When You Can't Measure What Works: The Quiet Data Gap in Senior Care

By Jon Warner, CEO, CareAxis

A new federal watchdog report says something that should give everyone in senior housing pause: one of the most valued programs in affordable senior living can't prove how well it works - not because it doesn't work, but because nobody can see the data. In early June, the Government Accountability Office released its review of service coordinators in rural federally assisted housing. The finding underneath the headline is a familiar one to anyone who has tried to run a wellness program at scale: the value is real, the outcomes are believed in, and yet the system can't put a number on any of it.

The program everyone trusts - and can't fully account for

Service coordinators are the connective tissue of independent senior living. They are non-clinical staff who help older residents reach community resources, navigate benefits, and stay in their homes longer. The case for them is well established. A 2015 study by LeadingAge and the Lewin Group found that having an on-site coordinator reduced hospital admissions among residents by roughly 18 percent. Later research from the Harvard Joint Center for Housing Studies showed coordinators were central to helping residents navigate services and weather pandemic disruptions, and a 2024 industry study found that catching a change in a resident's needs in the moment lets communities respond before a small problem becomes a crisis.

So, the benefit isn't in doubt. What's in doubt is whether the people funding and overseeing these programs can actually see it.

According to the GAO, the Department of Housing and Urban Development doesn't have uniform procedures for properties to enter data into its systems. The practical result is striking: HUD can't reliably say how many properties even employ a service coordinator, let alone confirm whether those programs meet federal requirements. And while coordinators are required to file regular performance reports, the agency doesn't routinely analyze what comes in. The reports are collected. They just aren't turned into knowledge.

Other agencies that help fund rural rental housing - the USDA and Treasury, among them- don't administer coordinator programs at all, which means there's no data from those properties to begin with. LeadingAge put it plainly: HUD should be able to answer the simple questions of how many coordinators there are and what the data it already collects says about the residents they serve.

The gap isn't effort. It's infrastructure.

It's worth being precise about what's broken here, because it isn't the program and it isn't the people running it. The information exists. Coordinators are doing the work and writing it down. The breakdown happens in the space between data being recorded and data becoming useful, the absence of a consistent way to capture it and a deliberate way to make sense of it.

That's a recognizable failure mode, and it isn't unique to federal housing. Any organization caring for older adults eventually hits the same wall: front-line staff hold an enormous amount of insight about residents, but that insight lives in scattered notes, individual memories, and reports that get filed and never read again. When it's time to demonstrate impact - to a funder, a regulator, a board, or simply to your own leadership, trying to decide where to put the next dollar, the evidence is technically present but practically unreachable.

The GAO's two recommendations land exactly on this point. First, standardize how data is entered, so it's consistent and comparable across properties. Second, build a real process to intake, analyze, and act on the performance reports that are already being submitted. In other words: create a system of record, then build intelligence on top of it.

What "knowing" actually requires

A system of record gives every observation a consistent home, the same fields, the same definitions, the same shape, so that what one coordinator records in one community can be read alongside what another records somewhere else. That consistency is what makes a number trustworthy enough to act on.

But a record on its own is just a tidier filing cabinet. The second half is the part HUD is missing today: the layer that watches the data as it accumulates, surfaces when a resident's risk is changing, and turns thousands of individual entries into a picture of how a program is performing. That's the difference between collecting performance reports and understanding them, between knowing you have data and knowing what it's telling you.

This is the thinking behind how we've built CareAxis. Holistic resident wellness generates a constant stream of signals: about health, daily function, social connection, and the small changes that precede big ones—and most of those signals are lost simply because there's no structured place to put them and no engine to interpret them. A system that captures wellness data consistently and scores risk from it doesn't replace the coordinator's judgment; it gives that judgment a memory and a vantage point, so a community can spot a declining resident earlier and show, with evidence, that its program is doing what everyone already believes it does.

The opportunity in the gap

There's an optimistic reading of the GAO report, and it's the right one. The hard part, proving that proactive, relationship-based support keeps older adults healthier and at home longer, has effectively already been done by the research. What remains is an infrastructure problem, and infrastructure problems are solvable.

The communities that close this gap won't just satisfy a federal recommendation. They'll be able to walk into any funding conversation with outcomes in hand, catch resident decline before it becomes a hospital stay, and make the case for service coordination not as a leap of faith but as a measured, defensible result. The value has always been there. The point now is to be able to see it.

The article was written by Jon Warner, CEO of CareAxis and Decision-support Architect for Innovation, Technology, Digital Health, and Aging populations, where a ‘System 2’ Mgt thinking approach is critical