Does the use of the Advanced Medical Priority Dispatch System affect cardiac arrest detection?

Comparing emergency medical service systems--a project of the European Emergency Data (EED) Project

AIM

The aim of this prospective study was the comparison of four emergency medical service (EMS) systems-emergency physician (EP) and paramedic (PM) based-and the impact of advanced live support (ALS) on patients status in preclinical care.

METHODS

The EMS systems of Bonn (GER, EP), Cantabria (ESP, EP), Coventry (UK, PM) and Richmond (US, PM) were analysed in relation to quality of structure, process and performance when first diagnosis on scene was cardiac arrest (OHCA), chest pain or dyspnoea. Data were collected prospectively between 01.01.2001 and 31.12.2004 for at least 12 month.

RESULTS

Over all 6277 patients were included in this study. The rate of drug therapy was highest in the EP-based systems Bonn and Cantabria. Pain relief was more effective in Bonn in patients with severe chest pain. In the group of patients with chest pain and tachycardia ≥ 120 beats/min, the heart rate was reduced most effective by the EP-systems. In patients with dyspnoea and S(p)O(2) <90% the improvement of oxygen saturation was most effective in Bonn and Richmond. After OHCA significant more patients reached the hospital alive in EMS systems with EPs than in the paramedic staffed (Bonn = 35.6%, Cantabria = 30.1%; Coventry = 11.9%, Richmond = 9.2%). The introduction of a Load Distributing Band chest compression device in Richmond improved admittance rate after OHCA (21.7%) but did not reach the survival rate of the Bonn EMS system.

CONCLUSIONS

Higher qualification and greater training and experience of ALS unit personnel increased survival after OHCA and improved patient's status with cardiac chest pain and respiratory failure.

How many emergency dispatches occurred per cardiac arrest?

BACKGROUND

The Medical Priority Dispatch System (MPDS) is an emergency medical dispatch (EMD) system that is widely used to prioritize 9-1-1 calls and optimize resource allocation. Calls are assigned an MPDS determinant, which includes a number (1-32) representing chief complaint and priority (Alpha through Echo) representing acuity.

OBJECTIVE

This study evaluates the number of emergency dispatches per cardiac arrest (NOD-CA) in cardiac arrest and non-cardiac arrest MPDS determinants.

METHODS

All patients assigned a determinant by MPDS from January 1, 2008 to June 30, 2009 in a large metropolitan area were included. Prehospital electronic patient care records were linked with dispatch data. For each MPDS determinant, the number of calls for which the paramedic impression was listed as "Cardiac Arrest - Non-Traumatic" was tabulated. The NOD-CA was calculated for each cardiac arrest and non-cardiac arrest MPDS determinant. Non-MPDS calls with cardiac arrests were analyzed separately.

RESULTS

A total of 101,642 patients were included. Among them, 555 had "Cardiac Arrest - Non-Traumatic" listed as the paramedic impression. The Cardiac/Respiratory Arrest/Death protocol had the highest number of cardiac arrests (285), followed by Breathing Problems (99) and Unconscious/Fainting (76). Overall, 183 dispatched occurred for each cardiac arrest, 131 of which resulted in a lights and sirens response. The NOD-CA was 7 in the Cardiac Arrest/Death protocol, 122 in Breathing Problems, and 104 in Unconscious/Fainting. 31 Cardiac arrests occurred in non-MPDS dispatch categories (N=62,989), most of which were calls for medical assistance from police or fire units.

CONCLUSIONS

MPDS was designed to detect cardiac arrest with high sensitivity, leading to a significant degree of mistriage. The number of dispatches for each cardiac arrest may be a useful way to quantify the degree of mistriage and optimize EMS dispatch. This large descriptive study revealed a low NOD-CA in most cardiac arrest MPDS determinants. We demonstrated significant variability in the NOD-CA among non-cardiac arrest MPDS determinants, and few cardiac arrests in non-MPDS dispatch categories.

