In the transition to value-based care, many providers have made a significant investment in population health to better support their patients. Nonetheless, the ability to achieve consistent patient outcomes varies significantly from one organization to the next.
Based on Evolent’s experience working with nearly 100 provider-led organizations, a key indicator for an organization to generate a return on investment (ROI) in population health is their ability to identify, engage and intervene effectively with the most “impactable” patients. Evolent uses the term “impactability,” to describe patients “likely to incur a specific, future adverse event that can be prevented through an optimized, evidence-based intervention.” If provider organizations can successfully implement the three pillars of impactability (identify, engage and intervene) they can provide the right level of intervention to the right person at the right time.
This two-part series examines impactability, broken down into its three pillars, through the case of a 66-year old woman named Jane with diabetes and coronary artery disease and through a Midwest Accountable Care Organization. Their stories reflect the positive patient and provider outcomes that can be achieved by using a best practice model.
Part I: Jane’s Story – Identifying Impactable Patients
Jane lived alone in a low income, predominantly Spanish speaking area located in a food desert. She was on nine unique medications, including a prescription for anxiety that was never filled. In addition, she had a recent change in HbA1c, a strong indicator of diabetes. She also had no inpatient admissions in the last 12 months.
In current-state models Jane wasn’t identified as an at-risk patient because she had no recent hospital visits, her HbA1c level was still in an acceptable range, and her social and physical environment were not assessed. Current population health intervention models are solely designed to predict financial risk and tend to overlook key warning signs. In other words, they identify which patients are likely to cost health systems money based on a narrow data set.
A best practice model digs deeper and understands a person’s clinical, social and behavioral backgrounds, in addition to historical cost and utilization. In Jane’s case, the combination of these diverse determinants of health would have identified her as an impactable patient prior to admission.
Engaging Impactable Patients
Engaging a patient can be a difficult process for physicians and providers, but plays a vital role in the prevention of health issues down the line. Most provider organizations currently use generic phone calls and voice messages to encourage a patient to engage in care management. This does not take into account factors that affect communication preferences, such a native language.
In a best practice model, the goal is to customize the approach based on a person’s predicted engagement level. To do so, it’s critical to consider factors such as a patient’s living conditions, ideas about health, native language, preferred mode of communication and the individuals who care for them.
In Jane’s case, the likelihood that she would engage in care management is low for several reasons: English isn’t her native tongue, her health literacy is low, she lives alone and her son is her primary caregiver. A generic cold call in English to describe her care program would be unlikely to motivate her to take control of her health.
The best practice model looks at Jane holistically and tailors the outreach strategy to fit her needs. An in-person visit with her primary care physician and son, explaining the care program in Spanish and immediately connecting her to reliable food sources in her community is a better approach to boost Jane’s engagement. If care teams can personalize their outreach in this way, the chance of engaging Jane, and subsequently preventing an adverse health event is much higher.
Implementing Interventions for Impactable Patients
For providers to fully realize a positive ROI, the right selection and delivery of interventions must occur. In the current state of population health programs, most physicians manually select the best intervention and method of delivery for a patient. The subjective choice by providers often results in a poor match between the patient and the intervention and a high degree of variability in delivery.
Best practice organizations have dealt with this problem through an automated intervention selection process driven by clinical evidence and an intervention delivery process influenced by standardized workflows. This allows the care team to consistently deliver evidence-based interventions to patients. Consistent delivery also allows clinical leadership to assess the effectiveness of each intervention.
In order for Jane to receive the best intervention, it’s important to take her clinical and social determinants into account. A manual selection and delivery process won’t properly weigh these factors, resulting in a generic referral to a diabetes educator to receive educational materials about lowering her HbA1c levels.
Comparatively, a best practice model considers Jane’s missed anxiety prescription refill, limited access to healthy food and her HbA1c reading to automatically determine the best course of action. An integrated self-management program not only puts Jane in contact with a behavioral health specialist for her anxiety, but also a social worker to improve her access to food and a pharmacy specialist to address medication adherence and barriers to medication access. This intervention will therefore get to the root of Jane’s health issues that were leading her down the path of a hospital admission.
Olivia C. Tran, MPH, is the Director of Health Outcomes Research at Evolent Health. She can be reached at OTran@evolenthealth.com. For more detailed insights on best practices across the dimensions of impactability, download “Why Do Population Health Results Vary? An Introduction to “Impactability” here: http://go.evolenthealth.com/Why_Do_Population_Health_Results_Vary_-_An_Introduction_to_Impactability.
Anita Cattrell is the Senior Vice President of Research and Development at Evolent Health. She can be reached at ACattrell@evolenthealth.com