Gender-Specific Ambulance Priority and Delays to Primary PCI
In Norway, most STEMI patients are admitted to hospital by ambulance after a telephone call (no. 113) to the EMC service or after contacting the general practitioner on-call. In many cases, it is the relatives or other companions who contact the EMC service. Special trained dispatch nurses at EMC service base their decisions on taking a structured medical history indicated in the Norwegian Index of Emergency Medical assistance (Index). The Index is a Norwegian version of the US Criteria Based Dispatch system. The severity grades considered are urgency ("Red" response), acutely ("Yellow" response), or nonurgent ("Green" response). If, in the telephone conversation with the patient, the presented symptoms correspond to a "Red" response alert, an ambulance or physician manned vehicles should be sent immediately. Trigger words for "Red" alert are unconsciousness, no breathing, chest pain (with or without radiation, diaphoresis, or dizziness), and dyspnea. Gender or the patients' ages are not trigger words. All emergency calls are handled by 2 operators (dispatch nurse and ambulance coordinator). It is always the nurses who set the criteria code and decide ambulance priority, and they always hold the line until the ambulance arrives at the scene.
For all inquiries to the EMC service, digital audio logs are recorded of the conversation between the caller and EMC service nurses, and between the nurses and the prehospital services. All ambulances are manned by authorized paramedics capable of taking and sending electrocardiogram (ECG), to activate the percutaneous coronary intervention (PCI) team and to perform primary medical treatment. An anesthesiologist-manned vehicle or helicopter in addition to the general practitioner on-call is activated routinely if AMI is suspected by the EMC service. Distance to the hospital, traffic congestion, and weather condition decide whether helicopter is used. The prehospitally identified STEMI patients are fast tracked directly to the catheterization laboratory. The PCI team are on-call during night time and weekends.
The Stavanger University Hospital is the only hospital in South Rogaland, Norway, and has a primary catchment population of approximately 295,000. From January 1, 2004, all patients admitted with STEMI around the clock underwent urgent coronary angiography and primary PCI if appropriate. We included all STEMI patients from January 1, 2004, to December 31, 2007, whose first contact with the health care system was by a 113 phone call to the Stavanger University Hospital EMC service. Data for all patients undergoing coronary angiography or PCI at our hospital (indication, findings, treatment, and follow-up plans) are entered in to a GE Centricity Xi2 database immediately after the procedure. All events and equipment used during the procedure are automatically registered or entered by the scrub nurse (eg, TIMI II-III flow time). We used the Centricity statistical module to initially identify all STEMI/primary PCI patients. In case of missing or unclear data, we reviewed the digital angiography film, where every scene is time stamped. We thereafter excluded STEMI patients not coming from our primary catchment area covered by our EMC center, by consulting the digital medical records and the hospital information system. The records for the remaining potential eligible patients were thereafter scrutinized at the EMC center, where patients not calling 113 directly were excluded.
We also investigated whether there were any overseen STEMI patient among the late comers (STEMI patient arriving at the catheterization laboratory >12 hours after symptom debut) who called 113 directly. However, no such patient were found as all STEMI patients calling 113 are transported by ambulance/helicopter, and are without delay transferred directly to the catheterization laboratory. Reperfusion therapy with thrombolytics was not used during the study period.
The included patients all had documented STEMI (ST elevation ≥0.1 mV in II, II, aVF [inferior STEMI], ≥0.2 mV in V2-V4 [anterior STEMI], ≥0.1 mV in I, aVL, V5, V6 [lateral STEMI], or new left bundle-branch block) on the presenting ECG, ischemic symptoms, and a subsequent typical rise or fall in serum troponin levels above the AMI decision limit.
