Michael Anne Kyle
The recent $157 million commitment from the Centers for Medicare and Medicaid Innovation (CMMI) for a new “Accountable Health Communities” test model is most welcome. This is major step for the agency in recognizing the significance of social determinants of health in improving outcomes and costs. A New England Journal of Medicine article accompanying the funding announcement does an excellent job of highlighting the extent to which social conditions affect health outcomes and costs.
The program will invest in 44 communities over five years in three progressively advancing tracks: “increasing awareness”, “providing assistance” and “aligning partners”. Evaluation (perhaps proof of concept is more apt) is an important aspect of the model: the goal is not only to find out whether social service linkages affect health outcomes, but what types of interventions work. The awareness and assistance tracks each involve randomizing patients to usual care or an intervention; in the case of awareness, this is information about relevant social services, and in the case of assistance, the patient is provided navigation to facilitate the connection. The alignment track provides navigation, and will not involve randomization; instead, outcomes in these communities will be measured against a matched control site.
The CMMI vision of AHCs (another new acronym, gulp!) reflects emerging trends in health care and antipoverty work. The funding announcement credits initiatives like Health Leads for inspiring the low-touch (e.g., awareness) pathways. The alignment track, meanwhile, aligns very nicely with the work of emerging Medicaid Accountable Care Organizations in states like Minnesota, Colorado, and New Jersey.
Looking towards social programs more broadly, Accountable Health Communities are the latest in a trend towards collective impact models, which acknowledge the multidimensional nature (and mutually reinforcing strands) of disadvantage. By gathering a broad platform of stakeholders, collective impact approaches try to tackle entrenched challenges from multiple angles at once, aiming to turn vicious cycles into virtuous ones. For instance, the Promise Neighborhoods initiative from the Department of Education is modeled on the Harlem Children’s Zone, and seeks to provide “cradle to career” wraparound support for kids. HUD has the Sustainable Communities Initiative, which aims to improve the built environment and foster jobs and economic opportunity. Private foundations have also advanced collective impact models, such as Living Cities’ Integration Initiatives and the Robert Wood Johnson Foundation’s Culture of Health.
I’m thrilled that social determinants of health are being addressed in a serious way, and agree that vulnerable patients’ needs should be viewed from a “whole person” perspective. That is, organizations should work together to relieve some of the very real time costs, financial costs, and emotional burdens of having to coordinate multiple, and often conflicting or duplicative, systems. However, with the proliferation of coordination efforts, it seems critical to avoid replacing agency-level silos with community-level ones. This is an area where intra- and inter-agency coordination has a crucial role to play, which was the most glaring weakness in the AHC test model.
The AHC test model wants to encourage social service linkages, yet in reading the CMMI funding announcement, I would have had no idea that many of the social programs they wish to bridge are actually housed within the very same Department of Health and Human Services. It seems that Departments, like Education and HUD, have similar goals – it would be nice to see dialogue facilitated here as well. When agencies work together, the impact can be tremendous: the HUD-VASH program made strides in ending veteran homelessness by harmonizing HUD’s housing vouchers and the VA’s supportive services. To make AHCs truly successful, CMMI must engage its own stakeholder community to enable the coordination and collaboration that is needed to truly make a difference for our most vulnerable communities.