BLUF (Bottom Line Up Front)
Stroke recognition in EMS is harder than most curricula admit, and the consequences of a miss are permanent. The Cincinnati Prehospital Stroke Scale identifies the majority of anterior circulation events but consistently underperforms on posterior circulation strokes and large vessel occlusions. The prehospital evidence that BE-FAST outperforms FAST remains weak — the only prospective field comparison found no improvement, and the 2025 systematic review found insufficient evidence to conclude otherwise. Secondary LVO screening tools (VAN, RACE, LAMS) are protocol-dependent but increasingly standard in progressive EMS systems. Prenotification is the single most impactful act a crew can perform after recognition. It is not optional — the outcome data have been unambiguous since 2012, and the 2026 AHA guidelines anchor prehospital stroke systems of care as the foundation of timely treatment. And none of this matters if your providers have only rehearsed the textbook presentation.
The Call Nobody Gets Wrong. Until They Do
A 78-year-old female. Her husband called 911 because she is acting confused. You arrive and she is seated at the kitchen table. She knows her name. She answers some of your questions. Her speech is a little slow. Her blood pressure is elevated. No facial droop. She moves all extremities to command. Blood glucose is 96.
That call gets missed all the time. Not because paramedics are careless. Because posterior circulation strokes and subtle large vessel occlusions do not look like the textbook picture. They look like altered mental status. They look like a psych call. They look like someone who is just tired.
The consequences of a missed stroke are irreversible. The window for acute intervention is narrow. And the paramedic's assessment in the first five minutes of that call determines what the neurology team can do.
What the Cincinnati Prehospital Stroke Scale Actually Tests
The Cincinnati Prehospital Stroke Scale (CPSS) evaluates three findings: facial droop, arm drift, and abnormal speech. One abnormal finding out of three indicates high probability of stroke. It is quick, reproducible, and explicitly listed in the NREMT's National Continued Competency Program education guidelines as core knowledge for EMS recertification. (NREMT NCCP EMT Education Guidelines)
That being said, the CPSS has a known ceiling. The American Stroke Association's EMS training data reports that paramedics demonstrated a sensitivity of only 61–66% for stroke recognition without specific assessment tool training. (Stroke Training for EMS Professionals, American Stroke Association) The CPSS improves on that baseline, but it was designed for anterior circulation events. The classic hemisphere strokes. Posterior circulation strokes affecting the brainstem and cerebellum frequently present without facial droop, arm drift, or speech abnormality.
A 2022 PubMed Central review titled "Not so FAST" documented this directly: posterior circulation strokes are systematically missed by FAST-based tools. (PMC: Not so FAST. Pre-hospital posterior circulation stroke) The patient with isolated vertigo, sudden ataxia, or double vision is a stroke until proven otherwise. The CPSS will not tell you that.
FAST vs. BE-FAST: What the Evidence Actually Says
BE-FAST added Balance and Eyes to the original FAST mnemonic specifically to address the posterior circulation gap. The logic was sound on paper. The prehospital evidence has not caught up with it.
The only prospective prehospital comparison to date, published in 2018, found that adding coordination and diplopia assessments to face, arm, and speech did not improve stroke detection in the prehospital setting. (PubMed: Prognostic Value of BEFAST vs. FAST to Identify Stroke in a Prehospital Setting) A 2022 systematic review and meta-analysis found BEFAST carried higher overall diagnostic value than FAST — but that analysis was not prehospital-specific, and its authors noted it remains unclear whether BEFAST can be applied for screening in the prehospital setting. (PMC: A Systematic Review and Meta-Analysis Comparing FAST and BEFAST) A 2025 systematic review by Hilditch and colleagues, focused specifically on prehospital use, concluded that the data are limited: both tools perform reasonably, BE-FAST may be more sensitive, and there is insufficient evidence to conclude it improves prehospital stroke recognition. (Hilditch et al. 2025. BE-FAST vs FAST in Prehospital Stroke Recognition, British Journal of Community Nursing)
Why has the field data lagged behind the logic? Two reasons. Balance and eye findings are harder to assess reliably on a symptomatic patient in a chaotic scene environment. And adding elements to a mnemonic increases cognitive load at exactly the moment you can least afford it.
Follow your agency's protocol. If it specifies BE-FAST, use it. Clinically, FAST combined with a high index of suspicion for atypical presentations is what actually catches cases.
Large Vessel Occlusion: The Question That Changed What the Paramedic Has to Answer
Approval of mechanical thrombectomy for large vessel occlusion (LVO) stroke changed the scope of the prehospital assessment. It used to be: identify stroke, call the hospital, go. Now it has to answer a second question. Does this patient need a comprehensive stroke center with an interventional suite, or will the nearest primary stroke center do?
That determination can only happen in the field. And the CPSS, alone, is not designed for it.
Three prehospital LVO screening tools are used across EMS systems: the VAN (Vision, Aphasia, Neglect), the RACE (Rapid Arterial oCclusion Evaluation), and the LAMS (Los Angeles Motor Scale). A 2021 study in Prehospital Emergency Care (published online in 2020) compared the CPSS against RACE, LAMS, and VAN for LVO detection and found that no single tool is clearly superior across all parameters. The authors recommended that EMS agencies evaluate ease of use and integration with existing workflows before choosing one. (PubMed: Cincinnati Prehospital Stroke Scale Compared to LVO Prediction Scales)
The VAN assesses cortical findings: vision disturbance, aphasia, and neglect. The pattern of a large anterior circulation occlusion. The AHA has featured it in EMS quality-improvement materials as an LVO screening adjunct. (AHA: The Prehospital VAN Stroke Assessment) The LAMS, validated in a 2018 AHA Stroke journal study, focuses on three motor items. Face, arm, and grip. And showed reliable prehospital performance for LVO identification. (AHA Journals: Los Angeles Motor Scale to Identify Large Vessel Occlusion)
A 2024 single-center comparison of prehospital LVO scales confirmed that every tool carries trade-offs — sensitivity for posterior circulation LVO remains a weak point across scales — and that local protocol, transport geography, and receiving hospital capabilities determine which is right for your system. (PMC: Comparison of prehospital stroke assessment scales for acute ischemic stroke)
Know which tool your medical director has adopted. If your agency does not have an LVO screening protocol, that conversation is overdue.
