Comment / Turning data into information

When asked about the most valuable resource in delivering health and care services, most people would rightly say ‘the staff’. From frontline to support functions, NHS workers are typically united by their focus on helping others and their pride in being involved with the country’s most prized institution. But second on the list must be data – or at least it should be. The NHS produces data in huge quantities on everything from waiting times and clinical performance to procurement, human resources and finance data.

The problem is that, too often, the quality of data in the NHS has been too poor to use in any serious decision making. Or even where the data quality is fine, a reputation for poor quality data means that information is too easily dismissed. Decisions remain untaken and changes unimplemented. 

There is a good argument for the service being data rich, but information poor.

Costing data has had a particularly bad press. While validations built into costing workbooks have addressed some of the more spurious cost data in recent years, you don’t have to go back too far to hear stories of knee replacements apparently undertaken for less than the cost of a knee prosthesis. Often in costing, it has been core data at fault, not the number crunching process itself. Activity not counted, miscounted or wrongly classified will lead to meaningless cost data. Recent reference cost audits have highlighted there is still considerable room for improvement in the coding and classifying of data – an issue explored by an HFMA roundtable last month.

But the data problems go beyond costing data. There are wide-ranging examples of organisations recording or submitting data in returns in slightly different formats or to differing degrees of completeness. And while electronic staff records may provide an accurate picture of staffing levels at the organisation level, they won’t always reflect the exact services or departments those staff work in.

The current transformation agenda needs to be built on evidence and the NHS – on the face of it – has data in abundance to lay the foundations. There is a lot of data that is fine and could and should be used more. But there are also examples of implausible data that undermines everybody’s confidence in the apparent messages or disengages the very staff an organisation is trying to engage.

Despite lots of rhetoric and promises about data and moving towards a greater evidence-based model, data quality has tended to improve in small increments. Current initiatives in the service – RightCare and the Carter-inspired Model Hospital, for example – are a major attempt to ramp up the pace. Both initiatives feed back existing data – to commissioners in the case of RightCare and to providers for the Model Hospital – so they can compare performance and spending levels across wide-ranging activities. 

There are already examples of the data being put to good use to inform change, often providing hard data evidence of existing suspicions over improvement opportunities. But the initiatives also put the service on track to make a reality of its promises around evidence-based decision-making.

By putting the data out there – and showing relative performance – and starting to use it, the hope is that the data quality will improve rapidly and ‘I don’t believe the data’ excuses will become a thing of the past. With greater confidence in data, energies can be redirected into making decisions on the back of it. This won’t be achieved overnight but the prize of an information-rich NHS is surely well worth pursuing.

For more about the Model Hospital, see Leading by example. More on RightCare in Unlocking variation