What a Medication Pass Survey Actually Checks (and Where Your eMAR Fits)

F759 or F760: What's Actually the Difference?Copy link

Two tags cover medication errors, and they’re not testing the same thing.

42 CFR 483.45(f) sets two separate requirements: your medication error rate can’t be 5 percent or greater, and residents have to be free of any significant medication error, full stop. The first is F759 — a rate test. The second is F760 — an individual-error test, and it doesn’t care what your rate is. A single significant error gets cited at F760 even if your facility’s overall rate is spotless.

Here’s how CMS tells surveyors to do the arithmetic, per Appendix PP of the State Operations Manual: the rate is errors observed (significant and non-significant combined) divided by the total opportunities for error — every dose given, plus every dose ordered but not given — times 100. CMS is explicit that the rate "must be 5% or greater in order to cite F759," and that rounding up doesn’t count: a 4.6% rate stays 4.6%, not 5%. If your rate is 5% or higher, surveyors are also told to consider whether something systemic is going on, like your medication distribution system (F755 territory).

The practical takeaway for your floor: F759 is about volume across the whole observed sample. F760 is about any one error that’s bad enough on its own. You can fail one without the other.

What Actually Makes an Error "Significant"?Copy link

CMS doesn’t hand surveyors a checklist for this — it hands them judgment factors. Per Appendix PP, a "medication error" is any administration that doesn’t match the prescriber’s order, the manufacturer’s specifications, or accepted professional standards. Whether that error is "significant" (and therefore F760-citable regardless of rate) depends on three things CMS names directly: the resident’s condition, the drug category — including narrow-therapeutic-index drugs like warfarin and digoxin, where small deviations matter more — and how often the error happens. CMS calls this "a matter of professional judgment," not a formula.

Appendix PP’s own worked example shows how differently the same mistake can land. In its Examples of Medication Errors table, a missed dose of Metoprolol Succinate — a once-daily blood-pressure and heart-rate medication — is marked Significant. A missed dose of Trazodone at bedtime, in the same table, is marked Not Significant. Same error type (an omission), opposite rating — because the drug category and what a missed dose could do to that resident are different. That’s the "significant vs. not significant" line in practice: not a universal rule about any one drug, but the three factors applied to the specific resident and situation in front of the surveyor.

The 60-Minute Wrong-Time Rule Isn't a Flat RuleCopy link

This is the one most DONs have half-memorized wrong. Per Appendix PP, giving a before-meal dose after meals (or the reverse) always counts as an error, no exceptions. For everything else, CMS tells surveyors to count a wrong-time error at 60 or more minutes early or late — but only when that timing could cause the resident discomfort or jeopardize their health or safety. A long-half-life drug given 15 minutes late isn’t automatically an error just because it’s late; CMS specifically flags that as improper to cite. The scheduled time itself comes from your own facility’s dosing-schedule policy, and CMS notes residents’ right to choose their own schedule, plus the need for narrower windows on time-critical drugs.

So "the 60-minute rule" isn’t really a standalone rule — it’s a 60-minute threshold wrapped in a harm test, measured against a schedule your facility set. Worth rereading your own policy against that, rather than treating 60 minutes as the whole story.

How CMS Actually Runs a Medication Pass ObservationCopy link

Per CMS's 2013 memo setting the current protocol, surveyors observe a minimum of 25 medication administration opportunities, using Form CMS-20056 (which replaced the older CMS-677). An earlier rule that extended the observation by another 20-25 opportunities whenever errors were found was eliminated — the sample size is fixed going in. Once a pass starts, the surveyor documents everything for that resident at that time and won’t stop partway through. Appendix PP adds that the survey team should watch several different passes — different staff, different routes, different times of day — partly so omissions (not just wrong doses) get caught too.

Where Does Your eMAR Actually Come In?Copy link

Here’s the honest answer, and it’s less dramatic than you might expect: the F759 error count comes from direct observation reconciled against the prescriber’s order — not from your eMAR. Appendix PP says it plainly: "detection of blank spaces on a medication administration record does not alone constitute the occurrence of actual medication errors." A gap in your MAR isn’t an automatic citation.

What the MAR does do is trigger follow-up. If a record discrepancy surfaces during review — or a family reports something, or a resident’s condition changes — the surveyor follows up through chart review and staff interviews. An error found this way isn’t counted toward the F759 rate (it wasn’t from direct observation), but if it’s significant, it can still be cited at F760. So your eMAR’s real job in a survey isn’t to be the error count itself — it’s to be the document that corroborates or contradicts what the surveyor already saw, and the thing that can open a follow-up path on its own when something looks off in it.

What a Surveyor Expects an eMAR to Be Able to DoCopy link

None of this is medication-specific — it’s the general record standard, and it applies whether you’re on paper or electronic. 42 CFR 483.70(h) (tag F842) requires resident records to be complete, accurately documented, readily accessible, and organized per professional standards, and safeguarded against loss or unauthorized use. Appendix PP’s F842 guidance names medication administration specifically as an example of EHR use, and is direct about the limits of what gets reviewed: there’s no requirement to use an EHR at all, and surveyors are told not to evaluate the EHR system’s features — only how it’s actually being used.

For electronic signatures specifically, CMS lists out the safeguards a facility’s policy should cover: an individualized identifier for each signer, a timestamp from the system clock rather than a typed-in date, an entry that is not to be changed after it has been recorded, and role-based control over what each user can enter. Separately, controlled-drug records fall under F755 — 483.45(b) requires a system that lets you accurately reconcile and periodically account for controlled medications. And if your facility runs an EHR, CMS expects the survey team to get timely access to it — by the end of the first survey day — plus a staff member assigned to help them use it.

Where AI Fits — and Where It Doesn'tCopy link

Be specific here, because it’s easy to overclaim. Nothing about AI changes who counts as having made a medication error, how the 5% threshold gets calculated, or what makes an error significant — those are surveyor judgment calls against CMS’s own criteria, not something software decides for you.

What AI can reasonably do is help your staff get a faster, sourced answer to a documentation-compliance question instead of digging through Appendix PP themselves at 2am. Relic Care’s Compliance Assistant is built for exactly that: a plain-language Q&A tool grounded in CMS rules, state regulations, and your own facility policy — a nurse or DON can ask it something like "what counts as a wrong-time error under our policy" and get a sourced answer back. It doesn’t observe medication passes, verify doses, or decide whether a given miss is significant — and every answer it gives is, in the product’s own words, a check against your own judgment, never a substitute for it. That boundary matters more here than almost anywhere else in your documentation stack, because significance determinations are explicitly reserved for professional judgment, not something a tool should be allowed to quietly take over.

Where This Fits Into Your Facility’s Survey-Readiness PictureCopy link

This post sits alongside the SNF Survey-Readiness Checklist’s chapter-by-chapter coverage of other tags, and pairs with how Section GG documentation scoring and fall-risk and pressure-injury documentation treat the same underlying question: whether the chart matches what was actually observed, not just what was checked off.

Next check-in: April 2027. CMS revises Appendix PP periodically (most recently Rev. 225, August 2024), and a secondary source incorrectly reported a further April 2025 revision that this research directly checked against the live CMS document and did not find — worth another direct check of the F759/F760/F761/F755/F842 revision stamps if this post is still live by then.

Disclaimer: This post is informational and reflects CMS’s federal guidance as of the date above — it is not legal, clinical, or regulatory advice, and it does not cover state-specific medication regulations. Confirm anything that affects your facility’s policy or a specific resident’s care with your own clinical, compliance, or legal counsel.