We recently met with Professor John McNeil, a leading expert in clinical registries, to talk about the direction he thinks registries need to take to be successful. John is based at Monash University and was, until recently, the Head of School of Public Health and Preventive Medicine.
A passion for clinical registries
Outlining the long history of registries in Australia, John spoke highly of the Australia and New Zealand Dialysis and Transplant Registry, the first, and still one of the most prominent, registries in this country. The registry has been in operation since 1977 and almost all renal units in Australia and New Zealand contribute data to it.
There was a big variation in outcomes from dialysis and transplants when the registry was first set up, and a lot of donated kidneys were being lost prematurely. ‘It’s a tragedy, really, to transplant a kidney into someone and then lose it,’ said John.
With the data from the registry, units that weren’t performing well were given recommendations on how to improve. The standards of renal transplantation in Australia rose to be among the highest in the world. Explaining the impact, John said, ‘It really burned in my brain the importance of measuring and reporting outcomes, and feeding back, and sometimes telling people that they weren't getting as good results as their neighbours. And they took it seriously. It was very effective.’
The minimum dataset
Monash University’s School of Public Health and Preventive Medicine currently maintains 28 clinical registries, including the Victorian Cardiac Outcomes Registry and the Victorian State Trauma Registry, which are funded by the Victorian Government and have reporting arrangements in place with VAHI.
In addition to running the registries, the team at Monash have devoted considerable effort towards ‘registry science’, that is the essential features that make a registry effective.
‘After a lot of trial and error, we have developed our own approach at Monash, based around a minimum dataset and outcome measurement at a time when clinical recovery would be expected,’ John said.
John explained that when registries require clinicians to collect huge amounts of data, it makes them unsustainable. They have found that with most conditions, they can distil the data that clinicians need to collect down to the 10–20 things that you really need to know. This forms the minimum dataset.
In addition to what happened to a person during their stay in hospital, it’s also important for registries to record the outcomes of the treatment. This means contacting a patient once they are out of hospital. But again, John explained that they don’t need massive amounts of data here.
‘At the end of the day all we're trying to do is classify people as having a really great outcome, a really bad outcome, or a couple of levels in between,’ he said.
A vision for the future
We asked John about his vision for clinical registries and he had three clear ideas.
Firstly, he sees a need for more efficiency in the data collection, which can be driven by emerging technology. ‘With the evolution of the hospital electronic medical record, we should have minimum clinical datasets incorporated and coded for every common condition. This would greatly simplify the work of registries and keep costs to a minimum,’ he said.
Secondly, we need to develop a much more automated way of measuring outcomes of care. As he explained, ‘In this day and age, where everyone has either a computer or mobile phone, that should all be done by automated IT systems where you get an SMS with three or four questions to answer.’
Finally, John believes that a big factor holding registries back is the lack of a good funding model. Registries are variously funded by governments (state and federal) and the private sector, such as health insurers. The difficulty is that different parties want different things from the registry data but we have not yet managed to develop an effective cost-sharing model.
The role of registries in healthcare
On the role of registries in healthcare, John spoke about how they engender competition, which drives units to improve to be better than their peers. He explained how the incentive to be among the best performers has been shown around the world to improve performance.
‘That’s what creates continuous quality improvement,’ he said.
John sees public reporting of registry data at individual clinician level as inevitable at some time in the future but says it’s a complicated area. He stressed the importance of ensuring they are reporting the right measures, which clinicians trust and think have been risk-adjusted correctly.
‘The registries go through a period of evolution until you get to that sweet spot where the clinicians say, “yes, okay, we believe this, the risk adjustment is not bad, we're measuring the right outcomes, we think the data's right, we think that the registry's been audited carefully enough."
‘At that stage, I think most clinicians would say we're prepared to consider public reporting, but they will be reluctant to see public reporting until they get to that stage,’ he said.
Speaking about the role of health service CEOs and boards with clinical quality registries, John suggests that they need to be aware of the registries that their hospitals are contributing to and should take notice of what the registry reports are showing about their performance. And if they’re outliers, they should know how to deal with it.
On VAHI’s role, John was very clear.
‘VAHI should see registries as the most fundamental measure of quality and safety, the most trusted by clinicians, and the most capable of bringing about change,’ he said.