The paradox of policy information: No shortage of health data but much of it languishes unused

The health sector is awash with data but so much of it goes unnoticed and uninterpreted

Health policy junkies will have been intrigued by news of Te Whatu Ora’s publication of erroneous emergency department data on its website.

In case you missed it, some incorrect data on ED presentations and admissions for the months of November and December 2022 were posted on Te Whatu Ora’s website on 31 January. Basically, it was a data transcription error – for about half the districts, the numbers of presentations reported were actually the figures for a different district. The rows from one spreadsheet were pasted into the wrong destinations of another.

This incorrect data then affected the calculation of ED wait times – the proportion of patients who had been seen, treated and discharged within six hours.

This was clearly a case of human error, but also indicated that systems for checking data before publication had also failed.

Apart from giving opposition MPs a free hit in Parliamentary question time, other interesting consequences arose.

In response to questioning by RNZ on 9 March, health minister Ayesha Verrall was at pains to point out that no decisions about resource allocation were made on the basis of this erroneous information.

But the very idea that resource decisions could or would be made based on monthly data about ED attendances is patently far-fetched to anyone working within the health system.

I completely understand that journalists are trained to ask questions that citizens might be asking. But this line of questioning reinforces a misleading impression about the relationship between data and decision-making in government.

This incident is reflective of the fact that the health sector is awash with data but so much of it goes unnoticed and uninterpreted once collected.

It is telling that nobody noticed this supposedly egregious error for close to six weeks after its appearance on the Te Whatu Ora website.

Ask anyone who has had a long-standing focus on addressing inequities between Māori and non-Māori in health, and they will tell you that, while data on inequities are available, such data rarely make a difference to resource allocation or service design.

This weakness is not confined to the New Zealand health sector, but plagues governments more generally.

In their recent book, Reconsidering Policy, a group of Australian-based policy scholars noted more data are collected and processed worldwide than ever before. But, write University of Tasmania professor Kate Crowley and co-authors, it is disappointing that policy-making capacity has developed insufficiently to take full advantage of these resources. “Part of the problem is that more data does not necessarily mean more information. Information communicates: data (of itself) does not,” they say.

In the case of ED presentations, the data clearly show a situation of long-term deterioration, but cannot reveal the reasons for it. And the big questions here are much harder to answer. To what extent is this deterioration a consequence of workforce shortages? Is it simply a matter of ED resourcing? Are we seeing much longer-term demographic shifts – such as the changes in the ratio of working age population to retirees – finally biting? How much of it has to do with critical gaps in the provision of residential aged care? There are so many examples like this in which the necessary data that could be used to support informed policy processes are simply not available.

Then there are examples in which data exist, but their veracity and meaning are disputed. A recent case in point is the exchange reported in this newspaper, between General Practice Owners Association chair Tim Malloy and former health minister Andrew Little on multiple policy issues. These included whether primary care nurses are paid significantly less than their hospital colleagues.

Organisations such as GenPro and the New Zealand Nurses Organisation clearly have a role and an interest in producing policy-relevant data – the point of which is to support their broader stances. But the Government has been sceptical, reflecting its own structural and political imperatives to be cautious in spending taxpayers’ money. Mr Little said the evidence was missing. But, if that is the case, the key questions are why is it missing, and why isn’t this type of information prioritised as important by either the Ministry of Health or Te Whatu Ora?

To be sure, better information alone will not resolve issues like this, as there are still the broader issues of who should stump up the cash to address any disparity, and who has the right to see primary care data. But the issue of nurse pay parity between primary and hospital sectors, like so many others in the primary healthcare space, is hardly new or unexpected, so there really should be more established ways and processes for managing it.

And perhaps therein lies the biggest deficiency in healthcare policy in Aotearoa. Information, and the data that feeds it, require more predictable pathways to channel their flow. Without these channels, useful information that is available is simply lost to the system, and there is little imperative to search for useful information.

For so many of the big issues in primary healthcare – population-based funding formulae, workforce planning and development, infrastructure needs – these predictable pathways and processes either do not exist or are very immature. Perhaps the creation of Te Whatu Ora and Te Aka Whai Ora can be a gamechanger and fill the information vacuum in these areas. But that will only happen if there is a sustained focus, supported by ministers and senior leadership, on information capacity-building.

Tim Tenbensel is professor, health systems, in the Faculty of Medical and Health Sciences at the University of Auckland