Early diagnosis and consequent treatment of acute myocardial

Use of Emergency Medical Service Transport Among Patients With ST-Segment Elevation Myocardial Infarction Findings From the National Cardiovascular Data Registry Acute Coronary Treatment Intervention Outcomes Network Registry Get With the Guidelines Robin Mathews, MD; Eric D. Peterson, MD, MPH; Shuang Li, MS; Matthew T. Roe, MD, MHS; Seth W. Glickman, MD, MBA; Stephen D. Wiviott, MD; Jorge F. Saucedo, MD; Elliott M. Antman, MD; Alice K. Jacobs, MD; Tracy Y. Wang, MD, MHS, MSc Background Activation of emergency medical services (EMS) is critical for the early triage and treatment of patients experiencing ST-segment elevation myocardial infarction, yet data regarding EMS use and its association with subsequent clinical care are limited. Methods and Results We performed an observational analysis of 37 634 ST-segment elevation myocardial infarction patients treated at 372 US hospitals participating in the National Cardiovascular Data Registry Acute Coronary Treatment and Intervention Outcomes Network Registry Get With the Guidelines between January 2007 and September 2009, and examined independent patient factors associated with EMS transportation versus patient self-transportation. We found that EMS transport was used in only 60% of ST-segment elevation myocardial infarction patients. Older patients, those living farther from the hospital, and those with hemodynamic compromise were more likely to use EMS transport. In contrast, race, income, and education level did not appear to be associated with the mode of transport. Compared with self-transported patients, EMS-transported patients had significantly shorter delays in both symptomonset-to-arrival time (median, 89 versus 120 minutes; P 0.0001) and door-to-reperfusion time (median door-to-balloon time, 63 versus 76 minutes; P 0.0001; median door-to-needle time, 23 versus 29 minutes; P 0.0001). Conclusions Emergency medical services transportation to the hospital is underused among contemporary ST-segment elevation myocardial infarction patients. Nevertheless, use of EMS transportation is associated with substantial reductions in ischemic time and treatment delays. Community education efforts are needed to improve the use of emergency transport as part of system-wide strategies to improve ST-segment elevation myocardial infarction reperfusion care. (Circulation. 2011;124:154-163.) Key Words: emergency medical services myocardial infarction outcomes Early diagnosis and consequent treatment of acute myocardial infarction (MI) is associated with improved clinical outcomes. This is particularly relevant for patients with ST-segment elevation MI (STEMI) in whom early reperfusion is associated with decreased morbidity and mortality. 1,2 Care processes to expedite emergency treatment of acute MI patients after presentation to the hospital have been well defined. 1,3 In recent years, the adoption of these care processes has led to significant reductions in the time from hospital door arrival to reperfusion. 1,2,4,5 Clinical Perspective on p 163 Yet, if the overall goal is to reduce the delay from symptom onset to treatment, then the scope of STEMI care must be broadened to include prehospital care. 3 The American Heart Association s Mission: Lifeline initiative aims to increase the number of patients with timely access to reperfusion by addressing the continuum of care for STEMI, beginning with patient recognition of symptoms (http://www.heart.org). Emergency medical services (EMS) activation is critical not Received October 13, 2010; accepted May 3, 2011. From the Duke Clinical Research Institute, Durham, NC (R.M., E.D.P., S.L., M.T.R., T.Y.W.); University of North Carolina, Chapel Hill (S.W.G.); Brigham and Women s Hospital, Boston, MA (S.D.W., E.M.A.); University of Oklahoma Health Sciences Center, Oklahoma City (J.F.S.); and Boston University School of Medicine, Boston, MA (A.K.J.). Guest Editor for this article was Judith S. Hochman, MD. The online-only Data Supplement is available with this article at http://circ.ahajournals.org/cgi/content/full/circulationaha.110.002345/dc1. Correspondence to Robin Mathews, MD, Duke Clinical Research Institute, Duke University Medical Center, 2400 Pratt St, Durham, NC 27705. E-mail robin.mathews@duke.edu 2011 American Heart Association, Inc. Circulation is available at http://circ.ahajournals.org DOI: 10.1161/CIRCULATIONAHA.110.002345 154

Mathews et al Use of EMS in Patients With STEMI 155 Figure 1. Population flow diagram. Initial study population is broken down (after exclusions) into an overall study population and then further divided into cohort of patients with ZIP code data available. GWTG indicates Get With the Guidelines; STEMI, ST-segment elevation myocardial infarction; and EMS, emergency medical services. only for rapid transport to the hospital, but also to provide the opportunity for early assessment and treatment, as well as activities such as evaluation of hemodynamic stability, prehospital medical stabilization, prehospital ECG in certain areas, and expedited communication with the accepting hospital. Despite community educational efforts, studies from the 1990s found that a significant proportion of STEMI patients failed to use EMS. 6,7 The National Cardiovascular Data Registry (NCDR) Acute Coronary Treatment and Intervention Outcomes Network Registry Get With the Guidelines (ACTION Registry GWTG) offers a unique opportunity to examine contemporary patterns of EMS use nationwide and to probe some of the sociodemographic and clinical factors that contribute to a patient s decision to use EMS. Therefore, the objectives for this study were to describe the prevalence of EMS transport compared with self-transport to the emergency department, to determine factors independently associated with an STEMI patient s mode of transport, and to compare differences in treatment times between STEMI patients transported via EMS and those who were self-transported. Methods Study Population The NCDR ACTION Registry GWTG is an ongoing quality improvement registry of consecutive patients with acute MI who are treated at participating centers across the United States. The registry is an initiative of the American College of Cardiology Foundation and the AHA, with partnering support from the Society of Chest Pain Centers, Society of Hospital Medicine, and American College of Emergency Physicians. Inclusion and exclusion criteria, data collection, and variables have been described previously. 