trends in mortality of acute myocardial infarction after...

Trends in Mortality of Acute Myocardial Infarction AfterDischarge From the Hospital

William J. Kostis, PhD, MD; Yingzi Deng, MD, MS; John S. Pantazopoulos, MD;Abel E. Moreyra, MD; John B. Kostis, MD; for the Myocardial Infarction Data Acquisition System

(MIDAS14) Study Group

Background—We assessed trends in the prognosis of patients with acute myocardial infarction hospitalized in New Jerseyhospitals. In recent decades, in-hospital mortality has declined markedly but the decline in longer-term mortality is lesspronounced, implying that mortality after discharge has worsened.

Methods and Results—Using the Myocardial Infarction Data Acquisition System (MIDAS), we examined the outcomesof 285 397 patients hospitalized for a first acute myocardial infarction between 1986 and 2007. Mortality at dischargedecreased by 9.4% from 16.9% to 7.5% (annual change, �0.44; 95% confidence interval, �0.49 to �0.40), but thedecrease at 1 year was less pronounced (6.4%) because of an increase in mortality from discharge to 1 year afterdischarge (from 12.1% to 13.9%; annual change, �0.15; 95% confidence interval, �0.10 to �0.20). Mortality from 30days after discharge to 1 year, a measure not affected by length of stay, increased by 1.2% (annual change, �0.10; 95%confidence interval, �0.06 to �0.23). The effect was more evident in the older age groups and was due tononcardiovascular mortality, especially from respiratory and renal diseases, septicemia, and cancer. All effects remainedstatistically significant (P�0.0001) after adjustment for demographics, comorbidities, infarction type, complications,and interventions. Piecewise linear regressions confirmed these trends.

Conclusions—Postdischarge mortality of patients with acute myocardial infarction is increasing, primarily because ofhigher noncardiovascular mortality in the older age groups. (Circ Cardiovasc Qual Outcomes. 2010;3:581-589.)

Key Words: epidemiology � mortality � myocardial infarction

In recent decades, a marked decrease of in-hospital mortal-ity of patients with acute myocardial infarction (AMI) has

been documented in clinical trials, in prospective registries,and in epidemiological studies.1–5 Although longer-term mor-tality of AMI has also declined, in some studies this declineis less pronounced than mortality at discharge, implying thatmortality after discharge has worsened.1,4,6–9 In recent years,AMI patients are older and have more comorbidities, thelength of hospital stay has decreased, the diagnostic criteriahave changed, and better control of risk factors may haveresulted in smaller AMIs.4–6,9–12 Many improvements havebeen made in the management of AMI and in secondaryprevention with reperfusion, medications, revascularization,and emphasis on process improvement.1,3,4,6,7,13–18

Editorial see p 568The purpose of the present study was to examine mortalitytrends observed among AMI patients admitted to NewJersey hospitals while considering changes in patient

characteristics, comorbidities, complications, interven-tions, and length of stay.

MethodsData SourcesThe data for this study were obtained from the Myocardial InfarctionData Acquisition System (MIDAS) from January 1, 1986, to De-cember 31, 2008.19,20 Cases admitted in December and discharged inJanuary were assigned to the year of discharge. MIDAS containshospital discharge data from nonfederal acute care hospitals in NewJersey and includes all records with a primary diagnosis of AMI(International Classification of Disease 9th Revision, 410.0 to 410.9[ICD9]).

Out-of-hospital death information (date and cause of death) wasobtained by matching the MIDAS records to the New Jersey deathregistration files, using a public automated record linkage andconsolidation software (The Link King).21 In a study of 500 000cases with a blinded clerical review of a sample of 500 record pairs,the algorithm had a sensitivity of 96.7% and a positive predictivevalue of 96.1%.22 First AMIs were AMIs occurring for the first timein an individual during the duration of the study and without the

Received April 27, 2010; accepted August 25, 2010.From the Cardiology Division (W.J.K.), Massachusetts General Hospital, Boston, Mass; and the Department of Medicine (Y.D., J.S.P., A.E.M., J.B.K.),

UMDNJ-Robert Wood Johnson Medical School, New Brunswick, NJ.A portion of the data included in this article has been accepted for presentation at the annual scientific meeting of the American Heart Association in

November 2010.The online-only Data Supplement is available at http://circoutcomes.ahajournals.org/cgi/content/full/CIRCOUTCOMES.110.957803/DC1.Correspondence to John B. Kostis, MD, Department of Medicine, UMDNJ-Robert Wood Johnson Medical School, One Robert Wood Johnson Pl, New

Brunswick, NJ 08903-0019. E-mail [email protected]© 2010 American Heart Association, Inc.

