Linked In Leveraging predictive analytics to deploy clinical resources - Sedgwick
Latest Coronavirus updates from Sedgwick Back to business solutions

Leveraging predictive analytics to deploy clinical resources

The decision to deploy clinical resources on a claim can be crucial to the outcome of the claim. This decision is frequently made when the claim hits a specific dollar threshold or when the examiner determines a referral is needed. However, assigning clinical resources once the claim costs hit a certain threshold limits the effectiveness that a nurse can have on changing the course of the claim. Employers are looking for better ways to more accurately determine when to assign clinical resources while reducing costs. Predictive analytics technology is showing great promise in helping do just that. A decision optimization model leveraging this technology is proving that it is more effective than standard business rules.

Determining which claims would benefit from clinical intervention requires careful analysis of claim data and characteristics. A recent blog post on improved decision making outlined the primary aspects of data analysis that can have an impact on claims and the key elements in an effective decision optimization model.

Instead of standard business rules that focus on a dollar threshold as the main indicator that nurse case management is needed, more advanced decision optimization models examine hundreds of data fields, including diagnosis, age of the worker, jurisdiction, and comorbidities, to identify combinations that show the potential for complications and higher cost. It would be almost impossible for an examiner to recognize all the patterns and combinations that are identified using these algorithms. By continuously scanning claim data, a claim that meets the criteria can be immediately flagged. Once flagged, the claim can be automatically referred to a nurse case manager for a full evaluation, ensuring any false triggers are caught in advance. 

It is important to note that the goal of predictive analytics is not to increase the number of claims referred to nurse case management. Instead, the goal is to have the greatest impact on the outcome of the claim by reducing the number of false triggers requiring review and to engage a nurse earlier.

Compared to the standard business rules, decision optimization helps improve results in key areas. In a recent analysis, employers using a sophisticated clinical decision optimization model have seen the following results:

  • Earlier intervention Decision optimization starts evaluating claims at the first report of injury and then continuously looks for specific data points as the claims progress to ensure early intervention. Employers that switched to the decision optimization model saw a 31% decrease in the number of days for referral.
  • Fewer false triggers Decision optimization is better able to identify those claims that would benefit from nurse case management. In a recent comparison of referrals using decision optimization versus those not using the model, accuracy in identifying claims needing a nurse increased and the number of false positives (claims triggered that did not require nurse engagement) decreased by 23%.
  • A reduction in medical and indemnity costs Average results for employers after using the decision optimization platform for one year include a 7% decrease in indemnity costs, an 8% decrease in medical costs and a 5% reduction in total incurred.
  • Faster return to work Average results for employers using the decision optimization platform saw their temporary total disability reduced by 7% on average.

The use of decision optimization is significantly improving the ability to identify when case management is needed. The decision is now based on methods that have been proven to make a positive impact on medical and indemnity costs for employers, and improve return-to-work results. 

Back to Blog
Back to top