`
`Application of the TIMI Risk Score
`for ST-Elevation MI in the National Registry
`of Myocardial Infarction 3
`
`David A. Morrow, MD
`Elliott M. Antman, MD
`Lori Parsons, BS
`James A. de Lemos, MD
`Christopher P. Cannon, MD
`Robert P. Giugliano, MD, SM
`Carolyn H. McCabe, BS
`Hal V. Barron, MD
`Eugene Braunwald, MD
`
`EFFECTIVE RISK STRATIFICATION IS
`
`integral to management of
`acute coronary syndromes.1
`Even among patients with
`ST-elevation myocardial infarction
`(STEMI), for whom initial therapeu-
`tic options are well-defined, patient risk
`characteristics impact short- and long-
`term medical decision making.2-4 Early
`risk assessment guides triage to alter-
`native levels of hospital care, deci-
`sions regarding therapeutic interven-
`tions, and application of clinical
`pathways that direct patient care and
`use of clinical resources. Despite well-
`characterized risk predictors,5-7 reli-
`able quantitative estimation of risk is
`challenging, as patients present with
`complex risk profiles requiring inte-
`gration of numerous elements of quali-
`tative and quantitative data. Thus, prac-
`tical tools that enhance clinicians’ ability
`to rapidly and accurately assess risk are
`of substantial interest.
`The Thrombolysis in Myocardial In-
`farction (TIMI) risk score for STEMI is
`a simple integer score that can be used
`at the bedside for risk stratification
`
`Context The Thrombolysis in Myocardial Infarction (TIMI) risk score for ST-
`elevation myocardial infarction (STEMI) is a simple integer score for bedside risk as-
`sessment of patients with STEMI. Developed and validated in multiple clinical trials of
`fibrinolysis, the risk score has not been validated in a community-based population.
`Objective To validate the TIMI risk score in a population of STEMI patients reflec-
`tive of contemporary practice.
`Design, Setting, and Participants The risk score was evaluated among 84029
`patients with STEMI from the National Registry of Myocardial Infarction 3 (NRMI 3),
`which collected data on consecutive patients with myocardial infarction (MI) from 1529
`US hospitals between April 1998 and June 2000.
`Main Outcome Measures Ability of the TIMI risk score to correctly predict risk of
`death in terms of model discrimination (c statistic) and calibration (agreement of pre-
`dicted and observed death rates).
`Results Patients in NRMI 3 tended to be older, to be more often female, and to have
`a history of coronary disease more often than those in the derivation set. Forty-eight
`percent received reperfusion therapy. The TIMI risk score revealed a significant graded
`increase in mortality with rising score (range, 1.1%-30.0%; P⬍.001 for trend). The risk
`score showed strong prognostic capacity overall (c=0.74 vs 0.78 in derivation set) and
`among patients receiving acute reperfusion therapy (c=0.79). Predictive behavior of the
`risk score was similar between fibrinolytic-treated patients (n=23960; c=0.79) and pri-
`mary percutaneous coronary intervention patients (n=15348; c=0.80). In contrast, among
`patients not receiving reperfusion therapy, the risk score underestimated death rates and
`offered lower discriminatory capacity (c=0.65).
`Conclusions Sufficiently simple to be practical at the bedside and effective for risk
`assessment across a spectrum of patients, the TIMI risk score may be useful in triage
`and treatment of patients with STEMI who are treated with reperfusion therapy.
`www.jama.com
`JAMA. 2001;286:1356-1359
`
`of patients at presentation with ST-
`elevation acute coronary syndromes.8
`Derived from 14114 patients enrolled
`
`in the InTIME II (Intravenous nPA for
`Treatment of Infarcting Myocardium
`Early) trial, the TIMI risk score is a ro-
`
`Author Affiliations: Department of Medicine,
`Brigham & Women’s Hospital, Boston, Mass
`(Drs Morrow, Antman, Cannon, Giugliano, and
`Braunwald and Ms McCabe); Ovation Research
`Group, Seattle, Wash (Ms Parsons); Donald W.
