Where the Record Breaks

How to read this paper

Every senior housing & care operator runs on a set of facts: who lives in the building, what care each resident needs, what was billed, who worked and what the owner sees. When those facts agree, leaders decide; when they don't, leaders reconcile. This paper names the ten places operating records most often fail to agree, why each persists, and what the executive team should govern first.

The ranking is SeniorCRE's judgment of combined consequence, drawing on operator surveys, NIC MAP data, state survey findings and REIT disclosures; it is not a statistical study. Prepared by the SeniorCRE Research Team; see the Methodology note.

Reading paths by role

The price of a wrong number has never been higher

The industry knows its information is fragmented. The expensive problem is narrower: two systems answer the same question differently, and no one has the authority to say which is right.

Senior housing & care is entering its most demanding operating period in a generation. Occupancy across NIC MAP's 31 primary markets reached 90.4% in the third quarter of 2026, the first reading above 90% since before the 2008 financial crisis and the 21st consecutive quarter of improvement [1]. Institutional owners are moving more of their portfolios into operating structures where they carry the results directly, and they are adding operators quickly [2, 3]. Residents arrive with higher needs: 81% of executives in a 2026 survey of 510 professionals said residents' care needs have increased [4], and nearly two-thirds of clinical leaders believe acuity is underreported in their own communities [5].

Each of these trends raises the cost of a wrong number. In a tight market, the decisions that matter turn on the last few points of occupancy. Under an operating structure, the owner sees what the operator reports, every month. As acuity rises, a care record that lags the resident becomes a revenue gap and a safety exposure at once.

Operators already know their information is scattered. In Argentum's 2025 technology survey, 77% of senior leaders ranked interoperability among their top three issues in implementing technology [6], and in the same survey only 26% said their organization had a clear and shared understanding of how to define and measure resident health and wellness outcomes [6]. But the problem leaders actually feel is narrower and more expensive than "data silos." It shows up at the moment two systems give different answers to the same question (how many residents, what level of care, what we billed, what the owner sees) and nobody holds the authority to say which answer is official.

Four findings

1. Records break at handoffs, not inside systems. Each system is usually right about its own job. The failures concentrate where a fact has to travel: from care to billing, from sales to clinical, from payroll to census, from operator to owner.

2. The most expensive failures are about definitions, not technology. An integration moves a number between systems. It does not decide which number is official. Without a governed definition, a faster interface simply delivers the disagreement faster.

3. Artificial intelligence raises the stakes. An AI model asked to find at-risk residents or explain a margin variance will read whatever records it is given. If those records disagree, the model produces confident answers from contested facts. Coherence has to come before intelligence.

4. The failures are already being litigated. Courts, juries and regulators have treated mismatches between assessments, staffing, billing and reporting as consumer fraud, elder financial abuse, negligence and securities matters. Section 02 cites the cases.

What leaders should do

• Treat the ten facts in this paper as executive property: name an owner and a written definition for each.
• For every fact held in more than one system, decide which source governs and what happens when sources disagree.
• Surface disagreements daily, in the department that created them, instead of discovering them at month-end.
• Report to the board and to owners from the same governed numbers used to run the business.
• Govern first. Replace a system only when it is the right choice, not because governance seemed to require it.

Exhibit 1The ten places the record breaks, at a glance

Source: SeniorCRE Research Team analysis. Ranking reflects judgment of combined financial, clinical, regulatory and capital consequence; it is not the result of a statistical study.

The Pattern: One Fact, Many Systems

Every system in the stack is right about its own job. The trouble starts when the business asks a question that more than one of them can answer.

A representative multi-community operator runs a CRM for sales, an electronic health record with a medication administration module for care, scheduling and timekeeping tools, a payroll system, a billing and receivables module, a general ledger and, increasingly, an owner portal or a monthly owner package. Each was bought to do one job well. None was bought to decide what the company's official numbers are.

Consider the simplest question in the business: how many residents did we serve on the last day of the month? Exhibit 3 shows how five systems can each give a different answer, and each can defend it.

Exhibit 3One question, five defensible answers: "How many residents did we serve at month-end?"

Source: SeniorCRE Research Team analysis of a representative operator configuration. Edge cases such as second occupants, hospital holds, respite stays and notice periods widen the gaps.

Five reasons records break

Definitions differ. Is a resident in the hospital still occupying the unit? Does a second occupant count as a resident, a unit or neither? Is a respite stay occupancy? Each system answers according to its purpose, and few companies have written the answer down once for all of them.

Timing differs. A move-in happens on a Tuesday, is charted on Wednesday, is billed from the first of the month and is reported at close. Event time, entry time and reporting time rarely line up.

Handoffs re-key. When a fact is typed into one system and then typed again into the next, every re-entry is a fresh chance for the two to disagree. Most disagreements start at a keyboard, not in a database.

Transitions reset. Acquisitions and operator changes import another company's definitions, history and habits. For a period after every transition, two sets of rules are in force at once.

No one holds authority. Systems have owners. Facts usually do not. When sales, clinical and finance disagree about census, there is often no rule, and no person, that settles which answer is official.

Integration moves a number. Governance decides which number is official.

The ten failures in Section 02 are all versions of this pattern. They differ in what is at stake: revenue in some, resident safety in others, the confidence of owners and lenders in the rest. Taken together, they explain why so many executive teams spend their operating reviews reconciling instead of deciding.

The Ten Places the Record Breaks

Ranked by combined consequence. Each entry sets out what breaks, why it persists, how to tell whether you have it, what to govern, the litigation it has drawn, and how SeniorCRE is designed to change it.

Read across a row and the map shows which systems carry the most risk. Read down a column and it shows where in a resident's stay the record is most fragile. Two patterns stand out. Most breaks begin in the CRM and the clinical record, where facts are first entered. Most surface at month-end, in billing and the general ledger, after the cheapest moment to fix them has passed.

1Where the record breaks · 1 of 10

Care changes that never reach billing

Felt first by: CFO · CCO

A resident's needs rise. Caregivers deliver the additional care. The invoice does not change, sometimes for months.