Ambulance records kept at the EMC center logged the key milestones during the transport to hospital. Timings and positioning for the ambulances were GPS tracked and synchronized with the EMC center. The dialogues between the callers and EMC service nurse were digitally voice recorded in real time. The voice recordings were audited independently by 2 experienced dispatch nurses of different sex who were blinded to all subsequent information about the enrolled patients. The milestones in the EMC center handling were stopwatch measured with seconds' accuracy by the auditing nurses. All other time intervals are recorded with minutes' accuracy. The following milestones were recorded: the time he/she called 113, the time the dispatcher actually and ideally could have initiated ambulance service (that was when a trigger word were first mentioned in the dialogue), the time the ambulance arrived at the scene and at the hospital with the patients, and the time TIMI 2 to 3 flow in the infarction-related artery was established. Based on these milestones, the following intervals in relation to system and nonsystems delays were calculated: patients delay (from the patient's recall of symptom debut to calling 113), dispatch time (from calling to ambulance sent out), response delay (from call to ambulance arrival at the scene), call-for-help-to-door time (from calling 113 to arriving at the hospital), door-to-balloon time (from arriving at the hospital to revascularization was achieved), and total ischemic time (from symptom debut to balloon). To measure the performance of the integrated EMC service and STEMI care system, the interval from call-to-balloon time was calculated (E2B). The number of questions-answers (dialogue passages) and the length of the call were counted until an ambulance priority was set and entered into the ambulance record log.
The study was considered an internal registry study and was approved by the Regional Ethics Committee. The study database was approved by the National Data Inspectorate.
Categorical variables were compared with the χ test. All continuous variables were skewed and are presented as median (interquartile range [IQR]) and compared with the Mann-Whitney U test. Multivariable analyses were performed with logistic regression tests. All patient demographics, medical history data, PCI data, and EMC center decisions were complete. Of the voice recordings, 97.6% were available and interpretable. The patients' recall of symptom debut was exact in 86.5% of the cases. Symptom debut was recorded as missing if we could not find an exact time in the voice recording or in the admission medical record. The analyses were repeated including or excluding the patients presenting with cardiac arrest and with and without multiple imputation of missing values. The results were essentially the same. Possible interobserver differences in the scoring of symptoms and the dispatcher time delays analysis were evaluated with the Pearson correlation coefficient for continuous variables or nonparametric tests as appropriate. No significant differences were found. P < .05 was considered significant. The analyses were performed with SPSS version 18 (SPSS Inc., Chicago IL) and R version 2.15.2 (ggplot2 package).
No extramural funding was used to support this work. The authors are solely responsible for the design and conduct of this study; all study analyses, the drafting and editing of the manuscript, and its final contents.
Methods
Dispatch System
In Norway, most STEMI patients are admitted to hospital by ambulance after a telephone call (no. 113) to the EMC service or after contacting the general practitioner on-call. In many cases, it is the relatives or other companions who contact the EMC service. Special trained dispatch nurses at EMC service base their decisions on taking a structured medical history indicated in the Norwegian Index of Emergency Medical assistance (Index). The Index is a Norwegian version of the US Criteria Based Dispatch system. The severity grades considered are urgency ("Red" response), acutely ("Yellow" response), or nonurgent ("Green" response). If, in the telephone conversation with the patient, the presented symptoms correspond to a "Red" response alert, an ambulance or physician manned vehicles should be sent immediately. Trigger words for "Red" alert are unconsciousness, no breathing, chest pain (with or without radiation, diaphoresis, or dizziness), and dyspnea. Gender or the patients' ages are not trigger words. All emergency calls are handled by 2 operators (dispatch nurse and ambulance coordinator). It is always the nurses who set the criteria code and decide ambulance priority, and they always hold the line until the ambulance arrives at the scene.
For all inquiries to the EMC service, digital audio logs are recorded of the conversation between the caller and EMC service nurses, and between the nurses and the prehospital services. All ambulances are manned by authorized paramedics capable of taking and sending electrocardiogram (ECG), to activate the percutaneous coronary intervention (PCI) team and to perform primary medical treatment. An anesthesiologist-manned vehicle or helicopter in addition to the general practitioner on-call is activated routinely if AMI is suspected by the EMC service. Distance to the hospital, traffic congestion, and weather condition decide whether helicopter is used. The prehospitally identified STEMI patients are fast tracked directly to the catheterization laboratory. The PCI team are on-call during night time and weekends.
