Small data creates backbone for medical homes

November 1, 2013

Medical home providers gain tailored prescription data

Community Care of North Carolina (CCNC), a medical home public-private partnership, recently teamed up with drug maker GlaxoSmithKline on a program intended to help providers and payers analyze medication data in real time.

 

The pilot aims to direct resources to where they will have the most impact. Counter to the common trend toward “big data,” the program uses a “small data” approach, allowing providers access to targeted information.

“What we found with small data is there are still a lot of barriers to being able to apply data,” says John Easter, senior director in the public policy department at GSK. “Providers may not have right data set or may be using legacy systems that can’t integrate with each other. What this group was able to do is create these models in a lightweight way so they only needed a small amount of data to predict the correct outcomes.”

A modelling tool combines different types of medical and business data and feeds it through a logistics engine that triggers results based on criteria. A data visualization component makes the results easier for providers to understand.

CCNC’s vice president of pharmacy programs, Troy Trygstad, says the “small data” is different than “big data” because it focuses on macro-level trends. Small data refers to specific information, such as a patient’s prescription fill history and hospital admission and discharge data. While this information is widely available in claims data and EHRs, it’s not often put to direct use.

“Traditionally, I’d run a patient’s information through a rule set,” Trygstad says. “I might see that ‘Yes, there’s a care gap or a therapeutic duplication.’ What we need is ‘Yes, there’s therapeutic duplication, and this particular caregiver can resolve it in this particular way.’ Rather than just saying there’s a problem, it says this is a problem, and this is who can solve it and when.”

Prescriptive modeling

The service uses what GSK calls “prescriptive modeling” to analyze the data and determine suggested interventions to appropriate healthcare providers. The goal of the project is to assist healthcare professionals in determining-and responding to-medication-related problems that often lead patients to use their medicine incorrectly, or to not adhere to the treatment regimen prescribed by their doctors. 

CCNC has begun using the system across its existing network. CCNC and GSK are also testing the approach as part of a Community Medication Management Collaborative within the hospital, primary care, home health and behavioral health settings in Bloomington, Ind.

“We really are in learning mode right now,” says Easter. “We’ve been seeing all the dynamic changes in the healthcare landscape and watching the emergence of ACOs, new payment methodologies and new coordinated care models. This is an opportunity for GSK to take a proactive step and learn how these new models are involved in care, and how they can improve medication management so the right person is getting the right medication at the right time.”

There are no immediate plans to commercialize the system, however CCNC and GSK do plan to expand it to other areas in 2014.

“The biggest benefit for us is efficiency,” Trygstad says. “We need to maximize who, when and how we’re touching patients, and how providers are coordinating [care]. There are precious few dollars devoted to medical management activities. We want to use them wisely.”

CCNC is a not-or-profit organization representing 14 community care networks that support the community in creating medical homes. As an emerging care model, medical homes are gaining tractional nationally.

The National Committee for Quality Assurance (NCQA) in September recognized its 6,000th medical practice earning medical home certification. There are now 29,505 recognized clinicians in 49 states, according to NCQA.