Here's How Healthcare Analytics Can Help Employers Understand Population Health

Population health has been a big buzzword in the news in 2020, in part thanks (or no thanks) to the COVID-19 pandemic. Everyone is drawing on public and private healthcare data, whether it's the percentage of positive COVID-19 tests from public testing centers or the number of high-risk individuals in an employee population. There are so many ways to measure population health, so many metrics to track, and so many uncontrolled factors that contribute to it. 

Employee benefits leaders find themselves at the forefront of tracking, measuring, and improving population health because approximately 50% of Americans access healthcare through their employers. Benefits teams are very focused on population health, as it helps them achieve a number of goals: 

• Accurately forecasting the cost of employee benefits 
• Finding and addressing gaps in care 
• Providing better benefits programs to employees 
• Improving employee health and well-being 
  

It’s not always easy to track and measure population health, but there are solutions for benefits leaders. Healthcare analytics solutions are key to finding insights and taking action on population health. 

What is population health? 

Population health is the health outcomes and indicators of a group of people, as opposed to a single individual. For the purposes of employee benefits, population health usually refers to the overall health and well-being of everyone enrolled in an employer’s benefits plan. That means employees, retirees, and dependent family members. While each individual on the plan is going to have a distinct risk score, unique health conditions, and differing benefits needs, employers will often measure population health by including every individual in their measurement strategy. This ensures both confidentiality for each member on the plan and a more holistic view of how healthy these members are. 

Population health measurement is a useful tool for benefits professionals who need to understand larger trends and risks. Benefits analysts will also break the population into smaller “cohorts” to understand the health of a smaller group of employees compared to another group. For example, an analysis might look at the costs and risk scores of a group who has been diagnosed with diabetes compared to those who do not have that disease. Or a group participating in a fitness challenge compared to those who didn’t sign up. 

By measuring population health, benefits leaders and their advisors can forecast costs for next year, check if interventions like care management programs are working, or find out if members are getting access to the right care in the right place at the right time. 

Population health dashboards

Employee benefit leaders with access to an innovative data warehouse will likely be able to view population health in an easy-to-read dashboard. Let’s look at Artemis Health’s population health Standard Story, an out-of-the-box report that’s available to every client. 

This population health dashboard starts by looking at risk scores across the population. You can see in this sample data set that these are mostly holding steady, but are slightly up from the prior analysis period (last year). Artemis Health recently made the switch from Johns Hopkins ACG risk methodology to Milliman Advanced Risk Adjusters (MARA). MARA allows Artemis to give clients a look at both concurrent risk (what’s happened over the last 12 months) and prospective risk (what is likely to happen in the next 12 months).

Next, the dashboard provides a look at risk bands across the population to give us an idea of the number of members in each category. 

Our sample data shows that just 188 members are in the “very high” risk band, but that their risk scores currently sit at 32. However, our prospective risk (what’s likely to happen next year) is 23. This methodology suggests that these very high risk members are likely to see their risk score go down next year, which is good news. The chart also shows that the largest group in our population, with 16,000 people, is in the very low risk category.

This dashboard also displays a bubble chart showing the per member per month costs for members by their risk band. Each member in the “very high” category incurs costs of approximately $17,000 per month, while those in the “very low” risk band cost the plan about $32 per month. You can see why a benefits manager would find this information valuable; they are often tasked with cost management and need to see opportunities for reducing risk scores and lowering benefits costs. 

Additionally, we see a grid chart called the “health continuum” that gives us a look at both prior and current risk statuses for members so we can see which way the population health is trending. 

Finally, this population health dashboard dives into one area that is often a clear indicator of population health: chronic condition prevalence. We calculate: 

• How many members have been diagnosed with a chronic condition
• The percentage of members in the population with a chronic condition 
• Costs for chronic vs. non-chronic conditions 
• The most costly chronic conditions diagnosed on our plan 
• The most costly non-chronic conditions on our plan 
  

It’s fascinating to see how these chronic conditions compare to other conditions being treated in the population. For example, musculoskeletal conditions that are considered chronic are costing this plan approximately $4 million more than temporary musculoskeletal. This gives the benefits team an actionable insight: they need to keep musculoskeletal injuries from becoming chronic conditions in order to contain costs. They might do this with better access to physical therapy, ergonomic work stations, fitness/exercise wellness programs, and opioid formulary limits. 

This population health dashboard is just one way employee benefits experts are keeping an eye on the risk, costs, and opportunities in their healthcare and benefits data. 

Want to learn more about healthcare analytics? Watch our 2-minute demo video for a sneak peak.