Home health operations & revenue reporting
Margin by payment period, LUPA risk while there is still time to act on it, visit utilization against the plan of care, HHVBP and OASIS timeliness, and the revenue still sitting behind a late Notice of Admission — modeled from your Homecare Homebase, Axxess, WellSky, MatrixCare or Alora data. Most builds ship in 3–5 business days.
Representative dashboard — sample data
The agency's view
Under the Patient-Driven Groupings Model, Medicare pays your agency in 30-day periods, and almost everything that decides whether a period earns money is settled inside the first two weeks of it — the case-mix that comes out of the OASIS, whether the visit count will clear the LUPA threshold, whether the Notice of Admission went out on time. Reporting that arrives after billing tells you what happened. It does not change anything.
We build the model that puts the period, the visit schedule and the money in one place: case-mix weight and margin per 30-day period, open periods ranked by how far they sit below their LUPA threshold, visits delivered against visits ordered by discipline, clinician productivity and travel, referral source through to admission, recertification rate, and the unbilled dollars behind each missing document. The data is in your EHR already, one screen at a time — joining it once is the whole job.
What we build for agencies
Buy a block of prepaid Flex hours and, if you're unhappy within the first 4 hours, we refund you in full — no questions asked. Hours are deducted only for work you've reviewed and approved, there's no project minimum, and your banked hours never expire.
Why agencies call us
You send the request; a senior engineer models your EHR, payroll and billing data around the way episodic home care actually gets paid.
Each 30-day period carries its own case-mix weight, admission source, timing and comorbidity level. Visit cost is allocated to the period it falls in, so the second period of a certification is judged on its own economics instead of being averaged into the first.
Thresholds vary by case-mix group, so each open period is measured against its own. Periods short of the threshold are ranked by days remaining, which turns a billing surprise into a scheduling decision the branch can still make.
Skilled nursing, physical therapy, occupational therapy, speech, aide and social work visits compared to what the plan of care called for, per patient and per discipline — so missed visits and over-service both surface as numbers rather than anecdotes.
Visits per clinician per day, weighted by visit type, alongside mileage and travel time by territory. A clinician covering a rural county and one covering four city blocks stop being compared on a single raw visit count.
Referrals by hospital, skilled nursing facility, physician office and community source, tracked through accepted, non-admit and the reason why, with time from referral to start of care. Liaison effort gets pointed at the sources that actually convert.
Periods with no final claim, quantified in dollars by branch and aged, each one tagged with what is holding it — an unsigned order, an unlocked OASIS, a missing face-to-face note, a Notice of Admission filed outside the five-day window.
Home health is paid per 30-day period under PDGM, and the payment for each period is set by a case-mix group derived from admission source, period timing, clinical grouping, functional impairment level and comorbidity adjustment — CMS documents the mechanics on its Home Health Prospective Payment System (opens in new tab) pages. The 60-day certification period still governs the plan of care and the recertification, which means one certification normally spans two payment periods with different timing and often different case-mix.
That single fact breaks most inherited reporting. An episode-level report blends a first period and a second period that were paid differently; a monthly revenue report shows cash that belongs to periods opened weeks earlier. Getting this right means the period is the grain of the model, with the certification, the patient, the visit schedule and the claim all hanging off it — see Power BI dashboard development for how a build like this runs.
Agencies watch four things that a general healthcare report will not give them: whether each period is profitable, whether it will clear its LUPA threshold, whether quality and assessment work is being done on time, and whether completed care has actually been billed. Here is how those group up, and what each one is read against.
