Provider Performance Analytics an important Regulatory Tool at ReturnToWorkSA
4 May 2023
● News and mediaThis guest editorial was written for the ANZSOG/National Regulators Community of practice monthly newsletter, highlighting new additions to the Regulation Policy and Practice collection on APO. The RP&P collection brings together a range of practical resources from national, local and state/territory governments, regulatory agencies and external institutions conducting monitoring, inquiries and reviews. You can receive this newsletter by joining the ANZSOG/National Regulators Community of Practice (membership is free) or subscribe to the newsletter directly.
By Samantha Jones, Leader Enforcement, ReturnToWorkSA
In 2021, as Leader of Enforcement in ReturnToWorkSA’s Regulatory group, I initiated a journey toward using data analytics to identify potential dishonesty among our service providers. The ReturnToWork Scheme like other workers compensation schemes relies heavily on clinicians, allied health care professionals, lawyers and non-medical contractors to support its purpose. Naturally related expenditure represents a very high proportion of overall costs to the Scheme.
At that time no structured analysis occurred over organisational data sets to flag potential fraudulent scheme activities among providers. There was a high reliance on controls that focused on self-regulation, deterrence, review of billing at case level and reactive investigation. A strategic priority had been set to improve the Scheme’s economic performance through a reduction in provider misconduct.
We partnered with KPMG to explore an analytical approach to understanding the nature and impacts of provider fraud. Through that work it was identified that research suggested fraud accounted for between 3-10% of health insurance expenditure, with providers estimated to be responsible for 80% of that fraud. What we came to understand is that we didn’t understand the problem or its extent in our Scheme, an issue compounded by inherent difficulties in detecting provider fraud or its precursors due to the ability for providers to shield their conduct through understanding Scheme processes.
Proposed was a forensic analytics approach that first examines behaviour and then explores the providers’ data footprint to show different types of provider fraud as well as trends, red flags and high-risk behaviours. In response we embarked on building a data analytics tool to help deliver early leading indicators of provider fraud. The decision to develop and build the tool in-house meant it was a slow burn, two key resources both with high level analytical capabilities collaborated on its development alongside our usual business, and broader change priorities.
The tool was finalised at the end of 2022. Using existing organisational data it incorporates analysis across anomalies between worker and provider locations, travel allowance fluctuations for comparable services and incidences of multiple different claims where common legal or medical providers are used; and analysis of incompatible treatments including irregular service types or levels relative to approved service plans and comparable providers (with reference to injury types, service codes, fee limits, services duration and frequency) and high volume or service patterns by common providers.
The tool also provides for network analysis identifying irregular concentration of activity and billing in provider relationships; unbundling and upcoding of injury types where service plans or history includes bundled service codes or reduced use compared to other providers; identification of undisclosed conflicts of interest through relationship matching between providers (addresses, bank accounts etc); irregular claim duration; impossibility of service; and numeric indicators such as round numbers, statistical anomalies, sequential and duplicate invoices.
While the initial focus was on detecting provider fraud, the tool was soon realised as being able to support broader regulatory impact. It offers opportunity for identification of early leading indicators of provider deviance and intervention opportunities; a basis for consistency of assessments across our impairment assessors; information to assist us to identify where tighter and more preventative risk controls can be put in place; information about breakdowns in controls; a better understanding of how providers have been able to engage in fraud or misconduct; and of course fraud detection.
Since the tool’s finalisation in December 2022, we have embarked on training staff in its use to support them to identify and test their concerns and to inform them in their activities with providers. This allows our people to add their experiential insights to understanding the data outputs and in identifying and crafting interventions with the added benefit of using data as a core capability. We have been able to identify providers suspected of overbilling and billing for no service, reveal indicators of claims soliciting, explore relationships between legal practitioners and medical providers, and produce data for strategic programs. We have informed and initiated investigations, informed practice and process change, identified demographic factors in claims soliciting and discounted concerns of collusion between providers.
Although in its early days of use this tool is a welcome addition to our Regulatory kit. It would not have been possible without the insights and guidance from our trusted partner in KPMG and the dedication and nous of two very clever analysts with sound technical and subject matter expertise. On reflection spending to co-develop the tool with KPMG would have resulted in a quicker and cheaper output having regard to staff time spent and appreciating anticipated savings from an earlier completion date. Nevertheless, the work has proved invaluable to influencing an understanding of the benefits data analysis has for our regulatory business and contributed to the development of a technical services team for the group. As this function matures so too will our work with the Provider Performance Analytics tool.
Our data is rich and I am excited about what lies ahead for us with this important work in ReturnToWorkSA and about sharing our experiences in the development and operationalisation of the Provider Performance Analytics tool with the National Regulators Community of Practice on 15 May 2023: at the Data, data, everywhere: Challenges, lessons and opportunities for regulators Webinar.
Related news and media
Extending regulatory public value in Australia
Improving regulatory language: Its contribution to regulatory capability
ANZSOG report into governance of the Queensland Building and Construction Commission
Extending regulatory public value in Australia
Improving regulatory language: Its contribution to regulatory capability
Creating opportunities for regulators to collaborate – reflections from the COVID-19 frontline