What we heard at MGMA: "The problems are shared, the records still aren’t."

SUMMARY: 

MGMA’s centennial conference in San Antonio ran under the theme “From Here Forward,” and the through-line in the conversations we had there was that medical groups are planning forward against records that only look backward. Workforce is taking the largest share of new budget dollars heading into the year. It is also the area where the operational record is most fragmented, which means groups are funding the function they can see least clearly. A hundred years of tooling has changed nearly everything about running a practice except the seams between the systems.

A hundred years on, operators still travel to compare notes on the same mutual problems.

MGMA dates its founding to 1926, when 32 clinic managers from 21 medical groups met at the Madison Club in Madison, Wisconsin to compare notes on what the association still calls the “mutual problems” of running a practice. Acting chief executive Akash Madiah opened the centennial meeting by naming what 1926 did not have: no modern reimbursement system, no EHR, no AI, no telehealth. What it did have was a room full of people responsible for making sure patients got care. Madiah, who came to the association from practice operations, told the room “I’ve sat in your seat.”

I have spent a good part of this year in conversations with workforce leaders across health systems and staffing organizations, and San Antonio mostly confirmed what those conversations had already been pointing at. The reason operators still get on planes to compare notes is not nostalgia. It is that the mutual problems are still genuinely mutual, and most of them still live between functions rather than inside any one of them. A room of a thousand people will solve scheduling, credentialing, and payroll a thousand competent ways. Almost nobody has solved the space between them, so that is what everyone ends up talking about in the hallway.

Essential coverage needs more than spreadsheets and memory.

One conversation has stayed with me more than any other. I spoke with an operations leader for a medical group in Alaska, where locum tenens coverage is not a supplement to the staffing model but a structural part of it. Their group has no system for managing locums at all. Everything lives in spreadsheets or in the heads of the individuals who happen to own a given relationship, and the result is that communication and process vary from person to person and from engagement to engagement. They knew exactly what they wanted, and they said it plainly: a single source of truth. Just as pressing was the financial side. They had no visibility into what they were spending on locums or how that spend compared to budget, which means one of the largest and least predictable lines in their workforce costs was being managed largely from memory.

Workforce is taking the largest share of new dollars, and it carrie the least connected record.

The budget picture coming into the conference was already clear. In an MGMA Stat poll of 213 medical group leaders, 37% said workforce investments would top their budget priorities, ahead of health information technology at 30%, with revenue cycle and patient access each at 12%. Adding staff, adjusting compensation, and improving retention were the moves leaders named.

Those are reasonable allocations made by people paying close attention. Compensation is the largest line most groups carry, and if you are going to defend margin and access at the same time, workforce is where the leverage actually sits. The complication is that workforce is also the function running on the most fragmented operational record. Coverage decisions get made in scheduling. Eligibility lives in credentialing. Hours land in timekeeping. Rates resolve in payroll. Each system is accurate about its own piece and blind to the other three.

So a group can fund workforce seriously and still be unable to answer, in the room, what the workforce actually cost and whether the coverage it bought was eligible on the day it was worked. The dollars go into the function with the least connected record, and the reporting that comes back is assembled after the fact from four exports. My colleagues have written about why leaders think they understand staffing when they don’t, and that gap was audible in San Antonio in the way people asked their questions.

The recurring question was not what to buy. It was what to conenct.

This is the part that struck me most, and it runs against how a conference exhibit hall is usually assumed to work. Very few of the leaders I spoke with were shopping for another system. Most of them had plenty. What they kept circling was narrower and harder: the handoffs. How a filled shift becomes a worked shift becomes an eligible worked shift becomes a correctly paid one, and where in that chain the record stops being trustworthy.

That framing matters because it changes what a good answer looks like. The problem is not a tooling gap, it is a connection gap, and those call for different responses. A tooling gap gets closed by buying. A connection gap gets closed by architecture, by deciding which systems share a live operational record and which ones will keep handing each other exports. Groups that own excellent software and still cannot answer a coverage question on the spot are not under-tooled. They are under-connected.