| Dashboard area | Metrics we model | Grain & comparison | Primary source |
|---|---|---|---|
| PDGM period margin | Case-mix weight, period revenue, visit cost by discipline, contribution per period | Per 30-day period, split by early/late timing, institutional vs community admission source, clinical grouping | EHR billing tables + payroll |
| LUPA exposure | Visits made and scheduled vs the threshold for that case-mix group, days remaining, periods at risk in dollars | Per open period, rolled to clinician, branch and case-mix group | EHR scheduling + case-mix |
| Visit utilization | Ordered vs delivered visits by discipline, missed visit rate and reason, visits per period vs branch norm | Per patient per period, compared to the plan of care, not to an agency average | EHR plan of care + visit log |
| HHVBP & OASIS quality | Measure results as recorded in your system, OASIS submission and assessment timeliness, acute-care transfer and discharge disposition | Per branch and clinician, trended between CMS reporting cycles | OASIS exports + claims data |
| Clinician productivity | Visits per clinician per day, weighted visit points, mileage and travel time, overtime and PTO cover | Per clinician per week, grouped by discipline and territory | EHR visit log + payroll |
| Referral & census flow | Referrals, non-admit reasons, referral-to-SOC days, active patient count, recertification rate, discharge reason | Per referral source and branch, with month-over-month trend | EHR intake / referral module |
| Unbilled & NOA lag | Days from start of care to Notice of Admission, periods past the five-day window, periods with no final claim, aged unbilled dollars by blocking reason | Per period and branch, aged in buckets, totalled in dollars | EHR billing + clearinghouse |
A Low Utilization Payment Adjustment applies when the visits delivered in a 30-day period fall below the threshold for that period's case-mix group, and the period is then paid per visit instead of at the period rate. The thresholds are not uniform — they differ by case-mix group — which is exactly why a single agency-wide "minimum visits" rule of thumb keeps producing both unnecessary visits and avoidable LUPAs.
We store the applicable threshold on each period, count completed and scheduled visits against it, and surface the open periods that are short, ordered by how few days are left. The same model shows LUPA rate after the fact by branch, clinician, case-mix group and referral source, so the pattern behind the worklist is visible too. Row-level security keeps each branch manager in their own patients — Microsoft documents the mechanism in row-level security for Power BI (opens in new tab).
Nearly all of it starts in the home health EHR. Patient, certification and period records, OASIS assessments, orders, the plan of care, the visit schedule and the visit log, the referral and intake record, and the billing tables behind RAPs, Notices of Admission and final claims all live there. Payroll and timekeeping add worked hours, overtime and mileage; the clearinghouse or remittance feed adds paid, denied and pending status; the general ledger adds branch overhead so contribution can become a real margin.
Home health platforms differ sharply in what they will let you extract. Some editions offer a reporting database or a warehouse feed, some a documented API, and some realistically only scheduled exports — and the answer depends on your contract and edition, not just the vendor's name. We establish the route in the scope estimate before any hours are used, then set up scheduled refresh in the Power BI service, which Microsoft covers in its guide to data refresh in Power BI (opens in new tab).
| Source | What it contributes | Dashboard it feeds |
|---|---|---|
| Home health EHR (Homecare Homebase, Axxess, WellSky, MatrixCare, Alora) | Patients, certifications and 30-day periods, case-mix, OASIS, orders, plan of care, visit schedule and visit log | Period margin, LUPA exposure, visit utilization, recertification rate |
| Intake / referral module | Referral source and type, referral date, accept or non-admit decision and reason, start-of-care date | Referral source conversion, referral-to-SOC days, forward census |
| Billing & clearinghouse | RAP and Notice of Admission submission dates, final claim status, payments, adjustments and denials | Unbilled revenue, NOA lag, cash collected per period |
| Payroll & timekeeping | Worked hours, per-visit and salaried pay, overtime, mileage and travel time by clinician | Visit cost, clinician productivity, cost per visit by discipline |
| OASIS exports & quality feeds | Assessment dates and submission timestamps, measure results as recorded, transfer and discharge disposition | HHVBP and OASIS timeliness, quality trend by branch |
| General ledger | Branch revenue, overhead and controllable expense, budget by cost centre | Branch P&L, margin after overhead, budget vs actual |
Since the Notice of Admission replaced the Request for Anticipated Payment, a period's payment is reduced for each day between the start of care and a late NOA submission, so a filing that slips by a week costs real money on work already delivered. The delay is rarely billing's fault — it is usually an order awaiting a physician signature, an OASIS not yet locked, or a face-to-face document that has not arrived.
Putting every unbilled period in one list, in dollars, aged, with the blocking item named and assigned, makes that a manageable queue instead of a month-end scramble. If your bigger problem is what happens after the claim goes out — payer denials, reason codes, appeals and 835 remittance — that is a different data set and lives on our Power BI denial management dashboard page, which many agencies run alongside this one.
You buy prepaid Flex hours that never expire, at published prices: Starter $2,500, Growth $4,600, or Power $8,000, with senior US-based work from $100/hour. A first build covering period margin, LUPA exposure and unbilled revenue typically fits inside one block, and most builds ship in 3–5 business days once you approve the scope estimate.