`Reynolds Cardiovascular Clinical Research Center,
`University of Texas Southwestern, Dallas (Dr de
`Lemos); and Department of Medicine, University of
`California, San Francisco, and Department of Medi-
`cal Affairs, Genentech Inc, San Francisco (Dr
`Barron).
`Financial Disclosure: Drs Antman, Giugliano, and
`
`Braunwald and Ms McCabe have received research
`grant support from Genentech. Dr Cannon serves as
`a consultant to Asahi Chemical Co and Bio-
`Technology General Corp; is a member of the speak-
`ers bureaus for Centocor and Genentech; and has re-
`ceived honoraria for preparation of educational
`materials from Centocor.
`Corresponding Author: David A. Morrow, MD, Car-
`diovascular Division, Brigham and Women’s Hospi-
`tal, 75 Francis St, Boston, MA 02115 (e-mail: damorrow
`@bics.bwh.harvard.edu).
`Reprints not available from the author.
`
`1356 JAMA, September 19, 2001—Vol 286, No. 11 (Reprinted)
`
`©2001 American Medical Association. All rights reserved.
`
`Downloaded From: https://jamanetwork.com/ on 06/18/2022
`
`1
`
`APPLE 1078
`Apple v. AliveCor
`IPR2021-00971
`
`
`
`bust clinical tool for mortality risk pre-
`diction in fibrinolytic-eligible patients
`with STEMI.8 Although it is docu-
`mented to perform well among pa-
`tients receiving fibrinolytics in clini-
`cal trials,8 the TIMI risk score has not
`been validated in a general population
`of patients with STEMI, including those
`treated with primary coronary revas-
`cularization or not receiving any form
`of acute reperfusion therapy. Since pro-
`spective validation in a data set reflec-
`tive of contemporary practice is impor-
`tant prior to widespread application of
`any prediction rule, we evaluated the
`prognostic performance of the TIMI risk
`score in a heterogeneous population
`treated in US hospitals for acute myo-
`cardial infarction (MI) and entered into
`the National Registry of Myocardial In-
`farction 3 (NRMI 3).
`
`METHODS
`The third NRMI is a prospective, ob-
`servational database of demographics,
`practice patterns, and health out-
`comes among patients with acute MI.9
`Data were collected on consecutive pa-
`tients with MI from 1529 hospitals be-
`tween April 1998 and June 2000. All
`treatment decisions were made at the
`discretion of the treating physicians.
`The present analysis included pa-
`tients with ST elevation or presumed
`new left bundle-branch block who com-
`pleted their stay at the admitting hos-
`pital and were not in cardiogenic shock
`at the initial evaluation.
`The TIMI risk score for STEMI is a
`weighted integer score based on 8 clini-
`cal risk indicators that can be easily as-
`certained at presentation (TABLE 1).8
`For each patient, the score is calcu-
`lated as the arithmetic sum of the points
`for each risk feature present (range,
`0-14). The TIMI risk score was devel-
`oped using multivariable methods
`among patients from the InTIME II trial,
`a phase 3 trial of lanoteplase vs al-
`teplase reperfusion therapy.8 The risk
`score was derived based on mortality
`through 30 days after presentation but
`showed stable prognostic perfor-
`mance across multiple time points, in-
`cluding time to discharge (c=0.78).8
`
`THE TIMI RISK SCORE AND MORTALITY RISK PREDICTION
`
`Evaluation of the TIMI risk score was
`based on NRMI 3 patients with com-
`plete baseline data (89%). The prognos-
`tic discriminatory capacity of the TIMI
`risk score was expressed as the c statis-
`tic, representing the area under the re-
`ceiver operating characteristic curve for
`prediction of in-hospital death.10 Differ-
`ences in event rates with increasing risk
`scores were assessed using the 2 test for
`trend. Model calibration was assessed by
`construction of plots of predicted vs ac-
`tual death rates across the entire spec-
`trum of predicted risk. Testing for dif-
`ferences in mortality gradients among
`groups with different treatment modes
`was performed using logistic regres-
`sion analysis with interaction terms. The
`prognostic contributions of variables not
`
`included in the risk score were as-
`sessed by stepwise logistic regression. Pa-
`tients who did not receive any reperfu-
`
`3
`2
`1
`
`Table 1. Elements of the TIMI Risk Score*
`Clinical Risk Indicators
`Points
`Historical
`Age, y
`ⱖ75
`65-74
`History of diabetes, hypertension,
`or angina
`Examination
`Systolic blood pressure ⬍100 mm Hg
`Heart rate ⬎100/min
`Killip class II-IV
`Weight ⬍67 kg
`Presentation
`Anterior ST elevation or left
`bundle-branch block
`1
`Time to reperfusion therapy ⬎4 h
`14
`Total possible points
`*TIMI indicates Thrombolysis in Myocardial Infarction.