Evaluation of an algorithm for estimating a patient's life threat risk from an ambulance call

Background

Utilizing a computer algorithm, information from calls to an ambulance service was used to calculate the risk of patients being in a life-threatening condition (life threat risk), at the time of the call. If the estimated life threat risk was higher than 10%, the probability that a patient faced a risk of dying was recognized as very high and categorized as category A+. The present study aimed to review the accuracy of the algorithm.

Methods

Data collected for six months from the Yokohama new emergency system was used. In the system, emergency call workers interviewed ambulance callers to obtain information necessary to assess triage, which included consciousness level, breathing status, walking ability, position, and complexion. An emergency patient's life threat risk was then estimated by a computer algorithm applying logistic models. This study compared the estimated life threat risk occurring at the time of the emergency call to the patients' state or severity of condition, i.e. death confirmed at the scene by ambulance crews, resulted in death at emergency departments, life-threatening condition with occurrence of cardiac and/or pulmonary arrest (CPA), life-threatening condition without CPA, serious but not life-threatening condition, moderate condition, and mild condition. The sensitivity, specificity, predictive values, and likelihood ratios of the algorithm for categorizing A+ were calculated.

Results

The number of emergency dispatches over the six months was 73,992. Triage assessment was conducted for 68,692 of these calls. The study targets account for 88.8% of patients who were involved in triage calls. There were 2,349 cases where the patient had died or had suffered CPA. The sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio and negative likelihood ratio of the algorithm at predicting cases that would result in a death or CPA were 80.2% (95% confidence interval: 78.6% - 81.8%), 96.0% (95.8% - 96.1%), 42.6% (41.1% - 44.0%), 99.2% (99.2% - 99.3%), 19.9 (18.8 - 21.1), and 0.21 (0.19 - 0.22), respectively.

Conclusion

A patient's life threat risk was quantitatively assessed at the moment of the emergency call with a moderate level of accuracy.

Statewide Survey of 911 Communication Centers on Acute Stroke andMyocardial Infarction

OBJECTIVE

We conducted a statewide survey of communication centers regarding practices, training, and outreach for stroke and myocardial infarction (MI) and explored differences for those that were Emergency Medical Dispatch (EMD) certified or not.

METHODS

A survey was mailed to all 115 centers that dispatched for emergency medical services (EMS) in North Carolina. Seventy-four percent (85/115) returned the survey.

RESULTS

For 31% of centers, all telecommunicators were EMD-certified, 28% of centers had some personnel certified, and 41% had no personnel certified. Forty-four percent of centers used dispatch guides or algorithms to aid telecommunicators. If telecommunicators suspected a stroke, 47% of the centers provided prearrival instructions and if they suspected an MI, 49% provided prearrival instructions. In the past 2 years, 27% of the telecommunicators received stroke-specific training and 29% received MI-specific training. Stroke or MI training for telecommunicators was more common among the EMD centers (51% stroke; 51% MI) than among the non-EMD centers (5% stroke; 9% MI). Only one center conducted a community outreach program about stroke or MI in the last 6 months.

CONCLUSIONS

Our results suggest that the development of statewide telecommunication training program to improve knowledge and care for suspected stroke or MI is needed in North Carolina. Dispatching for stroke and MI could be enhanced by requiring all communication centers to be EMD-certified and by creating consistent and standard dispatching practices across the state, using triage algorithms. Implementing these changes could improve rapid response and care for acute stroke and MI patients.

Low Acuity EMS Dispatch Criteria Can Reliably Identify Patients without High-Acuity Illness or Injury

OBJECTIVE

This retrospective study evaluated the appropriateness of requests assigned the alpha determinant at the time of dispatch by Emergency Medical Dispatchers using the Medical Priority Dispatch System (MPDS).

METHODS

The primary end point was the proportion of EMS calls assigned the alpha dispatch determinant that resulted in a high-acuity patient encounter. Patient care reports from January 1 to June 30, 2004, were eligible for inclusion; reports with known errors in data entry or those that were not the result of caller interrogation via the 9-1-1 emergency medical dispatch center (EMDC) were excluded. High-acuity patients were defined as those who met trauma triage criteria or received treatment for acute coronary syndrome, respiratory distress, altered mental status, acute stroke, allergic reaction, or abnormal vital signs. Secondary end points included call-processing time, the proportion of included patients who were transported emergently to hospital, and the adherence of the EMDC to National Academy of Emergency Dispatch (NAED) quality assurance guidelines.

