Konekt introduces referral triaging to improve outcomes

Workplace related injuries cost the Australian economy an estimated $60.6bn each year.1 A large portion of this cost can be mitigated by identifying high risk injuries and ensuring effective management processes are in place early – before high costs are incurred.

The 80/20 rule can often be applied! The use of predictive models to assist claims managers in effectively managing the 20% of claims that drive 80% of costs has become more common across compensation schemes nationally.

In Australia, Worksafe Victoria provides an example of where a statistical case estimate model has been implemented using a range of claims variables to assist in predicting the claims outcome. Some of the variables included in the model were:

  • Diagnostic: location, means and type of injury
  • Weekly compensation details: time off work, amount of payments
  • Demographic: Age, salary
  • Medical: medical costs, the dates of medical reports, days in hospital
  • Time based: Time since claim lodged, time to finalisation, duration since last payment.2

Also in Victoria, the Transport Accident Commission (TAC) has developed a claims segmentation algorithm to help predict claims likely to progress to a common law claim, and to assist in the streaming of claims into appropriate claims management teams to ensure effective management practices.

Sociological, medical and accident related variables captured through claims lodgement determine the segmentation outcome.3

The benefits of identifying high risk injuries early and assigning them to an appropriately experienced and qualified consultant are considerable. Konekt has recently developed a risk profiling application that imports data at referral and uses interpretive algorithms to determine if a referral is red, amber or green. A red rating represents a ‘high risk referral’ or one with a high probability of having a longer duration of service intervention and a higher cost. A green rating represents a ‘low risk referral’ or one with a low probability of having a longer duration of service intervention and service cost.

The variables used in Konekt’s risk profiling application are based on the Reed Groups Medical Guidelines and New Zealand’s Accident Compensation Corporation’s (ACC) Injury Statistics – definitions and classifications (2006). The variables include age; work type; mechanism of injury; injury type; body location of injury; and injury severity. Using the algorithms and logic built into the application, a combined recovery score in weeks is determined for each return to work referral. Based on the score reached, a classification is assigned.

The referral is then matched to a consultant using the risk profile classification and the qualifications, skills and experience of a consultant. This quick link to a consultant skilled in dealing with cases of similar complexity is designed to help ensure a quicker recovery and return to work.

Konekt’s risk application is currently being trialled in our Sydney and Melbourne CBD office locations.

1 Safe Work Australia. 2012. The Cost of Work-Related Injury and Illness for Australian Employers, Workers and the Community: 2008-09.
2 Hugh Miller, The Rise and Rise of Hybrid Modelling presentation delivered at 17th General Insurance Seminar, 7 – 10 November 2010.
3 Pocock, N., Holdenson, S. and Gifford, D. (2011). TAC Claims Management Transformation. 13th Accident Compensation Seminar of the Institute of Actuaries of Australia.
4 Neil Smithline and Nora Blay, Using Artificial Intelligence to Predict High Cost Workers’ Compensation Cases White Paper, January 2001, www.riskinfo.com