What breaks

A resident begins to need more help with bathing, transfers, medication management or escorts to meals. Caregivers provide it, and often note it. But for the community to be paid for that care, the change has to travel: from caregiver notes to a reassessment, from the reassessment to a new care level, from the care level to the rate, and from the rate to the invoice. Each step is a handoff, and at each handoff the change can stall. The resident receives the care; the community does not bill for it.

Why it persists

Acuity is rising faster than the records that describe it. In the 2025 clinical leadership report from the American Seniors Housing Association and August Health, nearly two-thirds of clinical leaders said acuity is underreported, and pointed to overdue assessments and intentional under-scoring as the main reasons; the report links under-scoring to avoiding rate-increase conversations with families [5]. The billing team cannot bill what the assessment has not scored. The nurse rarely sees the billing consequence of a late reassessment. Executive dashboards report care mix from whichever system they read. Many EHR and billing vendors now market tools aimed specifically at this gap, which is itself a measure of how common it is.

Exhibit 6Illustrative arithmetic: what a slow care-to-billing handoff can cost

Source: SeniorCRE Research Team illustration. All inputs are hypothetical and chosen for clarity; this is not a benchmark or a measured result. Substitute your own care-level pricing, change frequency and lag.

The arithmetic captures only the revenue. The same lag means caregivers are working from a plan that no longer describes the resident, which is a care-quality and survey exposure in its own right (see No. 06).

Signals you have it

• Care-level mix stays flat while caregivers report heavier workloads
• Reassessments cluster around annual dates rather than changes in condition
• Care charges are added retroactively, or credited after family disputes
• Direct-care hours per resident rise while care revenue per resident does not

What to govern

• Define the documented events that require a reassessment, and the days allowed
• Make the assessment the governing source for care level; billing reads it rather than re-keying it
• Measure the lag from change in care to change in rate, by community, monthly
• Separate clinical scoring from pricing conversations, so under-scoring is visible as a decision

Litigation risk

When fees are priced from resident assessments but staffing and billing do not follow those assessments, residents have sued under California consumer-protection and elder-financial-abuse statutes, arguing they paid for assessed care that the operator's own records show was never resourced.

• Winans v. Emeritus Corp. (N.D. Cal., No. 3:13-cv-03962). Residents alleged assessment-based fees were charged while staffing was set without reference to the assessments. Settled for $13.5 million in 2016. [L2]
• Carnes v. Atria Senior Living, Inc. (N.D. Cal., No. 3:14-cv-02727). Similar allegations about how care assessments were used in setting staffing. Settled for $6.4 million in 2016. [L3]
• Lollock v. Oakmont Senior Living, LLC (Alameda Cnty. Super. Ct.). Residents alleged care-point fees did not correspond to the staff time actually provided. Settled for $9 million. [L1]

Record lesson If care level sets the price, the record must show the same care level setting staffing and billing.

Outcomes to measure

• Days from a change in care to a change in rate, by community
• Share of reassessments triggered by events rather than the calendar
• Retroactive care charges and credits issued after family disputes

Illustration: in Exhibit 6, cutting the lag from 45 days to 5 reduces unbilled care from $64,800 to $7,200 a year per community, and from $1.62 million to $180,000 across 25 communities. This is arithmetic on hypothetical inputs, not a measured result.

2Where the record breaks · 2 of 10

Occupancy and census

Felt first by: CFO · CEO

The most important number in the business exists in several versions, and the operating meeting spends its first twenty minutes choosing one.

What breaks

Sales counts signed move-ins. Clinical counts residents in the building. Billing counts paying residents. Finance counts occupied units at close. Second occupants, hospital stays, respite residents, notice periods and pending move-ins are handled differently by each. Every department arrives at the operating review with its own percentage, and the board deck and the owner report may carry two more.

Why it matters now

When occupancy was recovering, a point of disagreement was noise. It is not noise now. NIC reports that six in ten properties across its primary markets operate above 90% occupancy [13], and the sector average cleared 90% in the third quarter of 2026 [1]. At these levels, the decisions that matter, such as pricing, waitlists, unit turns and staffing, turn on the last few points. Occupancy is also the denominator in almost every other ratio the company reports: revenue per occupied unit, labor per resident day, care mix. A disputed census contaminates all of them.

Exhibit 8Illustrative arithmetic: what one point of occupancy disagreement represents

Source: SeniorCRE Research Team illustration with hypothetical inputs; not a benchmark. The figure is the revenue whose existence depends on which system is asked, not a loss estimate.

Signals you have it

• Different occupancy figures in the operating review, board deck and owner report
• A standing footnote explaining "physical" versus "financial" occupancy
• Month-end census adjusted after the books close
• Sales shows a unit available that clinical shows occupied

What to govern

• One written definition of occupancy, on unit and resident bases, with every edge case decided
• A governing source for each component: presence from clinical, payment from billing
• A daily reconciliation that flags disagreements as they occur
• One number in every report, with the bridge to any other view shown rather than hidden

Litigation risk

Occupancy is a covenant in leases and loans and a headline metric for investors. A misstated occupancy figure or covenant certificate can bring securities claims as well as owner and lender defaults.

• In re Assisted Living Concepts, Inc. Securities Litigation (E.D. Wis., No. 12-cv-884). Investors alleged the company misstated its compliance with lease covenants, including a minimum 82% occupancy requirement on eight leased residences, and did not disclose uncorrected regulatory citations. The court largely denied motions to dismiss; the case settled for $12 million in 2013. [L4]

Record lesson The occupancy number in a covenant certificate must be the governed number, with its definition and trail.

Outcomes to measure

• One occupancy figure across the operating review, board deck and owner report
• Census exceptions open longer than 48 hours
• Census adjustments made after the books close

What changes in Exhibit 8: the one-point gap stops depending on which system is asked, because one governed definition answers the question.

3Where the record breaks · 3 of 10

Owner and REIT reporting through operator transitions

Felt first by: CEO · CFO

Owners now carry operating results directly. They receive those results in each operator's own definitions, and transitions reset the history.