Last Known Normal: The Only Time That Matters
Last known normal (LKN) time is not a documentation detail. It is the decision point for thrombolysis and thrombectomy eligibility. Every minute beyond LKN erodes the treatment window. Extended scene time. Searching for paperwork, delayed packaging, repeated assessment loops. Directly costs the patient viable brain tissue.
Get the LKN from the first reliable source: the patient, a family member, a neighbor, a security camera timestamp if necessary. Document it clearly on your PCR. Include it in your prenotification call. It is the first number the neurology team asks for.
Regardless of how complex the scene is, that piece of information comes before anything else.
If the patient was last seen normal eight hours ago, the door has not closed. Selected patients with large vessel occlusion remain candidates for endovascular thrombectomy from 6 to 24 hours after last known normal, and imaging-selected patients may still be eligible for IV thrombolysis beyond the classic window (AHA/ASA acute ischemic stroke guideline). Document the last-known-normal time precisely and route per your regional stroke plan — the extended-window patient is exactly the one a sloppy timeline disqualifies. If the LKN is 45 minutes ago and your transport time is 12 minutes, every action on that call carries urgency.
Prenotification Is Not Optional
A 2012 American Heart Association study demonstrated that EMS hospital prenotification is associated with significantly improved stroke outcomes, specifically through reduction in door-to-imaging and door-to-needle times. (AHA: EMS Hospital Prenotification Is Associated with Improved Stroke Outcomes) A subsequent PubMed Central study confirmed that prehospital notification improves stroke outcomes by reducing door-to-needle time. And that the strategy succeeds or fails primarily based on EMS execution, not hospital systems. (PMC: Prehospital Notification Procedure Improves Stroke Outcome)
The 2026 AHA Guidelines for the Early Management of Patients with Acute Ischemic Stroke reinforce prehospital stroke systems of care — EMS recognition with validated screening tools, coordinated transport destination decisions, and pre-arrival coordination with the receiving hospital — as the foundation of timely stroke treatment. (AHA: 2026 Guideline for the Early Management of Acute Ischemic Stroke)
The prenotification call takes 30 seconds. It activates the stroke team before you pull into the bay. It means the CT scanner is cleared, the neurologist is on the line, and the pharmacist has tPA drawn when the patient arrives.
Your call should include: suspected stroke, last known normal time, current deficits, relevant history (anticoagulants, prior stroke, current blood pressure), and ETA. That is the whole report.
If your agency does not require prenotification for stroke, that protocol needs to change.
Where Simulation Closes the Gap
Knowing the scales and applying them under pressure are two different things. A two-hour CE module on stroke recognition will not build the pattern recognition that holds when you are on scene at 02:00 with a family member yelling and a patient who does not fit the textbook.
A 2022 Frontiers in Neurology study examining interactive EMS stroke training found it was associated with a robust improvement in stroke recognition accuracy, faster onset-to-transport times, and a trend toward improved in-hospital outcomes. (Frontiers in Neurology: Interactive Training of the Emergency Medical Services Improved Stroke Recognition) The difference between didactic instruction and repeated scenario-based practice is not marginal. Recognition speed and clinical accuracy both improve when providers have worked through realistic presentations repeatedly before the real call comes.
EMS-MedSim's scenario library includes prehospital stroke cases across the full clinical spectrum. Anterior circulation, posterior circulation, hemorrhagic stroke, and hypoglycemia as a stroke mimic. Each scenario runs you through CPSS, LVO screening decisions, last known normal documentation, transport destination selection, and prenotification. The goal is pattern recognition that is automatic by the time it matters.
Run the Prehospital Stroke Scenarios →
Dr. Chet's Take
I have been on both sides of this problem. On the EMS side, I have watched crews do everything right. Rapid recognition, immediate prenotification, fast transport. And the patient does well. I have also seen the other call: a provider who was confident the subtle presentation was not a stroke, extended scene time trying to sort out "just AMS," and a patient who arrived 40 minutes later than he should have. The difference between those two outcomes often came down to whether the provider had ever actually worked through an atypical stroke scenario before.
The posterior circulation call is the one that catches people off guard. Isolated vertigo. Sudden ataxia. Slurred speech with no arm drift, no droop. Providers miss it because they have only rehearsed the classic presentation. If you have never had to make that recognition call in a high-fidelity scenario before it happens in real life, the on-scene cognitive load is enormous and the clock does not care.
That being said, the tools are there. CPSS, VAN, LAMS, RACE. Validated, accessible, protocol-ready. The 2026 AHA guidelines double down on prehospital stroke systems of care. The evidence on simulation training is clear. The question is whether your training has given you enough repetitions that the decision-making is automatic when the patient in front of you looks nothing like the textbook. That is what EMS-MedSim is built for.
Frequently Asked Questions
What does 'last known normal' actually mean?
The last time the patient was seen at their baseline — not when symptoms were discovered. It is the timestamp that anchors thrombolysis and thrombectomy eligibility.
Why is stroke prenotification not optional?
Because it moves the hospital before the patient arrives — the stroke team assembles and CT clears, compressing door-to-imaging and door-to-needle times.