RESULTS

There were 23,939 dispatches; 2,703 were recorded as alpha dispatches in the electronic patient care report (ECR), of which 582 were excluded. Twenty-one of 2,121 calls (<1%) meeting inclusion criteria met high-acuity criteria and were considered as inappropriate alpha dispatches. Fourteen of 2,121 (<1%) were transported emergently to the hospital, eight of whom also met the high-acuity criteria. The call-processing time at the 90th percentile was 2 minutes and 29 seconds. The EMDC demonstrated 99% compliance with NAED quality assurance standards.

CONCLUSION

The use of standard MPDS protocols can successfully identify patients who do not demonstrate high-acuity illness or injury more than 99% of the time.

VALIDATION OF USING EMS DISPATCH CODES TO IDENTIFY LOW-ACUITY PATIENTS

OBJECTIVE

To validate the predictive ability of previously derived emergency medical services (EMS) dispatch codes to identify patients with low-acuity illnesses.

METHODS

This prospective descriptive study was conducted in Rochester, New York. An expert panel reviewed and modified a previously derived set of low-priority EMS dispatch codes. Patients assigned these 21 codes between July 2002 and June 2003 were included for further analysis. Dispatch data and level of EMS care were recorded for each dispatch code. The proportion of low-acuity patients (i.e., those who received only basic life support (BLS) care or those who were not transported using lights and sirens) was determined using previously established definitions. Codes were defined as associated with low-acuity patients if the lower bound of the 95% confidence interval (CI) exceeded 90%. Medical records for patients identified as high-acuity were reviewed to evaluate whether the advanced life support (ALS) level care that was provided had a clinical impact.

RESULTS

Emergency medical services cared for 43,602 patients during the study, and 7,540 were dispatched as low-priority. We found that 7,197 (95%; 95% CI: 95-96%) of these patients met low-acuity criteria and that 11 of the evaluated codes were validated, with low-acuity care provided at least 90% of the time. Of the 343 patients identified as high-acuity, 62 (18%; 95% CI: 14-23%) were determined to have received interventions that had a clinical impact.

CONCLUSIONS

This study prospectively validates 11 EMS dispatch codes as being associated with low-acuity patients. These codes could be used to triage EMS patients based on dispatch information.

Can emergency medical dispatch codes predict prehospital interventions for common 9-1-1 call types?

OBJECTIVE

The Medical Priority Dispatch System is an emergency medical dispatch (EMD) system that is widely used to categorize 9-1-1 calls and optimize resource allocation. This study evaluates the ability of EMD and non-EMD codes (calls not processed by EMD) to predict prehospital use of medications and procedures.

METHODS

All transported prehospital patients placed in an EMD or non-EMD category that exceeded 500 total calls from January 1, 2004, to December 31, 2006, in a suburban California county were matched with their prehospital electronic patient care record. These records (N = 69,541) were queried for the following prehospital interventions: basic life support (BLS) care only, intravenous line placement only, medication given, and procedures. Advanced life support (ALS) interventions were defined as the administration of a medications or a procedure. The numbers of medications and procedures that were performed on patients in each EMD code were measured.

RESULTS

Thirty-one of 141 EMD and non-EMD codes met inclusion criteria and comprised 73% of all calls during the study period. Non-EMD codes accounted for 48% of all calls in this study. Patients with shortness of breath, chest pain, diabetic problems, and altered mental status received the most medications. High rates of medication administration were also seen in the following codes: 17A (fall, 27%), 17B (fall, 14%), EMDX (unable to complete EMD process, 22%), MED (medical aid requested--details to follow, 26%), and MED3 (medical aid requested by police--code 3, 18%). Procedures were performed on only 0.9% of all calls, of which 75% were related to advanced airways. Higher rates of ALS interventions in higher-acuity categories (Alpha, Bravo, etc.) were seen in a number of EMD categories, including seizure, laceration/hemorrhage, sick, and traffic accident, but not seen in many categories, including abdominal pain, falls, and chest pain.

