LUPA doesn't usually come up in the same conversation as coding accuracy, they get treated like separate problems handled by separate teams. In practice, they're much more connected than that, and a coding gap that seems unrelated to visit thresholds can be exactly what tips a period into a Low Utilization Payment Adjustment.

Here's how the two actually connect.

What LUPA risk actually depends on

A period becomes subject to LUPA when the number of visits falls below the threshold set for that specific case mix group, and gets paid per visit instead of at the full 30-day rate, typically at a steep discount to what the full episode would have paid. Those thresholds vary by clinical group, functional level, and comorbidity adjustment, which means the case mix group a patient actually lands in determines the visit bar they need to clear.

That's the connection most agencies don't think about day to day: coding accuracy doesn't just affect the payment rate for a period, it affects which LUPA threshold applies to that period in the first place.

Where the risk actually shows up

If a functional item gets scored one level lower than the documentation actually supports, or a comorbidity adjustment gets missed because a secondary diagnosis never made it onto the chart a patient can land in a case mix group with a lower LUPA threshold than the one their actual acuity would have placed them in. A visit count that would have comfortably cleared the correct threshold can end up falling short of a threshold the coding never should have assigned in the first place.

That's a case where accurate coding doesn't just mean a fairer payment, it means the difference between a full 30-day payment and a steeply discounted per-visit one, on a period that clinically deserved the higher rate all along.

Why this connection gets missed

Coding and scheduling usually live in different parts of an agency's workflow, coders focused on getting diagnosis and functional scoring right, schedulers focused on hitting visit frequency targets. Neither side is necessarily thinking about how the other's work interacts with LUPA thresholds, because the connection isn't obvious until you actually trace it through.

What actually reduces this risk

Accurate functional scoring and complete comorbidity capture do more than support fair reimbursement on their own terms, they put a patient's case mix group where it should be, which sets the LUPA threshold where it should be. 

A QA process that checks coding accuracy specifically with an eye toward downstream LUPA exposure, not just general correctness, catches this kind of risk before a visit schedule gets built around the wrong threshold entirely. Each quarter, our QA team audits thousands of randomly selected charts from the hundreds of thousands we complete and we publish the results. Our accuracy rates came in at 97.5% overall across coding, OASIS, and Plan of Care, which is the kind of consistency this specific risk actually depends on. A single missed item is often all it takes to shift a threshold. Olli Health publishes our accuracy scores every quarter - the real question is why doesn’t your current coding vendor do the same?

See how this could affect your own LUPA exposure

We'd like to walk through how coding accuracy connects to LUPA risk in your own case mix distribution, at roughly half the cost of a legacy vendor and with turnaround measured in hours, not days.