5 examples of using social determinants to improve population health
Adding social and demographic data to claims and clinical data sets can help healthcare providers, payers and policymakers gain insights to improve public health.
Optimizing population health requires understanding the lives of patients outside of their medical record. The social determinants of health (SDoH) – such as access to health care, health equity, employment, neighborhood safety, education level and housing – can also affect health outcomes and well-being.1
Bringing together multiple data sources can reveal insights that healthcare providers, payers and policymakers can use to improve population health. Adding hyper-localized SDoH data to existing clinical and claims data sets can help organizations understand more about the people they serve.
Here are five examples of how some of our clients have used SDoH data to better understand their populations and close gaps in care:
1. Address COVID-19 vaccine hesitancy.
To help identify at-risk communities, public health agencies can use tools such as the CDC Social Vulnerability Index.2 Advanced analytics can offer more precise detail about people’s behaviors, preferences and interests on a more granular level than zip code.
For example, these targeted analytics can help government agencies roll out the COVID-19 vaccine and overcome different reasons for vaccine hesitancy. The IBM Watson Health PULSE® Survey found that rural and lower income areas are less likely to get vaccines, even if those populations received the flu vaccine.3 Adding hyper-localized segmentation data to these findings can help further determine preferences that can help organizations shape effective outreach activities, for example, which populations prefer contact by mail, text or phone.
2. Launch a new telemedicine program.
Telemedicine is proving to be an affordable, convenient option for organizations to help improve access to care.4 Using SDoH, one of our clients mapped its population into three groups with different communication preferences to more effectively reach them to describe a new telemedicine program. Hyper-localized data and a Motivational Index methodology also helped the organization define the best geographic area – the one with people most likely to use telemedicine services – to launch the new program.
3. Examine high emergency room (ER) utilization.
One study estimated that the average cost of treating non-urgent conditions at an ER is 12 times higher than at a physician office, and that two-thirds of hospital ED visits are avoidable.5 By applying SDoH data, a health plan was able to better understand the subpopulations that were higher ER users and understand what was driving the ER usage, such as households that look for convenience in health care.
4. Increase enrollment in an Accountable Care Organization (ACO).
ACOs are designed to deliver more coordinated, high-quality care to Medicare patients, which should also help reduce healthcare costs.6 One client used SDoH data to better understand its ACO population in a specific Metropolitan Statistical Area (MSA). With the goal of increasing ACO enrollment for the following year, the organization identified people who were more likely to become members. The client then identified four sub-groups with different behavior patterns and communications preferences and developed a tailored outreach approach for each one.
5. Support mental health.
Before the pandemic, the world was facing another public health crisis, with as many as one in ten people suffering from a mental or substance use disorder.7 Technology and analytics can help organizations identify why people are not receiving mental health assistance – is it because they don’t have access to services? Do they feel there is a stigma attached to seeking help? Or is affordability the issue? Pinpointing the reasons can help increase utilization and provide much-needed intervention.
A more precise understanding of the social determinants of health can help inform decisions for clinicians at the point of care, in health insurance and benefit plan design, or for policymakers developing public health strategies. Using data to have a better understanding of a population’s attitudes, lifestyles and behaviors around healthcare can help them address population health challenges and improve overall health outcomes and well-being.