What breaks

Under operating structures, an owner bears the property's full financial results rather than collecting fixed rent, so the owner depends on the operator's monthly reporting. Owners are adding operators quickly. Ventas reported 44 operators in its operating portfolio in the first quarter of 2026, supported by an internal data platform [2]. LTC Properties described building a data science team as its operating partners grew to 11, nine of them new since it launched the segment in 2025 [3]. Each operator delivers performance in its own definitions, formats and timing. When communities move from one operator to another, the incoming operator inherits records built under someone else's rules; one REIT attributed lower-than-expected occupancy and revenue in a recent quarter to an ongoing portfolio transition [8].

Why it persists

Owner packages are typically assembled by hand at month-end from the operator's systems, in spreadsheets that sit outside the systems of record. Definitions such as occupancy, revenue per occupied unit, labor and NOI adjustments vary by operator and sometimes by community. Few transitions are run as data projects, so history rarely survives them intact. For operators, the consequence is strategic: reporting that an owner can trust is becoming a deciding factor in who wins and keeps management contracts.

Signals you have it

• The owner asks the same clarifying questions every month
• Prior months are restated after the package is sent
• The owner package is built in a spreadsheet no system can reproduce
• Data after a transition is treated as unreliable for a quarter or more

What to govern

• Agree metric definitions with each owner in writing, at contract
• Produce owner reporting from the same governed numbers used internally
• Keep lineage so any reported figure can be traced to its source records
• Run every transition as a data project: definition mapping, opening balances, history cutover

Litigation risk

Owner and operator disputes turn on reporting, audit access and how costs are charged to properties. When an owner cannot verify what it receives, the usual remedy is litigation over termination of the management agreement.

• HCP, Inc. v. Sunrise Senior Living, Inc. (Del. Ch., complaints filed June 29, 2009). HCP and its tenants alleged breaches of management agreements on 64 communities, including failing to provide financial and marketing reports needed to monitor performance, charging inappropriate pass-through costs and obstructing contractual audits. Sunrise denied material breach and counterclaimed. The dispute was resolved when HCP paid $50 million for revised contracts that limited certain fees. [L5]

Record lesson Reporting and audit rights are only as strong as the operator's ability to produce a traceable record.

Outcomes to measure

• Owner clarification requests per month
• Restatements of previously reported months
• Time to a reliable owner package after a transition

4Where the record breaks · 4 of 10

Medication records

Felt first by: CCO

This is where record failures touch residents most directly, and where regulators look first.

What breaks

Orders arrive from outside physicians by fax, phone and portal. Pharmacies dispense. Staff administer and document. The current order for a given resident can exist in several places, and they do not always agree. Minnesota's 2026 summary of common assisted living survey findings includes current orders that were missing or unavailable, expired orders not renewed, incomplete administration documentation, missing staff signatures or credentials, and refusals or missed doses left undocumented [9]. The pattern is national. An analysis of state survey data covering 38,080 assisted living and memory care communities in 43 states found that 45% received at least one deficiency in 2023 and 2024 [10]. In Washington, where 72% of facilities were cited, medication and treatment requirements and assessment practices led the list [10], and the state provider association summarized its 2025 citation trends in two words: medications and staff files [11].

Why it persists

Medication changes enter through many channels, and returns from the hospital bring new orders that must be reconciled against the old ones. Research summarized by the Agency for Healthcare Research and Quality finds that up to 90% of patients experience at least one medication discrepancy moving from hospital to home, and that those patients are almost twice as likely to be readmitted within 30 days [12]. Staffing pressure compounds the risk: as acuity rises, communities have expanded medication technician roles and added staff to absorb the work [5], while only 12% of caregivers and community leaders say they always have enough team members on shift to provide the care residents deserve [4]. Medication governance is a clinical-safety discipline first. It should be owned by a licensed clinical leader, and no technology decision should come before the clinical one.

Signals you have it

• Survey citations for medication documentation
• Recurring discrepancies between pharmacy records and the administration record
• Order changes processed by more than one path
• Blank administration entries with no documented reason

What to govern

• One authoritative current order per resident, with a defined path for every change
• Named responsibility for who receives, verifies and enters each change, and how fast
• Medication reconciliation at every transition of care
• Refusals and omissions recorded as events, never left blank

Litigation risk

Medication record failures carry some of the most severe liability in the sector. Elder-abuse statutes in several states allow punitive damages, and the records themselves become the central evidence.

• Eskaton FountainWood Lodge verdict (Sacramento Cnty. Super. Ct., 2019). A jury awarded $42.5 million, including $35 million in punitive damages, after the death of a memory care resident; plaintiffs alleged repeated administration of a sedative, and the operator's chief executive acknowledged a medication error. [L6]
• PruittHealth verdict (Gwinnett Cnty. State Ct., Ga., August 2026). A $23.6 million verdict in a nursing home case alleging unprescribed medication was given. The company disputes the verdict and is pursuing post-trial and appellate options. Documentation of medication access and administration was central to the trial. [L7]

Record lesson When the order, the pharmacy record and the administration record disagree, the disagreement becomes the evidence.

Outcomes to measure

• Order discrepancies caught before the next medication pass rather than at survey
• Time from an order change to a reconciled record
• Medication documentation citations

5Where the record breaks · 5 of 10

Labor against census

Felt first by: COO · CFO

Labor is the largest controllable cost, and every labor ratio divides by a census figure that may be disputed.

What breaks

Labor per resident day, overtime and agency usage are calculated from payroll and time records divided by census. Every census disagreement (No. 02) therefore becomes a labor disagreement. Schedules built on a planning census that lags the real one either overstaff, which costs margin, or understaff, which costs care. When acuity is under-recorded (Nos. 01 and 06), staffing cannot be matched to actual need at all.

Why it persists

Compensation averaged 56.2% of operating budgets among providers in Ziegler's 2026 workforce survey, and the share of respondents using agencies often for direct-care vacancies rose from 16.8% to 24.8% in a year [7]. The staffing model is being rebuilt while its denominator is contested. In the 2025 clinical leadership report, 43% of communities had added caregivers and 38% had expanded medication technician roles in response to higher acuity [5]. Scheduling, timekeeping and payroll systems each categorize labor their own way, and the census used for staffing often lives in a spreadsheet that no other system reads.

