One of the pain points we often discuss with potential customers is how hard it can be to track the success of their benefit programs. Rolling out a new program or improved messaging is easy – but how do you know if it’s working?
We’ve got a few ideas that can help you, and today we’ll take a closer look at the Cohorts App. This tool will help you look at lists of member IDs (de-identified members) with something in common: i.e. they had orthopedic surgery, they’re participating in your wellness program, or they visited the ER. You can use Cohorts to track these members over time and compare them to other groups in your organization.
Here’s an example: let’s say you wanted to find out more about your members who were being treated for depression. You’d start by creating a cohort of members who were taking antidepressants. Within the Cohorts App, you can then use filters to see:
Tip: Artemis experts suggest saving cohorts instead of just applying filters to get this info. It’s more accurate to save members this way, plus you can track them over time.
You get the idea – it’s a powerful tool for looking at a member population. And it’s actionable, too. For example, if you’re seeing that members from the Philadelphia office being treated with antidepressants aren’t utilizing the employee assistance program, but those from the San Diego office are, you know what to do next. Create some messaging, roll it out, and track whether member behavior changes as a result.
There are two main ways to create a cohort using the Artemis Platform. You can do it within the Cohorts App by clicking “New Cohort”:
Or if you’re kicking around in the Explore App and find something you’d like to dive into, you can click “Save Members as a New Cohort”:
For this customer, the Artemis team helped build two cohorts:
These two cohorts allowed the customer to see those using opioids for post-surgery vs. those on the drugs for longer periods.
Once these members were grouped in Artemis, this customer created a “Story” showing the correlation between opioid use and disability claims. Somewhat surprisingly, the data showed that 302 employees on disability were outside the opioid use cohort, but just 202 of those on disability were inside the cohort.
If the data analysis had shown a strong correlation between opioid use and disability claims, the customer could have created communications to target the cohort with customized messaging. Because it didn’t, the customer is looking at other ways to track disability claims and find meaningful correlations in their member population.