Circ Cardiovasc Qual Outcomes is available at http://circoutcomes.ahajournals.org DOI: 10.1161/CIRCOUTCOMES.110.957803

581

by guest on July 14, 2018http://circoutcom

es.ahajournals.org/D

ownloaded from

by guest on July 14, 2018

http://circoutcomes.ahajournals.org/

Dow

nloaded from



ownloaded from



Dow

nloaded from



ownloaded from



Dow

nloaded from



ownloaded from



Dow

nloaded from



ownloaded from



Dow

nloaded from



ownloaded from



Dow

nloaded from



ownloaded from



Dow

nloaded from



ownloaded from



Dow

nloaded from



ownloaded from


















diagnosis of prior or old AMI among the discharge diagnoses. Toobtain an estimate of deaths occurring outside the state of NewJersey, 7308 randomly selected MIDAS from the years 1986 to 2004cases were matched to the National Death Index. Only 14 deaths(0.19% of the sample) occurred out-of-state within 30 days ofdischarge and 119 within 1 year of discharge representing or 0.01%and 0.04%, respectively, of the patients admitted. Deaths occurringin hospice or home hospice (n�540) were included in the in-hospitaldeaths.

WHAT IS KNOWN

● A marked decrease of in-hospital mortality of acutemyocardial infarction (AMI) patients has occurred inrecent decades.

● Longer-term mortality of AMI has also declined, butin some studies this decline is less pronounced thanmortality at discharge.

● Many changes in the characteristics and managementof AMI have occurred during the same decades.

WHAT THE STUDY ADDS

● Studying 285 397 patients hospitalized in New Jer-sey for a first AMI between 1986 and 2007, weobserved an increase in mortality after discharge.

● The increase in postdischarge mortality was moreevident in the older age groups and was due tononcardiovascular mortality, especially respiratoryand renal diseases, septicemia, and cancer.

The study included 285 397 patients ages 35 years or older whowere discharged from New Jersey hospitals with a first AMI as theprimary reason for admission between 1986 and 2007. AMI patientsadmitted to nursing homes or to federal hospitals (amounting to�3% of all AMI patients) were excluded. Patients with an AMIlisted as a secondary diagnosis (ie, not the reason for admission)were also excluded (n�88 905). The New Jersey State InstitutionalReview Board and the University of Medicine and Dentistry of NewJersey Institutional Review Board approved the study.

Study VariablesIn-hospital mortality was calculated as the ratio of the number ofpatients who died during the AMI admission divided by the numberof AMI patients admitted. One-year mortality was calculated as theratio of the number of patients who died within 1 year from the dateof admission divided by the number of AMI patients admitted.Thirty-day and 1-year postdischarge mortality was computed as theratio of the number of patients who died between discharge and 30days (or 1 year) after discharge divided by the total number of thosedischarged alive. Interval after discharge mortality between 30 daysand 1 year after discharge was calculated as the ratio of the numberof patients who died between 31 days and 1 year after dischargedivided by the number of patients who were discharged alive.All-cause, cardiovascular disease (CVD), and non-CVD (NCVD)death after discharge mortality was calculated. The difference of themortality in 1986 and the mortality at each subsequent year of thestudy was calculated. CVD death included codes ICD9 390 to 459,ICD10 I00-I99. NCVD death included all other causes of death.

Covariates included patient demographics, AMI site, comorbidi-ties, and the use of invasive procedures during the admission. AMIsite was coded at discharge as follows: transmural (anterior ICD9410.0, 410.1; inferior ICD9 410.2, 410.4; lateral ICD9 410.3, 410.5;posterior ICD9 410.6), subendocardial (ICD9 410.7), and other/unspecified (410.8, 410.9). Comorbidities included anemia (281.9,

282.0, 280.0 to 281.3, 283.0 to 285.9), hypertension (401 to 405),liver disease (571.0 to 571.9), chronic renal disease (580 to 586),stroke (430, 431, 435, 436, 432.9, 362.3, 433.01, 433.11, 433.21,433.31, 433.81, 433.91, 434.01, 434.11, 434.91, 997.02), cerebro-vascular disease (430.0 to 438.9), heart failure (428), chronicobstructive pulmonary disease (490 to 496), cancer (140.0 to 208.0),invasive procedures including CATH (37.21 to 37.23, 88.52 to88.57), percutaneous coronary intervention (36.00 to 36.09, 00.66),and coronary artery bypass grafting (36.1, 36.2). Procedure identifi-cation was based on the AHRQ clinical classifications software(CCS 2008) as it pertained to each hospitalization year.23 Informa-tion on insurance (commercial, Medicare, Medicaid, HMO, andself-pay) was available only for patients discharged from 1994onward.

Statistical AnalysisStatistical analyses were performed using SAS v.9.2 software.Patient demographic and clinical characteristics were comparedacross hospitalization years using the Cochran-Armitage test orMantel-extension �2 test for categorical variables and ANOVA Ftests for continuous variables. A series of general linear models weregenerated to calculate the average annual change of mortalities andcorresponding 95% confidence intervals (CI). Goodness-of-fit of thelinear regression was evaluated by R2 and probability value for F teststatistic. We also investigated a possible nonlinear associationbetween mortality and index hospitalization year using piecewiselinear fits by hospitalization year (1986 to 1992, 1993 to 2001, and2002 to 2007). Multivariate logistic regression was used to measurethe effect of calendar year on in-hospital and 30-day mortality endpoints with the referent year of 1986, as well as with the calendaryear as a continuous variable, after adjusting for patient age, sex,race, AMI site, comorbidities, complications, and invasive cardiacprocedures performed during the admission. One-year mortality endpoints were analyzed using Cox regression adjusting for the variableslisted above.

