Welcome to the MedMetrics Blog

The MedMetrics blog provides comments and insights regarding the world of Workers’ Compensation, principally, issues that are medically-related. The blog offers viewpoints regarding issues affecting the industry written by persons who have long experience in the industry. Our intent is to offer additional fabric, perspective, and hopefully, inspiration to our readers.

Search The MedMetrics Blog

Sunday, February 27, 2011

WC Analytics Can't Live on Bill Review Data Alone

Analytics are all the rage in Workers’ Comp—finally. It’s been a long time in coming to this industry, but people are making serious moves to embrace and implement the concept. Analytics is the only untried option remaining that has the potential to control costs and create efficiencies in the industry. Intelligently crafted and applied in operations, analytics will recharge managed care programs, making them more efficient and effective, thereby measurably controlling medical costs.

Analytics (a fancy term for data analysis) fall into two basic categories: Descriptive Analytics and Predictive Analytics. Descriptive analytics are “the what”—what happened in the past described through reports, queries, and data drill downs to gain deep understanding of claim processes and participants. Descriptive analytics identify critical business issues, trends, and cost drivers. The approach is essential to understanding relationships, the business process, and provides the platform for asking the right business questions. Moreover, descriptive analytics are crucial to decision support and are the foundation for determining the right focus going forward.

Predictive Analytics, on the other hand, are used for forecasting, advanced reporting, and optimizing algorithms. Advanced mathematical and actuarial analyses are used to predict the future based on the past. If X is true, what is the probability Y will occur? Or when Y occurs, what are the factors that could have predicted it?

Using predictive analytics to more accurately estimate the cost of a claim, thereby setting accurate reserves, is one example of how predictive analytics takes an organization to a higher level of effectiveness. Intelligent use of predictive analytics can yield greater measureable cost savings and competitive advantage for an organization.

Both analytics approaches are important to optimizing the effects of claim and medical management, along with cost control. Still, there are challenging hurdles that must be overcome to effectively implement analytics in the Workers’ Comp industry. The challenges relate to how people perceive sufficient application of the process.

Crimes against analytics in Workers’ Comp relate to how the data is selected and applied. The Workers’ Comp industry has truckloads of data—quantity of data is not the problem. The trouble for analytics in Workers’ Comp is data collection, integration and, too often, narrow understanding of what type of data should be tapped for analytics.

A well-known fact is that Workers’ Comp data lives in separate silos. The fact that relevant and important data resides in disparate locations and sometimes in different companies is not a small problem, but it is very manageable problem technologically. Data from different sources can be readily transferred and integrated. Of greater concern is the resistance to gathering and integrating all the relevant data in order to perform adequate analytics, analytics that will enlighten operations. For instance, many think bill review data is enough.

One reason people rely on bill review data is that it is the most plentiful and accessible. Nearly all medical bills in Workers’ Comp are run through bill review systems. Conveniently, different bill review systems contain the same range of data, that which is generated from standardized medical billing formats. The formats are those required by CMS (Centers for Medicare and Medicaid Services) such as HCFA-1500 (Health Care Finance Authority) and the UB-04 (Uniform Billing-2004). Medical billers use those standard billing formats for Medicare and Medicaid and typically use the same formats for Workers’ Comp billing. In some places, the formats are required. Standardized formatted data are run through bill review systems, making the data not only prolific, but relatively uniform.

Bill review data is detailed and specific. ICD-9 diagnostic codes and other standardized charge codes, such as CPT codes (Current Procedural Terminology) are available. The treatment process, treatment providers, and recommended payment can be derived from bill review data. However, actual paid costs, non-medical costs, and treatment effectiveness measured in terms of actual outcomes cannot.

Analysis of the medical treatment process and provider performance in terms of claim outcome or the actual claim cost cannot be derived from bill review data. Claims level data is needed to shed light on provider performance in terms of return to work, indemnity costs, litigation, as well as the duration and outcome of a claim. At a minimum, claims level data should be combined with bill review data for meaningful analytics. Moreover, predictive analytics is without foundation when applied to bill review data only.

It’s true, one can use descriptive statistics of bill review data to capture and understand trends in injury types, treatment processes and billed costs, but little beyond that. Bill review data alone will not reveal comprehensive actual claim costs or illuminate treatment effectiveness, provider performance, total cost of the claim or outcomes because it represents only a portion of claim information. First class analytics, that which produces predictive, actionable information cannot be limited to bill review data. Those who say bill review data is enough are mislead.

View additional articles by Karen Wolfe under Blogs at www.medmetrics.org