More than 16 million Americans have coronary heart disease (CHD), and approximately 785 000 new cases occur each year.1 Coronary heart disease causes 1 of every 5 deaths in America.1 About every 25 seconds, an American will have a coronary event, and about every minute, someone will die of it,1 making CHD the No. 1 cause of death in the United States.1 Acute myocardial infarction (AMI) is the primary consequence of CHD and is associated with high morbidity and mortality.1 For example, people who survive the acute stage of a myocardial infarction have a chance of illness and death 1.5 to 15 times higher than that of the general population.1
Rapid arrival to the hospital for treatment of AMI improves long-term outcomes. A major determinant of the high morbidity and mortality after AMI is the amount of myocardial damage that occurs. The most important determinant of myocardial damage is the time from onset of patient symptoms to definitive treatment.2-6 As a consequence, rapid arrival to the hospital for early treatment is the most effective way of preventing long-term morbidity and mortality associated with AMI.7-9 Patients are most likely to survive AMI with a smaller area of damage when the endangered myocardium is reperfused quickly using urgent reperfusion therapy (fibrinolytic therapy and/or urgent percutaneous coronary intervention).7-11 In a study of 565 AMI patients who underwent acute angioplasty, those who received the first balloon inflation within 1 hour of arrival at the hospital had a 30-day mortality rate of 1.0%, but for every 15 minutes longer than 1 hour, the odds of death increased 1.6 times.7
Patient delay in seeking treatment for cardiac symptoms was the major source of delay in receipt of definitive treatment.2,5,12 However, whether prehospital delay time in seeking timely treatment is associated with short-term, in-hospital complications remains unknown, and our major aim in this study was to test this relationship. There are few studies of the relationship between prehospital delay and in-hospital outcomes among AMI patients. In those few studies, no relationship was found between delay and in-hospital outcomes. Caldwell and her colleagues13 found no significant differences in in-hospital outcomes and length of hospital stay for AMI patients with prehospital delay times of 6 hours or more as compared with those who presented in less than 6 hours.13 Failure to find an association in this study13 may be related to the definition of delay as a more appropriate cut-point dividing short and long delay times might be 1 to 2 hours.12,14-16
Examination of the impact of patient prehospital delay on in-hospital complications must also consider potential confounding covariates. Among the most important of these is anxiety, disease severity (or severity of cardiac dysfunction), sex, and age. Over the past 3 decades, studies have been conducted to identify determinants of increased patient delay in response to symptoms of AMI, and sociodemographic variables have been most studied.12,17 In this study, a theoretical model (Figure 1) was developed and tested based on current literature about prehospital delay, in-hospital complications, and length of hospital stay for AMI.7,18-20 Age and sex have been extensively studied, and it is clear that older age and female sex are strongly associated with increased delay in most studies.12,17,21 They are also strongly associated with worse prognosis after AMI.22-25
Anxiety is one of the most common symptoms experienced by AMI patients.18,19,26-28 The prevalence of anxiety has been reported to be as high as 50% to 70% among patients with AMI.28,29 State (and not trait) anxiety has been measured in most studies.19,28-31 State anxiety has been shown to be an important independent risk factor for in-hospital complications.19,31 Patients with higher anxiety when measured within 48 hours of patient arrival at the hospital were 4.9 times (95% confidence interval, 2.1-12.2 times) more likely to have subsequent in-hospital complications than those with lower anxiety.31
Not all investigators, however, have found anxiety to have only negative consequences. In a study of AMI patients, higher anxiety about acute cardiac symptoms was associated with shorter prehospital delays compared with patients who expressed little anxiety about their symptoms.32 Anxiety was measured by the Modified Response to Symptoms Questionnaire within 72 hours of admission to the hospital. Thus, it is not clear whether anxiety has positive or negative consequences in the context of AMI when prehospital delay is considered.
Severity of cardiac dysfunction by the Killip classification is another important predictor of in-hospital mortality and is associated with shorter prehospital delay.12,20 Because prehospital delay is a complex phenomenon, we aimed to determine the interactive relationships among age, sex, anxiety, Killip classification, prehospital delay, and in-hospital complications and length of stay in the hospital.
