Understanding HRA screening and the use of portable devices
Most employee wellness programs include a combination of health risk assessment (HRA) and relevant interventions. An ERS identifies risks to health and well-being at the individual level in order to educate employees about their condition and direct them to the appropriate resources. At the enterprise level, wellness screens are used to understand and track overall priorities and the burden of health risks for planning, program evaluation and risk management purposes. Another approach to wellness has been the use of wearable devices (mobile apps, smart watches, and fitness bracelets).
Numerous publications highlight key lessons learned from the long and extensive experience of employer wellness programs in developed markets. A common conclusion was that a well-being assessment should be carefully selected to appropriately match the program objective and the target population. As most HRAs are generally accepted as being evidence-based in their scoring, we conducted a sensitivity analysis on HRA data items that parallel the types of data that might be obtained through portable devices (for example, monitoring exercise and activity levels). We varied the different input metrics in three independent HRAs to see which metrics had the best predictive value for the overall outcome. We then studied the impact of the activities measured by the HRAs on the overall health scores reported by the HRAs. This has helped us understand the influence that activities measured by wearable devices can have on overall health levels, depending on how important HRAs consider these activities to be.
HEALTH RISK ASSESSMENTS
The United States Centers for Disease Control and Prevention (CDC) has defined an HRA as:
A systematic approach to collecting information from individuals that identifies risk factors, provides individualized feedback, and links the person to at least one intervention to promote health, maintain function, and / or prevent disease.
HRAs typically incorporate three key elements, as shown in Figure 1.
FIGURE 1: TYPICAL HEALTH RISK ASSESSMENT PROCESS
We tested the consistency between three HRAs and the relative weights they place on metrics in terms of predicting overall health levels. We compared the questions and results of Vitality’s publicly available HRAs, the UK National Health Service (NHS) cardiac risk assessment and the Rapid HRA tool developed by Milliman.
Each HRA also has a different type of output. The Vitality HRA produces an “age of vitality” which should be interpreted according to the age of the limb. A vitality age greater than the actual age of the limb indicates that the state of health of a limb is worse than expected based on the demographic profile, and vice versa. The NHS HRA produces a similar result, but with an emphasis on “the age of the heart”. The Rapid HRA produces a score out of 100 where a score of 100 indicates that the limb is in perfect health.
Figure 2 shows that there is substantial overlap in the question types for Vitality and Rapid HRA, while the NHS HRA has a narrower focus.
FIGURE 2: THE MAIN QUESTION GROUPS WITHIN EACH TYPE OF RHS
The full CDC-funded Thomson Reuters report providing this definition is available at https://www.cdc.gov/policy/hst/hra/HRAAWVGuidanceReportFINAL.pdf
We defined a base case for a 40-year-old male member with the health and demographic characteristics depicted in Figure 3 as representing the âaverageâ scenario. We also defined âvery healthyâ and âvery unhealthyâ scenarios for the 40-year-old member by varying the baseline scenario inputs to test the impact on HRA outputs. We used the baseline scenario to test the sensitivity of each HRA’s output metric to the input metrics that portable devices are able to capture. This is discussed in more detail in the following sections.
FIGURE 3: AN EXAMPLE OF THE BASE CASE MEMBER AND WHAT IS CLASSIFIED AS HEALTHY OR UNHEALTHY IN TERMS OF HEALTH OR AGE SCORE
HEALTH AND PORTABLE RISK ASSESSMENTS
Today’s portable devices measure a limited amount of information, and insurers can only use the data made available by each device. For this research, we tested the sensitivity of HRAs to metrics measured by portable devices, and then compared it to other component metrics of an HRA. This gives us additional insight into how HRAs and wearable devices interact with each other, as well as the value HRAs place on certain lifestyle behaviors.
Regarding exercise, our research has shown that, for Vitality Age, there is no marginal benefit from exercising more than four hours per week. However, exercising less than four hours per week has a negative impact on the limb’s vital age, as shown in Figure 4.
Figure 4: The impact of activity levels on the age of vitality (Vitality HRA)
For Rapid HRA, on the other hand, we saw that increasing activity levels compared to the base case can significantly improve the overall health score. Additionally, reducing activity levels has a negative effect on health score, as shown in Figure 5.
FIGURE 5: THE IMPACT OF ACTIVITY LEVELS ON THE HEALTH SCORE (RAPID HRA)
Another area we assessed was the impact of a changing body mass index (BMI) relative to overall health score or health age.
We found that for the Vitality HRA and Rapid HRA, any BMI value that strayed further from the “healthy” range had a significant impact on the overall health score or vitality age, as shown. Figures 6 and 8. The NHS HRA is less sensitive to changes in BMI, with a gradual change only occurring at a BMI of 30, as shown in Figure 7.
Figure 6: The impact of BMI on vitality age (Vitality HRA)
Figure 7: The impact of BMI on cardiac age (NHS HRA)
Figure 8: The impact of BMI on health score (RAPID HRA)
Finally, we looked at the sensitivity of HRAs to some key health metrics not measured by wearable devices. We looked at the effect on age or health score after varying each metric into âgoodâ and âbadâ values ââfrom the base case shown in Figure 3 above.
While we found that not exercising negatively impacted health score, it was not always the most important health factor. As shown in Figure 9, Figure 10 and Figure 11, parameters such as cholesterol, smoking and BMI can all have greater impacts on a limb’s vital age, NHS HRA cardiac age or the Rapid NRA health score, respectively.
For the Vitality HRA, it is only possible to improve the vitality age by improving cholesterol and eating habits from our defined base case (Figure 9). For the Rapid HRA, it is possible to improve the overall health score by improving various factors (Figure 11). For the NHS HRA, we found that it was not possible to improve cardiac age by improving certain factors, but worsening of certain factors could lead to worsening cardiac age (Figure 10).
Figure 9: Results of the sensitivity tests of the existing question groups on the age of vitality
Figure 10: Sensitivity test results of existing NHS cardiac age question groups
Figure 11: RESULTS OF SENSITIVITY TESTS OF EXISTING QUESTION GROUPS ON THE HRA Health RAPID SCORE
Our review of the factors that influence health risk assessments (HRAs) indicates that while wearable devices can encourage members to increase their activity levels, the implementation should be seen as part of an offering. complete well-being, as many of the most important factors that influence HRA scores, and (by implication) health status, are not captured by wearables.
A well-being assessment should be carefully selected to appropriately match the program objective and the target population. The evaluation campaign should have a clear objective and purpose. To achieve this goal, employers and wellness providers must carefully select the assessment approach that resonates with these best practices. Employer-centric analysis of employee risk profiles allows employers to tailor programs to their specific goals and employee risk profiles. To use HRAs to encourage respondent behavior change, the results must be responsive to behavior changes to help monitor the progress of employee and employer wellness interventions.
A version of this article was originally published by Milliiman in March 2020 and is available at https://www.milliman.com/en/insight/the-role-of-wearables-in-private-medical-insurance.
For any questions or discussions regarding HRA and the contents of this white paper, please send an email to [email protected]
Disclaimer: This article presents information of a general nature and is intended for educational purposes only. It is not intended as advice for a specific situation and cannot be relied on for any purpose. You should always consult with qualified professionals who are familiar with your situation before adopting any strategy or taking any action. Milliman does not guarantee the veracity, reliability or completeness of this white paper and has no responsibility for any damages which may have been caused by it.