`
`3
`2
`2
`1
`
`1
`
`Table 2. Baseline Characteristics of the NRMI 3 Validation and InTIME II Derivation Sets*
`NRMI 3 Validation Set
`
`All Patients
`(n = 84 029)
`
`Reperfusion
`Therapy
`(n = 40 214)
`
`No Reperfusion
`Therapy
`(n = 43 815)
`
`InTIME II
`Derivation Set
`(n = 14 114)
`
`69 (14)
`38.3
`40.5
`
`79 (20)
`28.0
`
`27.2
`27.2
`54.0
`
`23.3
`11.5
`9.5
`10.8
`
`48.6
`44.5
`25.1
`
`63 (13)
`21.1
`31.4
`
`83 (19)
`18.4
`
`37.7
`19.6
`47.7
`
`17.5
`8.8
`11.1
`7.1
`
`36.8
`59.7
`10.9
`
`86 (24)
`23.8
`
`141 (32)
`8.7
`
`78 (20)
`12.1
`
`140 (31)
`8.3
`
`. . .
`
`30.4
`
`. . .
`
`4 (2-6)
`
`3 (1-5)
`
`5 (4-7)
`
`Characteristics
`Demographics
`Age, y
`Mean (SD)
`⬎75, %
`Female, %
`Weight, kg
`Mean (SD)
`⬍67, %
`Risk factors, %
`Current smoker
`Diabetes
`History of hypertension
`Cardiovascular history, %
`Prior myocardial infarction
`Prior angina
`Prior PCI
`Prior CABG
`Presenting characteristics
`Infarct location, %
`Anterior or LBBB
`Inferior
`Killip class II-IV, %
`Heart rate, beats/min
`Mean (SD)
`⬎100, %
`Systolic blood pressure, mm Hg
`Mean (SD)
`⬍100, %
`Treatment
`Time to reperfusion
`therapy ⬎4 h, %
`TIMI risk score, median
`(interquartile range)
`*NRMI 3 indicates National Registry of Myocardial Infarction 3; InTIME II, Intravenous nPA for Treatment of Infarcting
`Myocardium Early; PCI, percutaneous coronary intervention; CABG, coronary artery bypass graft; LBBB, left bundle-
`branch block; ellipses, data not applicable; and TIMI, Thrombolysis in Myocardial Infarction.
`
`74 (13)
`54.1
`48.8
`
`75 (20)
`36.9
`
`17.6
`34.2
`59.8
`
`28.5
`13.9
`8.0
`14.2
`
`59.4
`30.5
`38.2
`
`93 (25)
`34.5
`
`141 (33)
`9.1
`
`61 (12)
`13.4
`24.8
`
`78 (14)
`20.0
`
`45.0
`14.1
`30.4
`
`16.0
`21.6
`4.4
`2.7
`
`42.7
`57.3
`12.2
`
`76 (18)
`7.6
`
`139 (22)
`2.5
`
`24.4
`
`3 (1-4)
`
`©2001 American Medical Association. All rights reserved.