CONCLUSIONS

This study demonstrated only a modest ability of the EMD system to predict which patients would require ALS intervention. There were limited differences noted in the ALS rates between the different codes (Alpha, Bravo, etc.) in the same complaint category, bringing into question the utility of the multiple subgroups. Non-EMD codes made up a large portion of calls (48%) and should be included in future studies.

Ability of the medical priority dispatch system protocol to predict the acuity of "unknown problem" dispatch response levels

OBJECTIVE

To determine if Medical Priority Dispatch System's (MPDS's) Protocol 32-Unknown Problem interrogation-based differential dispatch coding distinguishes the acuity of patients as found at the scene by responders, when little (if any) clinical information is known.

METHODS

"Unknown problem" situations (i.e., all cases not fitting into any other chief complaint group) constitute 5-8% of all calls to dispatch centers. From the total patient encounters (n=599,107) in the aggregate data of one year (September 2005 to August 2006), we examined 3,947 (0.7%) encounters initially coded as "unknown problem" by the London Ambulance Service Communications Center for the scene presence of cardiac arrest (CA) and paramedic-determined high-acuity (blue-in [BI]/"lights and siren") findings. Odds ratios (ORs) with 95% confidence intervals (95% CIs) and p-values were used to assess the degree of associations between determinant codes and case outcomes (i.e., CA/BI).

RESULTS

Statistically significant association between clinical dispatch determinant codes and case outcomes was observed in the "life status questionable" (LSQ; DELTA-1 [D-1]) and the "standing, sitting, moving, or talking" (BRAVO-1 [B-1]) code pair for the CA outcome (OR [95% CI]: 0.11 [0, 0.63], p=0.005) and for the BI outcome (OR [95% CI]: 0.47 [0.28, 0.77], p=0.003). The LSQ and all three code pairs (i.e., B-1; "community alarm notifications" [B-2]; and "unknown status" [B-3]) also demonstrated significant associations both with the CA outcome (OR [95% CI]: 0.43 [0.23, 0.81], p=0.010) and with the BI outcome (OR [95% CI]: 0.74 [0.56, 0.97], p=0.033). All the determinant code levels yielded significant association between BI and CA cases.

CONCLUSION

This dispatch protocol for unknown problems successfully differentiates dispatch coding of low-acuity and non-CA patients only when specific situational information such as the patient's standing, sitting, moving, or talking can be determined during the interrogation process. Also, emergency medical dispatcher (EMD) reliance on caller-volunteered information to identify predefined critical situations does not appear to add to the protocol's ability to differentiate high-acuity and CA patients. LSQ proved to be a better predictor of both CA and BI outcomes, when compared with the BRAVO-level determinant codes within the "unknown problem" chief complaint. The B-3 (completely unknown) determinant code is a better predictor of severe outcomes than nearly all of the clinically similar BRAVO determinant codes in the entire MPDS protocol. Hence, the B-3 coding should be considered-in terms of its predictability for severe outcome-as falling somewhere between a typical DELTA and a typical BRAVO determinant code.

Evaluating the effectiveness of dispatch-assisted cardiopulmonary resuscitation instructions

OBJECTIVES

To determine the frequency of agonal breathing during cardiac arrest (CA), its impact on the ability of 9-1-1 dispatchers to identify CA, and the impact of dispatch-assisted cardiopulmonary resuscitation (CPR) instructions on bystander CPR rates.

METHODS

A before-after observational study enrolling out-of-hospital adult CA patients where resuscitation was attempted in a single city with basic life support with defibrillation/advanced life support tiered emergency medical services. Victim, caller, and system characteristics were measured during two successive nine-month periods before (control group) and after (intervention group) the introduction of dispatch-assisted CPR instructions.

