Your Dashboard Is Only as Good as the Systems Feeding It: Why Data Reconciliation, Not More Reporting, Is What U.S. SNFs Actually Need

Key Takeaway: CMS calculates a skilled nursing facility’s staffing hours per resident day, and therefore its Five-Star staffing rating, by combining Payroll-Based Journal data with resident census figures that CMS derives independently from MDS assessments rather than from PBJ itself. A facility’s payroll data can be perfectly accurate and its public staffing rating can still be wrong if the census feeding into that calculation, submitted through an entirely separate clinical system by a different department, does not match reality on the same date.

Table of Contents

• Why “More Data” Isn’t the Same as Better Decisions
• The PBJ and MDS Split: A Case Study in Why Systems Disagree
• Where Else This Same Pattern Shows Up
• Building a Data Reconciliation Discipline, Not Just a Dashboard
• How LTCPro Supports Data Integrity Across Systems
• FAQ

A dashboard that blends payroll, billing, and clinical data into one screen feels like progress. It often is. But a single view built on top of numbers that already disagree with each other does not resolve the disagreement, it just displays it faster and in better color contrast. For U.S. skilled nursing facilities, the harder and more valuable work isn’t building a prettier report. It’s making sure the systems feeding that report actually agree with reality and with each other before the numbers ever reach a screen.

Why “More Data” Isn’t the Same as Better Decisions

Most SNF operations run on a handful of systems that were never designed to talk to each other: an EHR for clinical documentation and MDS assessments, a payroll or time-and-attendance system for staffing hours, a separate billing and revenue cycle platform, and often a standalone system for accounts payable. Each one is the authoritative source for its own slice of the operation. None of them was built on the assumption that another system down the hall would need to independently agree with it on a shared number, such as how many residents were actually in the building on a given day.

A centralized dashboard pulls from all of these sources and presents them as one coherent picture. That is genuinely useful for visibility. It is not, by itself, a fix for the underlying problem: if two source systems report different numbers for the same fact, a dashboard doesn’t know which one is right. It just shows you both, or worse, silently picks one without telling you that the other one disagreed.

The PBJ and MDS Split: A Case Study in Why Systems Disagree

The clearest, most consequential example of this problem is baked directly into federal policy. Facilities submit direct care staffing hours, including agency and contract staff, through the Payroll-Based Journal system every quarter, on a 45-day deadline after each fiscal quarter closes. CMS explicitly does not use PBJ-submitted census data to calculate a facility’s staffing measures. Instead, CMS derives the daily resident census from Minimum Data Set assessments, an entirely separate clinical submission system maintained on its own schedule by clinical staff rather than payroll or business office staff (CMS, PBJ Policy Manual).

Hours Per Resident Day, the figure that drives a facility’s public Five-Star staffing rating, is calculated by combining these two independently submitted numbers: staffing hours from PBJ, divided by census derived from MDS. That means a facility’s payroll department can submit flawless, fully reconciled staffing hours, and the public staffing rating can still come out wrong if the MDS-derived census for the same dates is off, whether from a late assessment, a discharge that wasn’t updated promptly, or a data entry error in an entirely different department. Neither system is technically “wrong” on its own. They simply were never required to check against each other before the number that matters publicly was calculated.

The same fragility shows up inside PBJ submissions themselves. Contract and agency hours are one of the most common failure points, since a facility’s own payroll system often has no visibility into an agency CNA’s shifts until an invoice arrives, well after the fact, and someone has to manually match it to the correct dates and convert it into reportable hours.

Reporting scheduled hours instead of actual hours worked, or misassigning an overnight shift to the wrong calendar date under PBJ’s midnight cutoff rule, creates the same kind of quiet mismatch, one that a dashboard displaying “total staffing hours” will never flag, because the dashboard has no way of knowing the underlying number was built incorrectly in the first place.

Anonymized case scenario: A multi-facility operator’s corporate dashboard showed strong, consistent staffing hours across its portfolio for two consecutive quarters. Its Five-Star staffing rating at one facility still dropped a full star. The payroll data behind the dashboard was accurate.

The issue traced back to a stretch of late MDS assessments at that facility during a nursing leadership transition, which understated the facility’s documented census for several weeks, inflating the HPRD denominator’s error margin even though staffing hours themselves never changed.

Nobody on the payroll or business office side had visibility into MDS timeliness, and nobody on the clinical side had visibility into how that timeliness fed into a public financial and quality metric.

Where Else This Same Pattern Shows Up

The PBJ and MDS split is the clearest example because it is baked into federal policy, but the same structural problem, two systems independently reporting on the same underlying fact without a required cross-check, repeats across a facility’s financial operations.

Clinical documentation and billing are in close parallel. Therapy minutes logged in an EHR, pharmacy orders routed through a consultant pharmacist, and diagnostic services ordered from an outside provider all generate their own records, separate from the claim eventually submitted for that care.

Our ancillary departments guide covers this specific handoff between clinical departments and billing in depth. The pattern is the same one described here: two systems, two owners, and a gap that only shows up once someone goes looking for it. Newer care models raise the same question of which system records what and how it reaches billing. Our guide to virtual nursing billing and reimbursement in long-term care walks through that for telehealth documentation.

