Printer Friendly

Work records: advanced analytics can help workers' compensation writers to become proactive about claims.

Challenging economic conditions, including company downsizings, business closings, rising unemployment, record low interest rates and escalating medical loss costs have made the workers' compensation line a tougher competitive environment.

Moreover, the uncertain impact of national health care reform and the unknown pace of the economic recovery are further pressuring the workers' compensation market, its constituents and participants.

As the insurance industry deals with these business challenges, market leaders are increasing the speed of innovation and their ability to adapt to the new market conditions.

Using advanced analytics throughout the claims management process is one way these first movers ate choosing to defend their bottom lines.

Stated another way, these insurers are turning their claims' life cycles into a competitive advantage by aggressively attacking the largest spending category for a property/ casualty insurer: claims payouts and expenses.


Analytics--the use of statistics and technology to solve business problems--has been used in insurance for many decades. However, its usage has been disproportionately confined to the underwriting and pricing functions.

Companies are now actively expanding their analytics capabilities into newer areas, notably the claims management space.

The use of analytics in claims management has allowed claims professionals to gain insights and a deeper understanding of the claims life cycle: from claim occurrence and first notice of loss to claim resolution and closure.

Advanced analytic methods have helped organizations reveal areas of operational deficiencies, inconsistencies in claims handling practices, and subjective decision making rooted in traditional organizational biases.

Areas needing improvement were then remedied to foster proactive and intelligent decision making. The largest efficiency gains were observed to be around claims triage and duration curtailing, by focusing the right resource on the right claim at the right time. Some organizations have expanded these analytics into the domains of fraud, litigation and recovery to achieve an even-higher return on investment.

Beyond Triage

Traditional rules-based triage offers useful information on how to classify and manage claims.

However, these rules are typically based on limited or outdated data and often are insufficient to deliver meaningful segmentation. As a result, companies are experiencing high reassignment rates and increased overall claims durations, which directly impact the bottom line.

If only claim handlers could possess and leverage better insights earlier in the claims life cycle, they would be able to segment claims from low to high.

Analytics incorporate internal data (such as claimant and payroll information, first-notice-of-loss claim data, adjusters' notes and police reports) coupled with a wealth of external third-party data (like lifestyle data, financial variables, geodemographic information and physician data) to test a large number of potential predictors and retain those with the necessary predictive power.

Because an analytics algorithm can sift through thousands of scenarios quickly and rigorously, it can identify characteristics of intricate claims that require the attention of a company's most experienced specialty services, as opposed to those that require minimum-to-no intervention.

Therefore, resource deployment can be improved on each claim as soon as it is reported. Claim management systems can fast-track the more straightforward and lower-cost claims, ultimately settling them with minimal cost and resource involvement.

A Solution for Fraud

At the same time, claims with a higher propensity to become more complex and costly over time can be assigned to the most-experienced adjusters and specialty resources at the outset.

This early, intelligent decisionmaking approach can help to curtail claims duration and decrease claims costs by focusing the right resource on the right claim at the right time.

With hard and soft fraud costing U.S. companies $80 billion a year, according to the Coalition Against Insurance Fraud, businesses are looking for new ways to proactively identify and deter fraud.

Traditionally, Special Investigative Units' efforts have been reactive in cases that already displayed suspicious activity--often months after the claim was reported and the investigation had begun. By that time, losses have been paid, expenses have been incurred and resources have been mobilized to handle and adjust the claim.

Organizations are now turning to analytics to avoid these out-of-pockets costs by identifying the monthlong suspicious activity at first notice of loss so that early intervention can occur.

Using analytics, businesses can process, mine and analyze a wealth of information to quickly spot unusual claim patterns that are the manifestation of soft fraud (claim padding).

Because analytics can interpret thousands of data points to identify soft fraud at first notice of loss, suspicious cases can be identified promptly so that SUIs can focus their efforts on more-targeted referrals and pursue investigative actions with more confidence.

Hard fraud is typically more challenging for SIUs to recognize until the new fraudulent schemes becomes widely reported and recognized--typically months or years after the fact.

However, other forms of complex analytics, such as machine learning and cluster analysis techniques, have been designed to feed off new patterns that accumulate and become detectable in the data. It is a form of analytics that can monitor and adjust itself to new patterns so that hard fraud schemes are recognized and dealt with much more quickly than before.