8 The individual institutional review board of each reporting hospital approved participation in ACTION Registry GWTG. All data were abstracted retrospectively and anonymously; therefore, informed consent was not required. All patients admitted with STEMI who reported to the ACTION Registry GWTG from January 1, 2007, to September 30, 2009 were included in the initial study population (Figure 1). A diagnosis of STEMI was defined as persistent ST-segment elevation 1 mmin 2 contiguous ECG leads or an STEMI equivalent such as new or presumed new left bundle-branch block or an isolated posterior MI. For this analysis, we excluded patients with missing mode of transportation (n 970). We also excluded patients who were transferred into an NCDR ACTION Registry GWTG hospital from another acute care hospital (n 19 258) because the processes of care are inherently different in patients who require transfer between facilities for reperfusion. Study Variables and Definitions Participating hospitals collected detailed information on baseline demographic and clinical characteristics, processes of care, and in-hospital outcomes using a standardized set of data elements and definitions. 9 Patient data were captured retrospectively via chart review and entered by sites into a Web-based data collection form. Data were screened on entry; only those meeting predetermined criteria for completeness and accuracy could be entered into the database. Each quarter, sites received a report summarizing any observed data quality issues, thus iteratively improving the overall quality of the database. On the data collection form, EMS transport was defined as transportation to the hospital by ambulance, mobile intensive care unit, or air transportation. Self-transport was defined as any mode of

156 Circulation July 12, 2011 transportation that did not involve EMS services; self-transported patients took taxis or public transportation, drove themselves, were driven by others, or walked to the hospital. All other variables have been previously described. 8 Statistical Methods The study population was categorized according to their primary mode of transport (EMS or self-transport) to the presenting hospital. Patient characteristics and outcomes in each group were described using frequencies and percentages for categorical variables and median (with 25th and 75th percentiles) for continuous variables. Baseline characteristics of patients in the 2 groups were compared by use of SDs. Missing data were rare ( 4.25%). In multivariable modeling, missing values of the continuous variables were imputed to the median value, and missing values of the categorical variables were imputed to their most common value. Generalized estimating equations logistic regression modeling with adjustment for within-hospital clustering was used to determine factors associated with the decision to use EMS transport. Patientlevel covariates examined in this model included age, sex, race, Hispanic ethnicity, insurance status, history of heart failure, MI, stroke, coronary artery bypass grafting surgery, percutaneous coronary intervention, hypertension, diabetes mellitus, chronic lung disease, atrial fibrillation or flutter, peripheral artery disease, requirement for hemodialysis, body mass index, systolic blood pressure, heart rate, heart failure on presentation, shock, dyslipidemia, and time of presentation (ie, working hours, off hours). The ACTION mortality risk score, previously developed and validated in this data set as a risk prediction tool, was used to compare overall risk balance between the 2 groups. 10 The generalized estimating equation method was implemented with a compound symmetrical working correlation matrix and empirical (sandwich) SE estimates. Additional sociodemographic and hospital-level variables, such as distance from patient home to hospital, urban versus rural residence, educational level, income level, and hospital region, may play a role in patient selection of transport mode, but are not typically captured in large registry databases. Version 2 of the ACTION Registry GWTG data collection form (implemented in July 2008) started collecting patient ZIP code information. Therefore, as a second step, we included only version 2 patients entered between July 2008 and September 2009 for whom patient ZIP code information was collected, excluding patients with missing ZIP code information (n 308). The distance from a patient s home was determined from the geographic distance between the patient s and hospital s ZIP codes. Additional socioeconomic variables were extrapolated from US Census data using patient ZIP codes linked to Federal Information Processing Standard county and state codes. These variables were added as additional covariates to the generalized estimating equation model described above. Patient characteristics were similar between those with ZIP code information and the overall cohort (see the Appendix in the online-only Data Supplement). We then compared time delay from symptom onset to hospital arrival, as well as time delay from arrival to reperfusion, according to travel distance stratified by tertiles. Generalized estimating equations logistic regression modeling was also used to examine the association between mode of transport and key reperfusion metrics, including rate of reperfusion use, door-to-first-ecg time within 10 minutes, and door-to-balloon time within 90 minutes, and