Case Study: Home Health Agency Reduces A/R Days from 58 to 23 with AI Revenue Cycle Automation

Home health billing operates under constraints that most other healthcare settings don't face. The Patient-Driven Groupings Model (PDGM) introduced in 2020 fundamentally changed how Medicare reimburses home health services, replacing the old episodic payment system with a case-mix-adjusted model based on clinical groupings, functional levels, and comorbidity adjustments. Agencies that haven't adapted their billing operations to PDGM's requirements carry higher A/R days, lower clean claim rates, and more cash flow volatility than agencies that have optimized for the new payment model.

This case study examines how a mid-size home health agency — serving 800+ active patients across a metropolitan service area — transformed its revenue cycle by deploying AI-powered automation across eligibility verification, claims management, and payment posting. The result: A/R days dropped from 58 to 23, clean claim rate improved from 82% to 96.7%, and monthly cash collections increased 34%.

Note: Metrics in this case study represent composite outcomes based on aggregate customer data. Individual results vary based on agency size, payer mix, and baseline operational maturity.

Results at a Glance

Metric	Before	After	Change
A/R days	58	23	-60%
Clean claim rate	82%	96.7%	+14.7 pts
Monthly cash collections	$2.8M	$3.75M	+34%
Billing staff FTEs	14	8	6 redeployed
Claim rejection rate	14.3%	2.8%	-80%
Average days to first payment	41	16	-61%
Annual write-offs	$1.1M	$320K	-71%
Payment posting lag	5.2 days	0.4 days	-92%

The Challenge: 58 A/R Days and a Cash Flow Crisis

The Agency's Operating Environment

The agency provided skilled nursing, physical therapy, occupational therapy, speech therapy, medical social work, and home health aide services to a predominantly Medicare population. The payer mix was approximately 62% Medicare (traditional and Medicare Advantage), 18% Medicaid, 14% commercial insurance, and 6% other (VA, workers' comp, private pay).

The agency employed 14 billing and revenue cycle staff: 4 eligibility and intake specialists, 6 billers/claims processors, 2 payment posting specialists, and 2 A/R follow-up specialists. Total annual revenue cycle labor cost was approximately $980K.

Why A/R Days Were at 58

An A/R days figure of 58 in home health is not unusual — the industry average hovers around 45-50 — but it was high enough to create genuine cash flow pressure. The agency's analysis identified five root causes:

PDGM complexity creating claim errors. The Patient-Driven Groupings Model requires accurate assignment of clinical groupings, functional levels, and comorbidity adjustments to determine the correct case-mix weight and payment amount. The agency's billing staff made errors in PDGM classification on 11% of claims — sometimes assigning the wrong clinical grouping based on the primary diagnosis, sometimes missing comorbidity adjustments that would increase the payment, and sometimes failing to correctly identify the admission source (community vs. institutional) that determines one of PDGM's four case-mix dimensions.

Each PDGM error generated either a rejection (if the classification was clinically illogical) or an underpayment (if the classification was valid but suboptimal). Rejections added days to the revenue cycle. Underpayments were often undetected — the agency didn't have a systematic way to identify claims paid below the correct PDGM amount.

Eligibility verification gaps. The agency verified patient eligibility at the start of each 30-day billing period, but eligibility changes that occurred during the period — patients losing coverage, switching plans, or exhausting benefits — weren't detected until the claim was denied. With 800+ active patients and billing periods starting on rolling dates, the eligibility verification workload was continuous, and the manual process couldn't keep pace.

Approximately 6.2% of claims were denied for eligibility-related reasons, with each denied claim requiring investigation (why did eligibility change?), patient notification, and either rebilling to the correct payer or initiating patient financial responsibility processes. The average resolution time for an eligibility-related denial was 32 days.

Slow and inconsistent claim submission. Claims were batched and submitted weekly rather than daily. The weekly batching was a workflow artifact — the billing team needed time to compile documentation, verify coding, and prepare claims — but it added an average of 3.5 days to every claim's lifecycle. For claims submitted on the last day of the weekly batch, the delay could be seven days from service completion to claim submission.

Manual payment posting. The two payment posting specialists manually posted payments from ERA (Electronic Remittance Advice) files and paper EOBs (Explanation of Benefits). The posting lag averaged 5.2 days from payment receipt to posting, and during high-volume periods (month-end, quarter-end), the lag could extend to 8-10 days. Delayed posting meant delayed identification of underpayments, delayed secondary billing, and inaccurate A/R aging reports that hampered follow-up prioritization.

