Population Health Management Requires a Holistic View of Patient Data

By Jerry Shultz, President, Lightbeam Health Solutions
Twitter: @LightbeamHealth

To succeed under value-based reimbursement, accountable care organizations (ACOs) and clinically integrated networks (CINs) must be able to manage population health. One key to effective population health management is a holistic view of patient data.

The data in a single physician practice’s EHR is insufficient to create the comprehensive view needed to help clinicians reach quality targets and cut costs significantly. Instead, the ACO or CIN must build an infrastructure that can automate the aggregation and normalization of clinical data from its members’ disparate EHRs and other sources, as well as claims data from its payers. It must be able to analyze this data in an electronic data warehouse (EDW) and use a health information exchange (HIE) to share the key data with its members.

This important data must be actionable to move the needle on health spending. Providers must be able to use it to impact the outcomes of and costs of caring for high-risk patients, and to prevent low- and medium-risk patients from becoming sick or sicker. That requires injecting the insights derived from the holistic view into the EHR workflow of clinicians. Physicians will not leave their workflows to go a website and log in to see outside information on their patients.

From a cost and utilization standpoint, a comprehensive view of patient care can help physicians avoid ordering redundant tests and procedures. It can also enable care teams to intervene with patients to prevent hospital admissions and emergency department visits.

Here are some of the use cases for the holistic patient view:

Risk stratification and care planning
For starters, ACOs and CINs must be able to stratify their populations by health risk and identify who needs what kind of care. If a patient has a BMI over 30 and four or more chronic conditions, chances are good they need care management to minimize adverse events. If a patient has been in the hospital for a hernia operation, but has no chronic conditions, they’d be managed in a different way.

The combination of claims data, which is broad but not very timely, and clinical data, which is narrower in scope but rich and timely, can be very powerful in determining health risks. For instance, if somebody has behavioral health issues, has been taking pain medications, and has been seen in a dozen different EDs in the past couple of months, there’s a good chance that person is an opioid addict. Or just as important, social determinants of health, captured either from the EHR or patient-generated health data, may be used to further refine the risk stratification engine for personalized care plans. We often hear doctors who have managed patients for many years are surprised to learn from claims data that their patients have chronic conditions of which they were unaware.

Actionable data
After the holistic patient view has been constructed in an EDW, the clinically important data is selected and pushed out to the EHRs of member practices and hospitals. An HIE with two-way EHR interfaces can be used both to collect data and transmit information that can be used in patient care, including care gap summaries, medication updates, and admission-discharge-transfer (ADT) alerts. Patient and care manager responses can be inserted into scheduling applications, and appointments can be booked for needed care. Providers should also be able to click a button to display lists of their patients who have not received particular services.

Patient outreach
Holistic patient data can also be fed into automated patient outreach applications to increase compliance with care plans. These outreach solutions are typically based on registries that show which services patients have received and when they’re due for services. Based on nationally recognized protocols that local clinicians usually tweak, the software triggers automated messaging that urges patients to make an appointment with their providers.

In recent years, patient outreach software has become more sophisticated. For example, it can be set to send messaging to patients by text, email or interactive voice response, depending on their preferences. In addition, ACOs/CINs can use simple questionnaires to create psychographic profiles of patients that are used to fine-tune messaging and increase its effectiveness. When available, social determinants of health may also be used to tailor the message as a high impact intervention for significantly increased patient response.

Some organizations have pared lists of patients who need recommended care to focus on the most resistant people. One healthcare system did that for women who were overdue for mammograms. A mobile mammography unit was sent to their homes to ensure they got tested, and it found one woman with stage 2 breast cancer.

Closed-loop referral management
An HIE that connects primary care physicians to specialists in the network and perhaps outside of it can be used for closed-loop referral management. When the PCP chooses a specialist, the referral management software can generate a CCD clinical summary based on the holistic patient data, embed it in the EHR, and transmit it to the specialist through the HIE. The consultant’s report, including medication changes, flows back into the EHR through the same route. If the patient doesn’t see the specialist, the PCP is informed. A similar strategy is used to tell the doctor whether the patient filled a prescription or got a lab test that he’d ordered. This improved communication allows the ACO/CIN to establish optimal care networks and collaborate with payers to further move the needle on reducing out-of-network costs while ensuring highly rated providers are readily available.

In conclusion
A holistic view of patients is essential to population health management. Without knowing whether patients received particular services outside of a practice or a network, it is difficult for clinicians to make optimal medical decisions or steward healthcare resources wisely. Moreover, having a comprehensive view of patients’ tests and treatments, as well as their psychosocial factors, can help care teams identify and intervene with patients who are at risk of getting sick or being hospitalized.

An advanced health IT infrastructure is necessary to create holistic patient views at the scale required by an ACO or a CIN. But once the organization has put this infrastructure in place, it will be able to succeed under a wide variety of shared savings and risk contracts.