symptom-onset-to-arrival time within 120 minutes. Adjustment variables in these models included those baseline characteristics found to be significantly different (P 0.05) between the EMS and self-transport groups. All analyses were performed with SAS version 9.2 (SAS Institute, Cary, NC). Results Baseline Characteristics Among 37 634 STEMI patients, 22 585 (60.0%) activated EMS for transport to the hospital. Patients who used EMS were more frequently older and women; those who selftransported were more likely to be Hispanic (Table 1). Patients who used EMS transport were more likely to have comorbid conditions such as a prior history of MI, percutaneous coronary intervention, heart failure, or stroke. Although the severity of anginal symptoms was not captured, EMS-transported patients were often sicker on presentation than self-transported patients, with a higher incidence of cardiogenic shock (8.9% versus 2.7%) and signs of heart failure (14.0% versus 8.3%). We also noted that, on average, those who used EMS lived farther from the hospital than those who self-transported (median distance, 6.9 versus 5.5 miles); however, there was no distinction in urban versus rural patient origination (Table 1). Among patients who activated EMS, 43% also received an ECG for diagnosis before hospital arrival. Independent Factors Associated With Emergency Medical Services Use After multivariable adjustment, baseline sociodemographic and clinical factors independently associated with the decision to use EMS versus self-transport are presented in Table 2. Age was the strongest sociodemographic variable associated with EMS use, with older STEMI patients being more likely to activate EMS than younger patients. Male patients were less likely to activate EMS. Although Hispanic ethnicity was associated with less EMS use, the association with race was not as strong. Those with private insurance were least likely to use EMS compared with patients with governmentfunded or no insurance. Clinical signs and symptoms of hemodynamic instability (ie, heart failure or low blood pressure) were the strongest clinical variables associated with EMS use. Table 3 presents the model that incorporates the additional sociodemographic variables derived from patient home ZIP code data. Greater geographic distance from home to hospital was shown to be the second strongest factor associated with use of EMS services. Patients who lived 10 miles from their presenting hospital had a 64% greater odds of using EMS than those living within 4 miles. Neighborhood education and income levels were not significantly associated with mode of transport. Differences in Time to Care Between Emergency Medical Services Transport and Self-Transport When process-of-care measures were compared between groups, self-transported patients experienced longer delays to care compared with their EMS-transported counterparts (Table 4). Those patients who used EMS arrived at the hospital 30 minutes sooner after symptom onset than those who self-transported (120 versus 89 minutes; P 0.001). After multivariable adjustment, self-transported patients remained significantly less likely to arrive within 120 minutes of symptom onset (odds ratio, 0.55; 95% confidence interval, 0.50 to 0.60; P 0.001). Despite the younger age and fewer clinical comorbidities in the self-transported group, the rate of contraindication to reperfusion was similar to that in the EMS-transported group (15.7% versus 15.1%). Compared with EMS-transported patients, those who were self-transported were more likely to

Mathews et al Use of EMS in Patients With STEMI 157 Table 1. Baseline Characteristics Comparing Self-Transported Patients and Emergency Medical Services Transported Patients Self-Transport (n 15 049) EMS Transport (n 22 585) SD Demographics Age, y 59 (51 69) 62 (53 74) 24.0 Male, n (%) 10930 (72.6) 15034 (66.6) 13.2 Race, n (%) White 13494 (89.7) 20232 (89.6) 1.0 Black 1219 (8.1) 1860 (8.2) 0.5 Other 336 (2.2) 493 (2.2) 0.3 Hispanic ethnicity, n (%) 733 (4.9) 753 (3.3) 7.6 Insurance status, n (%) HMO/private 9408 (62.5) 12814 (56.7) 11.8 Medicare 2860 (19.0) 5620 (24.9) 14.2 Medicaid/military/VA 668 (4.4) 1174 (5.2) 3.5 Self/none 2113 (14.0) 2977 (13.2) 2.5 Median household income, $* 43518 (39381 51304) 43519 (39766 52684) 3.7 College education, n %* 24.9 (17.6 30.6) 25.1 (17.9 30.7) 2.4 Distance to hospital, miles* 5.5 (2.6 10.9) 6.9 (3.4 14.7) 1.8 Rural residence, n (%)* 955 (14) 1500 (13.5) 0.4 Geographic region in the United States, n (%) Northeast 1361 (9.0) 2286 (10.1) 3.7 Midwest 5807 (38.6) 6960 (30.8) 16.4 West 2162 (14.4) 3363 (14.9) 1.5 South 5719 (38.0) 9976 (44.2) 12.6 Clinical characteristics Diabetes mellitus, n (%) 3395 (22.6) 5109 (22.6) 0.1 Hypertension, n (%) 8992 (59.8) 14169 (62.7) 6.1 Prior MI, n (%) 2683 (17.8) 4877 (21.6) 9.5 Prior PCI, n (%) 2713 (18.0) 4880 (21.6) 9.0 Prior stroke, n (%) 586 (3.9) 1401 (6.2) 10.6 Prior CABG, n (%) 1095 (7.3) 1700 (7.5) 1.0 Prior HF, n (%) 590 (3.9) 1554 (6.9) 13.1 Dyslipidemia, n (%) 7418 (49.3) 11278 (49.9) 1.3 Creatinine clearance, 89 (65.7 114.7) 80.3 (55.3 107.5) 19.1 ml/min HF on presentation, n (%) 1252 (8.3) 3169 (14.0) 18.2 Cardiogenic shock on 407 (2.7) 2003 (8.9) 26.6 presentation, n (%) Systolic BP, mm Hg 148 (128 168) 132 (112 151) 51.6 Heart rate, bpm 80 (67 94) 77 (64 92) 9.2 ACTION mortality score 30 34 44.7 Reperfusion contraindication, 15.7 15.1 1.7 n (%) EMS indicates emergency medical services; HMO, health maintenance organization; VA, Veterans Affairs; MI, myocardial infarction; PCI, percutaneous coronary intervention; CABG, coronary artery bypass grafting; HF, heart failure; BP, blood pressure; and ACTION, Acute Coronary Treatment and Intervention Outcomes Network. Continuous variables are reported as median (interquartile range). Categorical variables are represented as frequency (%). *Derived from ZIP code information among patients with home ZIP code information. Among all patients in study population.