Reactive A/R follow-up. The two A/R follow-up specialists worked aged claims on a first-in, first-out basis, regardless of claim value or likelihood of resolution. A $200 claim that had been in A/R for 90 days received the same priority as a $4,500 claim that had been in A/R for 35 days. This lack of intelligent prioritization meant that high-value, recoverable claims aged unnecessarily while staff time was consumed by low-value claims that were often uncollectible.

The Cash Flow Impact

At $2.8 million in monthly revenue and 58 A/R days, the agency had approximately $5.4 million in outstanding receivables at any given time. The practical effect was a constant cash flow squeeze — the agency frequently needed its line of credit to meet payroll and operational expenses, incurring interest costs of approximately $85K annually.

The agency's CFO characterized the situation simply: "We're earning the revenue but not collecting it fast enough to run the business."

The Solution: QuickRCM End-to-End Revenue Cycle Automation

The agency deployed QuickRCM, QuickIntell's comprehensive revenue cycle management platform, configured for home health-specific billing requirements including PDGM classification, OASIS-driven clinical grouping, and episodic billing workflows.

Automated Eligibility Verification

QuickRCM replaced the manual eligibility verification process with continuous automated monitoring.

Real-time eligibility checks. Rather than verifying eligibility once at the start of each billing period, QuickRCM runs automated eligibility checks at three points: patient intake, start of each 30-day billing period, and three days before claim submission. This three-point verification catches eligibility changes that would otherwise result in denied claims.

Coverage change detection. The system monitors for coverage changes between verification points, alerting the intake team when a patient's insurance status changes. This allows the agency to update billing information, notify the clinical team, and adjust the patient's financial responsibility before services are rendered under invalid coverage.

Medicare Advantage plan identification. One of the agency's most common eligibility errors was billing traditional Medicare when the patient had enrolled in a Medicare Advantage plan — or vice versa. QuickRCM's eligibility engine distinguishes between traditional Medicare and over 600 Medicare Advantage plans, routing claims to the correct payer automatically.

The impact was immediate: eligibility-related denials dropped from 6.2% to 0.9% within 60 days of deployment.

PDGM Classification Engine

QuickRCM's PDGM engine addressed the agency's classification accuracy problem by automating the four dimensions of PDGM case-mix calculation:

Admission source determination. The system automatically classifies patients as community or institutional admissions based on claims history and admission data — eliminating the manual lookup process that was incorrect 4.3% of the time.

Clinical grouping assignment. Based on the primary diagnosis code and associated clinical documentation, QuickRCM assigns the correct clinical grouping (musculoskeletal rehabilitation, neuro rehabilitation, wounds, complex nursing interventions, behavioral health, or medication management, teaching, and assessment — MMTA). The system cross-references the primary diagnosis against PDGM's clinical grouping logic and flags cases where the assigned primary diagnosis may not optimize the clinical grouping.

Functional level scoring. Using OASIS assessment data, QuickRCM calculates the correct functional impairment level and verifies that the OASIS responses are internally consistent and supported by clinical documentation.

Comorbidity adjustment. The system identifies qualifying comorbidity diagnoses from the patient's complete problem list and ensures that secondary diagnoses that qualify for comorbidity adjustments are included on the claim.

PDGM classification accuracy improved from 89% to 98.2% — and the 1.8% error rate was concentrated in genuinely ambiguous clinical scenarios where the correct classification required clinical judgment beyond what the documentation supported.

More importantly, the system identified systematic underpayments from suboptimal primary diagnosis sequencing. In 8.7% of episodes, a different primary diagnosis — supported by the documentation and clinically appropriate — would have resulted in a higher PDGM payment. This wasn't upcoding; it was selecting the primary diagnosis that most accurately reflected the primary reason for home health services, which sometimes differed from the hospital discharge diagnosis carried forward by default.

Automated Claims Processing and Submission

QuickRCM transformed claims processing from a weekly batch operation to a continuous flow.

Daily automated claim generation. Claims were generated automatically from clinical documentation and billing data, with built-in edits checking for PDGM accuracy, diagnosis sequencing, OASIS consistency, and payer-specific requirements. Claims that passed all edits were submitted daily. Claims that failed edits were routed to billers with specific error descriptions and correction guidance.

Pre-submission claims scrubbing. Every claim passed through a multi-layer scrubbing engine: PDGM classification verification, Medicare medical review criteria, payer-specific edits, NCCI bundling checks, and historical denial pattern matching. The scrubbing engine caught errors that manual review consistently missed — particularly OASIS-to-claim inconsistencies that triggered Medicare ADR (Additional Documentation Request) reviews.

Automated secondary billing. When primary payer payments were posted, QuickRCM automatically generated and submitted secondary claims to the appropriate payer, eliminating the 8-12 day lag that previously occurred between primary payment posting and secondary claim submission.

Intelligent Payment Posting

QuickRCM's automated payment posting replaced the manual process that had created a 5.2-day posting lag.

