Diagnostic & Pharmacy Risk Models of High-Cost and Long-Term Nursing Home Entry

Predicting high-risk individuals that can benefit from high-touch medical management programs has been proven to improve quality of care, identify and address healthcare access issues, and control long term costs. Identifying these individuals in a Medicaid program setting is more challenging than in a commercial population because of the unique nature of the population’s enrollment patterns, disease prevalence, and demographics. In addition, the unique issues surrounding the dually-eligible Medicare-Medicaid population make accurate identification of individuals that can potentially benefit from appropriate programs even more difficult.

JAI has developed a Medicaid-specific predictive modeling system that can not only predict future costs, but can also predict long-term nursing home entry. The system can:

• Provide Medicaid programs with information to assist in targeting potential candidates for screening and evaluation for inclusion in a nursing home facility entry “prevention group”; and
  
• Provide detailed diagnostic and impairment profiles* of recipients and populations for use in the planning and implementation of medical management programs; and
  
• Maximize the use of available data by using medical and pharmacy data as well as pharmacy data alone (when medical data may not be available or is not timely) in determining risk..

The components of JAI’s predictive model were initially developed with funding provided by the Robert Wood Johnson Foundation's Medicare/Medicaid Integration Project (University of Maryland Center on Aging). The system is based on tested impairment categories * of disease and illness that are predictive of concurrent and future need for long term care services. Some of these categories include: minor ambulatory limitations, severe ambulatory limitations, cognitive developmental disability, chronic mental illness, dementia, sensory disorders, self-care impairment, general symptoms, cancer, chronic medical disease, pneumonia, renal disorders and other systemic disorders (e.g. septicemia).

The system combines high-predictive accuracy combined with clinically-rich patient profiles that can assist Medicaid programs to identify high-risk, “intervenable” patients and become efficient in determining potential treatment patterns to improve patient care.

* A tool designed by JAI that relies upon diagnoses information contained in the claims data for beneficiaries to assign risk to persons, and to conduct epidemiological studies.

Long-Term Care Analytics

Learn More