Accordingly, the purpose of this study was to evaluate the fit of a theoretical model where patient prehospital delay time was indirectly associated with hospital length of stay through in-hospital complications after AMI considering simultaneously demographic (ie, age, sex), clinical (ie, Killip class), and psychological (ie, anxiety)factors. An additional purpose was to determine whether in-hospital complications predicted length of hospital stay.
The current study was part of a larger prospective examination of patients' experiences in seeking treatment for AMI symptoms, their in-hospital anxiety levels, and subsequent in-hospital complications.19 In the current study, we used structural equation modeling (SEM) to determine whether prehospital delay time predicted in-hospital complications after AMI, and whether in-hospital complications predicted length of hospital stay while simultaneously considering appropriate demographic (ie, age, sex), clinical (ie, Killip class), and psychological (ie, anxiety) factors.
Sample and Settings
Patients were recruited and enrolled from the cardiac care units of 6 hospitals in diverse settings in the United States and Australia, described in detail elsewhere.19 The hospitals included university academic medical centers and community hospitals serving racially/ethnically and economically diverse populations. Inclusion criteria were (1) diagnosis of ST-segment elevation AMI confirmed by elevated cardiac enzymes, cardiac symptoms lasting longer than 20 minutes along with greater than 0.1-mV ST-segment elevation in 2 or more electrocardiogram (ECG) limb leads or greater than 0.2-mV ST-segment elevation in 2 or more contiguous precordial ECG leads; (2) free of obvious cognitive impairment; and (3) free of noncardiac serious or life-threatening comorbidities such as sepsis, shock, stroke, or acute renal failure.
Data Collection Procedures
Institutional review board approval was obtained at all sites, and all patients gave signed, informed consent. Demographic and clinical data were obtained by patient interview and medical record review. Within 72 hours of arrival to the hospital, patients were interviewed regarding anxiety and their prehospital experience of seeking treatment for their symptoms, as described subsequently. After discharge, the patients' medical records were reviewed and abstracted by specially trained registered nurses who were cardiac care specialists. Data on all in-hospital complications were abstracted by the trained nurses who were blinded to the symptom onset to hospital arrival times. These nurses abstracted information on all in-hospital complications that occurred subsequent in time to the patient interview by carefully reviewing every page of each patient's entire medical record. The nurses doing data collection remained blind to patients' delay times when extracting the complication information.
To ensure reliability of data collection, all raters were trained by the principal investigator. Training included assessment of the same 10 charts by all raters. Results were then compared and discussed, with additional chart review and comparison undertaken until there was greater than 95% agreement on data collected by all raters.
Prehospital Delay Time
In this study, prehospital delay time was defined as the time from reported AMI symptom onset until the patient reached the hospital for care. We used the time that the patient first noted symptoms or noted that something was wrong as the time of AMI symptom onset, and admission to the emergency department as the time the patient arrived at the hospital. This time was determined by trained research assistants using data obtained from patient interview. Although interview techniques are subject to recall bias, we have developed techniques that include questioning and discussion with patients of familiar events associated with their symptom onset to help patients pinpoint the time symptoms began.32 Briefly, we first asked patients to generally estimate the time their symptoms began (eg, day time or night time?). Then, the time was narrowed down by placing it relative to activities of daily life such as eating meals, preparing for work, or daily activities. If a family members were present, we also used their input to help determine time of symptom onset.
We also review the medical record to confirm time of symptom onset. When patient recall of time of symptom onset was different from that found in the medical record, we talked with the patient and family again. Differences in symptom onset were attributed to differences in the approach of our researchers and health care providers in determining symptom onset time in many cases. After discussing differences in medical record symptom onset and patient-stated symptom onset with the patient and/or family, we chose the patient-stated time of symptom onset when there remained differences in the 2 sources.
