Development of clinical-quality registries in Australia: the way forward

The final report of the National Health and Hospitals Reform Commission, released in 2009, recommended that, in promoting better quality care, “we have systems in place to provide comparative clinical performance data back to health services and hospitals, clinical units and clinicians.”1

Also in 2009, the Australian Institute of Health and Welfare (AIHW) released a report proposing 55 quality indicators for national reporting on safety and quality (Box).2 However, of the 31 hospital and specialised health service-specific indicators, only two indicators (22 and 25) are currently being reported nationally.3 Most cannot be measured in an epidemiologically sound manner (ie, reproducibly and accurately) within the available routinely collected data sources without extensive development of both the indicator and the systems to collect it.2 Most importantly, the AIHW indicators omit key outcome measures for hospital activities associated with high cost, such as cardiac surgery, trauma care and intensive care, where poor quality care can impose a significant ongoing burden on the community.

Given these deficiencies, attention has turned towards developing clinical registries from which robust quality indicators can be derived.4 In this article, we outline the rationale for developing registries, and a framework to ensure they are strategically developed to provide the best return on investment.

Clinical-quality registries collect a defined minimum dataset from people who undergo a particular procedure, are diagnosed with a disease, or use a health care resource.5 Quality indicators derived from registries and adjusted for differences in casemix are used to benchmark and improve performance across institutions. This information is, in general, credible to clinicians and effective at driving quality improvement.

More than 70 clinical registries have been developed in Sweden. The Swedish stroke registry has identified variations in care,6 characteristics of clinical units providing the best service7 and attributes of patients most likely to benefit from such care.8 Longer-term information has included quality of life and factors influencing capacity to return to work;9 and — through linkage with a prescribing database — adherence to medication such as statins 2 years after commencing treatment.10 Quality indicators routinely collected within the registry are reported back to hospitals; they allow comparisons at regional and hospital level and have driven development, evaluation and improvement of stroke services.11

In Australia, high-quality registries that collect outcome data from eligible populations are fewer. Only five national registries, encompassing joint replacement, renal dialysis and various forms of organ transplantation (kidney, bone marrow, liver and cardiothoracic organs), have national coverage. One of these, the Australia and New Zealand Dialysis and Transplant Registry (ANZDATA), collects data on patients receiving renal transplants or renal dialysis. It provides risk-adjusted outcomes to contributing organisations that can be compared within Australia and internationally. They can also be used to determine compliance with best practice guidelines12 and to identify problems with access and equity in the delivery of renal services.13

While other high-quality Australian registries exist, many lack high levels of coverage.14 For example, the Australian and New Zealand Intensive Care Society Adult Patient Database, which benchmarks performance, enables resource planning, monitors clinical practice and is a valuable research resource in identifying risk factors for poor outcomes,15,16 does not collect data from 30% of eligible intensive care units in Australia.

Increasing the number of nationwide clinical registries will provide an enlarged inventory of valid, reliable and clinically credible measures of quality that can be compared among different institutions within Australia and internationally. As with initiatives in other developed countries, this can be achieved by expanding several existing registries to become national in scope and by developing new registries in key areas where they currently do not exist. This calls for a framework for prioritising registry development that ensures they are developed in areas of greatest need.

Registries prioritised for development or expansion must be underpinned by robust governance structures and mechanisms to ensure transparency to stakeholders in terms of managerial decision making and oversight, data collection and analysis methodologies, and systems for reporting and disseminating results.28 This, in turn, requires consideration of a framework that ensures data are used responsibly to drive quality improvement and to analyse and address variation in clinical approach.5 Those reporting adverse outcomes that may be uncommon or rare need to use techniques of data display that will best motivate quality improvement, such as exponentially weighted moving average charts.29 More research into refining these methods is needed.

In future, it is likely that a broad range of medical conditions will be monitored using registries that collect data inexpensively from electronic medical records and administrative datasets. To reduce the burden and cost of registry data collection, registry science will need to evolve in ensuring that required data items are reliably captured within, and easily extracted from, these routinely collected electronic datasets.

Like most developed countries, Australia has few registries capable of benchmarking outcomes nationally. Given the increasing attention to quality in the health agenda both here and overseas, policymakers will be in need of a clinical registry prioritisation strategy which ensures that registry development targets three areas — conditions or procedures associated with large variations in processes or outcomes of care, which have a significant impact on overall health care costs and patient morbidity; areas where transition of care across health services influences optimal outcomes; and medium-term to long-term safety of new clinical interventions. Clinical registries can play an increasingly important role as a stimulus for quality improvement by providing high-quality data and analyses that are respected by clinicians.