Learning Collaboratives

Learning Collaboratives Support Practitioners, Enhance Practice
Sharing Knowledge, Building Capacity to Effect Change and Improve Lives

Increasingly, peer-to-peer knowledge sharing through dynamic learning collaboratives promotes understanding, builds the evidence base, and pushes policy forward to create lasting change and solve some of today’s pressing challenges. Breaking down silos and bringing groups together requires expertise in collaborative methods and tools and an understanding of the deeper policy issues at play. In work spanning health and human services, Mathematica’s experts bring stakeholders together to transcend organizational silos, learn from one another, and share best practices. Using virtual and in-person meetings, interviews, shared online workspaces, and data-driven learning tools such as dashboards, Mathematica is helping to create access points to a collective knowledge base. The combined expertise of program administrators, managers, and frontline practitioners builds capacity to move the needle on positive change.

Sharing Knowledge and Data to Protect Children and Families in Crisis

One learning collaborative led by Mathematica brings together local grantees working to prevent homelessness among youth and young adults with foster care involvement. The project, executed under a contract with the Children’s Bureau and the Office of Planning, Research, and Evaluation in the Administration for Children and Families (ACF), creates facilitated opportunities for innovation and information sharing through peer-learning experience events. In these facilitated sessions, grantees share their approaches to specific challenges and discuss how others facing similar challenges might apply lessons learned. These topic-based discussions, hosted via webinar with on-demand recordings posted online, are grantee-driven so that they cover issues that are most salient and timely.

In another project for ACF, known as Child Maltreatment Incidence Data, Mathematica convenes learning groups to help improve the linking of administrative data to better understand the incidence of child maltreatment and related risks. A key element of Mathematica’s work with local sites is to design and support a cross-site learning network. The knowledge gleaned from the network, which meets via webinar, will help inform future surveillance, prevention, and treatment efforts. Although the sites face very different challenges, the network will spur informal partnerships and support structures that might extend beyond the end of the project.

Helping States Implement Policy and Effect Health Care System Change

Since 2011, Mathematica has partnered with the Centers for Medicare & Medicaid Services (CMS) on the Medicaid and CHIP (Children’s Health Insurance Program) Learning Collaboratives project. For this project, Mathematica leads the Learning Collaboratives—work groups of state and federal officials meeting on a regular basis for a defined period of time—to (1) identify states’ needs for developing and clarifying policies, (2) encourage peer-to-peer learning among states about promising practices, and (3) develop tools for all states to support high-performing health programs.

The Learning Collaboratives have produced materials that directly support policy implementation:

  • Strategies to increase coverage and care coordination among Medicaid-eligible, justice-involved populations and former foster care youth
  • Tools to help states and CMS measure the accuracy of determinations based on modified adjusted gross income in their new eligibility systems
  • Medicaid and CHIP performance indicators
  • Disaster preparedness toolkits for states and CMS 
  • Medicaid technical assistance

Promoting Peer-to-Peer Learning for Accountable Care Organizations (ACOs)

Mathematica’s health experts operate four learning systems for ACOs participating in the Shared Savings Program, the Next Generation ACO Model, and the Comprehensive ESRD (End-Stage Renal Disease) Care Model to help them improve population health and quality of care while reducing costs. To facilitate peer-to-peer learning, Mathematica develops content for and hosts more than 70 virtual learning events (webinars, affinity groups, and action groups) per year, as well as 17 small regional meetings for all ACOs and one national conference for select models. These events bring together ACO presenters and subject matter experts and aim to address learning interests identified through annual, model-specific needs assessments. To complement learning system events, Mathematica also develops a range of data-driven products, such as dashboards, survey reports, and chartbooks that help ACOs compare their performance to peers to promote improvement and accelerate implementation activities.

In addition, Mathematica has released a series of case studies and a new toolkit highlighting best practices and innovative care coordination strategies. Some of the case studies and the new ACO Care Coordination toolkit are available on CMS’s website. The toolkit is the first in a broader series of resources that will explore different aspects of how ACOs operate to provide value-based care. The toolkits will compile insights gathered during learning system events sponsored by CMS and through focus groups with the ACOs. Through these toolkits, CMS aims to educate the general public about strategies used by ACOs to provide value-based care, and to offer action-oriented ideas to current and prospective ACOs to help them improve or begin operations.