Anxiety was defined as a feeling of nervousness or fear caused by a real or imagined threats.19 In this study, anxiety was measured using the total score from the 6-item anxiety subscale of the Brief Symptom Inventory (BSI). The BSI anxiety subscale was administered within 72 hours of patients' arrival at the hospital.33,34 The anxiety subscale measures state anxiety, and we instructed patients to respond on the basis of how they felt at the time of filling out the instrument.33,34 Participants rated their level of anxiety on a scale of 0 to 4 (0 = "not at all" and 4 = "extremely"), with higher values indicating higher levels of anxiety. The average score on the BSI anxiety subscale in the general population is 0.35.34 Prior studies, including those conducted with acutely ill AMI patients, have demonstrated the reliability of the BSI with Cronbach α ranging between .81 and .87.19,27,33 A study of 243 patients admitted with AMI demonstrated that the BSI is a valid and reliable (Cronbach α = .84) instrument when used to measure anxiety.35 In the current study, Cronbach α was .86, which indicted acceptable internal consistency. The validity of the instrument has been supported in patients with AMI when the BSI was correlated with the Spielberger State Anxiety Index.36
In-hospital complications were defined as the combined end point of 1 or more of the following: (1) acute recurrent ischemia as evidenced by new-onset chest pain with 1 or more of the following objective indicators of ischemia: (a) ST-segment elevation on bedside ST-segment monitor or 12-lead ECG, and/or (b) hemodynamic compromise evidenced by reductions in blood pressure, or elevations or reductions in pulse from baseline persisting more than 15 minutes and causing patient symptoms, and/or (c) nitrates and/or intravenous pain medication given for chest pain; (2) reinfarction as evidenced by recurrent positive cardiac enzyme (ie, CK-MB); (3) sustained ventricular tachycardia (>15 seconds) or ventricular tachycardia requiring pharmacological or electrical intervention due to hemodynamic compromise and/or chest pain; (4) ventricular fibrillation; or (5) in-hospital death.
Admission Killip Class
Killip classification is a method of categorizing cardiac dysfunction in patients experiencing AMI. Using this method, patients are grouped into 4 groups that reflect the degree of cardiac impairment secondary to their infarct. The classifications range from class I, in which patients have no signs or symptoms of cardiac dysfunction, to class IV, in which patients have signs and symptoms of cardiogenic shock. The full classification system is as follows: (1) class I, patients were free of rales, a third heart sound, shortness of breath, or other signs and symptoms of cardiac dysfunction; (2) class II, patients had rales up to 50% of each lung field and/or shortness of breath; (3) class III, patients had rales in more than half of each lung field and/or radiological evidence of pulmonary congestion and has signs and symptoms of pulmonary edema; and (4) class IV, patients were in cardiogenic shock.37 We assessed patients' admission Killip classification based on their presenting signs and symptoms in the emergency department.
Review of the data indicated that there was no violation of the following assumptions: normality, univariate outliers, linearity, multicolinearity, and homoscedasticity. Although there were very few missing data, a missing data analysis (ie, Little Missing Complete At Random test) was run to test the assumption that data were missing completely at random. In this study, the Little Missing Complete At Random test was nonsignificant (χ242 = 48.874, P = .216), indicating that data were missing completely at random. Therefore, the expectation-maximization imputation of values was used for missing data.38
Data analysis began with a descriptive examination of all variables using SPSS (Chicago, Illinois), version 16.0, including frequency distributions, means, SDs, medians, and interquartile ranges, as appropriate to the level of measurement of the variables. An α of .05 was set a priori. Because delay time is skewed, median was reported in sample characteristic session, and the value was log transformed to more nearly approximate a normal distribution. All analyses were done using the transformed value.
Structural Equation Modeling
Structural equation modeling is a statistical method that adopts a hypothesis-testing approach to the multivariate analysis of a structural theory. Structural equation modeling can incorporate not only observed but also unobserved (latent) variables to allow researchers the opportunity to extract measurement error.39 Most importantly, SEM allows the entire system of hypothesized relationships to be estimated simultaneously to determine whether the hypothesized structure is consistent with the data.40
In this study, an SEM was constructed using Amos (Analysis of Moment Structures) 16.0 software (Chicago, Illinois). The model parameters were estimated using maximum likelihood estimation. Structural equation modeling isolates measurement error in models by incorporating error into equations. This technique enabled us to specify, estimate, and evaluate the fit of the theoretical model (Figure 1) where prehospital delay time was indirectly associated with hospital length of stay through in-hospital complications. The measurement model was a confirmatory factor analysis of indicators used to represent the single latent construct, the measure of anxiety, with multiple indicators. To identify latent constructs, the scale of each observed variable was set to the metric of the first indicator variable. Latent variance in the measurement model was fixed at 1.0.
