top of page
worker's compensation insurance.jpg

Workers Compensation & General Liability

Our casualty risk analytics practice is focused on materially reducing clients' total cost of risk through the development and deployment of loss experience-tailored, claim-level predictive analytical systems, designed to support daily claim management efforts of clients who self-administer their claims, as well as those who work with third-party administrators (TPAs).

Our Evidence-Based Casualty Cost Management Process, built around advanced predictive modelling and text mining capabilities, enables claim handlers to more rationally allocate their time and economic resources, so that specific claims that exhibit the greatest probability of cost escalation receive appropriate level of scrutiny and oversight. In a typical casualty claim pool, relatively few claims tend to account for a disproportionately large share of the total cost - in our cross-industry experience, 70% to 90% of total (indemnity + medical + expenses) claim related expenditures tend to be concentrated in 10%-30% of claims.


In spite of that well known asymmetry, claim processing workflows tend to expose most, if not all incoming claims to the same level of due diligence - stepped-up claim triage efforts are usually triggered by incurred costs exceeding a pre-determined threshold, such as $5k to $10k for smaller or $25k to $50k for larger organizations. Under that scenario, during the critical early lifescycle stages claims are effectively allowed to self-determine their cost trajectories, which commonly leads to unnecessary cost inflation. By harnessing the predictive content of readily available claim characteristics, our Evidence-Based Casualty Cost Management Process enables organizations to assign the expected-future-cost-suggested type and amount of claim handling resources, which ultimately results in significant cost savings delivered through more intelligent utilization of current staffing and resource levels.


Our Evidence-Based Casualty Cost Management Process is further supported by a set of specific data-drived templated capabilities:

Initial Resource Allocation

Organization specific multivariate statistical claim scoring algorithm, which utilizes claim characteristics, such as cause of accident, body part, nature of injury, and claimant demographics, and claim-derived attributes, such as reporting lag, to make a processing protocol determination: For example, a claim that is expected to exceed a pre-determined cost threshold, such as $5k, will be assigned a Protocol 1 handling; a claim expected to exceed a higher threshold, such as $25k, will be assigned a Protocol 2 handling, etc.


The core intent of the Initial Resource Allocation System is twofold: First, to make an objective assessment of development; second, to substantiate optimal claim management resource allocation, so that the amount of claim-level effort and oversight is proportional to the expected future cost of individual claims. The end result - and the ultimate benefit - of deploying the Initial Resource Allocation System is a material, systematic and sustainable reduction in the total cost of risk, accomplished through a combination of more aggressive management of claims exhibiting heightened likelihood of significant cost escalation, and claim risk assessment automation, which in turn enables organizations to optimize their claim management related staffing needs.


The Cumulative Claim Cost Index

Another key claim management challenge is the difficulty of correctly discerning year-over-year changes in residual claims, or those with duration greater than 1 year. Due to the fact that both the mix of surviving (duration > 1 year) claims and claim count are ever-changing, it is difficult to pinpoint the specific root causes of an increase or a decrease in the aggregate cost of those claims.


We developed the Cumulative Claim Cost Index (CCCI) to offer a methodologically sound solution to the problem of discerning and quantifying root causes of observed cross-time aggregate cost changes in residual claim pools. CCCI distils those changes to variability in 'open claim residuals', 'claim carrying cost', and 'average claim cost'; in addition, the Index also summarizes all cost-related claim pool attributes into a single standardized measure to help managers understand and address aggregate factors exerting the strongest impact on the total cost of risk, on residual claim pool-wide basis.


Depending on claim volume, the CCCI can be computed more frequently (for high claim volume organizations), such as monthly, or less frequently (for lower claim volume organizations), such as quarterly or annually. Lastly, the Cumulative Claim Cost Index also offers an easy to understand, 'state of the claim management' summary metric for communicating the key aspects of claim management performance to the organization's senior management.


Loss Prevention Analytics

To gain further insights into potential cost saving remedies we also offer a more open-ended Loss Prevention Analytics, the goal of which is to pinpoint specific pre- and post-loss factors that exert abnormally high impact on either the frequency or the severity of claim-precipitating accidents. Our goal is to extract the maximum amount of decision-guiding insights out of the available data, which includes structured/numeric data as well as unstructured adjustor notes and other text data. Our Loss Prevention Analytics offering is adaptable to a combination of informational needs, data circumstances and the level of data analytical sophistication of individual organizations.


When focusing on pre-loss (e.g., safety and related) insights, our analyses not only identify sources of both the frequency and the severity of accidents, but even more importantly, bring forth the often elusive cross-factor interdependencies. When mining the available data for post-loss related insights, Loss Prevention Analytics is focused on the identification of claim management process improvement opportunities.


The key aspect of our offering in this context is delving into normally not utilized data, such as adjustor notes, with the goal of finding 'action triggers' that could be used as forward-looking indicators in the future. Another key aspect of our post-loss offering is that of linking stand-alone data sources, such as claim records, adjustor notes, medical treatment and/or prescription metrics, via a process known as multi-source analytics, to gain yet additional insights into otherwise unknowable aspects of claim management, all with the goal of deepening and broadening the efficacy of data-derived, objective knowledge that is available to claim professionals.

bottom of page