Earlier this month on the blog, we talked about the wealth of information gathered in healthcare and the need to collaborate to integrate the siloes of patient data.  That’s because data has great utility in establishing relationships between these islands of information to support proactive and patient-centered care.

Patient-centered care has been the focal point as healthcare organizations are shifting to value-based payment systems across the globe. According to an  IHI article, “one size does not fit all anymore” and segmenting our patients, a historical practice in other industries like marketing and finance, is key to delivering care that meets the heterogeneous needs of patients while being cost-effective for the organization. 

Healthcare consumer segmentation

So what does segmentation mean? At its core, segmentation means dividing out populations based on shared criteria. Segmentation can draw relationships between demographic and clinical data, and social determinants of health such as socioeconomic status and health care access barriers.  With segmentation, the goal is to establish a more holistic view of a patient and redesign care to meet their unique needs.

There are many methods to segment patients.  One of the most standard data analysis methods leveraged in numerous frameworks is cluster-based segmentation. In cluster analysis, you let the data speak for itself.  This means, rather than going in with predetermined groups in mind, you let the data speak for itself and allow the algorithm to cluster patient into distinct groups. Segments arise from the traits shared by the group members and these distinct traits have some utility in predicting behaviors. 

Segmentation is not a new principle. The concept has been featured in studies and research journals for over a decade.  In 2008, the Deloitte Center for Health Solutions began collecting data on the US healthcare landscape. In 2012, they identified six distinct segments in the healthcare consumer market.  Each segment outlines the consumers’ diverse approaches with their health, their view on the overall health care system as well as health insurance and payment systems.

Use cases

H&HN, a publication by AHA, outlines how incorporating consumer segmentation has benefited specific U.S. healthcare organizations. Novant Health, a multi-site health system, is using segmentation for patient-centered care and population health management.  Understanding patient segments better provides insights into how patients want to be engaged so that they are more aligned with their needs.  It also enables them to provide better care to their patients and ultimately, the health of the communities.  This knowledge allows teams to make strategic decisions about access points to give patients the service they need.

Kaiser Permanente details their segmentation method to determine the characteristics of the subgroups in the top 1% high-need, high-cost (HNHC) patients in their healthcare system. Although some subgroups were clinically similar, nine subgroups were identified because they were divergent in their spending and health care utilization. Kaiser is using these segmentation findings to develop new care models for the costly 1% of patients who often have the most complex needs. These models will focus on delivering greater value and avoid the need for resource-intensive treatments.

Segmentation helps us achieve a better understanding of our unique patient segment profiles. This fuels new strategies and approaches to engage patients and provide care according to their specific needs. Decision makers and front-line staff alike achieve a longitudinal view of their data and the ability to drive improvements that truly impact population health. 

At the end of the day, segmentation provides an answer to an evolving question in healthcare. We’re no longer asking “how do we get more data?” but are instead asking how can we use the data we have to deliver proactive care?