158 Circulation July 12, 2011 Table 2. Independent Factors Associated With Emergency Medical Services Transport Among ST-Segment Elevation Myocardial Infarction Patients (n 37 634)* Adjusted 2 OR 95% CI P Sociodemographic variables Age (per 5-y increase) 179 1.06 1.05 1.08 0.0001 Hispanic ethnicity 48 0.69 0.63 0.78 0.0001 Insurance status (vs 33 0.0001 private) Medicare 1.09 1.02 1.15 Medicaid/military/VA 1.32 1.19 1.47 Self/none 1.19 1.11 1.27 Male (vs female) 29 0.87 0.87 0.92 0.0001 Race (vs white) 10 0.0067 Black 0.86 0.78 0.95 Asian/American Indian 0.87 0.75 1.01 Working hours 2 1.03 0.98 1.08 0.2048 Clinical variables Systolic BP (per 10-mm Hg 862 1.15 1.15 1.16 0.0001 decrease) HF symptoms (vs none) 75 0.0001 HF only 1.25 1.14 1.39 Shock only 1.96 1.69 2.33 HF and cardiogenic shock 1.82 1.56 2.13 Prior HF 27 1.35 1.20 1.51 0.0001 Prior stroke 25 1.30 1.18 1.43 0.0001 Prior CABG 20 0.82 0.75 0.89 0.0001 Prior PCI 18 1.16 1.09 1.25 0.0001 Hypertension 12 1.09 1.03 1.14 0.0007 Prior peripheral artery 11 1.18 1.06 1.28 0.0010 disease Dyslipidemia 11 0.93 0.89 0.97 0.0010 Diabetes mellitus 10 0.92 0.87 0.97 0.0016 Prior MI 5 1.08 1.01 1.16 0.0331 Currently on dialysis 2 0.87 0.71 1.07 0.1902 Body mass index 1 1.01 0.99 1.02 0.4303 Heart rate 0 1.00 0.98 1.01 0.5100 OR indicates odds ratio; CI, confidence interval; VA, Veterans Affairs; BP, blood pressure; HF, heart failure; CABG, coronary artery bypass grafting; PCI, percutaneous coronary intervention; and MI, myocardial infarction. *Model c index 0.67. Monday through Friday, 8 AM to 5 PM. have MI symptom onset 12 hours (10.2% versus 4.4%) and an anatomy not suitable to primary percutaneous coronary intervention (31.9% versus 27.1%) as the primary reason for reperfusion contraindication. Among patients without contraindications to reperfusion, self-transported STEMI patients were less likely to receive reperfusion therapy compared with EMS-transported patients (adjusted odds ratio, 0.77; 95% confidence interval, 0.68 to 0.87; P 0.0001). Compared with EMS-transported patients, self-transported patients also experienced greater delays to in-hospital care with longer time to first ECG, longer time to first balloon inflation, and longer time to fibrinolytic administration (Table 4). After multivariable adjustment, self-transported patients remained significantly less likely to receive an initial ECG within 10 minutes of arrival (odds ratio, 0.52; 95% confidence interval, 0.46 to 0.59; P 0.001) or to undergo primary percutaneous coronary intervention within 90 minutes of arrival (odds ratio, 0.43; 95% confidence interval, 0.37 to 0.50; P 0.001). As shown in Figure 2A and 2B, EMS activation consistently shortened and reduced the variance of presentation time delays and door-to-balloon times regardless of the distance traveled. Discussion Although EMS activation is recommended as a key component in the chain of survival for STEMI patients, 11 up to 40% of contemporary STEMI patients still self-transport to the hospital for MI care. Patients who did not use EMS were likely to be younger, men, Hispanic, and live closer to proximity to the hospital. Compared with those who used EMS, self-transported patients had significantly longer delays to hospital presentation after symptom onset and experienced significant delays to STEMI care and reperfusion. Underuse of Emergency Medical Services Transport Our study finds a persistent underuse of EMS transport among a contemporary cohort of STEMI patients in the ACTION Registry GWTG, with rates comparable to those found in a survey of the National Registry of Myocardial Infarction conducted several years ago. 7 Canto et al 7 noted an EMS use rate of only 53.4% among acute MI patients throughout the mid to late 1990s. The reasons for the persistence of this finding are not entirely clear, but cost, fear of false alarm, reluctance to bother or burden the medical community, lack of EMS benefit awareness, and other psychosocial factors, such as lack of trust in others have been implicated. 12,13 In contrast to the National Registry of Myocardial Infarction survey 7 and other previous work, 14,15 race was no longer a significant factor implicated in the decision regarding mode of transport in our contemporary population; however, patients of younger age, male sex, and Hispanic ethnicity persistently underuse EMS and therefore should remain targets of community education campaigns. 7 Interestingly, we found that education level was not a major predictor of mode of transport, and in contrast to prior studies, 16 we found that rural residence was also not a major predictor. However, those patients with private insurance used EMS less frequently than those with Medicare, Medicaid, or even selfinsurance or no insurance. Insurance status may be an imprecise surrogate for measuring economic hardship in that the financial liability associated with EMS use varies substantially across plans and geographic regions. In areas without a publicly funded EMS, charges can range from $390 to $900 for ambulance transport. 17 Impact on Process of Care Measures Compared with self-transport, EMS use was associated with at least a 30-minute reduction in the delay from symptom onset to presentation. Although patients who live in close

