A large Midwest county that employed 1,400 full-time workers struggled with an internally managed health and wellness program. The county wanted to lower the health risks of its employee population and, ultimately, curb its growing medical plan costs.
Enter big data. The county plan sponsors had the opportunity (for the first time) to explore their medical and pharmacy claims using sophisticated data analytics. A review of preventive screening service utilization, in particular, revealed that many of the services they had been promoting with their internally managed health and wellness program (i.e., mammograms, colonoscopies, immunizations) were significantly underutilized compared to benchmarked expectations. The municipality agreed that, in spite of their best efforts, they needed to be more aggressive in how they promoted and incentivized the utilization of preventive screening services.
As a result of the health data analysis, the municipality was able to justify the cost of an onsite solution where a regional health care provider administered these services using mobile units, and communicated results with treating providers. The county was able to incorporate these services in their data analytics and track reduction in health risks and costs going forward.
Precision Risk Stratification Enables Precision Prevention
Big data capabilities for medical and pharmacy claims data often include powerful algorithms that enable sophisticated risk stratification methodologies. An employer can better understand the risk stratification of its population as it compares to benchmark group populations.
In other words, plan sponsors can better understand whether they have greater proportions of “high health risk” or “medium health risk” in their covered group. This is important as greater proportions of higher-risk individuals suggest higher urgency around strategies that lower those risks. Preventive screening, in particular, is one such strategy.
With big data health analytics, employer plan sponsors can specifically understand gaps in care, such as:
- What percentage of members over 50 years of age have not had colonoscopies
- How many women over 20 years old have not had PAP smears in the last two years
- Proportion of women over 50 who have not had mammograms in the last year
Furthermore, best practices for these types of findings include normative data for group health populations. This helps employers understand whether they have the usual issues that most employers face, or if they have greater rates of members not seeking preventive care.
Employers who are able to pair health risk assessment and biometric data with medical claims data can take this a step further. Examples include:
- The number of smokers who have not seen their physician in the last two years
- How many overweight and obese members also have back pain or arthritis
- Percentage of members with self-reported, high levels of stress who are also taking sleep or pain medication
It should be noted that a hallmark of health analytics is patient privacy. In the examples above, member information should be de-identified in accordance with the requirements of all applicable laws.
The capabilities of big data in employer sponsored-health plan analytics are powerful, and the potential to improve health outcomes of employees, spouses and dependents is profound. Employers can now identify specific wellness, preventive screening and health risk issues and develop appropriate, laser-focused strategies to address them.