHS boards across Scotland will be given a share of additional funding of up to £160,000 in 2014/15 to deliver further online patient feedback through the Patient Opinion website.

Now, as it happened, I was at a conference in Edinburgh last month listening to—amongst others—Gina Alexander, the Director of Patient Opinion Scotland, talking about how important it is to learn from patients’ stories. The importance of patient stories is a big current theme in the NHS.

Gina Alexander had some slides with quotations on them. Here’s one of them. You may have seen it before:

“Not everything that can be counted counts; and not everything that counts can be counted.”

It’s a beautiful quote. I seem to remember that it was attributed to Albert Einstein. But, for all its beauty, it triggered a vague sense of unease. When you’re a data analyst (as I am), stories can be threatening. Stories (or, to give them their pejorative name, anecdotes) are subjective, personal, emotive. Surely we shouldn’t be using stories as the basis for decision-making. No, decisions need to be based on data. Data is objective, impersonal, dispassionate. Stories can often seem as if they are the opposite of data.

I once worked with someone who used to say: “Beware when someone in a meeting starts a sentence with the words “For example…”. What they actually mean is: “For exception…”. It’s the fact that it was an exception that made it memorable, that made it worth relating. We can’t possibly use anecdotes about exceptional events as the basis for making decisions without first checking how representative those events are.

Or can we?

I’ve recently been thinking about this distinction between data and stories. And I’ve reached the conclusion that it’s not as straightforward as comparing fact and fiction. Stories aren’t the opposite of data. They’re not even different from data. They are actually just another way of presenting data.

Let me give you three examples of what I mean.

The first thing that got me thinking along these lines was a video about an improvement project at Sheffield Teaching Hospitals Trust. I came across it on this website. There are three videos on the page but the one you need to watch is the second one (called Making change across organisations). It’s only 2’36” long so it won’t take all day, and it may just change the way you work.

In the video, people are self-consciously talking about how you need to get stories and data to work together. It’s not stories or data; it’s stories and data. And when they are telling a patient story (it starts at 00'43"), the details they invoke that make it a story (name, age, sex, previous diagnoses, marital status) are actually just data items. A patient story is often very little more than data items dressed up as something more palatable, something easier for people to relate to.

Paradoxically, when writers want to make a scene or an episode more lifelike and memorable, what they typically do is add detail. They tell us about how old someone is, how grey their hair is, how late they were for the appointment, what type of coffee they ordered. Writers might call this kind of detail “colour”; I call it “data”.

The second reason why I’ve come to the view that stories are just another way of presenting data comes straight from the quotation mentioned a couple of paragraphs back. Gina Alexander used that quote“Not everything that counts can be counted”to remind us that we need stories when there are data black holes. If something went wrong for a patient that isn’t routinely measured or reported, then the only way of drawing people’s attention to the problem is by telling the story of what went wrong. Seen in this light, stories are a way of prodding us into either reporting data that has previously gone unreported or into developing and publishing new measures and indicators that do reflect those concerns.

Here’s the third reason why I’ve decided that patient stories are just another way of presenting data. The way that most of us number-crunchers think about data, what we are doing most of the time is using numbers to describe a big reality. We can use just one number to describe what happened to a lot of patients. The hospital might’ve discharged 60,000 inpatients last year but we can summarise their experience into just one number by saying that their average length of stay was 6.2 days. Maybe 20,000 of those inpatients were elective and we can summarise all of their individual waiting experiences by saying that 92% of them were admitted within eleven weeks of being listed.

We have grown so accustomed to thinking this way about data that we hardly ever pause to consider whether we could turn this on its head. What if we were to actually use data to find out a lot of numbers about just one patient? What if we were to use data to describe an individual patient journey from start to finish, across all the organisational boundaries that had to be crossed. What if we were to use data to describe what happened yesterday – in all its gory detail? First this happened, then that happened, and then because of that, this third thing happened.

Data about a patient. Data about a day. Instead of big data: small data. And when you do that, you realise just how close to a patient story you’ve got.

Gina Alexander showed another quotation slide at that conference last month.

"Statistics are human beings with the tears wiped off."

Well, if we can show people that data and patient stories are just two sides of the same coin, maybe we can report statistics with the tears left on.

[31 December 2013]


Make a comment on this blogpost:


Job Title:


Email address:


© Kurtosis 2013. All Rights Reserved.    

Comments on this article

31 December 2013:

It’s not just the Scots that are into patient stories, the Welsh are too. See http://www.1000livesplus.wales.nhs.uk/pp-driven-care

And I agree with your basic tenet that stories are just another way of collecting data. I like to think of the 2 sides of the coin you describe as ‘words’ and ‘numbers’

Happy Hogmanay!

Mike Davidge

Head of Improvement Methodology, 1000 Lives Improvement Service

31 December 2013:

Great thoughts, Neil.
Narratives do seem like a useful framework to hang appropriate data items in a meaningful way (The subject of History? Law?). I think one of the really important things it can do is highlight what is not being measured that is of importance or value to the patient. Getting this information in valid and reliable ways, then using it to drive improvement is the big conundrum

John Soong

RCP QI Research Fellow, RCPL/Imperial College London

31 December 2013:

Thanks for this fascinating and thought-provoking post - much here to ponder on.
I am immediately struck by your counterpoint to "big data" as small data = a story. I love that, and will recycle it shamelessly!

James Munro

Director of Research, Patient Opinion