RESULTS

There were 529 CAs between July 1, 2003, and December 31, 2004. Victim characteristics were similar in the control (n = 295) and intervention (n = 234) period; mean age was 68.3 years; 66.7% were male; 50.1% of CAs were witnessed; call-to-vehicle stop was 6 minutes, 37 seconds; ventricular fibrillation/ventricular tachycardia occurred in 29.9%; and the survival rate was 4.0%. Dispatchers identified 56.3% (95% confidence interval [CI] = 48.9% to 63.0%) of CA cases; agonal breathing was present in 37.0% (95% CI = 30.1% to 43.9%) of all CA cases and accounted for 50.0% (95% CI = 39.1% to 60.9%) of missed diagnoses. Callers provided ventilations in 17.2% and chest compressions in 8.3% of cases as a result of the intervention. Long time intervals were observed between call to diagnosis (2 minutes, 38 seconds) and during ventilation instructions (2 minutes, 5 seconds). Bystander CPR rates increased from 16.7% in the control phase to 26.4% in the intervention phase (absolute rate, 9.7%; 95% CI = 8.5% to 11.3%; p = 0.006).

CONCLUSIONS

This trial demonstrates an increase in bystander CPR rate after the introduction of dispatch-assisted CPR. Agonal breathing occurred frequently and had a negative impact on the recognition of CA. There were long time intervals between call initiation and diagnosis of CA and during mouth-to-mouth ventilation instructions.

Cardiac arrest predictability in seizure patients based on emergency medical dispatcher identification of previous seizure or epilepsy history

OBJECTIVE

To determine predictability of at-scene cardiac arrest from a dispatch determined patient history of seizure or epilepsy ("E" history).

DESIGN AND METHODS

A retrospective study of a 1 year dataset from the London Ambulance Service (LAS) National Health Service (NHS) Trust was undertaken. Each of the nine determinant codes on the Medical Priority Dispatch System (MPDS) seizure protocol [Heward A, Damiani M, Hartley-Sharpe C. Does the use of the Advanced Medical Priority Dispatch System affect cardiac arrest detection? Emerg Med J 2004;21:115-8.] was examined for the addition of the "E" suffix finding. The cardiac arrest predictability of cases with reported "E" history was compared to those without using a protocol process to detect the infrequent but predictable presence of seizures caused by anoxic cardiac arrest.

RESULTS

Only protocol codes 12-A-1, 12-D-2, 12-D-3, and 12-D-4 demonstrated significant associations between outcomes and determinant codes (p=0.016, 0.007, <0.001, and 0.048, respectively). These codes showed reduced risk of predicting CA with the "E" suffix protocol determinant codes (RD (95% CI): -0.0025 (-0.0044, -0.0005), chi-square p=0.009; RD (95% CI): -0.0024 (-0.0042, -0.0005), p=0.005; RD (95% CI): -0.020 (-0.029, -0.011), p<0.001; RD (95% CI): -0.01 (-0.017, -0.005), and p=0.034, respectively).

CONCLUSIONS

Knowing whether a seizure patient is an epileptic or has had previous seizures is of clinical value and relevant to dispatch. By improving the discernment of the seizure protocol regarding seizure associated with anoxic cardiac arrest predictability, this information may now be applied at the response level as well as to emergency medical dispatcher's (EMD) decisions to stay on the telephone to enhance the monitoring of these patients.

Out‐of‐hospital cardiac arrest in Victoria: rural and urban outcomes

OBJECTIVE

To compare the survival rate from out-of-hospital cardiac arrest in rural and urban areas of Victoria, and to investigate the factors associated with these differences.

DESIGN

Retrospective case series using data from the Victorian Ambulance Cardiac Arrest Registry.

SETTING

All out-of-hospital cardiac arrests occurring in Victoria that were attended by Rural Ambulance Victoria or the Metropolitan Ambulance Service.

PARTICIPANTS

1790 people who suffered a bystander-witnessed cardiac arrest between January 2002 and December 2003.

