By Dr Rob Findlay, Director of Strategic Solutions, Insource Ltd

“It has become normal to work around poor-quality, disorganised data”, said Professor Sir Ian Diamond, the UK’s National Statistician and Alex Chisholm, Chief Operating Officer for the Civil Service in The Government Data Quality Framework. “Where quality problems have been identified, the symptoms are often treated instead of the cause”, they added.

It’s a chilling statement and it’s hard to disagree. Ask almost any NHS manager what their data quality is like, and they will probably pull a face. And when the data must be assembled from several systems that don’t talk nicely to each other (as referral-to-treatment (RTT) data is, to give just one example) then the problems are magnified, and the effort required to give a true picture is magnified too.

#datasaveslives

But accurate, joined-up data is not a ‘nice to have’, it is critical. The Care Act 2014 made it a criminal offence for Trusts to supply RTT and other data that is false or misleading, [DoH The False or Misleading Information Offence] and individual managers can face an unlimited fine and up to two years imprisonment. That may sound harsh, but is it? Bad data really can cost lives. Indeed the New HSIB report shows how patients can be ‘lost’ in follow-up appointment gaps.

The amount of data that managers are responsible for is only going to increase, and for good reason. The DHSC Data Strategy [Data saves lives: reshaping health and social care with data] outlines how data has saved lives during the pandemic, and launched with the declaration that – ‘our mission is to unleash the unlimited potential of data in health and care, while maintaining the highest standards of privacy, ethics, and accountability.’

Fixing data problems is essential

As well as laying out the penalties for getting it wrong, the Care Act also explained what was expected of managers, saying “It is a defence for a care provider to prove that it took all reasonable steps and exercised all due diligence to prevent the provision of false or misleading information” [Care Act 2014] so finding and fixing data problems is essential.

And here the National Statistician and Civil Service COO have some helpful advice around root cause analysis. [Data Quality Root Cause Analysis]. They highlight the complexity of today’s data systems, and the difficulty in tracking data back to source to find out where the errors are creeping in, which is essential if the causes of data error are ever going to be fixed.

They also highlight the many ways that data can degrade as it passes through multiple stages of processing. This is especially true in the NHS, as busy analysts build on existing data whenever they create new reports and new definitions, and these processes accumulate over the years until there are hundreds of overlapping steps which nobody really understands because the original authors are long gone. Are the RTT figures that come out of the other end accurate? How could you even know?

Data errors surfaced with PP+

So, it is unsurprising that, when Trusts implement Patient Pathway Plus (PP+) and create strong and consistently linked RTT data for the first time, all kinds of problems with their previous data come to light. Sometimes these are easy to explain, like open pathways left over from a previous PAS. Sometimes they are a nettle to be grasped, like the 5,000 open pathways found in one trust, which managers were previously unaware of.

But the clarity is always welcome. And because every entry can be traced quickly and easily with PP+ all the way back to its source, everybody can see where the data comes from and how any problems are creeping in.

Error-prone spreadsheets are not the answer

Technically, achieving this clarity is a surprisingly demanding task. Some trusts attempt it in spreadsheets, but that doesn’t work because they are notoriously prone to error. Most attempt it with database queries, but we have already seen how the hand-coded queries build up over time until nobody understands what they are all doing.

Which is why Insource have created highly structured platforms, using NHS data definitions, configurable to local requirements, which automatically unify data from multiple sources across the hospital or system. So, you get the clean and consistent data you need every day, and we make sure everything is done the way you want it.

Clean data saves lives

Clean data saves lives and keeps you the right side of the law. And on a daily basis, it just makes life easier.

As one of our clients said: “We now have evidence and auditability of how our data is produced. If anything fails, we can identify the root of the problem in five minutes rather than the hour or more that was previously the case. This complete transparency provides us with confidence in our data through a consistent output and one version of the truth.”