A number of success stories underscore the effectiveness of leveraging analytics to reduce fraud. For example, a number of claims with high fraud scores have been swiftly referred to SIUs at first notice of loss. After single in-person visits with the claimants, some of these claims were quickly withdrawn and the payouts driven down to zero. In other cases, a simple phone call was sufficient to cause the claimant to withdraw the claim.

Litigation Management

Recent surveys and business reports confirm sharp increase in the number of litigated cases and associated defense costs. Traditional reactive approaches to litigation have become obsolete as litigation losses continue to mount.

Claims are litigated for various reasons, some outside the control of organizations and others within their control.

In cases of avoidable litigation, analytics are being used to recognize the characteristics of those with a high propensity to move toward litigation. Business actions, including focused customer attention and deployment of specialty services, are then taken to prevent the case from becoming disputed. By doing so, legal expenses are often reduced, and in some cases avoided completely, ultimately curtailing duration.

For cases where litigation is unavoidable, analytics can be leveraged to help drive the appropriate defense strategy.

A clever combination of claim, claimant case characteristics and social media data, leveraged by appropriate technologies, are helping organizations steer their defense strategies and manage overall litigation more effectively--with a direct impact on the bottom line.

Maximizing Recovery

From salvage to third-party subrogation, opportunities exist to improve recovery amounts and retain revenue.

With a limited number of resources to work on returning revenue, the pool of claims on which to focus is of critical importance. Typically, referrals for third-party recoveries are based on simple rules and traditional insights.

However, these rules are typically biased and allow for a less-than-optimal pool of referral.

Using an analytical approach to determine the cost-effectiveness of each recovery opportunity, insurers have had success in many lines of business, including workers' compensation.

A wide array of data is used for this solution, starting from intuitive claim and policy information to newer external market and psychodemographic data. Text analytics on case management notes or other unstructured data also have been used by organizations to extract valuable insights that can be used to determine the validity of the recovery case.

All this data in a predictive model can point to, say, the top 10% of opportunities that have a greater propensity for recovery and a larger expected returned amount.

In addition, because these solutions are generally available at first notice of loss or shortly afterward, the recovery lags can be diminished, thus increasing the value of the recovery.

Implement and Enhance

The end-to-end implementation of any analytical tool is as essential as the development of the tool itself.

Successful implementation combines an effective IT integration with a predefined and well-accepted business implementation plan.

Organizational change management is also essential to create buy-in from a company's employees and to build faith and trust in the new tools and methods.

Performance measurement and monitoring will help assess success metrics and areas of improvement.

Key performance measures--such as reassignment rates, fraudulent claim percentages and durations--are just a sample of important metrics that companies must continuously monitor and report on.

Lastly, the implementation of any analytic solution is rarely a "one and done" effort.

The model must be routinely examined, refreshed, improved, and expanded upon as additional external data and lessons learned are identified.

As claim professionals take business actions based on the model scores, the change in the mix of claims being triaged typically requires the model to be refreshed. Much like human beings, models are able to grow and become stronger over time.

Advanced analytic methods and tools in claims management are transforming traditional behaviors from reactive to proactive.

By focusing the right resource on the right claim at the right time, insurance companies can realize significant benefits, including expected savings from loss and expense of 4% to 8%, based on market experiences.

Given the challenging economic environment and the soft insurance market, there is no doubt that companies that turn their claims' life cycles into a competitive advantage can emerge as winners in the long run.

With the added benefits of improved customer service, reduced insurance costs and employees returning to work sooner, science can win the day.

Key Points

* The Background: The rate of the nation's economic recovery and the impact of national health care reform are placing pressure on the workers' compensation line.

* The Trend: Many workers' comp insurers are using analytics to change their claims' life cycles into a competitive advantage.

* The Payoff: An analytical approach to claims can reduce fraud, prevent litigation and improve recovery amounts.

Watch an interview with Amel Arhab at video. Digital readers: Hold cursor over icon for content.

Contributor Amel Arbab (FCAS, MAAA), is a manager with Deloitte Consulting LLP. She can be reached at aarhab@
COPYRIGHT 2011 A.M. Best Company, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Property/Casualty: Adjusters/Claims Package
Author:Arhab, Amel
Publication:Best's Review
Date:May 1, 2011
Previous Article:Making change: transforming claims systems can help insurers retain customers.
Next Article:Raising standards: U.S. commercial and residential building codes need to be upgraded.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters