Section GG is the part of the OASIS assessment that records what a home health patient actually does for themselves — not what they are capable of doing on a good day, and not what a clinician believes they could do with enough encouragement. Formally titled "Functional Abilities and Goals," it captures self-care and mobility performance on a six-point scale, and it is one of the most consequential sections a home health clinician completes. Scored accurately, it produces a fair picture of patient acuity that flows into both quality reporting and payment. Scored carelessly — and it is very often scored carelessly — it quietly understates the patient, understates the agency's work, and understates the payment the episode should earn.
This guide explains what Section GG measures, how functional scoring connects to Patient-Driven Groupings Model reimbursement, where agencies most commonly lose accuracy, and what a defensible Section GG process looks like.
What Section GG actually measures
Section GG asks a deceptively simple question for each activity: how much help did this patient need to complete it?
The answer is recorded using a six-level scale that runs from full independence to full dependence:
- 06 — Independent. The patient completes the activity with no assistance and no supervision.
- 05 — Setup or clean-up assistance. The helper sets up or puts away items; the patient does the activity itself.
- 04 — Supervision or touching assistance. The helper provides verbal cues, steadying, or contact guard.
- 03 — Partial/moderate assistance. The helper does less than half the effort.
- 02 — Substantial/maximal assistance. The helper does more than half the effort.
- 01 — Dependent. The helper does all the effort, or two or more helpers are required.
When an activity genuinely cannot be assessed, Section GG provides specific "activity not attempted" codes rather than a guess — including codes for a patient who refused, an activity that was not applicable, an activity not attempted because of environmental limitations, and an activity not attempted because of medical or safety concerns. Using the correct not-attempted code is far better than inventing a performance level, and CMS treats these as meaningfully different from a low score.
The activities themselves fall into two families. Self-care covers tasks such as eating, oral hygiene, toileting hygiene, and upper- and lower-body dressing. Mobility covers transfers and ambulation — rolling in bed, sit-to-stand, chair and toilet transfers, walking defined distances, and navigating stairs. The full item set and the official coding instructions live in the CMS OASIS Guidance Manual, which is the authoritative reference every clinician and reviewer should be working from rather than an internal cheat sheet.
The "usual performance" rule is where most errors start
Section GG is scored on usual performance, not best performance and not potential. This single distinction accounts for a large share of the scoring errors we see in audits.
Usual performance means what the patient did most of the time during the assessment time frame. If a patient transferred with contact-guard assistance on three occasions and independently once, usual performance is the assisted level — not the independent one. Clinicians trained in a rehabilitation mindset often instinctively record the patient's best effort, because that is how therapy goals are framed. Section GG asks the opposite question. It is a record of typical function, and coding it as capability systematically inflates independence and understates acuity.
The assessment window matters for the same reason. Section GG performance is observed across the assessment time frame rather than captured in a single moment, which is precisely what makes a one-visit snapshot unreliable. A patient assessed only at their most alert hour will look more independent than they are across a full day.
How functional scoring connects to PDGM payment
Under the Patient-Driven Groupings Model, every 30-day period is assigned a case-mix group built from several variables — admission source, timing, clinical grouping, comorbidity adjustment, and a functional impairment level of low, medium, or high. That functional level is derived from designated OASIS functional items, and it is one of the multipliers that determines the case-mix weight and therefore the payment for the period.
The practical consequence is direct: two patients with the same diagnosis and the same clinical grouping can land in different payment tiers purely because one was assessed as more functionally impaired than the other. When functional items are scored toward independence out of habit, periods drift toward the low functional impairment level, and the agency is paid as though its patients need less care than they actually receive.
It is worth being precise about one thing here, because agencies get tripped up by it: the specific set of OASIS items CMS uses to calculate the functional impairment level has changed across rule cycles, and it is not safe to rely on a training deck from a few years ago. Confirm the current item set against the CMS Home Health Prospective Payment System page and the current OASIS guidance manual before building any internal scoring rule. For the wider mechanics of how case-mix is assembled, our complete guide to PDGM walks through each variable in turn.
Section GG also feeds quality measurement through the Home Health Quality Reporting Program, including discharge function measures. That means inaccurate scoring is not only a payment problem — it distorts publicly reported quality data in both directions, and an agency that systematically over-reports independence at admission can end up appearing to deliver less functional improvement than it actually did.
Where agencies lose functional scoring accuracy
Across coding and OASIS reviews, the same failure patterns recur.
Capability recorded instead of usual performance
The single most common error, and the most expensive. A clinician observes one strong attempt and records independence. Because the error is systematic rather than random — it always pushes in the direction of more independence — it does not average out across an agency's caseload.
Assessment by interview rather than observation
Patients routinely overstate their own independence, sometimes out of pride and sometimes out of fear of losing autonomy. Family caregivers, who have been quietly compensating for months, often describe the same patient very differently. Scoring from a patient's self-report alone, without observation or a corroborating caregiver account, reliably understates impairment.
Activity-not-attempted codes used as a wastebasket
Applying a not-attempted code because an item was inconvenient to assess — rather than because it was genuinely unsafe, refused, or not applicable — throws away information the agency needs and invites reviewer scrutiny.
Documentation that does not support the score
This is the failure that turns a scoring problem into an audit problem. If the OASIS records substantial assistance for a transfer but the visit note describes the patient ambulating to the bathroom unaided, the record contradicts itself. Under a medical review — and especially under the Review Choice Demonstration — that contradiction is what a reviewer finds first.
No reconciliation between disciplines
Nursing and therapy frequently assess the same patient and reach different functional conclusions without anyone reconciling them. Whichever assessment happens to reach the OASIS becomes the record, with no clinical rationale for why.
Building a defensible Section GG process
Accurate functional scoring is a process problem, not a knowledge problem. Most clinicians know the scale; what breaks down is the system around it.
- Train to usual performance explicitly, and re-train. Make the capability-versus-performance distinction the centrepiece rather than a footnote, and use real patient scenarios from your own caseload.
- Require an observation basis. Every self-care and mobility item should be scored from observed performance or a corroborated caregiver account, not patient self-report alone.
- Reconcile across disciplines before submission. When nursing and therapy disagree, resolve it clinically and document the reasoning rather than letting the earlier assessment win by default.
- Check the narrative against the score. A second reviewer should confirm that the visit documentation actually supports each functional level recorded. This is the step that survives audit.
- Audit a sample continuously. Track the distribution of functional impairment levels across your periods. A caseload that is overwhelmingly low-impairment, in an agency whose clinicians describe demanding patients, is a signal worth investigating.
- Query rather than assume. Where documentation is ambiguous, close the gap before submission — never by coding upward on an assumption.
That last point deserves emphasis. The goal of functional scoring review is accuracy in both directions, not maximisation. Coding toward higher impairment without documentation to support it is exactly the pattern medical review is designed to catch, and it converts a revenue problem into a compliance problem. A defensible process simply records what the record supports — no more, and equally, no less.
Common questions about Section GG
Is Section GG scored on what the patient can do or what they usually do?
Usual performance. Section GG records what the patient actually did most of the time during the assessment time frame, not their best single effort and not their rehabilitation potential. Scoring capability instead of usual performance is the most common source of functional scoring error.
What happens if an activity cannot be assessed?
Use the appropriate activity-not-attempted code — for a patient who refused, an activity that was not applicable, an environmental limitation, or a medical or safety concern. These codes exist precisely so clinicians do not have to guess a performance level, and they are treated differently from a low score.
Does Section GG affect home health payment?
Functional assessment feeds the PDGM functional impairment level — low, medium, or high — which is one of the variables determining a 30-day period's case-mix weight and payment. It also feeds Home Health QRP quality measures. Confirm the current item set with CMS, as it has changed across rule cycles.
Can inaccurate functional scoring trigger an audit?
Yes — most often when the functional level recorded on the OASIS contradicts the clinical narrative in the visit notes. Reviewers look for exactly that mismatch, so internal consistency between assessment and documentation is the strongest protection.
Accuracy, not maximisation
Functional scoring is one of the few places in home health where clinical accuracy and financial accuracy are the same task. Getting Section GG right does not mean scoring patients as more impaired; it means scoring them as they actually are, and having the documentation to prove it.
If you want a certified second look at how your agency is scoring functional items, our coding and OASIS review service audits functional accuracy alongside diagnosis coding, and our clinical documentation review checks that the narrative supports what the assessment records.