TLeaders have plenty of data. They need a record that explains the spend.

If I had to compress the hallway conversations into a single phrase, it would be the one I heard in some form at nearly every stop: "Data, data, data." Leaders wanted reliable data that tells a story, specifically what they are spending and where that spend is concentrated. What the phrasing reveals is that the gap is not in collecting data, since every system they own collects plenty of it. The gap is in a connected record that can explain spend in operational terms, which is exactly the seam between scheduling, credentialing, and finance.

The people in the seats are doing the work. The record is what is failing them.

The opening session made this point better than any vendor could. The emcee, describing two years of his own care as a patient, called the administrators and executives in the room the chaos coordinators who keep practices running, and Madiah framed their work as the reason patients get outcomes they never trace back to anyone. That is the accurate read. Nothing I heard in San Antonio suggested a workforce problem caused by inattention.

I want to be direct about this because it gets miscast constantly. The operations leader watching the same coverage pattern return after a deliberate fix is not failing to manage. The finance leader whose variance report keeps surprising them is not failing to forecast. They are making the best available decision against the information in front of them, and the information in front of them is partial by construction. When credential status, privilege state, and scheduling availability sit in three systems on three timelines, the person making the coverage call at four o’clock on a Thursday is working with what the architecture gave them.

Internal scheduling is moving forward while external workforce management is staying behind.

What surprised me most was how many operators are still running external labor the way they would have in 1999. Plenty of groups had invested in scheduling technology for their employed staff, but almost none had anything in place to manage locums or other external labor spend. That left them with no visibility into what was happening and, just as striking, no control over who inside the organization was contacting which outside agencies to engage coverage. At a conference celebrating a hundred years of progress in practice management, the external workforce was the corner of the operation that time seemed to have skipped.

The operational view we built Kimedics around

Kimedics is the clinician workforce operations platform built by healthcare operators. Scheduling, credentialing, time, and pay run on one connected operational record, so the people making coverage decisions can see eligibility and cost while the decision is still open rather than reconstructing both after the quarter closes. That design came out of the same rooms MGMA has been convening for a century, from people who had already spent years reconciling these seams by hand.

The organizations I have seen make real progress on this did not begin by replacing their systems. They began by connecting the ones they had, and the questions in their quarterly reviews got easier to answer on the spot. A century in, operators are still gathering to compare notes on mutual problems. The work ahead is making sure their systems compare notes too.

Q&A

Q: What was the main theme at MGMA 2026?

The published theme of the centennial conference was “From Here Forward,” and the programming leaned into planning: budgeting, cost containment, workforce and retention, and data and technology investment. The operational conversation underneath it was about connection, specifically whether the systems groups already own can answer the questions their 2027 plans depend on.

Q: Why does workforce spending keep growing without matching visibility?

Because spending and visibility are governed by different systems. Compensation decisions run through budget and payroll, while the operational facts that explain the spend live in scheduling and credentialing. Groups can increase workforce investment and improve retention while still being unable to trace what coverage cost at a given site, because no single record carries the decision, the eligibility, and the rate together.

Q: Is this a problem for medical groups specifically, or for staffing organizations too?

Both, and the shape is the same on either side. Medical groups see it inside their own walls. Staffing organizations see it across client sites, which usually sharpens the seams rather than softening them, since a client relationship is riding on the reconciliation. The real cost of incomplete eligibility and credentialing data shows up on both sides of that relationship.

Q: Do groups need to replace their existing systems to fix this?

Usually not. Most of the groups I spoke with had capable scheduling, credentialing, and payroll systems already. What they lacked was a shared operational record those systems could write to and read from. The useful question is architectural, about which functions run on a connected record, rather than about which product to buy next.

Q: How does Kimedics help?

Kimedics is the clinician workforce operations platform built by healthcare operators. Scheduling, credentialing, time, and pay share one live operational record, so coverage decisions carry eligibility and cost context at the moment they are made instead of being reassembled from exports afterward. Health systems and staffing organizations run on the same record, which is what shortens the reconciliation conversation at the end of the month.