A person’s life outside the healthcare system significantly impacts their ability to understand, prioritize, and act on clinical recommendations. By better understanding and then innovating to address SDOH, we will improve population health in more substantial and sustained ways.
COVID-19 Brings SDOH to the Forefront
In a multi-year study published by Health Affairs in February 2020, only 9.1% of hospitals had a plan to address SDOH. In the aftermath of COVID-19’s widespread, disparate impact on specific communities, the healthcare industry has an imperative for a more comprehensive, systematic approach to address this issue. While COVID-19 does not discriminate, greater vulnerability to exposure is rooted in the societal structures that create and reinforce racial/ethnic health disparities.
With Black and Hispanic populations disproportionately affected by COVID-19, new light has been shed on health disparities in the U.S. A recent American Heart Association study found that these populations made up nearly 60% of COVID-19 hospitalizations. An Economic Policy Institute Report outlined the historic and ongoing social and economic injustices that drive this disparate racial impact, including:
- The prevalence of racial/ethnic minorities who are frontline or essential workers
- Wage or benefits gaps that preclude the option of sick days or paid time off
- More single-earner or single-parent homes that bear the burden of childcare and distance learning
- More densely populated or multigenerational housing living environments
- Reduced access to healthcare services
As data collected has increased in abundance and complexity, making sense of it is the next challenge to effecting change. With the right approach, health systems can address these gaps and meet the healthcare needs of high-risk populations, transforming one-size-fits-all approaches into hyper-personalized behavioral interventions.
Make SDOH Data Actionable
To better address patients’ SDOH, health systems need both SDOH data and the capabilities to effectively leverage this data to benefit their patients. Collecting insights that enable targeted action for a specific population requires data scientists to make sense of the data, behavioral designers to develop behavioral interventions that address SDOH, and an AI-driven platform that can match the right communication intervention to the right patient, at the right time.
Research suggests there is a unique opportunity to leverage existing data sets to make SDOH highly actionable in prioritizing care strategies. According to a study recently published by JAMA, SDOH data was used to improve precision in algorithms identifying high-risk Veterans Affairs (VA) patients for targeted intervention resources. Researchers followed 4,685 VA patients with 1-year risk at or above the 75th percentile for hospitalization or death and found integration of certain patient-reported SDOH measures improved estimates of 90-day and 180-day hospitalization risk. According to the findings, “incorporating these types of factors into the EHR could also assist with population management and health system decisions, for example by highlighting the need for partnerships with certain community agencies.”
If your health system integrates SDOH data with your EHR, the next step is to make that data actionable. Lirio’s AI-driven platform can help you do this. It serves as the intelligence layer between systems of record and systems of engagement, meaning it can translate data from your EHR and apply it to systems like your patient portal and communication tools that support your population health goals. This ensures you get real value from your SDOH data and it doesn’t stay static in your EHR.
Engage Hard-to-Reach Populations
As healthcare becomes increasingly digital, there are many opportunities to create equity and expand access. This is reflected in the significant increase in Medicare claims for telehealth during the pandemic, offering patients with conditions like diabetes to seek safe outlets for care, or new moms without childcare support to maintain critical postpartum care. The ability to engage with all segments of the population is key to ensuring no one is left behind in this digital revolution.
Understanding how people communicate, receive information, and make decisions is best done in an ongoing framework that compensates for changes in life stage, environment, and health status. Behavioral science can provide strategies to overcome the barriers and biases patients encounter in their health journeys, which can be paired with AI to tailor messaging to an individual’s responses over time.
In one client campaign, Lirio’s combination of behavioral science and AI through Precision Nudging demonstrated a significant lift in engagement among high-risk, high-cost patients. For example, Lirio’s approach was 19% more effective than a one-size-fits-all approach for patients with commercial insurance and 48% more effective for patients with public insurance. Lirio’s Precision Nudging approach was also more effective for Asian Americans (92% greater), White/Caucasians (23%), Black/African Americans (19%), and Other, Mixed-Race (45%) patients.
Behavior Change AI Powers Population Health
With the Lirio platform integrating your system through Precision Nudging, you’ll be able to harness the power of SDOH data to systematically match unique interventions to each patient within your population.
Here’s what this looks like in the context of a Lirio Behavior Change Journey, which aims to improve interaction and engagement with primary care physicians among people with diabetes.
1. Behavioral science informs the problem.
We encode behavioral science expertise to identify the problem that needs to be solved, such as attending a routine chronic care exam with a patient’s primary care physician.
2. Behavioral interventions are matched with behavioral profiles.
We develop interventions comprised of behavioral nudges, or techniques that address the problem and use behavioral feedback to match them with individuals’ data that includes, but is not limited to, health and lifestyle factors (or SDOH) for patients. For example, emphasizing information about health consequences to address perceived cost-benefits.
3. AI enables continuous learning.
Our platform delivers interventions and our machine learning agent continuously optimizes communication for each patient. It identifies the right behavioral intervention to engage and activate each individual, based on behavioral feedback from the individual and others with similar and dissimilar profiles. This approach augments and accelerates the personalization of interventions for each, unique patient with diabetes.
4. Behavioral profiles are analyzed over time.
Lirio behavioral scientists analyze behavioral profiles to find insights that help refine solutions. The profiles of patients with diabetes continuously evolve to reflect changes in the environment, allowing interventions to address a patient’s current barriers to scheduling and attending routine appointments.
5. The platform scales behavior change.
Through continuous learning, health systems can elicit behavior change from more patients across populations. This means Lirio helps initiate and sustain change that improves the long-term health outcomes of entire populations.
Behavior change AI makes it possible to employ a population health strategy that is informed by SDOH data, rooted in a holistic approach that meets each patient where they are, and moves individuals toward better health.
Other readers viewed:
Want to learn more about how Lirio’s behavioral engagement solution utilizes behavioral science and machine learning to help organizations motivate the people they serve to achieve better outcomes?