Using BIG Data to forecast Medicare Set Asides

Is it possible to forecast Medicare Set Asides using data analytics? How does an analytics approach compare to conventional methods? And what is the value proposition for making the switch?

These are great questions that we will attempt to address in this article, as well as our case for the use of data analytics to forecast Medicare Set Asides. The foundation for this report begins with a journey, a quest to validate existing methods and determine if a better solution might exist that will allow primary payers to settle claims while also protecting Medicare’s interests.

How did we get where we are today?

That story began in 2001 when the CMS regional offices asked CMS to explain how the offices were to review Workers’ Compensation settlements for future medical to protect Medicare’s interests at the time of settlement. Mr. Parashar B. Patel responded in a memo, dated July 23, 2001, known today as simply the “Patel Memo”, stating that a “Medicare Set Aside Arrangement” should be used for commutation cases. The memo both separates and defines past and future payments to protect Medicare as follows,” Workers’ Compensation commutation cases are settlement awards intended to compensate individuals for future medical expenses required due to work-related injury or disease. In contrast, Workers’ Compensation compromise cases are settlement awards for an individual’s current or past medical expenses that were incurred because of a work-related injury or disease. Therefore, settlement awards or agreements that intend to compensate an individual for any medical expenses after the date of settlement (i.e., future medical expenses) are commutation cases.” While the Patel Memo recommended a Medicare Set Aside Arrangement, it did not offer any specific methodology.

Now flashback to the disability rights movement of the 1980’s, following the passage of the Rehabilitation Act of 1973, when activists lobbied hard for people with disabilities to have access to public transportation and public education. After decades of campaigning, the Americans with Disabilities Act (ADA) was passed in 1990 providing equal treatment and equal access of people with disabilities to employment and public accommodations. Under the ADA, businesses were mandated to provide reasonable accommodations to people with disabilities and full participation, inclusion and integration of people with disabilities. With this the emergence, the field of Life Care Planning evolved, entering the litigation space, to offer valuations of the injury costs to a person’s life and livelihood, in the event of an accident or injury causing disability. The Life Care Planning process was developed by Vocational Rehabilitation experts responsible for working with individuals with disabilities to restore and/ or maximize their functional capacity to perform independently both at home and at work. Life Care Planning methodology has become the standard for forecasting medical treatment over an individual’s lifetime and has been used extensively by both plaintiff and defense sides in litigation and jury trials to valuate medical damages.

Life Care Planners found an opportunity to provide a solution to the dilemma of forecasting medical exposure for “Medicare Set Aside Arrangements” by streamlining and hybridizing their methodology to forecast care that is Medicare covered, related to an injury in the form of a “mini” Life Care Plan that could be used to satisfy CMS requirements involving a Workers Compensation commutation settlement. The concept was brilliant, in the absence of any existing methods, and was quickly adopted by the Workers Compensation industry as the gold standard for meeting the requirement of future obligations to protect Medicare. Suddenly, a cottage industry was born and in 2005 the trade group, National Association of Medicare Set Aside Professionals (NAMSAP) was organized to meet the demand for the development of a consistent practice in this field, including credentialing as a Medicare Set Aside Consultant Certified (MSCC) professional. As the organization grew, so did the competition in this space, and early entrants aggressively pushed best practices to incorporate the submission of a Medicare Set Aside to CMS for a review and approval, when a Medicare Set Aside meeting the CMS review threshold was met. Primary payers rapidly adopted these best practices, and so did the Centers for Medicare and Medicaid Services (CMS). When the first Workers Compensation Review Contractor (WCRC) was established in 2003 to evaluate Workers Compensation Medicare Set Asides (WCMSA), the “mini” Life Care Plans became the standard form of accepted methodology by CMS.

Since that time, the Medicare Secondary Payer process has become increasingly more cumbersome with the addition of SCHIP MMSEA Section 111 Reporting, Conditional Payment research and resolution, and finally Medicare Set Aside development and Submission for CMS review and approval. Costs to settle a claim involving a Medicare beneficiary have increased. The first NCCI report regarding Medicare Set Asides was delivered at the Annual Issues Symposium, May 2014 and an Issue Brief released in September 2014, for purposes of quantifying the exposure of MSAs for actuaries and underwriters to the Workers Compensation claims environment. The brief, available here details the CMS review turnaround times and costs.

