The cost and human suffering related to chronic diseases in the United States is staggering.
Hypertension (high blood pressure) costs the United States about $131 billion each year.1 It is directly linked to heart disease and stroke—the two biggest killers each year combined.2 Similarly, more than 34 million people in the United States have diabetes and, of those, 90% to 95% have a type 2 diagnosis,3 which the Centers for Disease Control and Prevention (CDC) notes can be prevented or delayed through lifestyle changes.4 It costs approximately $327 billion yearly to cover medical expenses and lost work and wages for this population.5
Considering that the CDC reports, “90% of the nation’s $3.8 trillion in annual health care expenditures can be attributed to those living with chronic disease,” it’s clear that to lower costs and improve health, we must focus on preventing and managing chronic disease.6
Using data visualization to manage chronic disease
Using data visualization is one of the simplest ways we can begin to improve the care and outcomes for populations living with chronic diseases. To understand this better, let’s look at hurricane forecasting as an analogy for preventing, or at least managing, chronic disease care.
Figure 1: Hurricane Dorian projection, August 29, 20197
Hurricane forecasts begin with collecting data. Satellites, radar, and other instruments are used to capture atmospheric conditions. As the hurricane develops, the information is then used to inform the public of potential risks, creating models to track the speed, intensity, and path the storm may take. Every day, every data point gives meteorologists the ability to tighten up their predictions right up to the event. By the time the storm makes landfall, the forecast has given residents enough warning to secure their homes and head toward safer ground.
The regular input and tracking of vital signs play a similar role to our weather analogy above. By taking daily blood pressure or blood sugar readings, individuals and their care providers can begin to visualize data, identifying trends and patterns. Even if a health “storm” cannot be entirely averted, the data may provide forewarning to mitigate the landfall of a stroke, heart attack, or the onset of type 2 diabetes.
To further illustrate the point, the chart in Figure 2 displays an individual’s blood pressure readings over the course of a year (as tracked through the Milliman HealthIO app). An individual and their care provider might make several observations from this data visualization, including:
- A steady increase in blood pressure over time, as indicated by the blue line
- A recent downward trend in blood pressure, as indicated by the red line
- An overall cyclical nature to the individual’s blood pressure readings, as viewed by the dot placements of each reading
Armed with this data, the individual and care provider can work together to craft an appropriate intervention, heading off a potential negative health event.
Figure 2: Measuring blood pressure over time
Source: HealthIO system-generated report for a de-identified individual in a client’s population.
Putting it into practice
Milliman HealthIO’s work with a community health organization shows the power and promise of data visualization to manage at-risk populations.
Individuals were given “smart” health devices to track their basic vital signs. A scale, blood pressure cuff, glucometer, and pulse oximeter synced both daily and weekly data points for each person in the program. Over time, the reports provided to the clinic identified opportunities for intervention among the individuals who showed elevated, out-of-range readings. The data visualization allowed the clinic’s medical staff to make informed decisions more quickly. The clinic followed up with its at-risk population, getting ahead of potential health crises and their associated costs.
In one example, a member of the community health organization began to monitor his blood pressure each day. He then added blood glucose measurements to the daily monitoring schedule. Both the member and his care manager noticed a trend of elevated blood pressures and blood glucose.
The member’s physician was notified and reviewed the vitals data, leading to a medication change for antihypertensives and adjustments to the member’s insulin and oral hypoglycemic treatment.
In another example, a member used the HealthIO devices and app to complete daily pulse oximeter measurements. The member’s data revealed a trend: in the early morning, she continued to have pulse oximeter levels less than 90%. Her physician was notified, further testing was completed, and the member began a treatment protocol with continuous oxygen during sleep.
In each case, longitudinal trend data enabled both members and the care team to catch problems early on and intervene before a more serious health event occurred.
Data visualization has a long history of helping decision-makers take action across industries. For payers and health systems, these powerful, visual dashboards are helping drive action for vulnerable populations. For employers that have gotten a taste of population health management through the COVID-19 crisis, the opportunities to empower better health and manage financial risk may be a game changer. Partnering with a digital health provider can be a solid first step in using data visualization to predict, prevent, and manage chronic disease.
1 CDC. Facts About Hypertension. Retrieved November 29, 2021, from https://www.cdc.gov/bloodpressure/facts.htm.
2 CDC. Health and Economic Costs of Chronic Diseases. Retrieved November 29, 2021, from https://www.cdc.gov/chronicdisease/about/costs/index.htm.
3 CDC. Diabetes Fast Facts. Retrieved November 29, 2021, from https://www.cdc.gov/diabetes/basics/quick-facts.html.
4 CDC. Prevent Type 2 Diabetes. Retrieved November 29, 2021, from https://www.cdc.gov/diabetes/prevent-type-2/index.html.
5 CDC, Diabetes Fast Facts, op cit.
6 CDC, Health and Economic Costs of Chronic Diseases, op cit.
7 National Center for Atmospheric Research (August 29, 2019). Hurricane Dorian (AL05). University Corporation for Atmospheric Research. Retrieved November 29, 2021, from http://hurricanes.ral.ucar.edu/realtime/plots/northatlantic/2019/al052019/track_early/aal05_2019082918_track_early.png.