Mathews et al Use of EMS in Patients With STEMI 159 Table 3. Independent Factors Associated With Emergency Medical Services Transport Among ST-Segment Elevation Myocardial Infarction Patients With Home ZIP Code Information (n 18 069)* Adjusted 2 OR 95% CI P Sociodemographic variables Age (per 5-y increase) 143 1.09 1.08 1.10 0.0001 Distance (vs 4 miles), miles 53 0.0001 4 10 1.14 1.04 1.25 10 1.64 1.47 1.82 Hospital region (vs Midwest) 20 0.0002 West 1.22 0.97 1.54 South 1.43 1.20 1.69 Northeast 1.54 1.18 2.04 Insurance status (vs private) 19 0.0001 Medicare 1.06 0.97 1.15 Medicaid/military/VA 1.35 1.16 1.59 Self/none 1.16 1.06 1.28 Hispanic ethnicity 17 0.74 0.63 0.85 0.0001 Male (vs female) 13 0.87 0.81 0.94 0.0003 Education (vs 19.8%) 7 0.0381 19.8% 28.7% 1.14 1.00 1.31 28.7% 1.01 0.85 1.18 Income level (vs $41 180) 7 0.0347 $41 180 to $48 887 1.02 0.89 1.16 $48 887 1.18 1.02 1.37 Race (vs white) 2 0.4457 Black 0.92 0.82 1.04 Asian/American Indian 1.00 0.82 1.23 Working hours 1 1.03 0.96 1.10 0.4435 Rural vs urban 0 1.01 0.87 1.17 0.8720 Clinical variables Systolic BP (per 10-mm Hg 591 1.16 1.15 1.18 0.0001 decrease) HF symptoms (vs none) 51 0.0001 HF only 1.25 1.09 1.43 Cardiogenic shock only 1.85 1.56 2.17 Heart failure and cardiogenic shock 1.96 1.52 2.50 Chronic lung disease 19 1.27 1.14 1.41 0.0001 Prior CABG 18 0.77 0.68 0.87 0.0001 Prior PCI 15 1.22 1.10 1.33 0.0001 Dyslipidemia 15 0.89 0.83 0.94 0.0001 Prior congestive heart failure 9 1.28 1.09 1.51 0.0028 Hypertension 5 1.08 1.01 1.15 0.0227 Prior stroke 6 1.22 1.04 1.42 0.0135 Diabetes mellitus 4 0.92 0.85 1.00 0.0398 Prior peripheral artery disease 5 1.18 1.02 1.37 0.0229 Atrial fibrillation/atrial flutter 4 1.17 1.00 1.38 0.0552 Prior MI 1 1.07 0.96 1.20 0.2358 Body mass index 1 1.01 0.99 1.04 0.3824 Currently on dialysis 0 0.99 0.72 1.35 0.9390 Heart rate 0 1.00 0.98 1.01 0.7769 OR indicates odds ratio; CI, confidence interval; VA, Veterans Affairs; BP, blood pressure; HF, heart failure; CABG, coronary artery bypass grafting; PCI, percutaneous coronary intervention; and MI, myocardial infarction. *Model c index 0.70. Percentage of population 25 years of age with at least 4 years of college education. Median household income divided into tertiles. Monday through Friday, 8 AM to 5 PM.

160 Circulation July 12, 2011 Table 4. Comparison of Reperfusion Details and Reperfusion Care Processes Between ST-Segment Elevation Myocardial Infarction Patients Self-Transported and Those Transported by Emergency Medical Services* Self-Transport (n 15 049), min EMS Transport (n 22 585), min Symptom-onset-tohospital-arrival 120 (60 285) 89 (57 163) 0.0001 time Time to ECG 8 (4 14) 5 (2 10) 0.0001 Door-to-balloon time 76 (61 93) 63 (48 80) 0.0001 Door-to-needle time 29 (18 51) 23 (13 36) 0.0001 EMS indicates emergency medical services. *Variables are reported as median (interquartile range). Among all patients undergoing primary percutaneous coronary intervention (self-transport, n 10 010; EMS transport, n 15 805). Among all patients undergoing fibrinolytic therapy (self-transport, n 486; EMS transport, n 553). proximity to the hospital may believe self-transportation is faster, 18 our results show that EMS use is associated with shorter delays to hospital arrival regardless of geographic distance from the hospital. If one considers that initial antiischemic treatment is often delivered by EMS, then EMS use would certainly lead to shorter delays to medical contact and treatment. 19,20 The magnitude of our findings is even more striking when they are considered in the context of the impact made by recent national campaigns to improve STEMI systems of care (eg, the Door-to-Balloon campaign was associated with a 14-minute reduction in door-to-balloon times). 21 Furthermore, use of EMS transport was associated with expedited in-hospital STEMI care. Compared with self-transported patients, the triage process is likely more efficient, resulting in earlier ECG evaluation and delivery of reperfusion for patients brought in by EMS. The availability of prehospital ECGs further enhances the triage efficiency and timeliness of reperfusion for patients transported by EMS. In an earlier analysis of the ACTION Registry, 22 27% of patients transported by EMS received prehospital ECGs. The use of prehospital ECGs was associated with significantly shorter door-to-needle and door-toballoon times (19 versus 29 minutes, P 0.003; and 61 versus 75 minutes, P 0.0001, respectively). As demonstrated in our study, the increased use of prehospital ECGs may further enhance the benefit of EMS use. The routine integration of prehospital ECGs into contemporary system-wide STEMI care protocols has been associated with reduced door-toballoon times. 23,24 Therefore, measures to improve EMS use, along with perhaps more ubiquitous use of prehospital ECGs, have the potential to reduce ischemic time substantially, thereby maximizing the potential for myocardial salvage after reperfusion. 5 Need for Targeted Community Education Community-level interventions that have targeted improving the use of EMS have yielded mixed results. 25 29 In the randomized Rapid Early Action for Coronary Treatment (REACT) Trial, Luepker et al 28 found that a community intervention to increase patient recognition of and action on MI symptoms increased EMS use by 20% but did not P significantly shorten the delay from symptom onset to hospital arrival. Conversely, Blohm et al 25 observed a reduction in presentation delay of 40 minutes that was associated with public media campaigns. However, this reduction in presentation delay was without any increase in EMS use rates. Such intensive community-wide media campaigns are often initially effective, but the effect seems to taper quickly over time. 29,30 This may explain, in part, the low EMS use rate that has persisted over the last decade. In March 2007, the AHA launched Mission: Lifeline, which is a national community-based patient-centric initiative designed to advance systems of care for STEMI patients, targeting the entire continuum of STEMI care from symptom onset to in-hospital reperfusion. 