The theoretical model (Figure 1) was based on current literature about prehospital delay, in-hospital complications, and length of hospital stay for AMI.7,18-20 Paths that were not significant were dropped, based on this empirical evidence and also on the theoretical meanings of various possible solutions. Each time a path was dropped, the model was reconstructed to examine the changes that resulted.
For the theoretical model, the χ2 goodness-of-fit test is traditionally adopted to assess the degree to which the proposed model fits the observed data.40 However, the χ2 test's significance is overly sensitive to a large sample size.40 Thus, based on the literature, the model fit was also assessed using several additional indices including the normed χ2 (NC) (χ2 / df), the Confirmatory Fit Index (CFI), and the root mean square error of approximation (RMSEA) that provided evidence that our sample size was adequate. A model was considered to have a reasonable error if approximation of the CFI was 0.90 or greater. Values of RMSEA in the range of 0.05 to 0.08 were used to indicate an adequate fit.40 The Akaike Information Criterion (AIC), another measure of model fit, was used to compare 2 nonhierarchical models for model fit; for 2 models estimated from the same data set, the model with the smaller AIC is preferred.40-52 In the measurement model, factor loading of greater than 0.50 indicated good fit.
Characteristics of Sample
A total of 536 patients with AMI were included in this study (Table 1). The mean age of patients in the sample was 62 (SD, 14) years. Two thirds of the sample was male. Most patients were white (86%) and married (69%) and had completed a high school education (73%). More than a quarter of patients had experienced a previous MI (27%). The average left ventricular ejection fraction was 50%, reflecting enrollment of patients with and without systolic dysfunction. Most patients in the sample were stable on admission; only 8% were classified as Killip class III/IV. The median prehospital delay time was 3.29 hours. Only 15% of the patients arrived at the hospital within 1 hour of AMI symptom onset. Patients reported the full range of anxiety symptoms, and the mean anxiety level in this sample was twice that of the mean described in the general population.34 More than a quarter of patients had more than 1 in-hospital complication. The most frequent complications were recurrent ischemia (20%), arrhythmias (28%), and heart failure (13%). The median length of stay in the hospital was 4.73 days.
The latent variable anxiety had 6 indicators (each of the 6 items comprising the BSI), so this single variable was tested for goodness of fit with confirmatory factor analysis. The factor loadings for anxiety were 0.77, 0.76, 0.82, 0.79, 0.69, and 0.53. The model produced a significant χ29 of 81.8 (P < .001). Other indices indicated a fair fit of the model to the data (CFI = 0.95; RMSEA = 0.123). To determine if the measurement model could be improved, the item with the smallest factor loading was dropped (item 6 = 0.53). After dropping that 1 item, the fit indices did not improve (CFI = 0.96, RMSEA = 0.147), nor did the reliability (0.86 to 0.86). Thus, the 6-item scale was retained for the final structural regression modeling analysis.
Structural Equation Modeling
Because SEM is inherently a large-sample technique, researchers have suggested that the ratio of cases to free parameters should number at least 10:1 (preferably 15:1-20:1).40 With a sample of 536 and 12 measured variables, the ratio of participants to free parameters was 44:1. Thus, the power was more than sufficient for this analysis.
The hypothesized model (Figure 1) included 5 exogenous variables (age, sex, Killip class, anxiety, and prehospital delay time) and 2 endogenous variables (in-hospital complication and length of stay in the hospital). The model produced a significant χ252 of 215 (P < .001). However, other indices indicated a fair fit of the model to the data (NC = 4.129; CFI = 0.904; RMSEA = 0.076).
One exogenous variable, sex, was not statistically significantly related to the endogenous variable and was deleted. The first alternative model produced a significant χ242 of 165 (P < .001). If the AIC (fit index best for comparison in nonnested models) drops when an exogenous variable is deleted, the subsequent model is better than the original model. The AIC of 290 in the original, hypothesized model and the AIC of 235 in the first alternative model indicated that the first alternative model was better than the original hypothesized model. The indices of the alternative model also indicated a fair fit to the data (NC = 3.939; CFI = 0.925; RMSEA = 0.074).
