Healthcare providers and health plans are embracing population-health management—a practice that focuses on activities designed to improve health outcomes by targeting and influencing behavioral, social, economic and clinical determinants of those outcomes at both the group and individual level.
The goal of population-health management is improved overall health of the targeted population, which in turn helps improve providers' bottom lines. To master the population-health challenge, some healthcare organizations are moving beyond traditional clinical, risk-based analytics to focus on patient engagement and behavior.
Traditional healthcare analytics are designed to understand and assess clinical risk, such as health risks, disease prevalence, disease severity and progression. This enables the organization to identify opportunities for clinical improvement, such as gaps in care, process of care, clinical quality and healthcare outcomes. They empower clinicians to improve care and to promote evidence-based medicine while also making it easy for clinicians to take action.
Traditional healthcare analytics, however, provide little understanding of the challenges patients experience in managing their health and maintaining positive clinical outcomes.
If we define healthcare engagement as an individual's emotional, financial, physical and social involvement or commitment with their health and the entire healthcare ecosystem, we find that at the core, engagement is one of the most important factors in many healthcare challenges.
Evidence of healthcare engagement manifests itself across all aspects of the healthcare system:
- Voluntary disenrollment and satisfaction. Members' level of engagement with the healthcare system (not only the health plan) is directly related to their disenrollment and satisfaction behavior: members with fewer physician visits, lower clinical compliance and adherence, and fewer calls in the call center are more likely to disenroll and provide a less favorable rating of the plan.
- Individual health and well-being. More-engaged members have higher health literacy and generally make better lifestyle and preventive-care choices.
- Healthcare delivery. A more engaged member also has greater health system literacy and makes more informed decisions about their choice of doctor and use of out-of-network providers.
- Clinical care. Sustained clinical compliance (such as consistent adherence to care plans, recommendations and prescriptions), as well as tertiary prevention (such as post-discharge practices) are a function of a member's sustained engagement with their health and clinical care.
Since improving healthcare engagement is mostly a nonclinical challenge, healthcare engagement analytics can augment traditional clinical analytics by providing behavioral insights into how individuals will behave and why. Also, since member behavior is nonlinear and multifaceted, creating meaningful engagement analytics requires deploying creative “big data” analytical techniques that use all available data sources in nontraditional ways.
With analytics focused on engagement and adherence, organizations can target both individual patients as well as entire populations. They can assess and predict engagement for health and wellness, preventive care, chronic care, system/network behavior, satisfaction and retention. They also can identify barriers to engagement, such as health literacy, system literacy, provider relationships, socio-economic conditions or physical ability. Engagement analytics can go even further by helping to prioritize patient communications and understand the most effective channel preference. Ultimately, deploying a population-health management system with an engagement analytics underpinning can improve quality of care, lower readmissions and improve retention and satisfaction. It can also help prevent unnecessary utilization and improve medication adherence.
For example, we have found that preventing readmissions is not only a clinical challenge, but also an engagement challenge. Our experience shows that a significant portion (more than 40%) of readmissions have significant patient engagement-related causes such as lack of support, inability to navigate the healthcare system, and inability to comprehend and follow instructions. Many of the engagement challenges exist even prior to the original hospitalization.
Readmission predictive models allow us to accurately predict more than 75% of readmissions that are attributable to patient engagement issues and enable us to deploy both population-based and point-of-care interventions. By focusing on patient behavior, predicting readmission and identifying specific barriers to engagement, healthcare organizations are able to not only reduce readmissions but also prevent the initial admission through proactive targeting and support.
The use of healthcare engagement analytics can provide a multidimensional, actionable view of individuals and populations, resulting in improved patient outcomes and bottom-line value for healthcare providers and payers. We are just beginning to see how effective these practices can be.
Dr. Dogu Celebi is chief medical officer at Boston-based Decision Point Healthcare Solutions.