ERA auto-posting. Electronic remittance advice files were automatically matched to claims and posted, with payment amounts validated against expected reimbursement (contracted rates or fee schedules). Payments matching expected amounts were posted without human intervention. Payments below expected amounts were flagged for underpayment review.

Variance detection. The system identified three types of payment variances: underpayments (paid below contracted rate), incorrect adjustments (payer applied incorrect contractual adjustment), and coordination of benefits errors (primary payer processed as secondary or vice versa). Each variance was quantified and routed to A/R staff with the specific discrepancy identified.

Paper EOB processing. For the declining number of payers still issuing paper EOBs, QuickRCM's OCR engine scanned and digitized the documents, matching payments to claims with 96.4% accuracy and routing the remaining 3.6% for manual review.

Payment posting lag dropped from 5.2 days to 0.4 days — effectively same-day posting for electronic payments.

AI-Prioritized A/R Follow-Up

QuickRCM replaced the first-in, first-out A/R follow-up workflow with AI-driven work queue prioritization.

Value-based prioritization. Claims were ranked by a composite score incorporating claim value, aging, payer payment patterns, and likelihood of recovery. A high-value claim at 35 days with a historically responsive payer was prioritized over a low-value claim at 90 days with a payer known for slow resolution.

Automated follow-up actions. For common denial reasons — missing documentation, eligibility issues, coding errors — QuickRCM generated and submitted corrected claims or appeal packages automatically. Staff intervention was reserved for claims requiring payer phone calls, complex appeals, or patient financial responsibility actions.

Predictive write-off identification. The system identified claims with less than 10% probability of collection based on aging, payer, denial reason, and historical resolution patterns. These claims were flagged for early write-off review rather than consuming staff time on low-probability follow-up.

Implementation: A 10-Month Deployment

Phase 1: Eligibility and Payment Posting (Months 1-3)

The first phase deployed the two components with the least workflow disruption and the fastest ROI: automated eligibility verification and payment posting.

Eligibility automation required integration with the agency's intake system and Medicare's HIPAA Eligibility Transaction System (HETS). The integration was completed in 45 days, with automated eligibility checks active for all patients by month 2.

Payment posting automation required ERA file routing and claim matching configuration. The agency received ERAs from 12 payers electronically, covering 78% of payment volume. Auto-posting was active for all electronic payers by month 2, with paper EOB processing added by month 3.

Phase 1 results: A/R days dropped from 58 to 47. Eligibility denials dropped 82%.

Phase 2: Claims Processing and PDGM Engine (Months 3-6)

The second phase was the most operationally disruptive, requiring the billing team to transition from manual claim preparation to AI-assisted workflow. The PDGM classification engine required calibration against the agency's historical claims data and specific payer contracts.

The critical success factor was parallel processing during the transition period. For months 3 and 4, every claim was processed both manually and through QuickRCM, with discrepancies reviewed daily. This parallel period identified 340 claims where QuickRCM's PDGM classification differed from the manual classification. Review of these discrepancies confirmed QuickRCM was correct in 78% of cases and revealed two systematic classification errors in the manual process that had been generating underpayments for over a year.

Phase 2 results: A/R days at 34. Clean claim rate at 93.1%.

Phase 3: A/R Optimization and Staff Restructuring (Months 6-10)

The final phase deployed intelligent A/R prioritization and completed the staff restructuring that the earlier phases had enabled.

The billing team was reduced from 14 to 8 FTEs through a combination of natural attrition (3 departures) and redeployment (3 staff moved to clinical documentation support, patient intake, and payer relations roles). The 8 remaining staff were reorganized into specialized roles rather than generalist billing positions: 2 eligibility and intake specialists, 3 claims and billing specialists (managing exceptions from automated workflows), 2 A/R and denial management specialists, and 1 analytics and quality specialist.

Phase 3 results (month 10): A/R days at 23. Clean claim rate at 96.7%. Monthly collections at $3.75M.

Results: The Complete Transformation

A/R Days: From 58 to 23

The 60% reduction in A/R days released approximately $3.2 million in working capital. The agency retired its line of credit within four months of full deployment, eliminating $85K in annual interest costs.

The A/R days improvement came from compounding gains across the revenue cycle:

• Faster claim submission (weekly batch to daily automated): reduced 3.5 days
• Higher clean claim rate (fewer rejections requiring rework): reduced 8 days
• Faster payment posting (5.2 days to 0.4 days): reduced 4.8 days
• Intelligent A/R follow-up (prioritized, partially automated): reduced 8.7 days
• Reduced eligibility denials (verification gaps eliminated): reduced 10 days on affected claims

Clean Claim Rate: From 82% to 96.7%

The 14.7-percentage-point improvement in clean claim rate meant that 96.7% of claims were accepted and paid on first submission. On approximately 2,800 claims per month, this represented roughly 410 fewer rejected claims per month — each of which would have required investigation, correction, and resubmission under the old process.