`
`(Reprinted) JAMA, September 19, 2001—Vol 286, No. 11 1357
`
`Downloaded From: https://jamanetwork.com/ on 06/18/2022
`
`2
`
`
`
`heart failure, previous MIs, and prior
`coronary revascularization proce-
`dures than patients in the derivation
`set.8 Among the NRMI 3 validation set,
`40214 patients (48%) were treated with
`pharmacological or mechanical reper-
`fusion therapy. Those who underwent
`primary percutaneous coronary inter-
`vention (n=15348) represented 38% of
`patients who received reperfusion
`therapy. Patients treated without reper-
`fusion therapy had more frequent high-
`risk features and a higher median TIMI
`risk score (Table 2).
`In-hospital mortality was 12.6%
`(n = 10 612). Application of the TIMI
`risk score in the overall population from
`NRMI 3 revealed a significant, nearly
`30-fold graded increase in risk be-
`tween patients with a score of 0 and
`those with a score of 8 or higher (range,
`1.1%-30.0%; P⬍.001 for trend). As-
`sessed by the area under the receiver op-
`erating characteristic curve, the risk
`score showed a strong prognostic ca-
`pacity (c=0.74) that was comparable
`with the risk score performance in the
`InTIME II trial (c=0.78).8
`Stratification of the population into
`those who were treated with vs with-
`out reperfusion therapy revealed sub-
`stantial differences in risk score dis-
`criminatory performance as well as in
`calibration. The prognostic capacity of
`the TIMI risk score among patients
`treated with acute reperfusion therapy
`(c = 0.79) was unchanged compared
`with InTIME II and was similar be-
`tween patients treated with fibrinolyt-
`ics and those who underwent primary
`percutaneous coronary interventions
`(c=0.79 vs 0.80), with no significant
`difference in the slope of the risk gra-
`dient between the 2 groups (P=.09 for
`interaction).
`The FIGURE shows the behavior of
`the TIMI risk score in terms of predict-
`ing death across the spectrum of ex-
`pected risk. The observed mortality
`rates for patients in NRMI 3 receiving
`reperfusion therapy were strongly con-
`cordant with risk estimates derived
`from InTIME II (r=0.99; Figure), in-
`dicating good model calibration. Among
`patients treated without reperfusion
`
`THE TIMI RISK SCORE AND MORTALITY RISK PREDICTION
`
`sion therapy were not assessed for the
`predictor variable of time to reperfu-
`sion therapy. Two-tailed P values ⬍.05
`were considered significant. Analyses
`were performed using SAS, version 8.0
`(SAS Institute Inc, Cary, NC).
`
`RESULTS
`The analysis included 84029 patients
`with STEMI. Baseline characteristics are
`summarized in TABLE 2. Patients from
`NRMI 3 tended to be older; were more
`often female; and tended to have more
`
`Figure. Prediction of In-Hospital Mortality With TIMI Risk Score for STEMI
`
`NRMI 3
`No Reperfusion Therapy
`Reperfusion Therapy
`
`InTIME II
`Reperfusion Therapy
`
`35
`
`32
`
`29
`
`22
`
`19
`
`25
`
`23
`
`24
`
`24
`
`21
`
`18
`
`10
`
`6.8
`
`2.1
`
`0.7 0.7
`
`3.5
`
`0.9
`
`1.3
`
`2.2
`
`1.9
`
`14
`
`15
`
`15
`
`12
`
`10
`
`6.5
`
`5.7
`
`3.7
`
`3.9
`
`0
`
`1
`
`2
`
`3
`
`4
`5
`TIMI Risk Score
`
`6
`
`7
`
`8
`
`>8
`
`40
`
`30
`
`20
`
`10
`
`0
`
`In-Hospital Mortality, %
`
`Total No. of Patients
`No Reperfusion Therapy, NRMI 3
`1258
`2945
`Reperfusion Therapy, NRMI 3
`4039
`7630
`Reperfusion Therapy, InTIME II
`1705
`3029
`
`2479
`
`6223
`
`2287
`
`3590
`
`5739
`
`2252
`
`5771
`
`5750
`
`1983
`
`6000
`
`4261
`
`1313
`
`5881
`
`2781
`
`824
`
`5295
`
`1765
`
`367
`
`4453
`
`1011
`
`209
`
`6143
`
`1015
`
`145
`
`STEMI indicates ST-elevation myocardial infarction; NRMI 3, the National Registry of Myocardial Infarction 3.