RESULTS

Bystander cardiopulmonary resuscitation was more likely in rural (65.7%) than urban areas (48.4%) (P = 0.001). Urban patients with bystander-witnessed cardiac arrest were more likely to arrive at an emergency department with a cardiac output (odds ratio [OR], 2.92; 95% CI, 1.65-5.17; P < 0.001), and to be discharged from hospital alive than rural patients (urban, 125/1685 [7.4%]; rural, 2/105 [1.9%]; OR, 4.13; 95% CI, 1.09-34.91). Major factors associated with survival to hospital admission were distance of cardiac arrest from the closest ambulance branch (OR, 0.87; 95% CI, 0.82-0.92), endotracheal intubation (OR, 3.46; 95% CI, 2.49-4.80), and the presence of asystole (OR, 0.50; 95% CI, 0.38-0.67) or pulseless electrical activity (OR, 0.73; 95% CI, 0.56-0.95) on arrival of the first ambulance crew.

CONCLUSIONS

Survival rates differ between urban and rural cardiac arrest patients. This is largely due to a difference in ambulance response time. As it is impractical to substantially decrease response times in rural areas, other strategies that may improve outcome after cardiac arrest require investigation.

Effect of protocol compliance to cardiac arrest identification by emergency medical dispatchers

OBJECTIVES

The objective of the study was to assess the effect of protocol compliance to the accuracy of cardiac arrest (CA) identification by the dispatchers.

METHODS

The study was conducted prospectively over a 1-year period in 1996. The calls categorized as non-traumatic CAs by the dispatcher and calls where the patient was in non-traumatic CA when ambulance crew arrived were included in the study. The data was collected from emergency call tape recordings and ambulance run sheets. The compliance to the protocol was defined as gathering information to two questions: (1) Is the patient awake or can she/he be awakened? and (2) Is she/he breathing normally?

RESULTS

The number of calls included in the study was 776 and the dispatchers identified 83% of the CAs. The protocol was adhered in 52.4% of calls, more often in witnessed than unwitnessed cases (72.3% versus 45.0%, P<0.001). In correctly identified CAs, the protocol compliance was 49.4%. The compliance was higher in cases of unidentified CAs (60.3%, P=0.0326) and in cases of wrongly identified as CAs (false positives, 61.9%, P=0.0276).

CONCLUSIONS

A high identification rate of CAs seems to be achievable despite poor protocol compliance by dispatchers.

Sensitivity and specificity of the medical priority dispatch system in detecting cardiac arrest emergency calls in Melbourne

INTRODUCTION

In Australia, cardiac arrest kills 142 out of every 100,000 people each year; with only 3-4% of out-of-hospital patients with cardiac arrest in Melbourne surviving to hospital discharge. Prompt initiation of cardiopulmonary resuscitation (CPR), defibrillation, and advanced cardiac care greatly improves the chances of survival from cardiac arrest. A critical step in survival is identifying by the emergency ambulance dispatcher potential of the probability that the person is in cardiac arrest. The Melbourne Metropolitan Ambulance Service (MAS) uses the computerized call-taking system, Medical Priority Dispatch System (MPDS), to triage incoming, emergency, requests for ambulance responses. The MPDS is used in many emergency medical systems around the world, however, there is little published evidence of the system's efficacy.

OBJECTIVE

This study attempts to undertake a sensitivity/specificity analysis to determine the ability of MPDS to detect cardiac arrest.

METHODS

Emergency ambulance dispatch records of all cases identified as suspected cardiac arrest by MPDS were matched with ambulance, patient-care records and records from the Victorian Ambulance Cardiac Arrest Registry to determine the number of correctly identified cardiac arrests. Additionally, cases that had cardiac arrests, but were not identified correctly at the point of call-taking, were examined. All data were collected retrospectively for a three-month period (01 January through 31 March 2003).

RESULTS

The sensitivity of MPDS in detecting cardiac arrest was 76.7% (95% confidence interval (CI): 73.6%-79.8%) and specificity was 99.2% (95% CI: 99.1-99.3%). These results indicate that cardiac arrests are correctly identified in 76.7% of cases.

CONCLUSION

Although the system correctly identified 76.7% of cardiac arrest cases, the number of false negatives suggests that there is room for improvement in recognition by MPDS to maximize chances for survival in out-of-hospital cardiac arrest. This study provides an objective and comprehensive measurement of the accuracy of MPDS cardiac-arrest detection in Melbourne, as well as providing a baseline for comparison with subsequent changes to the MPDS.