A second exogenous variable, age, was not statistically significantly related to the endogenous variable and was deleted. The second alternative model (Figure 2) produced a significant χ233 of 111 (P < .001). The AIC of 235 in the first alternative model and the AIC of 175 in the second alternative model indicated that the second alternative model was better than the first alternative model. The indices of the alternative model also indicated a fair fit to the data (NC = 3.377; CFI = 0.951; RMSEA = 0.067).
According to Kline,40 a model is considered to have a reasonable error of approximation if the NC is less than 5 and values of RMSEA in the range of 0.05 to 0.08 are indicative of a fair fit. Based on Kline's40 criteria, the second alternative model was more parsimonious, and the fit indices were better than those of the hypothesized model. Thus, the second alternative model was judged to be the best model.
The second alternative model demonstrated that prehospital delay, admission Killip class, and anxiety were the best predictors of in-hospital complications (P < .05) (Table 2, Figure 2). Patients who had more in-hospital complications had a longer prehospital delay, were more anxious (assessed 72 hours within admission), and had a more severe admission Killip class. The predictors of length of stay in the hospital were in-hospital complications, anxiety, and admission Killip class (P < .05). Those who were more anxious had worse cardiac function, as reflected by Killip class, had more in-hospital complications, and had a longer length of stay in the hospital. Anxiety and admission Killip class both influenced length of stay in the hospital directly and indirectly (through in-hospital complications).
A significant finding of this study was that prehospital delay time in seeking timely treatment for AMI symptoms predicted in-hospital complications, including recurrent ischemia, reinfarction, sustained ventricular tachycardia or fibrillation, and cardiac death. Likewise, the occurrence of in-hospital complications was related to length of stay in the hospital. Others have demonstrated that longer prehospital delay results in a delay in patients receiving both thrombolytic therapy and coronary intervention, with the result that global ejection fraction is reduced and regional hypokinesia develops.6 Thus, our finding of an independent relationship between longer prehospital delay time and a greater incidence of in-hospital complications may be related to the negative effects of delay on cardiac dynamics.
Previous investigators have shown that delay time is an important predictor of patient morbidity and mortality outcomes once out of the hospital.7-9,43,44 A delay of only a few hours can have a significant impact on a patient's outpatient prognosis.8 Few studies, however, have verified the association between prehospital delay and in-hospital morbidity. Our study confirms that delay in seeking timely treatment is also associated with more in-hospital complications and a longer hospital stay. These outcomes may also contribute to the high health care costs seen in the United States today. Thus, decreasing patient delay in seeking timely treatment for AMI symptoms could improve in-hospital and out-of-hospital outcomes and reduce health care costs.45,46
The median prehospital delay time in this study was 3.29 hours, far exceeding the goals of 1 hour for receipt of thrombolytic therapy and 2 hours for primary percutaneous angioplasty.12,15,16 This result is consistent with prior studies. Median prehospital delay time from symptom onset to arrival to the hospital ranges from 1.5 to 6 hours.12 Both in-hospital and postdischarge mortality- and morbidity-associated AMI can be reduced substantially by timely receipt of therapies to establish reperfusion to affected regions of myocardium by administration of urgent reperfusion therapy.12,14 In this study, only 15% of the patients arrived to the hospital within 1 hour from the onset of symptoms. Patient delay in seeking treatment for cardiac symptoms has been a long-standing and particularly resistant problem.12 Several extensive mass public education campaigns have failed to have any impact on patient delay.5,47,48 Unique approaches that include individual education of high-risk individuals, more involvement of health care providers in the education process, application of diagnostic technology in the home setting, and other creative strategies must be developed and tested if we are to decrease patient delay time.
