Areas of Research

The Center's goal is to identify, classify, track, and compare health systems in today's consolidated health care markets and to characterize the attributes of high-performing health systems. To pursue this goal efficiently and to maximize integration across all of its research efforts, the Center is organized around a Data Core, which has assembled and maintains a data library for Center-wide use. The Data Core also provides support to four independent study teams.

Data Core

The Data Core provides Center research teams with an integrated data library as well as methods, measurement, and analytic support. Center data come from multiple sources (federal, state, regional quality collaboratives, the private sector, primary data collection); data are housed at RAND with appropriate safeguards. The Data Core coordinates quantitative and qualitative analyses across all of the Center’s projects.

To date, the Data Core team has:

  • Defined "health system" (as a unit of study)
  • Defined what constitutes "high performance" in a health system
  • Developed interview protocols and conducted "virtual" site visits with c-suite executives in 24 health systems across 4 states
  • Constructed a multi-year integrated library of primary and secondary data on physician organizations, hospitals, and health systems
  • Created a novel health system database that maps physicians, physician organizations, and hospitals to health systems
  • Characterized the attributes of health systems and the health care markets in which they operate
  • Constructed measures of health system performance (including total cost of care, quality of care, low-value care, and primary care spending)
  • Constructed a composite measure of clinical performance that reliably differentiates high- versus low-performing health systems
  • Constructed a novel composite measure of health information technology (IT) capabilities to distinguish "superusers" of health IT from "underusers" of health IT

Characterizing Health Systems

We selected 24 health systems for comparative case study analysis. These systems were chosen from four states that have been at the forefront of collecting and publicly reporting standardized performance measures. Each hosts a health care improvement collaborative that agreed to partner with RAND to provide performance data and assist in recruiting health systems.

To choose health systems for the study, we identified all physician organizations (POs) publicly reporting performance data in those four states and mapped them to health systems. The sample of health systems we selected wasn’t random because we wanted to make sure that our sample included small, medium, and large health systems as well as some that were high, medium, and low performers.

We then organized “virtual” site visits: a series of 60-90 minute telephone interviews with 5-8 senior executives (CEO, CFO, Chief Medical Officer, etc.) in each system. We developed the interview protocols with the input of a technical expert panel of health system executives and researchers. The coded interview data were used to compare and contrast health systems and to produce a series of papers on health system organization and governance, clinically integrated networks, care delivery redesign within health systems, and the use of health IT to improve performance.

Study Teams

The Center is also home to four study teams. Each team has collected primary data to examine a specific aspect of health systems that may influence how uptake of evidence-based practices and performance affects quality, cost, and patient outcomes.

Health Information Technology (HIT)

Health information technology can promote adherence to clinical guidelines, improve care quality and patient safety, and help to reduce costs. However, implementation of health IT varies across health systems, and systems differ in the kind of health IT capabilities that they have and the extent to which they use them.

The Health IT study team has examined adoption and use of health IT. in 24 health systems, identified differential use of health IT functionalities (such as CPOE and CDS) by health systems and affiliated physician organizations, and tracked changes in health IT use over time. To understand the role of health IT in shaping health system performance, the team created a novel composite measure of health IT capabilities to examine how physician organizations use health IT.


The current health care environment is characterized by widespread experimentation with incentives. Incentives can be both financial (e.g., compensation) and non-financial (e.g., behavioral nudges, physician profiling). But despite the prominence of incentives, we lack a sound understanding of how to design and use them to improve health system performance.

The Incentives study team has cataloged financial and non-financial incentives used in 24 health systems (and affiliated physician organizations) across four states and is conducting analyses to characterize different compensation structures, use of behavioral nudges, and the relationship between different compensation arrangements and performance.


Fragmentation undermines the ability of health systems to deliver good patient care and to achieve good outcomes. A common strategy for addressing fragmentation is to adopt integrated organizational models such as ACOs. Other mechanisms of integration focus on the clinical level—for example, improving communication between primary and specialty physicians. However, studies have yet to demonstrate whether greater integration at the organizational level leads to greater integration at the clinical level and consequently better adoption of evidence-based practices and better outcomes.

The Integration study team has developed a conceptual framework to advance the study of integration , developed innovative measures of integration that assess the degree to which care delivery is integrated, and measured clinical integration within 24 physician organizations in 17 health systems across 4 states. The integration team is analyzing these data to assess the relationship between organizational and clinical integration and the relationship between clinical integration and performance.

Safety Net

Community health centers (CHCs) provide primary care to vulnerable and low-income residents. Unlike other types of physician organizations that have rapidly joined large health systems, most CHCs maintain only informal connections with other local health care providers and health systems. However, the expansion of new delivery models and value-based payment systems are providing new incentives for safety net providers to develop more integrated systems of care.

The Safety Net study team has examined strategies that CHCs use to achieve greater integration with three types of service providers (specialists, hospitals, and social service organizations). The team’s goal is to understand whether the strategies actually promote better integrated care and to determine whether integration affects the provision of evidence-based care.