Study Patient Selection
The Stavanger University Hospital is the only hospital in South Rogaland, Norway, and has a primary catchment population of approximately 295,000. From January 1, 2004, all patients admitted with STEMI around the clock underwent urgent coronary angiography and primary PCI if appropriate. We included all STEMI patients from January 1, 2004, to December 31, 2007, whose first contact with the health care system was by a 113 phone call to the Stavanger University Hospital EMC service. Data for all patients undergoing coronary angiography or PCI at our hospital (indication, findings, treatment, and follow-up plans) are entered in to a GE Centricity Xi2 database immediately after the procedure. All events and equipment used during the procedure are automatically registered or entered by the scrub nurse (eg, TIMI II-III flow time). We used the Centricity statistical module to initially identify all STEMI/primary PCI patients. In case of missing or unclear data, we reviewed the digital angiography film, where every scene is time stamped. We thereafter excluded STEMI patients not coming from our primary catchment area covered by our EMC center, by consulting the digital medical records and the hospital information system. The records for the remaining potential eligible patients were thereafter scrutinized at the EMC center, where patients not calling 113 directly were excluded.
We also investigated whether there were any overseen STEMI patient among the late comers (STEMI patient arriving at the catheterization laboratory >12 hours after symptom debut) who called 113 directly. However, no such patient were found as all STEMI patients calling 113 are transported by ambulance/helicopter, and are without delay transferred directly to the catheterization laboratory. Reperfusion therapy with thrombolytics was not used during the study period.
The included patients all had documented STEMI (ST elevation ≥0.1 mV in II, II, aVF [inferior STEMI], ≥0.2 mV in V2-V4 [anterior STEMI], ≥0.1 mV in I, aVL, V5, V6 [lateral STEMI], or new left bundle-branch block) on the presenting ECG, ischemic symptoms, and a subsequent typical rise or fall in serum troponin levels above the AMI decision limit.
Identification of System and Nonsystem Delays
Ambulance records kept at the EMC center logged the key milestones during the transport to hospital. Timings and positioning for the ambulances were GPS tracked and synchronized with the EMC center. The dialogues between the callers and EMC service nurse were digitally voice recorded in real time. The voice recordings were audited independently by 2 experienced dispatch nurses of different sex who were blinded to all subsequent information about the enrolled patients. The milestones in the EMC center handling were stopwatch measured with seconds' accuracy by the auditing nurses. All other time intervals are recorded with minutes' accuracy. The following milestones were recorded: the time he/she called 113, the time the dispatcher actually and ideally could have initiated ambulance service (that was when a trigger word were first mentioned in the dialogue), the time the ambulance arrived at the scene and at the hospital with the patients, and the time TIMI 2 to 3 flow in the infarction-related artery was established. Based on these milestones, the following intervals in relation to system and nonsystems delays were calculated: patients delay (from the patient's recall of symptom debut to calling 113), dispatch time (from calling to ambulance sent out), response delay (from call to ambulance arrival at the scene), call-for-help-to-door time (from calling 113 to arriving at the hospital), door-to-balloon time (from arriving at the hospital to revascularization was achieved), and total ischemic time (from symptom debut to balloon). To measure the performance of the integrated EMC service and STEMI care system, the interval from call-to-balloon time was calculated (E2B). The number of questions-answers (dialogue passages) and the length of the call were counted until an ambulance priority was set and entered into the ambulance record log.
Ethical Considerations/Consent
The study was considered an internal registry study and was approved by the Regional Ethics Committee. The study database was approved by the National Data Inspectorate.
Statistics
Categorical variables were compared with the χ test. All continuous variables were skewed and are presented as median (interquartile range [IQR]) and compared with the Mann-Whitney U test. Multivariable analyses were performed with logistic regression tests. All patient demographics, medical history data, PCI data, and EMC center decisions were complete. Of the voice recordings, 97.6% were available and interpretable. The patients' recall of symptom debut was exact in 86.5% of the cases. Symptom debut was recorded as missing if we could not find an exact time in the voice recording or in the admission medical record. The analyses were repeated including or excluding the patients presenting with cardiac arrest and with and without multiple imputation of missing values. The results were essentially the same. Possible interobserver differences in the scoring of symptoms and the dispatcher time delays analysis were evaluated with the Pearson correlation coefficient for continuous variables or nonparametric tests as appropriate. No significant differences were found. P < .05 was considered significant. The analyses were performed with SPSS version 18 (SPSS Inc., Chicago IL) and R version 2.15.2 (ggplot2 package).
No extramural funding was used to support this work. The authors are solely responsible for the design and conduct of this study; all study analyses, the drafting and editing of the manuscript, and its final contents.