We have confirmed, using SEM, our previous finding that anxiety was an independent predictor of in-hospital complications.19,31 The prevalence of anxiety is as high as 70% among patients hospitalized with AMI.29 Anxious patients hospitalized for an AMI event are at approximately 5 times higher risk for developing in-hospital complications than are nonanxious patients.31 It is recommended that clinicians use suitable measures to identify those who are anxious by evaluating every AMI patient on a routine basis. Likewise, it is necessary to implement interventions to reduce high levels of patient anxiety and thwart its detrimental consequences.
This study is consistent with prior studies about the relationship of Killip class, an indicator of ventricular function, on health outcomes. Killip class is consistently a significant predictor of morbidity and mortality.49-51 In this study, Killip class was a significant predictor of in-hospital complications (P < .001) and length of stay in the hospital (P = .004). Those who had worse cardiac function as reflected by greater Killip class should be considered at high risk for poor outcomes.
The conclusions that can be drawn from this study are limited by the use of secondary data analysis. For example, anxiety was measured 72 hours after admission in this study. Based on this 1-time measure, we do not know about patients' prehospital anxiety or about the course of anxiety during the hospitalization. This study does not have data to suggest biological mechanisms for the associations seen. In addition, we did not include all possible factors that might be related to prehospital delay, in-hospital complications, and length of stay in this study because they were not available. Further prospective studies are needed to replicate this finding.
We found that prehospital delay in seeking hospital treatment for AMI symptoms, together with state anxiety and Killip class (ie, presence of heart failure symptoms) on admission, is associated with the occurrence of more frequent serious AMI complications during the hospital stay. Further investigation is needed to test innovative strategies to reduce prehospital delay and assess whether reduction of anxiety acutely reduces the rate of arrhythmic and ischemic complications observed after AMI.
The findings from this study provide evidence of the impact of prehospital delay in seeking treatment for AMI symptoms on in-hospital clinical outcomes. Prehospital delay in seeking medical treatment is a substantial problem in most patients with AMI. It is essential that all clinicians caring for AMI and CHD patients teach them how to respond appropriately and quickly to acute cardiac symptoms, discuss the benefits of responding early, and discuss frankly the high rate of delay.52,53 Family members also should be educated. As the incidence of CHD continues to rise,1,54 it is essential that research and clinical efforts continue to focus on the complex and dynamic issue of reducing prehospital delay in patients with CHD.
Summary and Implications
- Prehospital delay, admission Killip class, and anxiety predicted in-hospital complications.
- The occurrence of in-hospital complications was related to length of stay in the hospital.
- Clinicians should teach all cardiac patients to seek treatment as quickly as possible for acute cardiac symptoms.
1. Lloyd-Jones D, Adams R, Carnethon M, et al. Heart disease and stroke statistics-2009 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation
2. Dracup K, Moser DK, Eisenberg M, et al. Causes of delay in seeking treatment for heart attack symptoms. Soc Sci Med
3. Canto JG, Shlipak MG, Rogers WJ, et al. Prevalence, clinical characteristics, and mortality among patients with myocardial infarction presenting without chest pain. JAMA
4. Zapka JG, Oakes JM, Simons-Morton DG, et al. Missed opportunities to impact fast response to AMI symptoms. Patient Educ Couns
5. Luepker RV, Raczynski JM, Osganian S, et al. 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
6. Meyborg P, Abdel-Wahab M, Herrmann G, et al. Relationship between therapeutic time intervals and intermediate term left ventricular systolic function in patients treated with facilitated percutaneous coronary intervention for acute myocardial infarction. Clin Res Cardiol
7. Berger PB, Ellis SG, Holmes DR Jr, et al. Relationship between delay in performing direct coronary angioplasty and early clinical outcome in patients with acute myocardial infarction: results from the global use of strategies to open occluded arteries in Acute Coronary Syndromes (GUSTO-IIb) trial. Circulation
8. Figueras J, Cortadellas J, Calvo F, et al. Relevance of delayed hospital admission on development of cardiac rupture during acute myocardial infarction: study in 225 patients with free wall, septal or papillary muscle rupture. J Am Coll Cardiol
9. Schuler J, Maier B, Behrens S, et al. Present treatment of acute myocardial infarction in patients over 75 years-data from the Berlin Myocardial Infarction Registry (BHIR). Clin Res Cardiol
10. Al-Mallah MH, Bazari RN, Khanal S. Delay in invasive risk stratification of women with acute coronary syndrome is associated with worse outcomes
