PCRTrend
EMS ANALYTICS
EMS + HOSPITAL INTEGRATION
Demo County EMS
Stroke EMS + Hospital Integration Dashboard
Pre-Hospital + Hospital Registry  ·  Policy PD# xxxx  ·  NEMSIS v3.5 × GWTG-Stroke  ·  2026-1Q
DashboardMedical Director FindingsHospital Integration
01
EMS SystemEvery stroke EMS touched. Source: ePCR export (StrokeData) · EMS data only
Stroke Transports
1,002
2026-1Q · all agencies
Positive Stroke Scales
831
82.9% of transports
Pre-Alert Compliance
97.6%
760 of 779 qualifying
▲ Above 95% target
Stroke Center Destination
84.6%
703 of 831 positives
▼ Below 95% target
Field Documentation Quality
MetricDocumentedTargetStatus
Stroke scale recorded96.4%95%✓
Blood glucose (BGL)92.1%95%✗
Symptom onset / LKW time88.3%95%✗
Last-known-well captured90.7%95%✗
Median symptom onset → EMS on scene: 47 min · community response gap.
Stroke Scale Classification
Positive
831
82.9%
Non-Concl.
107
10.7%
Negative
53
5.3%
Not Rec.
11
1.1%
Of 831 positive scales, 164 (19.7%) screened LVO-positive.
Stroke Scale Type & Score (EMS) EMS data
Scale Used (eVitals.30)n% of scalesMedian Score
Cincinnati (CPSS)61261.1%2 of 3
LAMS24824.8%3 of 5
RACE969.6%4 of 9
BE-FAST / Other464.6%—
Score documented on 96.4% of positive scales (eVitals.29). LAMS & RACE are the LVO-screening tools.
LVO Screening (EMS) EMS data
LVO-Suspected
164
16.4% of transports
To Thrombectomy Ctr
117
71.3% routed
Missed Routing
47
LVO+ to non-TSC
LVO suspicion flagged by LAMS ≥4 or RACE ≥5. Whether the field screen was right is answered in Section 02.
02
Matched Records — EMS Meets HospitalOne patient, both clocks. Source: merged EMS × Hospital registry · EMS + Hospital
Matched Patients
690
EMS records linked to a hospital record
Match Rate
93.0%
690 of 742 EMS hospital arrivals
Calculation Cohort
540
100% matches only · 150 fuzzy excluded
EMS Unmatched
52
no hospital record found
ⓘ
All calculations in Section 02 use 100% (perfect) matches only — n=540. Fuzzy matches (n=150) are partial-confidence links where the script could not confirm the EMS and hospital records are the same patient. They are counted toward the match rate for transparency but excluded from every clinical calculation below to avoid skewing times or metric accuracy.
End-to-End Pipeline — Symptom Onset to Needle (median) 100% matches · n=540
Onset → FMC
47 min
EMS
→
FMC → Scene Depart
14
EMS
→
Scene → Door
19
EMS
→
Door → Needle
47
Hospital
→
FMC → Needle
80
Full system
The only place this number exists is the merge — neither dataset can compute FMC → Needle alone. Computed on 540 perfect matches.
Did Pre-Notification Actually Move the Needle? 100% matches · n=540
EMS Pre-Alerted
41 min
median door → needle · n=410
No Pre-Alert
58 min
median door → needle · n=130
17 min faster to tPA when EMS Pre-Alerted Notification is received.
EMS Field Assessment vs Hospital Final Diagnosis 100% matches · n=540
EMS said stroke, hospital found…n%
Confirmed stroke (ischemic/hemorrhagic)42178.0%
TIA499.0%
Stroke mimic7013.0%
Field accuracy 78%. Mimics that triggered a stroke alert are a training signal.
EMS Stroke Score vs Hospital Diagnosis 100% matches · n=540
EMS LAMS ScorenConfirmed StrokeConfirmed LVO
0–1  (low)19858%4%
2–3  (moderate)21279%14%
4–5  (high)13091%47%
The score earns its keep: LAMS 4–5 confirms LVO 47% of the time vs 4% at 0–1 — a working field triage signal.
LVO: EMS Suspicion vs Hospital Confirmation 100% matches · n=540
 Hospital LVO+Hospital LVO−
EMS suspected LVO61 · true pos38 · false pos
EMS did not24 · false neg417 · true neg
Sensitivity 71.8% · PPV 61.6% · 24 confirmed LVOs the field screen missed — those ePCRs are extracted for QI.
03
Hospital System — Full Picture by Arrival ModeAll Cases. Source: Hospital stroke registry · Hospital data
Total Stroke Cases
1,240
all arrival modes
Median Door → Needle
52 min
all patients
D2N Under 60 min
68%
benchmark target
LVO Positive
19.4%
241 of 1,240
Arrival Mode Breakdown — How did the patient arrive at the facility?
Arrival ModeCases% of Total Median D2ND2N <60 minPre-NotifiedLVO+tPA Given
EMS / Ambulance arrivals74259.8%44 min79%89%21.0%18.2%
Transfers (from another facility)31225.2%71 min48%n/a22.4%9.6%
Walk-ins / Private vehicle18615.0%63 min54%0%11.8%12.4%
WHOLE SYSTEM1,240100%52 min68%53%19.4%14.8%
EMS arrivals to tPA 27 min faster than transfers — pre-notification is the measurable difference.
Final Diagnosis Distribution — whole system
1,240cases
Ischemic · LVO
241
19%
Ischemic · non-LVO
602
49%
Hemorrhagic
174
14%
TIA
112
9%
Stroke mimic
111
9%
Door-to-Needle by Arrival Mode
EMS arrivals
44 min
Walk-ins
63 min
Transfers
71 min
Whole system
52 min
Target: < 60 min.
Discharge Quality & Outcomes — whole system
Statin at D/C
88%
target 90%
Antithrombotic
94%
prescribed
Home Disposition
61%
vs SNF / rehab
In-Hospital Mortality
7.2%
89 cases

What This Integration Tells the Medical Director (sample)

EMS pre-notification saves 17–27 min to tPA — proven across matched records and arrival-mode splits.
LAMS 4–5 confirmed LVO 47% of the time — the field score genuinely predicts large-vessel occlusion.
78% field accuracy: 13% of EMS stroke alerts were mimics — a targeted training opportunity.
LVO screen sensitivity 71.8% — 24 confirmed LVOs were under-triaged; that is the review list.
Transfers are the bottleneck at 71 min D2N — the inter-facility process needs attention.
52 EMS records never matched — data-quality gap worth chasing in the run-number fields.
Stroke EMS + Hospital Integration · Demonstration data — figures are synthetic · Generated by PCRTrend