25 Mission: Lifeline identifies patient education and awareness as a critical and challenging area to address. In this regard, efforts not only have been directed toward patient and bystander recognition of ischemic symptoms but also have emphasized both the urgency and benefit of EMS activation. In addition to these patient education efforts, future healthcare reform initiatives that may help reduce this particular barrier to EMS use should include provisions for EMS coverage for patients with symptoms concerning for ischemic coronary artery disease. Limitations Some limitations in our study should be acknowledged. First, although observational data are key to examining patterns of resource use and the association between EMS use and care rendered is adjusted for numerous patient differences between groups, residual unmeasured confounding or bias in these observational data is possible. Second, information on patient income and educational level was available on only approximately half of our sample and was extrapolated from Census data. Although this methodology is commonly used, 31,32 this aggregate community measure may not accurately reflect these factors on an individual level. Third, the ZIP code derived distance to the hospital is a calculation of geographic distance and does not take into account local transportation infrastructure (eg, roads and highways); therefore, the ZIP code derived distance may not accurately reflect the actual distance an individual patient may have to travel. Finally, patient factors such as denial, awareness of available resources, individual thresholds for seeking medical attention, and the potential embarrassment of false alarms 18,33 were not collected but may influence the decision to utilize EMS. Conclusions Emergency medical services transport to the hospital remains underused among contemporary STEMI patients and is associated with substantial delays in presentation and treatment. Our findings indicate that policy interventions to increase EMS use have the potential to decrease total ischemic time substantially both by reducing delays to hospital presentation and by expediting in-hospital care. We believe that comprehensive changes in the structure of care are needed, as demonstrated by Mission: Lifeline s STEMI chain of survival. The challenge moving forward is to broaden this benefit to larger audiences, particularly to sociodemographic

Mathews et al Use of EMS in Patients With STEMI 161 Figure 2. Box plot. Comparison of care processes between self-transported and emergency medical services (EMS) transported ST-segment elevation myocardial infarction patients stratified by distance between home and hospital. Box plot represents median (line) and 1.5 times the interquartile range with outliers not represented. A, Symptom onset to arrival among all patients. B, Door to reperfusion among patients undergoing primary percutaneous coronary intervention.

162 Circulation July 12, 2011 groups known to underuse EMS. These education efforts should be integrated with system-wide strategies and initiatives to improve the timeliness of reperfusion therapy for STEMI patients. Acknowledgments We would like to thank Erin LoFrese for her editorial contributions to this manuscript. She did not receive compensation for her assistance, apart from her employment at the institution where the study was conducted. Sources of Funding The ACTION Registry GWTG is an initiative of the American College of Cardiology Foundation and the AHA, with partnering support from the Society of Chest Pain Centers, Society of Hospital Medicine, and American College of Emergency Physicians. The registry is sponsored by Bristol-Myers Squibb/Sanofi Pharmaceuticals. This work was supported by an award from the AHA Pharmaceutical Roundtable and David and Stevie Spina. This project was supported by grant U18HS016964 from the Agency for Healthcare Research and Quality (AHRQ). The content is solely the responsibility of the authors and does not necessarily represent the official views of the AHRQ. The funding source had no role in the design or implementation of the study, or in the decision to seek publication. Disclosures Dr Peterson has received research grants from the American College of Cardiology, Bristol Myers Squibb, Eli Lilly & Co, Johnson & Johnson, Merck & Co, Sanofi-aventis, Schering-Plough Corp, the Society of Thoracic Surgeons, and the AHA, as well as consulting fees from Bristol Myers Squibb, Merck & Co, Tethysbio, and AstraZeneca. Dr Roe has received research grants from Eli Lilly, Novartis, Merck-Schering Plough, Bristol-Myers Squibb, the American College of Cardiology, and the AHA, as well as consulting fees or honoraria from Glaxo Smith Kline, KAI Pharmaceuticals, Novartis, Eli Lilly, Bristol-Myers Squibb, Sanofi-aventis, and Astra- Zeneca. Dr Wiviott has received research grants from Eli Lilly, Daiichi Sankyo, and Merck-Schering Plough; honoraria from Astra- Zeneca, Novartis; Eli Lilly, and Daiichi Sankyo; and consulting fees from AstraZeneca, Sanofi-aventis, Portola, and Arena. Dr Jacobs has served as the chair for the AHA Advisory Working Group for Mission: Lifeline. Dr Wang has received research grants from Bristol-Myers Squibb/Sanofi and has partnerships with Schering Plough, The Medicines Company, Heartscape, Canyon Pharmaceuticals, and Eli Lilly/Daiichi Sankyo Alliance. The other authors report no conflicts. References 1. Kushner FG, Hand M, Smith SC Jr, King SB 3rd, Anderson JL, Antman EM, Bailey SR, Bates ER, Blankenship JC, Casey DE Jr, Green LA, Hochman JS, Jacobs AK, Krumholz HM, Morrison DA, Ornato JP, Pearle DL, Peterson ED, Sloan MA, Whitlow PL, Williams DO; American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. 