Turning data into decisions for rural hospitals

Summary

The challenge

Rural hospitals across the United States were under serious financial and operational strain. The COVID-19 pandemic intensified those pressures, accelerating closures and pushing already fragile systems closer to the breaking point.

Source: Cecil G. Sheps Center for Health Services Research, University of North Carolina; image modified from The Pew Charitable Trusts.

To address these challenges, CMMI launched the CHART model — a program designed to improve outcomes and expand access to care by funding new approaches to service delivery in rural communities.

Many of the hospitals CHART aimed to help couldn’t use data effectively, though. Critical information about cost, performance, and regional health dynamics was difficult to access, fragmented across systems, or required significant manual effort to analyze.

We send reports out to a third party to have them written. We don’t have a lot of funds to do that. It’s hard to get data out. We go with bare minimum reporting because system-generated reports aren’t useful.

Rural hospital executive

Providers often relied on incomplete or anecdotal information when making high-stakes decisions about services, staffing, and financial sustainability. CMMI needed to understand where those data breakdowns occurred — and find a path toward making information more accessible and actionable for rural providers.

The solution

We partnered with CMMI, the U.S. Digital Service (USDS), and Innovation Development and Operations Services vendors to develop a data strategy grounded in real user needs. We conducted in-depth research across nearly a dozen hospitals, engaging executives, analysts, and frontline staff. Through interviews and service design mapping, we uncovered how data flowed across systems, where breakdowns occurred, and what information providers actually needed to make decisions.

The research surfaced a consistent pattern: the data rural providers needed most — disease prevalence, peer cost comparisons, and hospital quality metrics — was either inaccessible, fragmented, or too labor-intensive to extract. We prioritized those gaps and prototyped dashboards using CMS financial and quality data, then tested and iterated with users to refine both the information hierarchy and the usability of each view. We also evaluated off-the-shelf tooling options against the needs we’d documented, giving CMMI a grounded analysis of build-versus-buy alternatives.

We don’t have access to this kind of data anywhere else right now. This is helpful for honing in on areas to improve care.

Rural hospital executive

Beyond the dashboards themselves, we helped CMMI stakeholders adopt user-centered design practices — continuous user research, usability testing, and iterative prototyping — so that future data solutions would stay grounded in real provider needs. The work didn’t just improve immediate access to data. It established a repeatable, user-centered approach to evidence-based decision-making within the CHART program.

The results

• Delivered the CHART model team’s first discovery sprints, conducting 11 in-depth interviews across hospital leadership and operational roles to map data flows, identify breakdowns, and establish a research-driven foundation for the program’s data strategy
• Prototyped and user-tested dashboards built on CMS financial and quality data, iterating on information hierarchy and usability with real providers
• Identified critical data gaps — including disease prevalence, peer cost comparisons, and quality metrics — and prioritized opportunities to make high-value information accessible
• Evaluated off-the-shelf tooling options against documented provider needs, giving CMMI a grounded analysis of alternatives
• Established user research as a core practice within the CHART program, embedding user-centered design methods into CMMI’s approach to data strategy