`Data for the Intravenous nPA for Treatment of Infarcting Myocardium Early (InTIME II) trial are from Morrow.8
`
`Table 3. Adjusted Mortality Risk Relationships in NRMI 3 vs InTIME II*
`NRMI 3
`
`Reperfusion
`Therapy,
`OR (95% CI)
`3.3 (3.0-3.7)
`2.4 (2.2-2.7)
`2.2 (2.0-2.4)
`1.7 (1.6-1.8)
`3.2 (2.8-3.5)
`1.0 (0.9-1.1)
`1.4 (1.2-1.5)
`1.1 (1.0-1.2)
`
`No Reperfusion
`Therapy,
`OR (95% CI)
`1.6 (1.5-1.7)
`1.5 (1.4-1.6)
`1.2 (1.2-1.3)
`0.93 (0.9-1.0)
`4.3 (4.0-4.6)
`. . .
`1.2 (1.2-1.3)
`0.94 (0.9-1.0)
`
`InTIME II,
`OR (95% CI)
`2.7 (2.2-3.2)
`2.3 (1.9-2.7)
`2.3 (1.9-2.8)
`1.6 (1.4-1.9)
`2.7 (1.9-3.8)
`1.4 (1.2-1.6)
`1.4 (1.2-1.7)
`1.4 (1.2-1.6)
`
`Points in TIMI
`Risk Score
`3
`2
`2
`1
`3
`1
`1
`1
`
`Risk Characteristics
`Age ⱖ75 y
`Killip class II-IV
`Heart rate ⬎100/min
`Anterior MI or LBBB
`SBP ⬍100 mm Hg
`Time to reperfusion therapy ⬎4 h
`Weight ⬍67 kg
`History of angina, hypertension,
`or diabetes
`1.3 (1.1-1.5)
`1.5 (1.3-1.6)
`1.6 (1.5-1.8)
`Nonsmoker
`1.3 (1.1-1.6)
`0.9 (0.8-1.0)
`†
`Prior MI
`1.5 (1.1-1.9)
`NA
`NA
`Peripheral vascular disease
`1.8 (1.1-2.8)
`NA
`NA
`Antiarrhythmic medication
`0.7 (0.5-1.0)
`NA
`NA
`Lipid-lowering medication
`1.2 (1.0-1.5)
`†
`1.4 (1.3-1.6)
`Female
`*Comparison of the full Intravenous nPA for Treatment of Infarcting Myocardium Early (InTIME II) multivariable model
`between data sets should be viewed with caution because of missing covariates in the National Registry of
`Myocardial Infarction 3 (NRMI 3). OR indicates odds ratio; CI, confidence interval; MI, myocardial infarction; LBBB,
`left bundle-branch block; SBP, systolic blood pressure; ellipses, data not applicable; and NA, data not available in
`NRMI 3.
`†Variable did not enter into model.
`
`1358 JAMA, September 19, 2001—Vol 286, No. 11 (Reprinted)
`
`©2001 American Medical Association. All rights reserved.