2009 Focused updates: ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction (updating the 2004 guideline and 2007 focused update) and ACC/AHA/SCAI guidelines on percutaneous coronary intervention (updating the 2005 guideline and 2007 focused update): a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Circulation. 2009;120:2271 2306. 2. Krumholz HM, Bradley EH, Nallamothu BK, Ting HH, Batchelor WB, Kline-Rogers E, Stern AF, Byrd JR, Brush JE Jr. A campaign to improve the timeliness of primary percutaneous coronary intervention: Door-to- Balloon: an Alliance for Quality. JACC Interv. 2008;1:97 104. 3. Jacobs AK. Regional systems of care for patients with ST-elevation myocardial infarction: being at the right place at the right time. Circulation. 2007;116:689 692. 4. Bradley EH, Nallamothu BK, Herrin J, Ting HH, Stern AF, Nembhard IM, Yuan CT, Green JC, Kline-Rogers E, Wang Y, Curtis JP, Webster TR, Masoudi FA, Fonarow GC, Brush JE Jr, Krumholz HM. National efforts to improve door-to-balloon time results from the Door-to-Balloon Alliance. J Am Coll Cardiol. 2009;54:2423 2429. 5. Antman EM. Time is muscle: translation into practice. J Am Coll Cardiol. 2008;52:1216 1221. 6. Goff DC Jr, Feldman HA, McGovern PG, Goldberg RJ, Simons-Morton DG, Cornell CE, Osganian SK, Cooper LS, Hedges JR. Prehospital delay in patients hospitalized with heart attack symptoms in the United States: the REACT trial: Rapid Early Action for Coronary Treatment (REACT) Study Group. Am Heart J. 1999;138:1046 1057. 7. Canto JG, Zalenski RJ, Ornato JP, Rogers WJ, Kiefe CI, Magid D, Shlipak MG, Frederick PD, Lambrew CG, Littrell KA, Barron HV; National Registry of Myocardial Infarction 2 Investigators. Use of emergency medical services in acute myocardial infarction and subsequent quality of care: observations from the National Registry of Myocardial Infarction 2. Circulation. 2002;106:3018 3023. 8. Peterson ED, Roe MT, Rumsfeld JS, Shaw RE, Brindis RG, Fonarow GC, Cannon CP. A call to ACTION (Acute Coronary Treatment and Intervention Outcomes Network): a national effort to promote timely clinical feedback and support continuous quality improvement for acute myocardial infarction. Circ Cardiovasc Qual Outcomes. 2009;2:491 499. 9. National Cardiovascular Data Registry. ACTION Registry: GWTG elements and definitions. http://www.ncdr.com/webncdr/action/ Elements.aspx. Accessed September 10, 2010. 10. Chin CT, Chen AY, Wang TY, Alexander KP, Mathews R, Rumsfeld JS, Cannon CP, Fonarow GC, Peterson ED, Roe MT. Risk adjustment for in-hospital mortality of contemporary patients with acute myocardial infarction: the Acute Coronary Treatment and Intervention Outcomes Network (ACTION) Registry Get With the Guidelines (GWTG) acute myocardial infarction mortality model and risk score. Am Heart J. 2011; 161:113 122.e2. 11. Jacobs AK, Antman EM, Faxon DP, Gregory T, Solis P. Development of systems of care for ST-elevation myocardial infarction patients: executive summary. Circulation. 2007;116:217 230. 12. Mensah GA, Hand MM, Antman EM, Ryan TJ Jr, Schriever R, Smith SC Jr. Development of systems of care for ST-elevation myocardial infarction patients: the patient and public perspective. Circulation. 2007; 116:e33 e38. 13. Sullivan MD, Ciechanowski PS, Russo JE, Soine LA, Jordan-Keith K, Ting HH, Caldwell JH. Understanding why patients delay seeking care for acute coronary syndromes. Circ Cardiovasc Qual Outcomes. 2009;2: 148 154. 14. Ting HH, Bradley EH, Wang Y, Lichtman JH, Nallamothu BK, Sullivan MD, Gersh BJ, Roger VL, Curtis JP, Krumholz HM. Factors associated with longer time from symptom onset to hospital presentation for patients with ST-elevation myocardial infarction. Arch Intern Med. 2008;168: 959 968. 15. McGinn AP, Rosamond WD, Goff DC Jr, Taylor HA, Miles JS, Chambless L. Trends in prehospital delay time and use of emergency medical services for acute myocardial infarction: experience in 4 US communities from 1987 2000. Am Heart J. 2005;150:392 400. 16. Solie C, Larson D, Duval S, Sharkey S, Menssen K, Unger B, Hokanson J, Satterlee P, Henry T. Urban versus rural utilization of emergency medical services by ST-elevation myocardial infarction patients. Catheter Cardiovasc Interv. 2010;75:S47 S162. 17. Siepmann DB, Mann NC, Hedges JR, Daya MR. Association between prepayment systems and emergency medical services use among patients with acute chest discomfort syndrome. Ann Emerg Med. 2000;35: 573 578. 18. Meischke H, Ho MT, Eisenberg MS, Schaeffer SM, Larsen MP. Reasons patients with chest pain delay or do not call 911. Ann Emerg Med. 1995;25:193 197. 19. Song L, Yan H, Hu D. Patients with acute myocardial infarction using ambulance or private transporting to reach definitive care: which mode is quicker? Intern Med J. 2010;40:112 116. 20. Hutchings CB, Mann NC, Daya M, Jui J, Goldberg R, Cooper L, Goff DC Jr, Cornell C; Rapid Early Action for Coronary Treatment Study. Patients with chest pain calling 9 1-1 or self-transporting to reach definitive care: which mode is quicker? Am Heart J. 2004;147:35 41. 21. Wang TY, Fonarow GC, Hernandez AF, Liang L, Nallamothu BK, Shah BR, Cannon CP, Peterson ED. The dissociation between door-to-balloon time improvement and improvements in other acute myocardial infarction care processes and patient outcomes. Arch Intern Med. 2009;169: 1411 1419.

