Agreement between emergency medical services and expert blood pressure measurements.

<p><b>BACKGROUND: </b>Emergency Medical Services (EMS)-measured blood pressures (BPs) are utilized for administering medications in the field and for triage decisions. Retrospective work has demonstrated poor agreement between EMS and Emergency Department (ED) BP but has lacked a valid, reliable reference standard.</p>
<p><b>STUDY OBJECTIVES: </b>To compare EMS BP measurements with those of trained research assistants (RA) and observe measurement technique for sources of error.</p>
<p><b>METHODS: </b>A prospective study was performed with a large urban EMS. BP measurements were made by RA within 5 min of patients presenting to the ED. EMS personnel were asked about technique. EMS personnel were then observed while RA simultaneously measured BP. Analysis was performed using methods outlined by Bland and Altman.</p>
<p><b>RESULTS: </b>There were 100 patients enrolled for each phase. In the first phase, the mean difference in systolic BP was -3.8 ± 18.6 mm Hg (95% confidence interval [CI] -8.3 to 0.59), and the mean difference in diastolic BP was 0.42 ± 13.8 mm Hg (95% CI -3.3 to 4.1). In the second phase, the mean difference in systolic BP was -4.6 ± 10.1 mm Hg (95% CI -6.6 to -2.6) and the mean difference in diastolic BP was -3.6 ± 10.6 mm Hg (95% CI -3.6 to -0.2). EMS personnel failed to properly place the cuff or deflate it 2-3 mm Hg/s in over 90% of the readings. They failed to properly inflate the cuff in 74% of the patients, and failed to properly place the stethoscope in 40%. EMS personnel demonstrated a significant preference for the terminal digit of "0" (p &lt; 0.0001).</p>
<p><b>CONCLUSIONS: </b>EMS and expert BP measurements showed smaller discrepancies than those previously noted, especially with simultaneous measurements. However, EMS demonstrated poor adherence to American Heart Association recommendations for measuring BP. EMS also showed terminal digit preference.</p>