`
`Downloaded From: https://jamanetwork.com/ on 06/18/2022
`
`3
`
`
`
`THE TIMI RISK SCORE AND MORTALITY RISK PREDICTION
`
`therapy, a significant graded relation-
`ship between the TIMI risk score and
`mortality was also evident (P⬍.001 for
`trend; Figure). Of note, however, the
`slope of the risk gradient in this group
`was less steep owing to a pattern of
`higher mortality among patients with
`risk scores in the low and middle range
`(P⬍.001 for interaction; c=0.65). As
`such, the quantitative mortality esti-
`mates from InTIME II underestimated
`the risk for patients treated without
`reperfusion therapy except those with
`the highest predicted death rates.
`Exploratory analysis was per-
`formed to identify additional impor-
`tant predictors in patients treated with-
`out reperfusion therapy. Among the
`variables considered, bleeding risk (ac-
`tive internal bleeding or recent surgery/
`trauma), uncertainty regarding diag-
`nosis, major organ failure, and chronic
`renal failure added significantly to the
`multivariable model including each of
`the risk score predictors (c = 0.746;
`P⬍.001). While history of smoking and
`prior stroke added further to the model,
`the improvement in discriminatory ca-
`pacity was small (c=0.750).
`
`COMMENT
`Results from risk prediction tools de-
`veloped in carefully selected patients en-
`rolled in clinical trials may not be gen-
`eralizable to heterogeneous, “real-
`world” patient populations. This analysis
`demonstrates the robust prognostic per-
`formance of the TIMI risk score in a gen-
`eral population of patients with ST-
`elevation acute coronary syndromes
`treated with acute reperfusion therapy
`in a diverse group of US hospitals. The
`strong predictive capacity of the risk
`score was evident among patients treated
`with either pharmacological or mechani-
`cal reperfusion therapy. These observa-
`tions establish the prognostic efficacy of
`the TIMI risk score in a large group of
`patients representative of contempo-
`rary clinical practice.
`Patients who were not adminis-
`tered reperfusion therapy showed a pat-
`tern of higher mortality risk. Al-
`though the difference in outcomes may
`be due in part to the established ben-
`
`efits of reperfusion therapy, we iden-
`tified several high-risk features not in-
`cluded in the TIMI risk score that are
`likely related to the decision not to ad-
`minister reperfusion therapy, and of-
`fer additional predictive information.
`While the quantitative mortality esti-
`mates from InTIME II do not apply to
`these patients, the TIMI risk score may
`aid in their categorization into groups
`of low, moderate, and high relative risk.
`Limited information is available regard-
`ing the performance of other vali-
`dated models stratified by use of reper-
`fusion therapy. Our data suggest that
`future work should include evalua-
`tion of existing and new models in these
`important subgroups of the overall
`population with acute MI (TABLE 3).
`The TIMI risk score was developed
`with the objective of creating a risk as-
`sessment tool that is both effective and
`convenient for use at patient presenta-
`tion. With just a few important clinical
`factors, the risk score captures the ma-
`jority of prognostic information avail-
`able from more complex models among
`patients treated with reperfusion
`therapy.8 The discriminatory capacity of
`the model could be increased by inclu-
`sion of additional variables or more com-
`plex modeling, but at the cost of hinder-
`ing practical application. Other well-
`validated models have been derived for
`the purpose of risk-adjusted analysis of
`hospital outcomes to direct quality im-
`provement efforts.7,11 Such models have
`incorporated data acquired during hos-
`pitalization to provide high-discrimina-
`tory capacity with respect to long-term
`outcomes. In contrast, the TIMI risk
`score is based on clinical information that
`is available at the time of hospital ar-
`rival and, thus, is suitable for early risk
`stratification at the bedside without the
`need for a computer. The impact of dif-
`ferences in treatment, along with non-
`invasive and invasive data accrued dur-
`ing the course of hospitalization, should
`be considered by clinicians in a continu-
`ous process of updating the initial as-
`sessment of risk offered by the TIMI risk
`score.
