Combating Fraud from the Back Office - Part 3

Graham Newman, European Product Marketing Manager, FINEOS

The last blog closed with a summary of the various technological arms in the arsenal of defence against fraud and homed in on Business Rules as being the most propitious place to start.

The First Line of Defence

So what can we do about it?

The first line of defence, the most important part of the infrastructure is the enterprise claims management system. The FINEOS Claims fraud solution handles at least 2/3 of the potential fraud by type and a higher percentage by volume and value.

For the best benefits start dealing with fraud at the point of entry, and deal with it in real time.

If you can automatically apply multiple configurable fraud rules, dynamically run at every touch point that then apply a fraud score to Red Flag a case you can detect a very large proportion of potential fraud activity as it appears in the system and at no extra demand on your hard-worked claims handlers.

The processing of the rules is therefore automated, configurable, controllable, and improvable.

This is the approach we have taken with FINEOS Claims.

The existing structure of FINEOS Claims forms a sound basis for the automatic application of fraud detection business rules.  A rich environment of information is already assembled through Case and Task Management, Party Management and Process Maps.  A Claims Professional can review claimant history and claim participant’s history e.g. cross check if claimant in one case is a witness in another.    We make use of the existing case management and task management structure of processing claims cases and create separate fraud investigation tasks that can be allocated to the appropriate people.

This enables the claims system to automatically create and track notifications to the senior or specialist staff or the Special Investigation Unit (SIU).  It can supply structured information claims data to predictive analytics tools such as SAS or IBM SPSS.  The configurable rules mean that payments may be restricted so they cannot be made if a claim is under investigation, they can provide “Four eyes verification” used to avoid internal fraud for payments and will automatically monitor the activity and payments of all Service Providers of a claim.

Using this as a foundation we now have built-in Fraud Rules that are automatically invoked.  The key here is to work first with what you’ve already got and ensure that it is properly used before you move on the more sophisticated and complicated solutions.

What is a Fraud Rule?

The framework is in place – there are pre-defined points in the claims lifecycle, for example at intake, update . . . and so on where triggers have been created that invoke rules to weigh and assess and subsequently notify through the configured channels.

A Fraud Rule in FINEOS terms starts with an Event Trigger that uses a set of relevant data, its own evaluating logic to produce a weighted fraud score.  This is then checked against full configurable logic that determines how this weighted score will be treated.  Will it just a warning?  Will it go to a supervisor?  Will it form an entirely separate investigation case and be routed to the SIU?

This is all fully integrated in to the case cycle and work path, creating a parallel task, or set of tasks with escalation to appropriate resource – whatever that is, e.g.:

  • Claims handler
  • Supervisor
  • SIU

With appropriate authority requirements and checks built in.

All of these are, as you would expect, Line of Business specific.

The benefits of this approach are enormous. It:

  • Reduces the amount of time taken at intake – no manual fraud check
  • Prompts the claims handler to look at specific claims and not spend time on others
  • Rejects false positives – and give specific reasons
  • Formal mechanism for transferring to SIU
  • Can build links to external databases very quickly, simply and cheaply.

It allows the insurer to gain maximum return on investment:

  • detect organised fraud more rapidly
  • repudiate fraudulent claims before settlement is made
  • avoid potentially non-productive recovery actions
  • reduce costs by avoiding duplication on investigation
  • better focus their resources for fighting crime
  • protect innocent policyholders.

Consider this:

10% to 20% of the claims you handle today may be fraudulent . . . and if they’re not up to those levels today – how long will it be before they are?

For the insurer:

  • Fraud losses weaken an insurer’s financial position, and undermine its ability to offer competitive rates and to underwrite reputable and potentially profitable business.
  • Fraud drains profits
  • Failing to deal with fraud as effectively as possible means competitive disadvantage

For the policyholder:

  • Higher premiums.

Make the most and best use out of the first line of defence.

You may also be interested in