Toward a more comprehensive understanding of people and populations: New tools and approaches

February 15th, 2017 / By L. Gordon Moore, MD

With HIMSS17 around the corner, I’ve been thinking about advances in technology and data and the history of silos and systems in healthcare delivery. Optimizing a part of a complex system in a way that undercuts overall performance is an all too common experience.  We’ve seen this in health care: Overly aggressive reductions in hospital length of stay can lead to increased rates of hospital readmission.  Lack of funding for substance abuse treatment can lead to increased rates of incarceration. Entities engaged in population health management and interested in improved population outcomes are often working to gain a more global understanding of the people they serve so that focused interventions may improve outcomes.

In a Health Affairs blog, the authors Romm & Ajayi describe the Commonwealth Care Alliance’s (CCA) experience integrating social factors into their global work for people dually eligible for Medicare and Medicaid. While their defined population may be at one end of the spectrum of needs, the lessons from CCA can be useful across the entire population.

They point out that non-medical factors are the largest drivers of health outcomes. Understanding and addressing these factors can therefore improve population health outcomes. They provide an example: If a single mom with an asthmatic child cannot attend an education session on effective asthma management due to lack of child care coverage, it may make sense to use “medical dollars” to cover babysitting so the mom can attend.

CCA defines this comprehensive care for people as truly person-centered care – they are addressing the key factors that get between that person and optimal health outcomes. For some it may be child care, for others transportation, an air conditioner for asthma, housing for a homeless person, etc.

The authors point out a set of barriers to this comprehensive approach:

  • Lack of awareness of the impact of social factors on health outcomes. ACOs (short hand here for any organization accepting responsibility for health outcomes and costs) focus almost exclusively on disease management. Some may address non-medical factors if they become an issue for a person, but rarely is this work systematic and comprehensive.
  • Fractured budgeting. CCA addresses issues that cross over Long Term Supports and Services, Medical and Behavioral Health budgets. If these budgets are managed by different organizations there is near certainty that each organization will work to optimize its own budget, which can have a deleterious impact on total spending.
  • Patience for results. It may take several years to demonstrate good ROI in a program for reducing homelessness. This reduces tolerance for interventions that may have dramatic but delayed impact and shift funding toward short-term improvements that might not have as significant an improvement in the long term.
  • Locus of control. CCA recognizes the expertize of community agencies, supports them financially and engages with them as partners. This differs from the typical health system approach that is less collaborative, more one-directional: the “refer out” approach.

“This lack of appreciation for the immense role of social determinants of health may prove to be the critical obstacle to the expansion and sustainability of fully integrated care and, ultimately, to the realization of improved health outcomes for society’s most vulnerable members.”

CCA is an example of a fully integrated system of care addressing multiple factors. This work can be facilitated and replicated.  Here are some enablers:

  • Global understanding of total cost of care. Although the data sets are immense, it is possible to combine behavioral, medical, LTSS and other budgets.  This avoids the common flaw of optimizing a part of a system in a way that detracts from overall system performance (e.g. improving a behavioral health budget in a way that increases medical costs).  This work requires a high degree of sophistication with data, claims, budgets, and modeling.
  • Health information exchange. With appropriate safeguards, a person benefits when their information is shared across all who contribute to that person’s health outcomes.  This work starts with comprehensive care plans securely shared across multidisciplinary teams.
  • Methodologies to identify medical and non-medical risk factors. People are more than their diseases. Outcomes are predicted by multiple factors beyond diagnosis, severity, age and sex. Regular assessment of all of these factors gives the person and their care team information that may help improve outcomes.

The healthcare delivery system has lived in silos for a long time. Technology and analytics are advancing at a pace that makes it possible to glean understanding and insights from unimaginably huge data sets. Those engaged in the work of population health management can use existing tools and technological advancements to achieve insights and outcomes that were not previously possible.

L. Gordon Moore, MD, is senior medical director for Populations and Payment Solutions at 3M Health Information Systems.