Mathews et al Use of EMS in Patients With STEMI 163 22. Diercks DB, Kontos MC, Chen AY, Pollack CV Jr, Wiviott SD, Rumsfeld JS, Magid DJ, Gibler WB, Cannon CP, Peterson ED, Roe MT. Utilization and impact of pre-hospital electrocardiograms for patients with acute ST-segment elevation myocardial infarction: data from the NCDR (National Cardiovascular Data Registry) ACTION (Acute Coronary Treatment and Intervention Outcomes Network) Registry. JAm Coll Cardiol. 2009;53:161 166. 23. Rokos IC, French WJ, Koenig WJ, Stratton SJ, Nighswonger B, Strunk B, Jewell J, Mahmud E, Dunford JV, Hokanson J, Smith SW, Baran KW, Swor R, Berman A, Wilson BH, Aluko AO, Gross BW, Rostykus PS, Salvucci A, Dev V, McNally B, Manoukian SV, King SB 3rd. Integration of pre-hospital electrocardiograms and ST-elevation myocardial infarction receiving center (SRC) networks: impact on door-to-balloon times across 10 independent regions. JACC Cardiovasc Interv. 2009;2: 339 346. 24. Le May MR, So DY, Dionne R, Glover CA, Froeschl MP, Wells GA, Davies RF, Sherrard HL, Maloney J, Marquis JF, O Brien ER, Trickett J, Poirier P, Ryan SC, Ha A, Joseph PG, Labinaz M. A citywide protocol for primary PCI in ST-segment elevation myocardial infarction. N Engl J Med. 2008;358:231 240. 25. Blohm M, Hartford M, Karlson BW, Karlsson T, Herlitz J. A media campaign aiming at reducing delay times and increasing the use of ambulance in AMI. Am J Emerg Med. 1994;12:315 318. 26. Gaspoz JM, Unger PF, Urban P, Chevrolet JC, Rutishauser W, Lovis C, Goldman L, Héliot C, Séchaud L, Mischler S, Waldvogel FA. Impact of a public campaign on pre-hospital delay in patients reporting chest pain. Heart. 1996;76:150 155. 27. Ho MT, Eisenberg MS, Litwin PE, Schaeffer SM, Damon SK. Delay between onset of chest pain and seeking medical care: the effect of public education. Ann Emerg Med. 1989;18:727 731. 28. Luepker RV, Raczynski JM, Osganian S, Goldberg RJ, Finnegan JR Jr, Hedges JR, Goff DC Jr, Eisenberg MS, Zapka JG, Feldman HA, Labarthe DR, McGovern PG, Cornell CE, Proschan MA, Simons-Morton DG. Effect of a community intervention on patient delay and emergency medical service use in acute coronary heart disease: the Rapid Early Action for Coronary Treatment (REACT) Trial. JAMA. 2000;284:60 67. 29. Dracup K, McKinley S, Riegel B, Riegel B, Moser DK, Meischke H, Doering LV, Davidson P, Paul SM, Baker H, Pelter M. A randomized clinical trial to reduce patient prehospital delay to treatment in acute coronary syndrome. Circ Cardiovasc Qual Outcomes. 2009;2:524 532. 30. Eppler E, Eisenberg MS, Schaeffer S, Meischke H, Larson MP. 911 and emergency department use for chest pain: results of a media campaign. Ann Emerg Med. 1994;24:202 208. 31. Krieger N. Overcoming the absence of socioeconomic data in medical records: validation and application of a census-based methodology. Am J Public Health. 1992;82:703 710. 32. Rao SV, Schulman KA, Curtis LH, Gersh BJ, Jollis JG. Socioeconomic status and outcome following acute myocardial infarction in elderly patients. Arch Intern Med. 2004;164:1128 1133. 33. Brown AL, Mann NC, Daya M, Goldberg R, Meischke H, Taylor J, Smith K, Osganian S, Cooper L. Demographic, belief, and situational factors influencing the decision to utilize emergency medical services among chest pain patients: Rapid Early Action for Coronary Treatment (REACT) study. Circulation. 2000;102:173 178. CLINICAL PERSPECTIVE Although current guidelines recommend activation of emergency medical services (EMS) for the expedited triage of patients experiencing cardiac chest pain, EMS has historically been underused. In this article, we explored the prevalence of EMS use among contemporary patients with ST-segment elevation myocardial infarction, the sociodemographic and clinical factors associated with a patient s decision regarding mode of transport, and the differences in treatment between those patients who self-transport and those who use EMS. Examining the ST-segment elevation myocardial infarction patients in the National Cardiovascular Data Registry Acute Coronary Treatment and Intervention Outcomes Network Registry Get With the Guidelines, we found that EMS remains underused, with only 60% of patients arriving to an emergency room via EMS. Patients who do not use EMS are more likely male, Hispanic, and younger than EMS users. The decision to use EMS was associated with at least a 30-minute reduction in symptom onset to arrival time, which is a substantial shortening of the total ischemic time. The decision to use EMS was also associated with reduced delays to in-hospital evaluation and reperfusion treatment. We believe our findings indicate a pressing need for community education efforts targeted at known demographic groups who underuse EMS transport. Furthermore, these efforts need to be integrated with system-wide strategies to improve reperfusion care.