. J Thromb Thrombolysis
11. Pohlen M, Bunzemeier H, Husemann W, et al. Risk predictors for adverse outcomes
after percutaneous coronary interventions and their related costs. Clin Res Cardiol
12. Moser DK, Kimble LP, Alberts MJ, et al. Reducing delay in seeking treatment by patients with acute coronary syndrome and stroke: a scientific statement from the American Heart Association Council on cardiovascular nursing and stroke council. Circulation
13. Caldwell MA, Froelicher ES, Drew BJ. Prehospital delay
time in acute myocardial infarction: an exploratory study on relation to hospital outcomes
and cost. Am Heart J
14. Faxon DP. Early reperfusion strategies after acute ST-segment elevation myocardial infarction: the importance of timing. Nat Clin Pract
15. Access to timely and optimal care of patients with acute coronary syndromes-community planning considerations: a report by the National Heart Attack Alert Program. J Thromb Thrombolysis
16. Krumholz HM, Anderson JL, Bachelder BL, et al. ACC/AHA 2008 performance measures for adults with ST-elevation and non-ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Performance Measures (writing committee to develop performance measures for ST-elevation and non-ST-elevation myocardial infarction): developed in collaboration with the American Academy of Family Physicians and the American College of Emergency Physicians: endorsed by the American Association of Cardiovascular and Pulmonary Rehabilitation, Society for Cardiovascular Angiography and Interventions, and Society of Hospital Medicine. Circulation
17. Cox JL, Lee E, Langer A, et al. Time to treatment with thrombolytic therapy: determinants and effect on short-term nonfatal outcomes
of acute myocardial infarction. Canadian GUSTO Investigators. Global Utilization of Streptokinase and + PA for Occluded Coronary Arteries. CMAJ
18. Moser DK. "The rust of life": impact of anxiety
on cardiac patients. Am J Crit Care
19. Moser DK, Riegel B, McKinley S, et al. Impact of anxiety
and perceived control on in-hospital complications after acute myocardial infarction. Psychosom Med
20. Zanini R, Aroldi M, Bonatti S, et al. Impact of prehospital diagnosis in the management of ST elevation myocardial infarction in the era of primary percutaneous coronary intervention: reduction of treatment delay and mortality. J Cardiovasc Med
21. Lefler LL, Bondy KN. Women's delay in seeking treatment with myocardial infarction: a meta-synthesis. J Cardiovasc Nurs
22. Berger JS, Brown DL. Gender-age interaction in early mortality following primary angioplasty for acute myocardial infarction. Am J Cardiol
23. Kosuge M, Kimura K, Kojima S, et al. Sex differences in early mortality of patients undergoing primary stenting for acute myocardial infarction. Circ J
24. Saleem MA, Kannam H, Aronow WS, et al. The effects of off-normal hours, age, and gender for coronary angioplasty on hospital mortality in patients undergoing coronary angioplasty for acute myocardial infarction. Am J Cardiol
25. Kanamasa K, Ishikawa K, Hayashi T, et al. Increased cardiac mortality in women compared with men in patients with acute myocardial infarction. Intern Med
26. Rawles JM, Metcalfe MJ, Shirreffs C, et al. Association of patient delay with symptoms, cardiac enzymes, and outcome in acute myocardial infarction. Eur Heart J
27. Moser DK, Dracup K, McKinley S, et al. An international perspective on gender differences in anxiety
early after acute myocardial infarction. Psychosom Med
28. De Jong MJ, Chung ML, Roser LP, et al. A five-country comparison of anxiety
early after acute myocardial infarction. Eur J Cardiovasc Nurs
29. Crowe JM, Runions J, Ebbesen LS, et al. Anxiety
and depression after acute myocardial infarction. Heart Lung
30. Huffman JC, Smith FA, Blais MA, et al. Anxiety
, independent of depressive symptoms, is associated with in-hospital cardiac complications after acute myocardial infarction. J Psychosom Res
31. Moser DK, Dracup K. Is anxiety
early after myocardial infarction associated with subsequent ischemic and arrhythmic events? Psychosom Med
32. Moser DK, McKinley S, Dracup K, et al. Gender differences in reasons patients delay in seeking treatment for acute myocardial infarction symptoms. Patient Educ Couns
33. De Jong MJ, Moser DK, An K, et al. Anxiety
is not manifested by elevated heart rate and blood pressure in acutely ill cardiac patients. Eur J Cardiovasc Nurs
34. Derogatis LR, Melisaratos N. The Brief Symptom Inventory: an introductory report. Psychol Med
35. De Jong MM, An K, McKinley S, et al. Using a 0-10 scale for assessment of anxiety
in patients with acute myocardial infarction. Dimens Crit Care Nurs
36. Abu Ruz MA, Lennie TA, Riegel B, et al. Anxiety
can be measured quickly and reliably in patients hospitalized for acute myocardial infarction [abstract]. Circulation
37. Killip T 3rd, Kimball JT. Treatment of myocardial infarction in a coronary care unit. A two year experience with 250 patients. Am J Cardiol
38. Scheffer J. Dealing with missing data. Res Lett Inf Math Sci
39. Bollen KA. Structural Equations With Latent Variables
. New York: Wiley; 1989.
