Leveraging the LACE Algorithm for Readmission Prediction
Better care for individuals, better health for patients and lower per-capita costs form the triple aim of the U.S. healthcare reforms for the Centers for Medicare & Medicaid Services (CMS). The number of unplanned and avoidable readmissions that occur within 30 days of discharge from a hospital, too, are under the scanner to maintain the quality of care.
The federal government has taken many steps such as incentivization and penalty under the Affordable Care Act (ACA) to optimize and increase the quality and efficacy of the healthcare system. Medicare’s Hospital Readmission Reduction Program (HRRP) imposes financial penalty on hospitals with excess readmissions.
As per a recent report by CMS, nearly 2,225 of the 5,700 hospitals in the U.S. will be subjected to penalties and payment reductions amounting to a total of $227 million in 2014. The study reveals that one in every five elderly patients is readmitted within 30 days of leaving a healthcare provider setting. Medicare Payment Advisory Commission (MedPac) statistics show that 12% of the readmissions are avoidable — thus costing Medicare $15 billion.Avoiding even one of every 10 of those readmissions could save Medicare $1 billion.
With maximum penalty increasing to 3% in 2015, it is imperative for healthcare providers to curb the avoidable 30-day readmissions. One of the foremost ways to achieve this goal is to have a methodology for predicting patients’ readmissions prior to their discharge.
This will help providers analyze the risk associated with patient readmission and take necessary steps to avoid them. Providers can mitigate this challenge by leveraging the LACE algorithm to predict the chances of readmission.
How does the LACE algorithm work?
Length of stay (L), acuity of admission (A), comorbidity (C) and emergency visits within the last six months (E) are the four variables associated with unplanned readmissions within 30 days.
LACE scores each of the variables based on various criteria for each patient encounter to arrive at the probability of the patient getting readmitted within 30 days of discharge. For example, a patient having a LACE score of 10 at the time of discharge has a 12.2% probability of getting readmitted within 30 days of discharge from the hospital. This score helps care delivery teams plan discharge to prevent readmission.
Predictive Analytics workflow using LACE model
Providers such as Parkland, North Shore and Texas have benefitted by stratifying risk for 30-day readmissions. Many others are looking to adopt similar predictive analytics capabilities, helping them increase financial margins, provide effective care to patients and attain the triple aim. This move in healthcare reforms, from volume-based reimbursements to value-based reimbursements, is bringing readmission analytical solutions such as LACE into the limelight.
Dr. Sumit Rai is a Business Analyst, Healthcare Provider. He has over four years of experience across clinical practice and healthcare IT. Having an in-depth understanding of the U.S. healthcare provider market, he is currently involved in business analysis on upcoming trends/reforms and developing new offerings.