Signals you have it

• Labor per resident day differs between operations and finance reports
• Schedulers maintain their own census
• Agency shifts are approved without visibility into census or acuity
• Overtime is explained after the pay period closes, not during it

What to govern

• One governed census as the denominator for every labor ratio
• Labor categories defined once across scheduling, timekeeping and payroll
• Acuity-informed staffing standards approved jointly by operations and clinical leaders
• Variances surfaced during the pay period, while they can still be corrected

Litigation risk

This is the most active class-action theory against operators: staffing set by labor budgets rather than assessed resident needs, while contracts and marketing tie care to assessments. Claims have been brought under unfair-competition, consumer-remedies, elder-financial-abuse and disability-access laws.

• Heredia v. Sunrise Senior Living, LLC (C.D. Cal., No. 8:18-cv-01974; 9th Cir., No. 22-55332). Residents alleged staffing was driven by budgets rather than assessments. A class was certified, the Ninth Circuit affirmed, and an $18.2 million settlement was approved in 2024. [L8]
• Morrison v. Aegis Senior Communities LLC (King Cnty. Super. Ct., Wash., No. 18-2-06326-4-SEA). Residents alleged staffing to predetermined labor budgets regardless of assessed care points. A $16.25 million settlement received final approval in 2021. [L9]
• Stiner v. Brookdale Senior Living, Inc. (N.D. Cal., No. 4:17-cv-03962). The court held the Americans with Disabilities Act applies to assisted living (2019). After class certification was partly denied in 2023 and subclasses later certified, an injunctive-relief settlement addressing staffing and other practices was approved in 2025. [L10]

Record lesson Courts look for evidence that the census and acuity in the record actually drove the schedule.

Outcomes to measure

• Agreement between operations and finance on labor per resident day
• Agency hours approved with census and acuity visible
• Overtime explained within the pay period

6Where the record breaks · 6 of 10

Assessments and service plans that lag a change of condition

Felt first by: CCO · COO

The resident changes on a Tuesday. The plan caregivers work from may not change for months.

What breaks

A fall, a hospital stay, a new diagnosis or a cognitive decline should trigger a reassessment and an updated service plan. Often it does not. State survey findings include reassessments that were missing or incomplete [9], and assessment practices rank among the most cited deficiency areas in state data [10]. Hospital returns compound the problem: discharge information is often incomplete, and medication discrepancies at that transition are common [12]. Each return becomes a reset of the assessment, the plan and the medication list at once.

Why it persists

Assessments are scheduled by calendar while change happens by event. Hospital discharge information arrives late or incomplete. Reassessments compete with direct care for scarce nursing time. This is the clinical twin of No. 01: the same lag that leaves care unbilled leaves caregivers working from a plan that no longer describes the resident in front of them.

Signals you have it

• Service plans dated only at annual review
• Hospital returns without a documented reassessment
• Repeat incidents for residents whose plans have not changed
• Families report changes the record does not reflect

What to govern

• Event-based triggers (fall, hospitalization, new order, behavior change) with set response times
• A return-from-hospital protocol with a reconciliation checklist
• A weekly list of residents whose condition changed but whose plan did not
• A named clinical owner for plan currency in every community

Litigation risk

When the plan lags the resident, negligence and elder-abuse claims follow, and the gap between the documented condition and the actual one becomes the plaintiff's case.

• Boice v. Emeritus Corp. (Sacramento Cnty. Super. Ct., 2013). A resident with Alzheimer's disease developed pressure wounds. The jury found elder abuse and wrongful death and awarded $23 million in punitive damages; the trial judge upheld the verdict and added $4.3 million in fees. The company said it would appeal. [L11]

Record lesson The date a record changes is itself evidence.

Outcomes to measure

• Residents whose plan predates their last change in condition
• Days from a hospital return to a documented reassessment
• Repeat incidents for residents on unchanged plans

7Where the record breaks · 7 of 10

The move-in and move-out handoff

Felt first by: COO · CFO

This is where most census disagreements begin: the same resident typed into three systems by three departments.

What breaks

Move-in date, apartment, rate, care level, deposit, payer and responsible party are entered in the CRM by sales, re-entered in the EHR by clinical staff, and re-entered in billing by the business office. Each re-entry can disagree with the last. Move-outs and transfers are harder still: notice dates, final charges, deposit refunds and unit-turn status all have to agree across the same systems, at a moment when the family is often under strain.

Why it persists

The handoff is owned by no one in particular. Sales considers its work finished at signature, clinical at admission, and the business office at first invoice. Each system's record is complete for its own purpose, so no one sees the disagreement until census, billing or a family dispute exposes it.

Signals you have it

• Residents billed from a different date than they moved in
• Deposits left unapplied or refunded late
• Units shown as available in sales while still occupied
• Disputed move-out charges

What to govern

• One move-in record with a single effective date
• An enter-once rule: each fact entered by its owner and read by everyone else
• A daily move-in and move-out reconciliation across CRM, EHR and billing
• Unit status treated as a governed field, not a sales convenience

Litigation risk

Admission paperwork decides who is bound, to what terms and in which forum. If the authority of the person signing is not documented, key terms can fail when they are needed most.

• Kindred Nursing Centers L.P. v. Clark (581 U.S. 246 (2017)). Family members holding powers of attorney signed arbitration agreements in admission paperwork. The Supreme Court held Kentucky's clear-statement rule violated the Federal Arbitration Act, enforced one agreement and remanded the other for a ruling on whether the power of attorney conferred the authority at all. [L12]
• Stiner v. Brookdale Senior Living, Inc. (9th Cir., 810 F. App'x 531 (2020)). The Ninth Circuit affirmed in part the refusal to compel arbitration for certain named plaintiffs, allowing their claims to proceed in the class action. [L10]

Record lesson Who signed, with what authority and on what date belongs in the governed move-in record.

Outcomes to measure

• Residents billed from a date other than their move-in date
• Deposits unapplied after 30 days
• Admission files with documented signer authority

8Where the record breaks · 8 of 10

Month-end close

Felt first by: CFO

Every upstream disagreement lands in finance at once, in the first week of every month.