40. Kline RB. Principles and Practice of Structural Equation Modeling
. 2nd ed. New York, NY: The Guilford Press; 2005.
41. Burnham KP, Anderson DR. Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach
. New York: Springer-Verlag; 2002.
42. Hu L, Bentler PM. Evaluating model fit. In: Hoyle RH, ed. Structural Equation Modeling: Concepts, Issues, and Applications
. 1st ed. Thousand Oaks, CA: SAGE Publications, Inc; 1995:76-99.
43. Lupon J, Valle V, Marrugat J, et al. Six-month outcome in unstable angina patients without previous myocardial infarction according to the use of tertiary cardiologic resources. RESCATE Investigators. Recursos Empleados en el Sindrome Coronario Agudo y Tiempos de Espera. J Am Coll Cardiol
44. Zahn R, Schiele R, Seidl K, et al. Primary angioplasty in acute myocardial infarction: differences between referred patients and those treated in hospitals with on-site facilities? J Invasive Cardiol
45. Kastanioti C. Costs, clinical outcomes
, and health-related quality of life of off-pump vs. on-pump coronary bypass surgery. Eur J Cardiovasc Nurs
46. Dimick JB, Pronovost PJ, Cowan JA, et al. Complications and costs after high-risk surgery: where should we focus quality improvement initiatives? J Am Coll Surg
47. Caldwell MA, Miaskowski C. Mass media interventions to reduce help-seeking delay in people with symptoms of acute myocardial infarction: time for a new approach? Patient Educ Couns
48. Dracup K, McKinley S, Riegel B, et al. A nursing intervention to reduce prehospital delay
in acute coronary syndrome: a randomized clinical trial. J Cardiovasc Nurs
49. Jiang SL, Ji XP, Zhao YX, et al. Predictors of in-hospital mortality difference between male and female patients with acute myocardial infarction. Am J Cardiol
50. Mehta RH, Marks D, Califf RM, et al. Differences in the clinical features and outcomes
in African Americans and whites with myocardial infarction. Am J Med
. 2006;119:70 e71-e78.
51. Thielmann M, Neuhauser M, Marr A, et al. Predictors and outcomes
of coronary artery bypass grafting in ST elevation myocardial infarction. Ann Thorac Surg
52. Antman EM, Hand M, Armstrong PW, et al. 2007 focused update of the ACC/AHA 2004 guidelines for the management of patients with ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol
53. Antman EM, Hand M, Armstrong PW, et al. Focused update of the ACC/AHA 2004 Guidelines for the Management of Patients With ST-Elevation Myocardial Infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines: developed in collaboration with the Canadian Cardiovascular Society endorsed by the American Academy of Family Physicians: 2007 writing group to review new evidence and update the ACC/AHA 2004 Guidelines for the Management of Patients With ST-Elevation Myocardial Infarction, writing on behalf of the 2004 writing committee. Circulation
54. Rosamond W, Flegal K, Furie K, et al. Heart disease and stroke statistics-2008 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation