Improving Health Outcomes – Whole Person Approach to Data |… | MOBE

AI Using Data Header AI Using Data Header sm

Driving Better Health Outcomes With a Whole-Health Approach to Data

Traditional health care data analysis has long been rooted in vertical data models based on condition. In fact, the health care industry has invested substantially in condition-specific programs designed to engage and motivate people based on their disease diagnoses. Unfortunately, these types of programs and solutions have dismal engagement rates, ranging on average between 4% and 6%. More specifically, many digital point solutions have failed to deliver much engagement at all, often due to barriers unrelated to the chronic conditions they focus on.

Despite overwhelming evidence of the relationship between better engagement and improved health outcomes, engagement rates sometimes seem little more than a box to check for many programs and solutions.

There’s a lot behind this problem. Finding cardiac patients, people with diabetes, and others based on a chronic condition is relatively simple. While many health care organizations and vendors claim to use data-driven insights to design programs and solutions, from a data perspective, most are managing their populations based on these vertical silos. As a result, they’re missing opportunities to engage people with the highest potential for improved health outcomes and reduced costs.

Modeling data based on whole health.

The Veterans Health Administration (VA) defines whole person health as “an approach to care that empowers and equips a person to take charge of their health and well-being.” While the concept of whole person health, or whole-person care, isn’t new, it has evolved considerably over the past several years. More recently, government organizations like the VA have embraced programs and solutions based on whole person health. Across the board, this approach has simultaneously lowered costs and improved health outcomes for this large health system.

The case for a data approach based on the concept of whole person health is compelling for several reasons:

  • Social determinants of health (SDoH) play a significant role in determining health outcomes. Food insecurity, for instance, can lead to poor nutrition, exacerbating chronic conditions and reducing overall well-being. Lack of stable housing can result in stress and insecurity, impacting mental health. Recognizing and addressing these factors is essential for improving health equity and reducing health care disparities.
  • Focusing on the “whole person” allows health care organizations to craft more effective interventions. By understanding an individual's needs, preferences, and social context, health care providers can motivate people more effectively to take action to improve their health.
  • Whole person health typically incorporates behavioral data, powerful information that drives more meaningful messaging and outreach programs. Examples include health plans that use behavioral data to identify individuals who would respond well to a targeted smoking cessation program.

Most health experts agree that an integrated, multi-dimensional approach is essential to whole person health. Instead of creating vertical silos of data based on conditions and costs, data based on whole person health views health populations from a horizontal perspective that considers each individual’s care utilization patterns, lifestyle, and health preferences. From there, health plans can focus programs more effectively by finding people with the highest opportunity for engagement that results in improved health outcomes.

Uncovering data that matters.

Data models based on a horizontal view of a person’s health typically combine multiple data sources and measures, including de-identified data ranging from common demographics to claims and, increasingly, SDoH factors. Behavioral data is a big part of this multi-faceted view of health.

The behavioral patterns of health care consumers encompass a wide range of actions, from care-seeking behavior to medication adherence and lifestyle habits. Behavioral data is an important part of a person-centered view of populations, helping plans and employers identify more of those people who may be trending toward more significant health care utilization and cost.

People may not stand out for any specific conditions or comorbidities, but they have certain behaviors that predict how well programs like MOBE’s can improve health outcomes while lowering costs. That helps us find individuals for whom we can make the biggest difference. That’s the power of looking at data through the lens of whole person health.

— Travis Hoyt, Chief Analytics Officer, MOBE

MOBE’s whole-person approach includes lifestyle and medication guidance for people with certain identifiable behaviors regarding health care and their lifestyle. Interestingly, these are often people who don’t typically engage with well-being programs. They can also be people with multiple conditions who use a high volume of low-intensity health care services. At MOBE, we’ve found these people don’t respond to outreach that identifies them by their disease but do engage with programs that address them as a whole person.

To create a deeper view and segment people with the highest propensity to engage, MOBE’s expert data analysts work with highly sophisticated, proprietary machine-learning and predictive algorithms considering more than 10,000 variables. These systems combine attributes from three common data model types:

  • Predictive cost and utilization models identify behaviors (like care-seeking) that may generate a probability of higher costs in the future. Typically, this uncovers people who have yet to find the help they need.
  • Similar in many ways to Johns Hopkins and Medicare ACC models, MOBE’s Comorbidity Index is a machine-learning model that analyzes the severity of each chronic condition, how they interact, and how likely they are to progress.
  • Episodic suites examine episodic health issues that are often important to a whole-health view. MOBE’s model identifies musculoskeletal surgery risks, triggers related to pharmacy drug interactions, and other issues of this nature.

The result is an aggregate population profile with the highest potential for reducing costs and improving health outcomes. With this data, MOBE’s proprietary algorithm helps plans and employers target the right people and craft more meaningful interventions.

Measuring success with the SF-36.

The Short Form Health Survey, commonly known as the SF-36, is a validated tool allowing individuals to self-report their health improvements. Based on eight health domains, it is an essential tool for informing health professionals and solutions providers of changes to an individual's health status. The SF-36 is a proven standard for measuring quality-of-life improvements across targeted health populations.

Based on SF-36 reports, MOBE participants from a large employer program demonstrated significant improvements in their overall energy, vitality, and other emotional and physical benefits in just 12 weeks. The program significantly improved overall health outcomes, reducing utilization and health care costs at the same time. Read about it here.

Engaging the right people, the right way.

A whole-person view delivers more focused, personalized interventions with much better results in terms of boosting health outcomes and reducing costs. Whole person health offers more opportunities to create interventions that deliver the right message to the right person in the right way.

This type of “deep personalization” requires sophisticated data and analytics to build a more seamless health journey tailored to each person’s needs and preferences. It’s an approach that enhances engagement and boosts the likelihood of sustainable behavior change. Instead of dense education materials or frequent calls to action, this might include:

  • Sending a cookbook designed for specific dietary needs to make healthy eating more accessible and appealing
  • Offering transportation or online access to pharmacies to improve medication adherence
  • Providing incentives to encourage important prenatal visits to improve the health of both parent and baby
  • Creating feedback programs for individuals who want to track progress toward their personal goals

More meaningful interactions between health plans and their members do a much better job of changing behavior. Creating more effective programs starts with the right selection algorithm and a whole-person approach, which is the heart of our work here at MOBE.

— Kurt Cegielski, Chief Commercial Officer, MOBE

Moving from disease to whole health.

Disease-specific health programs are reactive and often reach people who are already engaged with their provider, plan, or employer. Focusing on a robust data set that combines behavior, utilizations, and other variables creates a whole-health view of individuals, informing more effective interventions with much more sustainable, impactful results.

You may also like:

A Whole-Person Approach to Supporting Optimal Mental Well-Being

While mental health is becoming a bigger part of the benefits conversation for employers, many still consider mental health in isolation from physical health. Fortunately, that view has started to change.

In conversation with: Leslie Helou

See the difference MOBE’s leading-edge clinical pharmacy approach is making in the well-being of MOBE participants in this conversation with Leslie Helou, MOBE’s Vice President, Medication Strategies.

The Dangers of Algorithm Bias

Gain a greater understanding of AI-generated bias within the health care system and its impact on people.