Timely Initiation of Care Is 96.51% Nationally. The Misses Are All Scheduling Calls.

Timely Initiation of Care Is 96.51% Nationally. The Misses Are All Scheduling Calls.

Across Medicare-certified home health agencies, care begins on time 96.51% of the time. That is the national value for the measure How often the home health team began their patients’ care in a timely manner, published in the CMS Home Health Care — National Data file. It sits near the top of the publicly reported home health measure set.

It is also one of the easier measures to pass, because of how it is defined. It does not ask whether you reached the patient quickly. It asks whether you reached them by a date that was already written down.

What the measure actually counts

Per the CMS Home Health QRP quality measure specifications, an episode counts as timely when the start-of-care or resumption-of-care date equals the physician-ordered start-of-care date — or, when no physician date was specified, when care began within two days of the referral date (or within two days of inpatient discharge, if the discharge came after the referral). The calculation runs off OASIS items you submit yourself: M0030, M0032, M0102, M0104, M1000 and M1005.

Two consequences follow, and operators should sit with both.

• If a physician ordered a start-of-care date of next Tuesday and you started next Tuesday, the episode is timely. The clock in the measure is the ordered date, not the referral.
• The measure is built from your own assessment data. It reflects what your intake team recorded, which is not always what happened.

So 96.51% is a floor on process discipline, not a ceiling on responsiveness. An agency can score well on this measure and still be slow in every way a referral source notices.

The Conditions of Participation are stricter in one specific respect

The quality measure and the regulation are not the same instrument, and conflating them causes real trouble in survey.

42 CFR 484.55(a)(1) requires that the registered nurse conduct the initial assessment visit “within 48 hours of referral, or within 48 hours of the patient’s return home, or on the physician or allowed practitioner-ordered start of care date.” That is a condition of participation, not a star-rating input. It attaches to the initial assessment visit, and it runs in hours rather than calendar days.

Separately, 42 CFR 424.22 conditions payment on a face-to-face encounter that “occurred no more than 90 days prior to the home health start of care date or within 30 days of the start of the home health care,” with the certifying practitioner documenting the date of that encounter. That is a documentation condition, not a start-of-care deadline — but in practice it is one of the most common reasons an intake coordinator holds a case rather than scheduling it.

None of these rules requires you to record a call. They require you to be able to show what was assessed, when, and on whose order. How you produce that evidence is your choice.

This post is general information about regulatory requirements, not legal advice. Medicaid home health programs and Medicare Advantage contracts frequently set their own initiation timeframes, and those vary by state and by plan; check your specific payer agreements rather than assuming the Medicare standard governs.

Where the delay actually happens

If your timely-initiation rate slips, the cause is rarely clinical. Nobody refuses to see a patient. The episode stalls in a sequence of conversations that nobody owns end to end.

The referral call that never got a date

A discharge planner calls at 4:40 p.m. on a Friday. Your intake coordinator takes the demographics and the diagnosis, and does not pin down an ordered start-of-care date. There is now no date in the record, which means the two-day referral clock is running and nobody knows it.

The staffing callback that went one direction

Scheduling calls three nurses, leaves three voicemails, and marks the case pending. Two of those nurses call back on Saturday, when nobody is logged into the scheduling system. Monday morning the case looks untouched.

The patient who was not home

The patient is discharged to a daughter’s house across town. Intake has the hospital address. The first visit attempt fails, and the clock does not reset just because the address was wrong.

The order that was never actually confirmed

Intake believed the physician’s office confirmed a start date over the phone. The office believed it was still waiting on the referral packet. Both are documented in a disposition code that says “pending MD.”

Every one of these is a conversation. None of them is visible in your EMR until after the deadline has passed.

What this looks like in your call data

Failure mode	What it sounds like on the call	Where it surfaces later
No ordered SOC date captured	Referral call ends without anyone stating a date out loud	Episode judged against the two-day referral clock
Staffing gap	Repeated outbound calls to the same three clinicians, no confirmation	Case sits in pending status past day two
Wrong location	Caller mentions a family member’s home, agent repeats the hospital address	Failed first visit, no reset of the clock
Unconfirmed physician order	Agent says “we will wait to hear back” with no callback owner named	Certification and face-to-face documentation gaps at billing
After-hours referral	Voicemail or answering service intake with no same-day handoff	Referral effectively starts a day late

Do this week

You do not need a new system to move this number. You need to treat the first 48 hours as a call workflow with named owners.

• Make the ordered start-of-care date a required verbal confirmation. The intake agent says the date back to the referral source and gets agreement before the call ends. If no date is given, the agent says so explicitly, which puts the two-day referral clock on the record.
• Pull your last 20 late episodes and listen to the referral call. Not the notes — the call. You are looking for the moment the date should have been set and was not. Twenty calls is roughly two hours of work and it will tell you which of the four failure modes above is yours.
• Assign a callback owner by name, on the call, every time. “We will get back to you” is how cases stall. “Dana will call you before noon tomorrow” is a commitment you can audit.
• Check your after-hours and weekend intake path. Count how many referrals arrived outside business hours last month and how many of those started on time. If the gap is meaningful, your problem is coverage, not clinical capacity.
• Reconcile OASIS M0104 against your phone system timestamps for a sample of episodes. If the recorded referral date is routinely later than the first inbound call, your measure is flattering you and a surveyor or auditor may see it differently.

That last item is worth dwelling on. A measure built from self-reported dates rewards clean data entry. It does not protect you if the underlying conversation says something else.

The coaching loop matters more than the dashboard

Intake and scheduling performance moves when specific agents get specific feedback on specific calls, on a predictable cadence. We wrote up a workable rhythm in coaching contact-center agents with call intelligence, and the same structure applies here: one recurring session, a small number of real calls, one behavior to change.

The same logic governs the other end of the episode. The 72 hours after discharge are a call workflow too, with the same tendency to fail quietly. And if you are rebuilding your intake documentation standards from scratch, the home care call compliance checklist covers what to capture, redact and retain.

What the 96.51% does not tell you

Three stars is the national quality-of-patient-care star rating, and only 3.83% of agencies hold five stars. Timely initiation is not what separates them, because nearly everyone clears it. The separation happens in the part of the episode this measure cannot see: whether the referral source got a call back, whether the family knew who was coming, whether the case that stalled got escalated by someone who noticed.

Consider a hypothetical agency running 94% on timely initiation and unable to explain the six. Its EMR shows six episodes with a late start and a disposition code. Its phone system holds six referral calls in which the ordered start-of-care date was never stated aloud. Only one of those two records explains anything.

SurfacerIQ is in the category of tools that read the conversation layer — every intake, scheduling and callback call — and surface the commitments, missing dates and unclosed loops that operational systems record only after the fact. That is the gap between what your measure says and what your week actually looked like.

If your timely-initiation rate is fine and your referral sources still say you are slow, both things can be true. Talk to us about what your calls would show.