Analytics Enable Action on Social Determinants of Health Data

By Dr. Kevin Keck, Chief Medical Officer, SCIO Health Analytics
Twitter: @SCIOanalytics

Healthcare organizations are looking to tap into the potential value of social determinants of health (SDoH) data. Using this data, companies can identify previously unaddressed gaps in care, the most effective communication channels, and the best approach to delivering the care members need.

However, SDoH data on its own won’t generate the results healthcare companies are looking for – improved member health, reduced costs, and increased revenue. SDoH must be part of an organization’s larger data strategy and should be combined with the other data organizations analyze, such as pharmacy, medical claims, and demographics data. Organizations should look to analytics to enable the insights from this valuable data. Predictive and prescriptive analytics capabilities turn this information into insights for determining and prioritizing which members to engage with, what care to deliver, and how to best deliver that care.

Going beyond traditional data
Creating a 360-degree view of health plan members can be a difficult prospect. Organizations often find themselves hobbled by a lack of information. Health plans may have insight into the clinical side of their membership, such as conditions or medications, but lack a holistic view that includes social determinants of health, behavioral history, etc.,

For example, they may see that a diabetic patient regularly misses getting their insulin prescription filled. Closing this care gap would greatly benefit the patient through improving their health, as well as the health plan by cutting down on the chance that this patient will require costly, preventable care in the future. SDoH data can help determine what factors are causing this patient to not regularly fill their insulin prescription. With this powerful information, the health plan can determine the most meaningful message and channel to use that will be the most effective in engaging the patient/member and strategies for closing this care gap.

Intervenability and Impactability
The deep, patient-centric view SDoH data enables can help health plans determine how to best allocate resources based on which members are most impactable and intervenable.

Impactable members are those who have serious conditions that can be remedied through intervention. For instance, the aforementioned diabetic patient that regularly misses their insulin would be considered highly impactable – if they regularly take their insulin, their condition will improve.

Just because a member’s condition can improve with intervention doesn’t necessarily mean that the member will comply with suggested interventions. This is where the intervenability score comes into play. By measuring a member’s psychosocial situation, lifestyle, and other factors, health plans can determine how likely they are to change their behavior when presented with the right message and resources.

Putting SDoH data to work
Once a health plan has prioritized which members have the most important closable gaps in care and the motivation and resources to improve their health, they can use SDoH data in combination with other information to determine the best way to intervene with each individual member. For instance, some members may prefer receiving text messages rather than emails. Tailoring the communication channel to the member gives health organizations a better chance of making sure their messages stick.

Additionally, SDoH data lets health plans choose the most effective interventions by holistically considering their members rather than relying on clinical data. This can include looking at factors such as whether a member has access to transportation, if they have family members that live with them who can assist with providing care, and their level of education.

In the case of the diabetic patient who struggles to fill their insulin prescription on time, SDoH data could be used to determine that they don’t have any form of transportation to regularly travel to a pharmacy. Finding a different way for this member to fill their prescription could drastically improve their health, lower costs, and improve patient satisfaction without requiring the plan to expend significant resources.

Outcomes take analytics
For most health plans, getting access to SDoH data isn’t too much of a struggle. Many vendors now offer databases that can provide information on member lifestyles, habits, demographics, and motivations.

Having a lot of data isn’t the same as being able to use a lot of data, however. Without the talent, tools and training required to analyze SDoH information, plans won’t be able to produce the insights needed to lower healthcare costs and improve patient outcomes.

Building up an in-house predictive and prescriptive analytics program is a challenging prospect for healthcare organizations. It can be difficult to attract data scientists that can successfully combine new SDoH data with existing databases, build the modeling capabilities that can sort through all this information, and turn the results into easily understandable reports that power better decisions.

From insight to action
Health plans with the ability to gain insights from SDoH data will gain an edge in improving their revenue, patient satisfaction scores, and health outcomes. Whether this is accomplished by creating these capabilities in-house or choosing the right partner, health organizations must begin moving forward with developing SDoH programs.

This article was originally published on SCIO HealthAnalytics and is republished here with permission.