The remaining 3.3% rejection rate was concentrated in three areas: payer-specific edits not yet incorporated into the scrubbing engine (1.4%), patient demographic changes not captured in real-time (0.8%), and complex multi-payer coordination of benefits scenarios (1.1%).

Cash Collections: Up 34%

Monthly cash collections increased from $2.8M to $3.75M — a $950K monthly improvement, or $11.4M annually. This increase came from three sources:

Faster collection of the same revenue ($680K/month). The primary driver was simply collecting existing revenue faster. Revenue that previously sat in A/R for 58 days was now collected in 23 days. The agency wasn't earning more revenue — it was collecting what it earned more quickly.

Recovery of previously written-off revenue ($180K/month). Improved PDGM classification, fewer eligibility denials, and better A/R follow-up recovered revenue that would have been written off under the old process. Annual write-offs dropped from $1.1M to $320K.

PDGM optimization ($90K/month). Better primary diagnosis sequencing and comorbidity capture resulted in higher average PDGM payments on approximately 8.7% of episodes — legitimate revenue that was left uncaptured due to suboptimal classification in the manual process.

Staff Redeployment

The reduction from 14 to 8 billing FTEs generated approximately $420K in annual labor savings. But the more significant benefit was where the redeployed staff went:

• Clinical documentation support (1 FTE): Working with clinicians to improve OASIS accuracy and clinical documentation quality, which fed back into better PDGM classification and fewer medical review requests.
• Patient intake optimization (1 FTE): Focusing on insurance verification, benefits counseling, and patient financial responsibility — proactive work that prevented downstream billing problems.
• Payer relations (1 FTE): Dedicated to contract management, underpayment identification, and payer escalations — strategic work that had previously been neglected because billing staff had no bandwidth for it.

Compliance and Audit Performance

The agency underwent a Medicare ADR (Additional Documentation Request) review during month 8 of the implementation. The review covered 30 claims, and 29 were sustained without modification — a 96.7% pass rate compared to the agency's historical ADR pass rate of 84%. The improvement was attributed to better documentation-to-claim consistency enabled by the PDGM classification engine and pre-submission scrubbing.

Key Takeaways for Home Health Agencies

1. A/R Days Are a Symptom, Not the Disease

The agency's 58-day A/R was the result of multiple upstream problems — eligibility gaps, claim errors, slow submission, delayed posting, and unfocused follow-up. Addressing A/R days directly (by adding follow-up staff, for example) would have been treating the symptom. Addressing the upstream causes with automation treated the disease. Each upstream improvement generated compound benefits downstream.

2. PDGM Optimization Is Not Upcoding

The 8.7% of episodes where primary diagnosis resequencing increased the PDGM payment were not upcoded. They were correctly coded. The documentation supported the higher-paying clinical grouping — the manual process simply defaulted to the hospital discharge diagnosis rather than evaluating which primary diagnosis best reflected the reason for home health services. This distinction is critical for compliance: PDGM optimization means selecting the most accurate primary diagnosis, not the most lucrative one.

3. Payment Posting Speed Has Outsized Impact

The most surprising finding was how much impact same-day payment posting had on A/R days and operational clarity. When payments are posted 5 days after receipt, every A/R report is stale, every follow-up decision is based on incomplete information, and secondary billing is delayed by a week. Automating posting was technically straightforward (ERA file matching) and operationally transformative.

4. Home Health Cash Flow Is a Competitive Advantage

Agencies with strong cash flow can invest in clinical programs, recruit better clinicians, and weather payer payment delays without financial stress. Agencies with weak cash flow operate in survival mode — making decisions based on cash availability rather than strategic value. The difference between 58 A/R days and 23 A/R days isn't just financial — it changes how the agency's leadership thinks and operates.

5. Billing Staff Should Be Problem-Solvers, Not Data Entry Clerks

The agency's most experienced billers were spending 70% of their time on tasks that AI handled better — eligibility lookups, claim assembly, payment matching, routine follow-up calls. Redeploying them to CDI support, payer relations, and complex problem resolution leveraged their expertise in ways that manual billing never could. The remaining 8 FTEs were more effective than the original 14 because they were focused on the work that actually required human judgment.

---

This case study presents representative outcomes based on aggregate customer data from home health agencies using the QuickIntell platform. Individual results depend on agency size, payer mix, PDGM classification accuracy, and baseline A/R performance. To discuss how these results might apply to your organization, contact QuickIntell for a custom analysis.