`Sufficiently simple to be practical at
`the bedside and effective for risk as-
`
`sessment across a heterogeneous spec-
`trum of patients, the TIMI risk score
`may be clinically useful in the triage and
`treatment of patients with STEMI who
`undergo acute reperfusion therapy.
`
`Author Contributions: Study concept and design: Mor-
`row, Antman, Barron, Braunwald.
`Acquisition of data: Barron.
`Analysis and interpretation of data: Morrow, Ant-
`man, Parsons, de Lemos, Cannon, Giugliano,
`McCabe, Barron, Braunwald.
`Drafting of the manuscript: Morrow.
`Critical revision of the manuscript for important in-
`tellectual content: Morrow, Antman, Parsons, de
`Lemos, Cannon, Giugliano, McCabe, Barron, Braun-
`wald.
`Statistical expertise: Morrow, Antman, Parsons,
`Giugliano, Barron.
`Obtained funding: Barron.
`Administrative, technical, or material support:
`Barron.
`Study supervision: Antman, Cannon, Barron, Braun-
`wald.
`Funding/Support: The National Registry of Myocar-
`dial Infarction 3 is supported by Genentech Inc.
`
`REFERENCES
`1. Maseri A, Rebuzzi AG, Cianflone D. Need for a com-
`posite risk stratification of patients with unstable coro-
`nary syndromes tailored to clinical practice. Circula-
`tion. 1997;96:4141-4142.
`2. Fibrinolytic Therapy Trialists’ Collaborative
`Group. Indications for fibrinolytic therapy in sus-
`pected acute myocardial infarction. Lancet. 1994;
`343:311-322.
`3. Becker RC, Burns M, Gore JM, et al. Early assess-
`ment and in-hospital management of patients with
`acute myocardial infarction at increased risk for
`adverse outcomes. Am Heart J. 1998;135:786-796.
`4. Hochman JS, Sleeper LA, Webb JG, et al. Early re-
`vascularization in acute myocardial infarction compli-
`cated by cardiogenic shock. N Engl J Med. 1999;341:
`625-634.
`5. Lee KL, Woodlief LH, Topol EJ, et al. Predictors of
`30-day mortality in the era of reperfusion for acute
`myocardial infarction. Circulation. 1995;91:1659-
`1668.
`6. Jacobs DR Jr, Kroenke C, Crow R, et al. PREDICT:
`a simple risk score for clinical severity and long-term
`prognosis after hospitalization for acute myocardial in-
`farction or unstable angina: the Minnesota Heart Sur-
`vey. Circulation. 1999;100:599-607.
`7. Krumholz HM, Chen J, Wang Y, Radford MJ, Chen
`YT, Marciniak TA. Comparing AMI mortality among
`hospitals in patients 65 years of age and older. Cir-
`culation. 1999;99:2986-2992.
`8. Morrow DA, Antman EM, Charlesworth A, et al.
`TIMI risk score for ST-elevation myocardial infarc-
`tion: a convenient, bedside, clinical score for risk as-
`sessment at presentation: an InTIME II trial substudy.
`Circulation. 2000;102:2031-2037.
`9. Rogers WJ, Bowlby LJ, Chandra NC, et al. Treat-
`ment of myocardial infarction in the United States
`(1990 to 1993): observations from the National Reg-
`istry of Myocardial Infarction. Circulation. 1994;90:
`2103-2114.
`10. Hanley JA, McNeil BJ. The meaning and use of
`the area under a receiver operating characteristic
`(ROC) curve. Radiology. 1982;143:29-36.
`11. Tu JV, Austin PC, Walld R, et al. Development
`and validation of the Ontario acute myocardial
`infarction mortality prediction rules. J Am Coll Car-
`diol. 2001;37:992-997.
`
`©2001 American Medical Association. All rights reserved.
`
`(Reprinted) JAMA, September 19, 2001—Vol 286, No. 11 1359
`
`Downloaded From: https://jamanetwork.com/ on 06/18/2022
`
`4
`
`