Now, flash forward to 2018, the insurance industry is changing rapidly amidst a technology revolution that threatens an institution which has remained largely unchanged since the early 1900’s. Consumers and the millennial workforce demand a more streamlined and effective approach to purchasing and engaging with insurance forcing the industry to re-examine and improve risk and claims management outcomes, cost containment and efficiencies to compete in this changing environment. Which raises the question, where do we go from here? Is there a more effective, simpler path to settling claims involving Medicare beneficiaries, to meet the obligation to protect Medicare’s interests while still settling claims?

Research and Analysis of the Accuracy/Validity of Medicare Set Asides and Life Care Plans In 2013 we conducted an analysis of 100 Medicare Set Asides (MSAs) submitted to CMS for review and approval and compared the CMS future care results to actual treatment rendered, two-year post settlement. Upon review of the medical records, we identified that 94 of the 100 claimants treated less frequently and received less prescription drugs than what was allocated. Of the remaining 6 claims which closely mirrored the Medicare Set Asides forecasts, we identified that what these 6 claims all had in common is that they involved catastrophic or chronic, progressive conditions ie. spinal cord injuries, chronic obstructive pulmonary disease (COPD) related to respiratory exposures and/ or debilitating rheumatoid arthritis. We could not ignore that the overwhelming majority of Medicare Set Asides did not resemble the care forecasted in the first two years post settlement.

In 2014, we performed claimant interviews, medical record reviews and an analysis of seventeen Life Care Plans and compared the actual treatment rendered, ten-years post settlement, involving Medicare beneficiaries living in the Bradenton/ Sarasota, Florida region. The results were a non-starter once we identified that only one out of the seventeen beneficiaries knew they had a Life Care Plan. This single beneficiary who was aware of the existence of a Life Care Plan, did not have a copy of the original plan of care, and therefore did not follow the future care recommendations detailed in the Life Care Plan that were intended to maximize his functional independence and wellness.

Following an exhaustive quest to identify a Life Care Planner or Life Care Planning firm with post settlement measurements, we concluded that no empirical evidence exists to support that the methodology of Life Care Planning is accurate or valid. Members of the Life Care Planning community defended the lack of measures and results by stating that the practice itself defines a Life Care Plan as a “dynamic document”, a process characterized by constant change or activity, and therefore no claims are made as to the accuracy of the forecasts.

While we considered that the two-year post settlement analysis of 100 MSAs effects a limitation of time since the final settlement date and the ten-year post study of Life Care Plans is limited in sample size, the results of these analysis do raise legitimate questions about the validity and reliability of existing methodologies to forecast medical care and treatment as well as CMS’ capabilities to determine with any degree of accuracy or certainty, a primary payer’s future obligation to Medicare under the Medicare Secondary Payer Act. This is very disturbing as the sum of Medicare Set Asides produced on an annual basis, represent approximately $13.8 million dollars per year to Workers Compensation primary payers and to the Federal Government, vying for access to these funds under the Act.

A Discussion about CMS’ Use of BIG data Since our journey to validate Life Care Planning and Medicare Set Asides only raised more questions, we searched for examples of CMS’ use of BIG data, and we found a wealth of information about CMS’ commitment under the Affordable Care Act to embrace data analysis. In fact, CMS collects over two billion data points per year from hospitals, physicians, pharmacy benefit managers, durable medical equipment providers, home health care, enrollment information, quality metrics, patient assessment and survey data and information gleaned from phone calls to Medicare. With this robust amount of data, CMS is taking a lead in transforming healthcare. There is an excellent video of Niall Brennan, Chief Data Officer, explaining CMS’ role in leveraging data to drive access, quality, and cost containment under the affordable care act available here.

The use of big data is not a new concept for CMS, when a framework for monitoring quality of care and utilization of inpatient services was first developed at Yale in the 1960’s, and the framework was adopted by the State of New Jersey in the 1970’s Medicare followed its adoption in 1982. With the passage of the Tax Equity and Fiscal Responsibility Act, modified Section 223, Medicare hospital reimbursement limits were developed to include case mix adjustment based on diagnosis related groups (DRG). The following year, Congress amended the Social Security Act to include a national DRG-based hospital prospective payment system for all Medicare patients; establishing a fixed inpatient payment based on a list of diagnoses. With the final rule changes for 2018, the MS-DRG is based upon 462 diagnoses and risk factors and it represents the inpatient payment model for Medicare reimbursement.

