We collaborated with Humana on a pilot that enabled 100,000 Medicare Advantage beneficiaries to manage their chronic illnesses more effectively. Our approach aided Humana in closing gaps of care and enhancing Star ratings. We put a strong emphasis on patient education and engagement by integrating various datasets, including Social Determinants of Health (SDOH), motivation, and lifestyle data gathered through a game that members loved playing. This "health concierge" user experience tackled immediate health concerns while promoting long-term patient well-being and healthcare efficiency.
Humana, under its organization-wide Bold Goal initiative, partnered with us for a pilot within their Stars Group. This pilot was strategically designed to drive savings and craft an acquisition and retention strategy for Group Medicare Advantage plans. Leveraging a patient engagement intellectual property that we co-own, machine learning techniques, and comprehensive social determinants of health data, we engaged a Medicare Advantage population of 100,000 lives. This initiative delivered a 'concierge of health' experience through enhanced, personalized communication, aiming to improve health outcomes and elevate Star measures for chronic conditions such as diabetes and hypertension.
In the healthcare industry, businesses are immersed in a relentless, multi-trillion dollar endeavor to secure and enhance patient engagement—a challenge that intensified during the pandemic as payors lost touch with their populations due to ineffective communication channels. It is a known fact that the cornerstone of impactful personalization lies in the gathering and seamless integration of elusive datasets, including behavioral health and lifestyle information.
Sourcing Social Determinants of Health (SDOH) data from public sources presents a significant challenge. It requires collecting data from various sources and harmonizing it within a single data warehouse. In this project, we aimed to examine health outcomes through a comprehensive lens, encompassing both physical and social environmental factors. This approach also demanded a deep understanding of the diverse needs of our target population.
Source: Institute for Clinical Systems Improvement, Going BeyondClinical Walls: Solving Complex Problems (October 2014)
Missingness map showing spots in our database with missing values (black = missing)
Navigating data gaps and ensuring accuracy were significant challenges. We were tasked with managing vast amounts of member data, which occasionally lacked crucial contact information, such as email addresses and phone numbers. This absence of key details complicated our efforts to effectively engage with Humana's members and personalize their healthcare journey.
Employing an experimental design, two groups were constituted: a “test” group, which was exposed to standard Humana communications augmented with personalized content, and a “control” group, which was exposed solely to Humana communications. A chi-square goodness of fit test was employed to ensure the homogeneity of the groups, and a two-sample test of proportions was used to assess the statistical significance of the increase in success rate among program participants.
We established a KPI continuum as the foundation for our approach. This allowed us to measure progress across all pilot phases, from initial engagement, through data mashup insights and self-empowerment, to addressing Star Measures and ultimately quantifying cost savings.
We deployed a card game tailored to mine patients’ intrinsic motivation. Intrinsic motivation - engaging in activities for the sheer joy of it - is the purest form of motivation. Studies indicate that when motivation is externally induced, such as through rewards or avoiding guilt, it seldom leads to long-term health behavior change.
Caption: card-sorting game used to gather member’s motivations & lifestyle data
We employed clustering algorithms, specifically K-means, and silhouette analysis for dynamic member segmentation. By integrating intrinsic motivation data from our game, we discerned shared traits among participants. This enabled tailored communication, optimizing message impact by understanding what resonates with each cohort.
We established a Health Content Library in partnership with NYU Langone Health experts, facilitating bilateral communication for realizing a concierge of care experience with content targeted to the individual. The library comprises 108 communication tracks with embedded tone variations. The content encompasses common intervention communications for individuals aged 60 and above, addressing prevalent chronic conditions, general wellness, and other topics pertinent to this population.
We explored edutainment by incorporating diverse formats, including memojis, which we tested separately with resounding success. While the Humana team was keen, aligning this approach with the organization's established voice presented challenges, as it sparked intense discussions on incorporating 'fun' in health communications. Currently, Humana wasn’t ready to challenge their voice guidelines, but we remain optimistic about employing edutainment in a more conversational manner in the future.
Nudges—subtle tweaks in choice architecture—can positively sway behavior in healthcare. Our agency harnesses the BASIC toolkit, focusing on intrinsic motivation through engaging, game-like activities. We draw inspiration from the Penn Medicine Nudge Unit, a pioneer in embedding behavioral design within healthcare. The potential is vast; well-crafted nudges can revolutionize patient outcomes and healthcare delivery.
We developed a dashboard that tracked our performance across a spectrum of KPIs, ranging from basic engagement metrics like program onboarding to more advanced indicators such as 'health activation,' which reflects increased patient participation in their health improvement.
The pilot showcased the capacity to steer patients towards services essential for the prevention and management of chronic conditions. Through bilateral communication, it captured patients' challenges and barriers to better health. Integrating diverse datasets, real-time analytics facilitated informed decision-making. Notably, we devised a method to quantify the bottom-line impact, translating closed care gaps into estimated savings of $1,598 per engaged life per year. For instance, mammograms increased by 26.9% and colorectal cancer screenings by 22.1%.
Health insurers have long recognized the importance of engaging members to improve the value of care
The program served as a personalized "health concierge," offering bespoke attention and aiding patients in navigating their health plans and maximizing rewards. By tapping into intrinsic motivation, we encouraged health activation, getting patients actively involved in their own health. The card-sorting game played a pivotal role in engagement optimization and trust-building, while communications were health-promoting and tailored. This approach not only empowered patients but also fostered a sense of ownership over their health journey.