What breaks

Finance closes the books from operating systems that do not agree. The first days of each month go to reconciling census to billing, billing to cash, payroll to labor reports and operating reports to the general ledger. The close is slow, and late adjustments mean executives see one number on day five and another on day twelve. Dashboards built on operating data tell a different story from the financial statements, and the CFO spends board time explaining the bridge.

Why it persists

Finance is the last line of reconciliation, so finance absorbs the cost of disagreements created everywhere else (Nos. 01, 02, 05, 07 and 10). As communities are added, the work grows with them: every new community brings more sources to reconcile without changing how reconciliation is done.

Signals you have it

• Recurring top-side adjustments that recur by type
• Days to close lengthen as the portfolio grows
• Operating dashboards and financial statements disagree
• Results are "preliminary" for longer than anyone would like

What to govern

• Move reconciliation upstream: flag disagreements daily, assigned to the department that created them
• Governed cut-offs and a close calendar every community follows
• One set of definitions shared by operating and financial reporting
• Track the count and value of post-close adjustments as a management metric

Litigation risk

Reconciliation failures that reach the financial statements become restatements, and restatements draw regulators and securities plaintiffs.

• In re Sunrise Senior Living, Inc. Securities Litigation (D.D.C., No. 07-cv-00102). After accounting errors led the company to restate prior years' financial statements and the SEC opened a formal investigation, investors alleged securities violations based on accounting practices and disclosures, along with stock-option claims. The case settled in 2009, with insurers paying $13.4 million. [L13]

Record lesson Post-close adjustments are the early warning; a restatement is the expensive version.

Outcomes to measure

• Days to close
• Count and value of post-close adjustments
• Items in the monthly bridge between operating and financial reports

9Where the record breaks · 9 of 10

Incident and staff compliance records

Felt first by: CCO · COO

Surveyors review these records together. Most operators keep them apart.

What breaks

Incident reports, change-of-condition notifications, staff training, background checks, licenses and certifications live in different places: incident software, HR systems, learning platforms and, still, paper binders. Staff files are a leading source of citations: Washington's provider association named medications and staff files as the two dominant themes of 2025 assisted living citations [11], and staff training and hiring requirements recur in state deficiency data [10]. An incident that should trigger a reassessment (No. 06) often sits in a system that the care plan never reads.

Why it persists

Compliance records are owned by many people and governed by none. Turnover means constant onboarding, and credentials expire on individual dates rather than on a schedule anyone controls. The result is that readiness is discovered at survey rather than managed in advance.

Signals you have it

• Survey preparation becomes a scramble
• Expired credentials are found by surveyors, not managers
• Incident follow-up is not linked to care plan changes
• Compliance status is known community by community, not across the portfolio

What to govern

• A defined set of required records per employee and per resident event
• Expirations surfaced well ahead of their dates
• Incident, reassessment and plan update linked as one chain
• Compliance status reported to executives monthly

Litigation risk

Negligent hiring and retention claims turn on whether required screening was done and can be shown to have been done.

• Assisted living negligent-hiring verdict (Newport News, Va., 2009). A jury awarded $750,000 against an assisted living home and its owner after a staff member with a prior criminal record was accused of assaulting a resident; the suit alleged a failure to use reasonable care in hiring. [L14]

Record lesson If the background check, training and credential are not in the record, a jury may conclude they did not happen.

Outcomes to measure

• Credentials that lapse without notice
• Incidents without a documented reassessment decision
• Staff time spent preparing for survey

10Where the record breaks · 10 of 10

Rates, concessions and receivables

Felt first by: CFO · CEO

Owners increasingly ask for actual rate growth after concessions. Most operators can produce it only by hand.

What breaks

The rate a resident agreed to, the concessions granted at move-in, the annual increase, the care charges and the balance owed live in agreements, CRM notes, billing tables and receivables aging, and they are rarely reconciled to one another. A concession promised in sales may be coded differently, or not at all, in billing. Rate increases are applied with exceptions that no report captures. Delinquency is defined and pursued differently at each community.

Why it persists

Rates touch sales, finance and community leadership, and each keeps its own view. Without a single governed rate record per resident, in-place rate growth, concession use and collection performance must be rebuilt for every owner request, every budget cycle and every diligence process.

Signals you have it

• Sales and finance calculate in-place rate differently
• Concessions are discovered at audit or diligence
• Bad-debt surprises at year-end
• Collection policies applied unevenly across communities

What to govern

• One rate record per resident: base, care, concessions and effective dates
• Concession approvals logged and coded consistently
• Rate-growth and concession reporting produced from that record
• One definition of delinquency, reviewed on one standard

Litigation risk

Fee and refund practices draw elder-financial-abuse and consumer claims, especially over how costs are allocated into monthly fees and whether refundable amounts are protected.

• Richter v. CC-Palo Alto, Inc. (N.D. Cal., No. 5:14-cv-00750). Residents of a continuing care community alleged the operator transferred refundable entrance fees to its parent company and inflated monthly fees by allocating improper costs. Early complaints were dismissed with leave to amend; claims were repleaded and a class was certified in 2021. [L15]

Record lesson Every fee, concession and refund obligation needs one governed record that a resident, auditor or court can follow.

Outcomes to measure

• One in-place rate growth figure that sales and finance both use
• Concessions discovered at audit or diligence
• Time to answer an owner or lender rate request

Govern Before You Automate

Every one of the ten failures has the same shape: a fact that matters, more than one system holding a version of it, and no rule for which version governs.

The usual responses each help, and none of them settles the authority question on its own. A dashboard makes the disagreement visible. An integration moves it faster. A data warehouse stores every version in one place. Replacing a system trades one set of handoffs for another and adds a transition along the way. Exhibit 9 sets out what each response does and does not resolve.

Exhibit 9What common fixes resolve, and what they leave open

Source: SeniorCRE Research Team analysis.

Seven principles of a governed record

1. Name the questions that matter. Start with the ten in this paper. A company that cannot list its governed facts does not have any.

2. Define each answer once, in writing. Occupancy, care level, labor per resident day, in-place rate. Decide every edge case once, so no system decides it by default.

3. Assign authority between sources. For each fact, decide which source governs and what happens when sources disagree. The rule belongs to the operator, not to whichever vendor's screen is open.

