Enabling Caring Communities
Premiere EditionRobert | Family Medicine May 10, 2019
About the authors:
Michael S. Klinkman is Professor in the University of Michigan Department of Family Medicine and the Director of Great Lakes Research Into Practice Network.
Donald E. Nease, Jr. is Vice Chair for Community in the University of Colorado Department of Family Medicine.
Healthcare and medicine have recently taken up the banner of addressing social determinants of health (SDOH). This effort arises from the evidence that health outcomes are largely determined by factors that lie outside the usual domain of healthcare, such as housing, food security, social isolation, etc. Multiple efforts are underway using the standard medical approaches of screening, detecting and treating.With respect to SDOH, however, treating has largely meant referral to community service organizations. This standard approach has been successful when addressing medical, and to some degree behavioral health issues, when the resources are known and can be coordinated. The web of SDOH and community services is in contrast highly complex with a spectrum of ever-changing organizations of differing sizes, missions and capacities. Additionally, individuals with needs may already be engaging with one or more of these community service organizations, doing their best to traverse the web of services and manage their own needs.
We have observed across multiple communities that there is fragmentation and sequestration of information related to personal health at a local community level. This leads to significant gaps in shared understanding about the interplay of social determinants, behavioral health and chronic illness among those providers dedicated to serve community members. Several factors contribute to these problems, including the proliferation of health and social services provided to individuals, the distribution of services across multiple organizations, specialization in both medical and social services domains, the development of isolated digital platforms by service providers, and concerns about privacy and security of digital information. The end result is that those dedicated to serve the needs of individuals operate in largely dissociated care islands and unintentionally create complex pathways of care. There is a large gap (chasm) between community social services and the medical enterprise, and a variable gap between community behavioral health care services and the medical enterprise.
As illustrated by Millie’s case, the consequences are felt across the system.
- Persons (patients) must negotiate their own way across multiple service providers and integrate isolated and conflicting recommendations for care, resulting in poor adherence to treatment recommendations, unnecessary and expensive care, ‘inappropriate’ emergency services utilization, inadequate attention to social determinants and personal barriers, and poorer health outcomes.
- Service providers may provide only a single service or address a single problem or health issue (specialty care services). They routinely work from a limited understanding of the full scope of personal needs and other actors engaged by those they serve. This leads to incorrect or conflicting treatment decisions, duplication of services, poor coordination with other providers, and unmet service needs.
- Health care providers and health care delivery systems are working to integrate the ‘medical enterprise’ but have no visibility or access to “non-medical” information. Bringing clinical information together in the electronic health record creates a vast repository of detailed information, often without context or information regarding provenance. Improving the collection of ‘non-medical’ data without integration into the EHR simply adds a new silo of health-related data.
- Community service agencies offer important services that remain uncoordinated and ineffectively used. Most communities lack a services hub, so people enter the system by self-referral or ad hoc referral to specific service providers. In the absence of a shared and comprehensive intake process, poor matching between needs and services is common and needs may remain unaddressed even where services are available.
Our proposed solution: a socio-technical infrastructure to support Communities of Solution.
A Communities of Solution approach aims directly at the problems of fragmentation and sequestration. It seeks to develop community-tailored solutions through shared local conversations that identify ‘problem sheds’ and ‘asset sheds’. In addition to finding hot spots, a COSfinds the cold spots in the problem shed; places, neighborhoods, and communities where health care problems are not just individuals’ problems. And a COScan identify the asset shed, the individuals and groups who can collectively provide a local solution.
But to implement, sustain, and scale these local solutions will require shared community infrastructure that : provides the social network and governance structure to organize services and link people to services; and a community health information hub that provides the clinical and communications platform to link people to services, link service providers to each other, and link the community to the medical enterprise. This sociotechnical infrastructure will support multiple and dynamically emerging individual community-based use cases. It is ‘locally owned’ but can be scaled.
We propose a sociotechnical design process whereby community members, service providers, and technical experts work together to create this infrastructure, then learn how to refine and adapt it over time as conditions, priorities, and community goals change.
For the Hubs to work, it is critically important to establish both the social infrastructure and the technical infrastructure through engagement of community members, stakeholders, and health service providers.
The social infrastructure supports 2 essential functions: community governance to establish and maintain acceptable standards for the sharing and use of health information among community providers, and data stewardship principles to guide effective and secure use and dissemination of information to improve community health. Communities develop local advisory/governance boards that convene and broker roles, relationships and responsibilities between service providers and the community-at-large. Through this process, the ethics that guide community uses of information are developed and refined. These roles, relationships and responsibilities are then embodied in the local technical infrastructure of each Community Hub.
The technical infrastructure of the Hub consists of ‘middleware’ enabling disparate, proprietary software applications within a community to communicate and share information. Hubs support a lightweight and inexpensive set of services that provide secure data/document transmission and exchange, patient level views of integrated health information and community level dashboards. Hubs are built using standards endorsed by IHE, NIST, HIMSS, HL7, supported by major EHR vendors and Health Information Exchanges. This permits each organization to maintain its existing IT tools and simply link to the Hub. In brief, Hubs utilize an architecture built to include all healthcare enterprises and community services that have agreed to use a common set of policies and share a common infrastructure. Federation of Hubs enable intercommunity and statewide data linkages and connection with state health information exchanges, enhancing public health surveillance statewide.
Although much of the above was originally written in August of 2016, much of it has stood the test of time and our ongoing work in communities in Michigan, Colorado and elsewhere. This blog will take up the issues that we have stumbled upon, tripped over and attempted to work through in our efforts to advance this work. We do not pretend to have all of the answers, but we hope with the posts that detail our experiences to shed light on the issues and create a dialogue and community among those working to address the problems in this space.
Mike Klinkman – University of Michigan
Don Nease – University of Colorado