Financial close data tells a similar story at the portfolio level. Claims-level revenue data, general ledger entries, and AR aging reports often live in different systems updated on different schedules, which means a month-end close built by simply exporting each system’s own number, without a variance step comparing them against each other and against the prior period, can close clean and still be wrong.

Standard accounting practice addresses this with a variance analysis step before sign-off, comparing the current period against the prior month, the same month a year earlier, and the budget, with any unexplained movement investigated before the books are considered final. That discipline matters more, not less, once multiple disconnected source systems are involved.

Building a Data Reconciliation Discipline, Not Just a Dashboard

Facilities that avoid getting surprised by mismatched numbers build a few specific habits into how they manage data across systems, independent of which reporting tool eventually displays it. If you’re mapping who owns what across these systems, our guide to streamlining operational workflows in skilled nursing is a useful companion.

Name an owner for each source system, not just for the dashboard. PBJ staffing accuracy belongs to payroll or HR. MDS timeliness and accuracy belong to the clinical and nursing team. Billing accuracy belongs to the business office. A dashboard owner who did not generate any of the underlying data cannot be the person accountable for fixing a mismatch between two source systems they don’t control.

Build a cross-check into the calendar, not just into the reporting tool. Because PBJ submissions are due 45 days after each fiscal quarter closes, that same window is the natural checkpoint to confirm MDS-derived census data for the quarter looks reasonable against internal admission and discharge records, before the deadline rather than after a Five-Star rating drops.

Treat a dashboard as an alert layer, not a source of truth. A well-built dashboard should make disagreement between systems visible, flagging when two sources that are supposed to describe the same fact don’t match, rather than quietly averaging or defaulting to whichever system happened to load first.

Set a materiality threshold for investigation. Not every small variance needs a root-cause investigation. A defined threshold, whether a dollar amount, a percentage, or a specific data field like census count, tells staff when a mismatch is normal noise and when it needs to be chased down before it becomes a denied claim, a misreported staffing quarter, or a closed set of books that has to be reopened.

Revisit the reconciliation process whenever a source system changes. A new EHR, a new payroll vendor, or a change in how agency staffing invoices arrive each month resets the specific failure points worth checking for, since the old reconciliation habits were built around the old system’s quirks, not the new one’s.

Give clinical leadership visibility into the financial side of their own data. The MDS coordinator submitting an assessment a few days late rarely knows that timing feeds directly into a public staffing rating and, by extension, referral volume and payer perception. Closing that visibility gap, so clinical staff understand the downstream financial stakes of their own submission timing, tends to fix more mismatches than any dashboard feature does, because it moves the fix upstream to where the data actually originates.

How LTCPro Supports Data Integrity Across Systems

LTCPro works across payroll, billing, and financial reporting specifically to catch the gaps between systems before they become a denied claim, a misreported PBJ quarter, or a financial close that has to be reopened. That includes reconciling staffing hours against census data ahead of PBJ deadlines, checking clinical documentation against what actually gets billed, and running variance analysis on financial close numbers pulled from multiple source systems rather than treating each system’s own export as automatically correct.

Frequently Asked Questions

Why doesn’t CMS use PBJ-submitted census data for staffing calculations?

CMS made census data entry into PBJ optional and instead derives daily resident census from Minimum Data Set assessments, a separate clinical submission system. This means Hours Per Resident Day, the figure behind a facility’s Five-Star staffing rating, is calculated by combining two independently submitted data sources rather than one unified staffing-and-census file.

Can a facility have accurate payroll data and still get an inaccurate Five-Star staffing rating?

Yes. Because the census half of the HPRD calculation comes from MDS rather than from payroll, a facility’s PBJ staffing submission can be completely accurate while an MDS-derived census error, from a late assessment or an undocumented discharge, still produces an incorrect public staffing rating.

What is the most common cause of PBJ data errors?

Contract and agency staffing hours are one of the most frequent failure points, since a facility’s own payroll system often has no direct visibility into agency shifts until an invoice arrives after the fact. Reporting scheduled hours instead of actual hours worked and misapplying PBJ’s midnight shift-date cutoff are two other common sources of mismatch.

Does a centralized reporting dashboard fix data reconciliation problems on its own?

No. A dashboard that pulls from multiple source systems displays whatever those systems report, including any disagreement between them. Without a reconciliation step built into how data moves from source systems into the dashboard, the tool can present mismatched numbers cleanly formatted, which makes the underlying error easier to miss, not harder.

Does this data reconciliation approach need to account for differences between U.S. states?

The federal PBJ and MDS systems operate the same way nationwide, so that specific reconciliation point applies consistently across the United States. Financial close reconciliation, however, often needs to account for state-specific Medicaid cost report timing and requirements, which vary enough that a multi-state operator should build state-specific checkpoints into its close process rather than relying on a single national calendar.

LTCPro provides revenue cycle management, medical billing and accounts receivable, prior authorization, accounts payable, payroll, and bookkeeping services for skilled nursing and assisted living facilities across the United States, backed by proprietary long-term care financial software.