4. Keep the trail. Every reported number should be traceable to the records it came from, the rule that produced it and the person who approved any exception.

5. Surface disagreement early. Flag conflicts daily, in the department that created them, rather than leaving finance to discover them at close.

6. Keep people accountable for decisions. Software can recommend. Named leaders authorize. That matters more, not less, as AI enters the workflow.

7. Own the record. The operator, not the vendor, should control the definitions and be able to take the governed record, with its rules and history, wherever it goes.

Coherence before intelligence

Operators are optimistic about AI: 76% of senior leaders in the Argentum survey expect AI to have a significantly transformative or positive effect on managing communities over the next five years [6], and 88% of clinical leaders see earlier detection of at-risk residents as its most valuable outcome [5]. Both ambitions depend on the record. A model asked which residents are at risk will read whatever it is given. If the assessment lags the resident (No. 06) and the census disagrees with itself (No. 02), the model will be confidently wrong, and it will be wrong faster and at greater scale than any person. The order of operations matters.

AI in senior housing & care only earns trust when the operator governs the record it runs on.

The Executive Agenda

Governing the record is not an IT project. Each of the ten facts belongs to an executive, and each executive has three commitments to make.

The First 90 Days

Start with the denominators. Census and labor sit underneath nearly every ratio the company reports, so governing them first improves everything built on them.

The sequence below is a recommended approach for any operator, whatever systems it runs. It describes what the executive team should accomplish, not a product or service commitment. The aim of the first 90 days is not to govern all ten facts. It is to govern two of them completely, prove the method in one month-end close, and decide what comes next from evidence rather than from a plan.

Why census and labor first, and what comes next

Census is the denominator for revenue per occupied unit, care mix and labor per resident day, and labor is the largest controllable cost. Governing both first makes Nos. 02, 05 and 07 tractable immediately and removes much of the reconciliation that lands on finance at close (No. 08). With the denominators governed, most operators turn next to care level and rate (Nos. 01 and 10), which carry the most direct revenue consequence, and to owner reporting (No. 03), which rests on all of the above. Clinical records (Nos. 04, 06 and 09) follow the same principles under clinical ownership, on a timeline set by licensed clinical leaders rather than by finance or technology.

The Record Integrity Index

Twenty questions, two for each of the ten. Each "yes" is worth five points, for a score out of 100. Answer for the company as a whole, not for its best community.

About SeniorCRE and What This Paper Does Not Claim

About SeniorCRE

SeniorCRE builds operator-controlled operating infrastructure for senior housing & care. Its central construct, the Operator-Controlled Operating Record, is designed to sit above an operator's existing systems and govern the facts this paper describes: definitions written once, authority between sources set by the operator, lineage kept for every figure and disagreements surfaced as they occur, across care, census, labor and revenue.

Operators choose how to deploy it. Some govern the systems they already run; others choose to run selected domains on SeniorCRE. Govern first. Replace only by choice. In either case the operator, not the vendor, holds authority over the record, a principle SeniorCRE calls the Operator Authority Chain™, and SeniorCRE publishes its data-governance commitments, including data portability, at seniorcre.com/open-data-governance.

Request a 30-Minute Assessment

A working session with your executive team on where your record breaks, starting with the diagnostic in Section 06.

seniorcre.com

What this paper does not claim

• It does not report SeniorCRE results. No figure in this paper is a measured SeniorCRE outcome.
• Illustrative arithmetic (Exhibits 6 and 8) uses hypothetical inputs. It is not a benchmark, an estimate for any operator or a promise of savings.
• Composite scenarios are illustrations built from common industry patterns. They do not describe any specific operator, resident or event.
• The ranking of the ten is SeniorCRE's judgment, informed by the sources cited. It is not the result of a statistical study.
• Third-party survey and market figures are reported as published by their sources and checked against the original publications where they were accessible. SeniorCRE has not audited the underlying survey data.
• Where SeniorCRE capabilities are described, including the "With SeniorCRE" panels in Section 02, they are described by design intent. Nothing here states that a capability is in production at any operator, and the outcomes listed are measures to track, not results SeniorCRE has demonstrated.
• It does not provide clinical, legal, regulatory or financial advice. Medication and clinical-record practices should be set by licensed clinical leaders under applicable state rules.

Methodology

How this paper was prepared

Prepared by the SeniorCRE Research Team. This note sets out the scope, the evidence, how the ten were ranked, and the limits of what the paper can show.

Scope and question

The paper asks one question: where do operating records in senior housing & care most often fail to agree, and what should the executive team govern first? It covers assisted living, memory care and independent living operators, and the institutional owners who rely on their reporting. Evidence from other settings, such as research on hospital-to-home transitions, is used only where it bears directly on these operators.

Evidence base

The team drew on four kinds of evidence, all public:

• Operator and clinical surveys: Argentum and A Place for Mom (2025 technology report); Sage (The State of Care 2026; 510 respondents); the American Seniors Housing Association and August Health (2025 clinical leadership report; 68 clinical leaders); Ziegler (2026 CFO Hotline workforce report).
• Market data: NIC MAP occupancy for the 31 primary markets, as published by NIC and in trade coverage.
• Regulatory and research sources: the Minnesota Department of Health's 2026 survey findings; a national analysis of state survey data for 38,080 assisted living and memory care communities in 43 states, published by the Washington Health Care Association; that association's 2025 citation trends; and research summarized by the Agency for Healthcare Research and Quality on medication discrepancies at care transitions.
• Owner disclosures: public REIT earnings calls and coverage of REIT results, used only for facts the companies reported themselves.

To this the team added its own analysis of where operating facts cross systems in a representative multi-community operator. That analysis underlies Exhibits 3 and 4.

How the ten were identified and ranked

Candidate failure points were identified wherever a fact the business reports must pass between two or more systems, and then tested against the evidence above. The ten retained are those with a clear executive owner and support in published evidence or in the team's systems analysis. Nos. 07 (move-in and move-out), 08 (month-end close) and 10 (rates and receivables) rest primarily on that analysis, because public data on them is thin; they are the entries where operator interviews would add most. They were ranked by the team's judgment across four consequences: revenue and margin, resident safety and care, regulatory exposure, and the confidence of owners and lenders. The ranking is a structured judgment, not a statistical result, and reasonable operators may order the middle of the list differently.

