ACO offers population health pointers for beginners

'You really need to try to bring all your data together in one location'
By Erin McCann
04:33 PM
Share
Advocate Lutheran General Hospital

Advocate Health Care in Illinois has 13 hospitals and one of the largest accountable care organizations in the U.S., including a Medicare Shared Savings ACO, a combined full-risk HMO and a commercial ACO contract with Blue Cross & Blue Shield of Illinois. All in all, its accountable care footprint covers 528,000 lives. With population health management at the center of ACO quality metrics, the health system is well acquainted with data analytics. And they have some tips for those getting started with population health.

Rishi Sikka, MD, is one of the individuals in charge of it all. 

Sikka, the senior vice president of clinical transformation for Advocate Health Care, is charged with the oversight of clinical operations, business intelligence and big data initiatives at the health system. He will be speaking at the HIMSS Media and Healthcare IT News Big Data & Healthcare Analytics Forum in New York June 18, discussing population health strategies.

Looking to get started with population health at your organization? Sikka has some tips that may come in handy.

First things first. You need to establish the "why." It's basic, but an essential part of the population health foundation, and one often neglected, said Sikka: "Why are you on this journey for population health?" That answer will help establish the goals you want to get out of population health. Once you have them defined, then comes the strategy to achieve those goals.

Another key step? Before you get started with it all, put all the data together. "You really need to try to bring all your data together in one location centrally that is normalized, standardized and reconciled along the single master patient index," said Sikka. "Because if you continue to look at your data in silos, then your care will be fractured in silos, and you won't really achieve the goals of true continuum of care."

Know and talk about where both the internal and external data resides and how you'll bring it together in one place, he explained.

Next, don't be so quick to jump to the predictive models, said Sikka. You need to better understand the present conditions of your patient populations before you focus on the future.

"The very first thing you need to do from an analytics standpoint is just understand the current state of your population," he said. Ask, "What is happening today?" he added, and "you'll actually find that they're quite challenged to answer that."

Instead of thinking about predicting readmissions, for instance, do you know about the populations being readmitted today? This step is about "understanding a state of being."

Then, when you have this down, you can move onto predictive models. But remember: "If you just have all these models and data, it's just an academic exercise." To really improve the care of patient populations and address cost issues, "you need to take the model, build it into workflow and generate some type of action, an actionable impact on the patients you're caring for."

TABLE: 2013 results from Advocate Physician Partners ACO (Data via PulsePilot)

Description Measure Percentile Category
Description Measure Percentile Category
How Well Your Doctors Communicate ACO-2 90th+ Patient/Caregiver Experience
Patients' Rating of Doctor ACO-3 90th+ Patient/Caregiver Experience
Health Promotion and Education ACO-5 90th+ Patient/Caregiver Experience
Health Status/Functional Status ACO-7 90th+ Patient/Caregiver Experience
Risk Standardized, All Condition Readmissions ACO-8 90th+ Care Coordination/Patient Safety
Tobacco Use Assessment and Cessation Intervention ACO-17 90th+ Preventive Health
Depression Screening ACO-18 90th+ Preventive Health
Getting Timely Care, Appointments, and Information ACO-1 80-90th Patient/Caregiver Experience
Access to Specialists ACO-4 80-90th Patient/Caregiver Experience
Shared Decision Making ACO-6 80-90th Patient/Caregiver Experience
Medication Reconciliation ACO-12 80-90th Care Coordination/Patient Safety
Proportion of Adults who had blood pressure screened in past 2 years ACO-21 80-90th Preventive Health
Percent of beneficiaries with diabetes whose HbA1c in poor control (>9 percent) ACO-27 80-90th At-Risk Population Diabetes
Beta-Blocker Therapy for LVSD ACO-31 80-90th At-Risk Population HF
Percent of PCPs who Qualified for EHR Incentive Payment ACO-11 70-80th Care Coordination/Patient Safety
Falls: Screening for Fall Risk ACO-13 70-80th Care Coordination/Patient Safety
Percent of beneficiaries with hypertension whose BP < 140/90 ACO-28 70-80th At-Risk Population Hypertension
Percent of beneficiaries with IVD who use Aspirin or other antithrombotic ACO-30 70-80th At-Risk Population IVD
Adult Weight Screening and Follow-up ACO-16 60-70th Preventive Health
Mammography Screening ACO-20 60-70th Preventive Health
Patients who meet all the following measures: Low Density Lipoprotein (LDL) (<100 mg/dL), Aspirin Use, Tobacco Non Use, Hemoglobin A1c Control (HbA1c) (<8 percent), Blood Pressure (BP) < 140/90 ACO-23 ACO-26 ACO-25 ACO-22 ACO-24 60-70th At-Risk Population Diabetes
Percent of beneficiaries with IVD with complete lipid profile and LDL control < 100mg/dl ACO-29 60-70th At-Risk Population IVD
Influenza Immunization ACO-14 50-60th Preventive Health
Colorectal Cancer Screening ACO-19 50-60th Preventive Health
Patients who meet all the following measures: ACE Inhibitor or ARB Therapy for Patients with CAD and Diabetes and/or LVSD, Drug Therapy for Lowering LDL Cholesterol ACO-33 ACO-32 50-60th At-Risk Population CAD
ASC Admissions: COPD or Asthma in Older Adults ACO-9 40-50th Care Coordination/Patient Safety
Pneumococcal Vaccination ACO-15 40-50th Preventive Health