“Data is Like Garbage”

So this week I start to grapple with some of NHS England’s patient incident data. As I was doing my initial checks of the data one thing instantly grabbed me….. the incompleteness of it. Of over 500,000 incident records, a mere 87 records would be classed as ‘complete’. Now this is not the fault of the national patient safety team in NHS England, nor can it be seen as the fault of the very hard working staff on the wards and in the departments that regularly report the c2m records every year.

No, surely the issue with the data we collect is the way in which we collect it. If we leave fields open as ‘voluntary’ or ‘optional’ then we must expect for these fields to be not completed. We cannot surely expect staff who are under pressure due to high demand and reduced staffing numbers to complete an incident form that does not demand of the user the data it needs.

As far as the NHS England data, this then creates a bit of a problem. In all my research it is latent impacts or contributory factors that show where the true reason for incident occurring lies. Understanding the root cause of any incident is key in order to prevent it reoccurring. However, datasets have captured only 4% of contributory factors. There is however, one saving grace – volume. Even at 4% response rate, this still leave over 23,500 records to analyse.

If nothing else, the initial review of the data sets has reminded me of the importance of complete data, and ensuring that any systems that I set up in the future capture the data we require, and not leaving it down to the reporter to decide whether certain information will be needed in the future or not.

As Mark Twain was alleged to have said, “Data is like garbage. You’d better know what you are going to do with it before you collect it”.