Source standards

• Original publications were preferred. Survey figures were checked against the original report or the publisher's own release; the Argentum figures were checked against the full report, with page references in Sources.
• Where an original could not be accessed, the Sources entry says so and names the publication relied on.
• Two of the surveys are published by software companies (Sage and August Health). They are cited for their survey results only, never for claims about their products.
• Figures are reported as their sources state them, including their wording. Where a summary and a report body differ, the report body is used.
• NIC revises prior quarters; Exhibit 7 uses values as reported and may differ slightly from the current series.
• Cases were drawn from court records, SEC filings and legal and trade press, and are cited by caption and court. Each is described by what was alleged and what happened procedurally; allegations are not findings, and settlements do not admit liability. Party names come from public court records.

Illustrations and scenarios

Exhibits 6 and 8 use hypothetical inputs, labeled as such, to show scale. They are not benchmarks or estimates for any operator. Composite scenarios combine patterns common across the industry; no scenario describes a specific operator, community, resident or event. No operator data, client data or resident data of any kind were used in this paper.

Limitations

• The paper contains no original survey data. Its contribution is the framework, the ranking and the governance approach, built on published evidence.
• The surveys differ in size, population and timing, from 68 clinical leaders to 510 professionals, and all are self-reported. Their figures should not be combined or compared directly.
• State survey data vary by state in frequency, categories and enforcement, so national comparisons are indicative rather than precise.
• The cases illustrate legal theories; they are not a survey of litigation. Many come from California, whose elder-financial-abuse and consumer statutes drive class actions, and some come from nursing homes, an adjacent setting. Outcomes may change on appeal.
• Figures are current as of November 9, 2026.

Disclosure of interest

SeniorCRE, LLC builds operator-controlled operating infrastructure for senior housing & care, and benefits when operators invest in governing their records. The diagnosis, the seven principles, the executive agenda, the 90-day sequence and the Record Integrity Index are written to apply whatever systems or vendors an operator uses. SeniorCRE's own offering is described in Section 07 and in the panel marked "With SeniorCRE" in each of the ten entries. Those panels describe design intent and the outcomes an operator should measure; they do not report results.

How to cite this paper

SeniorCRE Research Team. Where the Record Breaks: The Ten Places Operating Information Fails Senior Housing & Care, and What the Executive Team Should Govern First. SeniorCRE, LLC, November 2026.

Sources

1. National Investment Center for Seniors Housing & Care (NIC), NIC MAP third-quarter 2026 data for the 31 primary markets, as reported by Senior Housing News (October 1, 2026) and IREI (October 2026): 90.4% occupancy; 21st consecutive quarter of improvement; first reading above 90% since before 2008. seniorhousingnews.com; irei.com
2. Ventas, Inc., first-quarter 2026 earnings call, April 28, 2026 (transcript, The Motley Fool): 44 operators in its operating portfolio, supported by an internal data platform. fool.com
3. Senior Housing News, April 23, 2026, on LTC Properties' operating-portfolio strategy: 11 operating partners, nine new since the segment's May 2025 launch; data science team build-out. seniorhousingnews.com
4. Sage, The State of Care 2026, released September 24, 2026 (PR Newswire): survey of 510 professionals (138 executive leaders; 372 community leaders and caregivers), July 9 to August 31, 2026. 81% of executives say residents' care needs have increased; 12% of caregivers and community leaders say they always have enough team members on shift. prnewswire.com; sagehealth.com
5. American Seniors Housing Association (ASHA) and August Health, 2025 clinical leadership report, 68 clinical leaders, and the publishers' release of December 9, 2025: 74% report higher acuity (50% somewhat, 24% significantly); 63% believe acuity is underreported (35% occasionally, 28% frequently), citing overdue assessments and intentional under-scoring, which the report links to managing rate-increase conversations with families; 43% added caregivers, 38% expanded medication technician roles and 32% added medication technicians; 88% see earlier detection of at-risk residents as the most valuable outcome of AI and predictive tools. augusthealth.com; businesswire.com
6. Argentum and A Place for Mom, The 2025 Argentum Technology Report (released July 1, 2025), survey of senior executives and technology leaders fielded in late 2024; nearly 1,000 leaders were invited. 77% ranked interoperability among their top three issues in implementing new technology and 74% ranked compatibility with existing infrastructure; 76% expect AI to have a significantly transformative or positive effect on community management over five years; 26% report a clear and shared understanding of how to define and measure health and wellness outcomes (pp. 5, 7, 22, 23). argentum.org
7. Ziegler, CFO Hotline 2026 workforce report (July 2026), primarily not-for-profit providers in 40 states: compensation 56.2% of operating budgets; respondents using agencies often for direct-care vacancies rose from 16.8% to 24.8%; agency labor a top staffing expense for 44.4%. ziegler.com
8. Senior Housing News, August 4, 2026, on Diversified Healthcare Trust's second-quarter 2026 results and ongoing portfolio transitions. seniorhousingnews.com
9. Minnesota Department of Health, Medication Management: Common Survey Findings (2026), assisted living and home care. health.state.mn.us
10. CarePrepare, national assisted living and memory care deficiencies analysis, published by the Washington Health Care Association (June 2025): state survey data for 38,080 communities in 43 states, January 2023 to December 2024; 45% of communities with at least one deficiency; 143,716 deficiencies cited; in Washington, 72% of facilities cited, led by medication and treatment requirements and assessment practices. whca.org
11. Washington Health Care Association, Assisted Living Citation Trends, January through July 2025: medications and staff files as the dominant citation themes. whca.org
12. Agency for Healthcare Research and Quality (AHRQ), Corbett et al., research on medication discrepancies in hospital-to-home transitions: up to 90% of patients experience at least one discrepancy; patients with discrepancies almost twice as likely to be readmitted within 30 days. ahrq.gov
13. NIC, "The Great Tightening: How Senior Housing Is Entering a New Occupancy Era," August 17, 2026: 61% of properties in the 31 primary markets operating above 90% occupancy. nic.org
14. NIC quarterly occupancy releases and coverage, 31 primary markets: 4Q24 87.1% (NIC, April 2025); 1Q25 87.3% (NIC, July 10, 2025); 2Q25 88.0% (NIC, October 2, 2025); 3Q25 88.7% (NIC, January 15, 2026); 4Q25 89.1% and 1Q26 89.5% (NIC, April 2026); 2Q26 89.8% (Senior Housing News, October 1, 2026). nic.org