Under the Affordable Care Act (ACA), there are eight primary goals. Two of them involve the use of data analytics and streamlining payer models, both ripe for improving MSP. As a data initiative, physicians and hospitals submit quality reporting data and participate in a quality payment program for clinicians. A current example of CMS’ use of big data is the implementation of the Medicare Access and CHIP Reauthorization Act (MACRA) of 2015. MACRA ended the Sustainable Growth Rate (SGR), a means by which Medicare controlled is expenditures for Medicare Part B physician reimbursements. There are 55 million Medicare beneficiaries who elect Medicare Part B outpatient benefits. Because of MACRA, physicians are paid under the Merit Based Incentive Program System (MIPS). Physicians are motivated by value-based incentives to achieve better health outcomes for patients. In other words, they are paid for performance. Performance is measured, and providers submit annual performance measures in March, they will receive feedback from CMS at the end of the year and are paid risk adjustment payments, for good performance, on January 1st of the following year. There are also Advanced Alternative Payment Models (AAPM) for physicians, who meet the criteria, to participate in risk sharing and shared savings programs. The underlying intent of these payment models is to align provider incentives with health outcomes, to improve healthcare results. The quality measures are driving evidence-based medicine and a more wholistic approach to care and patient engagement.

Using BIG data and Predictive Modeling to Forecast Medical Care Care Bridge International hosts a data warehouse of over 1 billion medical claim transactions, including Medicare Set Aside (MSA) data of CMS reviewed and approved MSAs. Using data algorithms and predictive modeling, the company can forecast Medicare Set Asides and medical reserves for Workers’ Compensation claims with a high degree of accuracy. An actuarial review of the company’s data, methods and processes was given a credible endorsement for its analytics platform. The following represent our steps and experience for the development of our model-generated medical treatment and cost predictions:

  • Identify data sources for outcome measures, predictors, and variables.

  • Understand the completeness of the available data set(s).

  • Prepare and clean the data for basic statistical analyses to ensure integrity of the data.

  • Generate/ validate predictive models using different sets of similar historical data.

  • Research, evaluate & determine indicators/ variables required for good model(s).

  • Develop forecast models by diagnosis.

  • Maintain and update the model(s) ongoing.

The patterns and behavior of the aggregate data can display very stable characteristics, even when individuals exhibit high degrees of randomness; pooled data provides a high degree of accuracy!

Analytic Powered MSAs vs. Conventional MSAs: Comparative Value In comparing an analytic, or data driven approach (as shown below) it is clear the use of data provides a valid and reliable foundation from which to forecast medical care.

It is easy to agree that with such accuracy and efficiency, that an analytic-powered model, built on logic and data algorithms, is a viable alternative to existing methods. But the obvious question is what is CMS Acceptance of Analytic-Powered MSAs?

CMS Submission of Medicare Set Asides: Discussion It is recognized that WCMSA Submission to CMS is a voluntary process. On May 11, 2011, CMS issued a memorandum reiterating guidance regarding the submission of WCMSAs and the memorandum states: “Submission of a WCMSA proposal to CMS for review and approval is a recommended process. There are no statutory or regulatory provisions requiring that a WCMSA proposal be submitted to CMS for review.” CMS submission has always been voluntary, meaning there is no legal requirement for CMS to review and approve a Medicare Set Aside Arrangement for a Workers Compensation commutation settlement.

Using our professional MSA software, MSAs may be submitted to CMS for review and approval, but if a primary payer chooses non-submission and desires to generate a Medicare Set Aside in minutes and settle claims 6-8 months faster, an Analytic-Powered MSA is a great alternative to today’s cumbersome MSA process.

Based on our analysis of over 300 Medicare Set Asides produced using Care Bridge’s Professional software vs. Analytic Powered MSAs, the following graph compares the results to CMS approved amounts.

CMS Acceptance of Data Driven MSAs:

Analytic-Powered MSAs are standardized using diagnosis, treatment and prescription drug coding as well as provider indexes for dashboard analysis. The data reveals that diagnostic tests and other outpatient services generally occurs with less frequency than what CMS or a credentialed professional might expect, and Analytic powered MSAs substantively correct for missed opportunities in claims handling as in the example of unauthorized or unrelated care and treatment that inadvertently creeps into the claim. Analytic-Powered MSAs forecast future care based upon state rules and guidelines and diagnoses identified as compensable to the injury claim. This capability may save a primary payer 20% or more on future care projections. Last year we saved our clients $2.6 million dollars in both truer MSA settlements and claims leakage with more efficient use of human capital! The Future of MSAs is Here!

Featured Posts
Recent Posts
Search By Tags
Follow Us
  • Facebook Basic Square
  • Twitter Basic Square
  • Google+ Basic Square