Compared with a traditional consultancy, there is no project minimum, no retainer you have to keep feeding, and no unused hours expiring at quarter end — hours deduct only for work you have reviewed and approved. See how prepaid Flex hours work, or our affordable Power BI consulting page for the full comparison.
Owners, administrators, directors of clinical services, branch managers, revenue cycle leads and CFOs at Medicare-certified home health agencies without an in-house BI team — single-branch agencies, regional multi-branch groups, and agencies that have grown by acquisition and inherited a second EHR along with the branch.
Several neighbouring needs are covered elsewhere on this site, deliberately. If your setting is a facility with beds rather than care delivered in the home, the reporting turns on resident census and the EHR extraction path instead, and our PointClickCare Power BI dashboard page is the right starting point — there is no facility census in home health, and no 30-day payment period in skilled nursing. If you operate buildings and need occupancy by care level, move-ins and NOI per community, see Power BI for senior living operators. For tool-agnostic clinical throughput across a provider organization — patient flow, length of stay, slot and bed utilization — Power BI for healthcare is the broader page. And if you already have home health reports that nobody trusts, a Power BI health check is the fastest way to find out why. Contact is form-only — send your request and we'll reply with a scope estimate.
business days, typical
hours, never expire
own everything
A first period-margin, LUPA and unbilled build usually fits inside one Flex block, and hours deduct only for work you approve. See Flex pricing →
How it works
Tell us which EHR your branches run, whether you have a reporting database, an API or only scheduled exports, how you define an active patient and a productive visit, and who should see which branches. We reply with a scope and hours estimate before any work begins.
A senior US-based engineer builds the period-grain model across your EHR, payroll, billing and ledger data, writes the DAX for case-mix margin, LUPA exposure and visit utilization, and reconciles period counts and revenue back to your billing reports — typically 3–5 business days.
We publish to your workspace, set overnight refresh so the LUPA and unbilled lists are current each morning, apply row-level security by branch or territory, and hand over the model and credentials. You own everything.
Representative perspective
We knew our LUPA rate at month end and could do nothing about it. Seeing open periods ranked by how far short they were, with the days left, moved the whole conversation into the scheduling meeting.Director of clinical services, multi-branch home health agency
Representative perspective
Everyone agreed there was unbilled revenue sitting out there. Nobody could say how much, or what was actually holding each one. One list with the blocker named changed that in a week.CFO, regional home care organization
Tell us which EHR your branches run and how you extract from it today. We'll scope the task and tell you the hours before any work begins — drawn from prepaid Flex hours that never expire, with no project minimums and no obligation.
FAQ
Yes, and it is normally the first thing we model. The 30-day payment period is held as its own grain, with case-mix weight, admission source, timing, clinical grouping and comorbidity level as attributes, and visit cost allocated to the period the visit falls in. A 60-day certification usually spans two periods, so the second one is modeled separately instead of averaged away.
LUPA thresholds vary by case-mix group, so we store the threshold that applies to each period, count the visits made and scheduled against it, and rank open periods by how far short they are with how many days remain. Scheduling and clinical managers get a worklist for the current week rather than a variance report after billing.
Homecare Homebase, Axxess, WellSky, MatrixCare and Alora are the platforms we are asked about most, and each exposes data differently — a reporting database or data warehouse feed for some, a documented API for others, and scheduled exports where neither is available. We confirm the extraction route your contract and edition actually allow in the scope estimate, before hours are used.
No. Skilled nursing reporting is built on facility census — beds occupied, admissions and discharges, resident days. Home health has no facility census at all: the unit of work is a 30-day payment period with a visit schedule attached, and the money is decided by case-mix, LUPA thresholds and timely billing. Facility-based reporting is covered on our PointClickCare Power BI dashboard page instead.
Yes. We track each period's start of care against its Notice of Admission submission date and flag the ones outside the five-day window, alongside periods with no final claim submitted and the reason they are stuck — an unsigned order, an unlocked OASIS, a missing face-to-face document. The output is a dollar figure per branch with the blocking item named.
We build the operational half of it: OASIS submission and assessment timeliness, recertification rates, visit patterns by discipline, and how your own measure results trend by branch and clinician between CMS reports. We report what your source systems record — we do not recalculate CMS scoring or predict your payment adjustment.
Book a free demo
Tell us the task. We scope it, you approve, and a senior consultant delivers — typically in 3–5 business days. Prepaid hours that never expire.
The fastest way to reach us is the form — tell us the task and we'll reply within one business day.