Cases and Legal Sources

1. L1 Lollock v. Oakmont Senior Living, LLC, No. RG17875110 (Alameda Cnty. Super. Ct.); removed as No. 4:17-cv-05912 (N.D. Cal.) and later returned to state court. Settlement of $9 million as reported by McKnight's Senior Living. truthinadvertising.org; mcknightsseniorliving.com
2. L2 Winans v. Emeritus Corp., No. 3:13-cv-03962-SC (N.D. Cal.). Settlement of $13.5 million (2016) as reported by McKnight's Senior Living. mcknightsseniorliving.com
3. L3 Carnes v. Atria Senior Living, Inc., No. 3:14-cv-02727 (N.D. Cal.). Preliminary approval April 6, 2016; settlement of $6.4 million. jonesday.com; mcknightsseniorliving.com
4. L4 In re Assisted Living Concepts, Inc. Securities Litigation, No. 12-cv-884-JPS (E.D. Wis.). Motions to dismiss largely denied June 21, 2013; $12 million settlement approved December 19, 2013. blbglaw.com; law360.com
5. L5 HCP, Inc. and tenants v. Sunrise Senior Living, Inc. and subsidiaries, complaints filed June 29, 2009 (Del. Ch.), with a related complaint filed July 24, 2009 (E.D. Va.); allegations as described in HCP's 2009 Forms 10-Q. Resolution with a $50 million HCP payment and revised contracts as reported by the Los Angeles Business Journal. sec.gov; labusinessjournal.com
6. L6 Jury verdict of $42.5 million against Eskaton FountainWood Lodge and affiliates (Sacramento Cnty. Super. Ct., 2019), as reported by McKnight's Senior Living and the Sacramento Bee. mcknightsseniorliving.com
7. L7 Jury verdict of $23.6 million against PruittHealth entities (Gwinnett Cnty. State Ct., Ga., August 21, 2026), as reported by The Atlanta Journal-Constitution and Expert Institute. ajc.com; expertinstitute.com
8. L8 Heredia v. Sunrise Senior Living, LLC, No. 8:18-cv-01974 (C.D. Cal.); class certification affirmed, No. 22-55332 (9th Cir.); $18.2 million settlement approved August 2024. uschamber.com; law360.com
9. L9 Morrison v. Aegis Senior Communities LLC, No. 18-2-06326-4-SEA (King Cnty. Super. Ct., Wash.); $16.25 million settlement, final approval August 23, 2021. topclassactions.com
10. L10 Stiner v. Brookdale Senior Living, Inc., No. 4:17-cv-03962-HSG (N.D. Cal.): 354 F. Supp. 3d 1046 (N.D. Cal. 2019); 810 F. App'x 531 (9th Cir. 2020); 665 F. Supp. 3d 1150 (N.D. Cal. 2023); injunctive-relief settlement approved 2025. rbgg.com; leagle.com
11. L11 Boice v. Emeritus Corp. (Sacramento Cnty. Super. Ct.): verdict March 2013, upheld June 2013 with $4.3 million in fees and costs, as reported by Senior Housing News, McKnight's and Insurance Journal. seniorhousingnews.com
12. L12 Kindred Nursing Centers L.P. v. Clark, 581 U.S. 246 (2017). law.cornell.edu
13. L13 In re Sunrise Senior Living, Inc. Securities Litigation, No. 07-cv-00102-RBW (D.D.C.); restatement, SEC investigation and 2009 settlement as described in Sunrise's Forms 10-K for 2006 through 2009. sec.gov
14. L14 Jury verdict of $750,000 against an assisted living home in Newport News, Virginia, and its owner (2009), as reported by Claims Journal, June 5, 2009. claimsjournal.com
15. L15 Richter v. CC-Palo Alto, Inc., No. 5:14-cv-00750-EJD (N.D. Cal.): orders granting motions to dismiss (2014, 2015); order granting in part and denying in part motions to dismiss (April 21, 2021); order granting class certification (August 3, 2021). justia.com

Important Notices

Purpose

This paper is published for general information by SeniorCRE, LLC. It does not constitute clinical, legal, regulatory, accounting, investment or other professional advice, and it should not be relied on as such. Readers should consult qualified advisors about their own circumstances.

Clinical matters

SeniorCRE, LLC is not a healthcare provider. Operators and their licensed clinical staff remain solely responsible for clinical decisions, medication administration, resident care and regulatory compliance. References to state survey findings are illustrative of common themes and are not a statement of any state's requirements.

Data and sources

Third-party data are drawn from the public sources listed and are reported as published. SeniorCRE has not audited them and makes no representation as to their accuracy or completeness. Illustrative calculations use hypothetical inputs and do not represent the results of any operator or of any SeniorCRE product.

Litigation references

Cases are cited from public records to illustrate legal theories associated with record failures. Descriptions summarize allegations and procedural outcomes; allegations are not findings of fact, settlements do not constitute admissions of liability, and verdicts may be modified on appeal. Nothing in this paper is a statement about the conduct of any party beyond what the cited records report, and nothing in it is legal advice. Operators should consult counsel about their own exposure.

Forward-looking statements

Statements about how a governed operating record is designed to work describe design intent and are not representations of current availability, performance or results at any operator.

Trademarks and affiliation

All other product and company names mentioned are the property of their respective owners. Their mention does not imply affiliation, endorsement, partnership, integration or certification. Organizations whose data or disclosures are cited have not reviewed or endorsed this paper.

© 2026 SeniorCRE, LLC. All rights reserved.

SeniorCRE and Operator Authority Chain™ are marks of SeniorCRE, LLC.