Sensitivity analysis was performed using the subset of patientsadmitted between 1994 and 2007. In this group (n�185 483), weexamined trends adjusting for insurance type in addition to thevariables used in the primary analysis. Regression lines were fitted toexamine trends of all cause, CVD, and NCVD mortality fromadmission as well as after discharge and starting from 30 days afterdischarge.

ResultsTrends in Patient CharacteristicsDuring the years of the study, the age of the patients with firstAMI increased by 1.9 years (Table A of the online-only DataSupplement). The percentage with commercial insurancedecreased by 4.9%, whereas HMO increased by 11.3%. Thepercentage of patients with subendocardial AMI increased by46.2%, from 16.6% to 62.8%, with hypertension increased by25.0%, with anemia by 5.4%, with chronic obstructivepulmonary disease by 4.1%, with chronic renal disease by14.4%, with diabetes by 4.4%, and with cancer by 1.5%. Thepercentage of patients who had percutaneous coronary inter-vention during the index hospitalization increased 26-fold,from 1.3% to 34.1%, whereas the length of stay decreased byhalf from 11.3 to 5.7 (P�0.0001 for all changes above).

In-Hospital and Postdischarge MortalityFrom 1986 to 2007, in-hospital mortality decreased by 9.4%from 16.9% to 7.5% (annual change, �0.44; 95% CI, �0.49to �0.40; R2�0.95; Table 1 and Figure 1). Mortality fromadmission to 30 days also decreased by 8.0%, from 18.2% to10.2% (annual change, �0.36; 95% CI, �0.41 to �0.31;R2�0.91), as did mortality from admission to 1 year (by 6.4%

582 Circ Cardiovasc Qual Outcomes November 2010



ownloaded from


Table 1. All-Cause Death, CVD Death, and NCVD Death Rates at Different Time Intervals by Year

In-Hospital30 Days From

Admission� (30-Day)(In-Hospital)

1 Year FromAdmission

30 Days FromDischarge

1 Year FromDischarge

All-cause death, %

1986 16.93 18.24 1.30 26.62 3.32 12.07

1987 16.47 17.89 1.42 25.87 3.35 11.77

1988 16.87 18.28 1.41 26.57 3.71 12.23

1989 15.45 16.75 1.30 25.36 3.72 12.12

1990 14.73 16.15 1.42 24.01 3.31 11.36

1991 14.10 15.39 1.29 23.31 3.12 11.11

1992 12.72 14.05 1.32 22.22 3.47 11.24

1993 11.85 13.25 1.40 21.34 3.22 11.17

1994 11.79 13.44 1.65 21.67 3.58 11.49

1995 12.34 14.13 1.79 22.17 3.52 11.54

1996 11.51 13.77 2.26 21.71 3.86 11.83

1997 10.69 13.33 2.64 21.42 4.01 12.21

1998 10.11 12.42 2.31 21.24 3.72 12.61

1999 10.52 12.99 2.47 22.70 4.08 13.80

2000 10.15 12.95 2.81 22.65 4.44 14.09

2001 9.90 12.70 2.80 23.15 4.54 14.93

2002 9.30 12.04 2.73 21.89 4.20 14.03

2003 8.63 11.34 2.71 21.65 4.18 14.45

2004 8.47 11.07 2.60 21.11 4.19 13.96

2005 8.41 11.19 2.79 20.99 4.34 13.97

2006 8.15 10.61 2.46 20.34 3.82 13.40

2007 7.53 10.24 2.71 20.19 3.92 13.90

Annual change,% (95% CI)

�0.44(�0.49 to �0.40)

�0.36(�0.41 to �0.31)

0.09(0.07 to 0.22)

�0.24(�0.32 to �0.17)

0.05(0.03 to 0.07)

0.15(0.10 to 0.20)

R2 0.95 0.91 0.82 0.66 0.56 0.62

P value �0.0001 �0.0001 �0.0001 �0.0001 �0.0001 �0.0001

CVD death, %

1986 12.18 13.64 1.46 20.32 3.03 9.99

1987 11.92 13.38 1.46 19.63 2.91 9.43

1988 12.48 14.05 1.57 20.53 3.29 9.91

1989 11.03 12.41 1.39 18.83 3.21 9.41

1990 10.60 12.14 1.54 17.60 2.81 8.40

1991 10.03 11.28 1.25 16.80 2.53 8.08

1992 9.30 10.55 1.25 16.39 2.84 8.30

1993 8.86 10.10 1.24 15.94 2.66 8.23

1994 8.36 9.99 1.63 15.56 2.97 8.23

1995 8.48 10.23 1.75 15.53 2.85 8.20

1996 7.72 9.64 1.93 14.96 3.06 8.30

1997 7.93 10.11 2.18 15.45 3.13 8.51

1998 7.23 9.10 1.87 14.61 2.81 8.32

1999 7.15 9.12 1.97 15.12 2.96 9.01

2000 6.75 8.90 2.16 14.78 3.25 9.06

2001 6.73 8.85 2.12 14.98 3.21 9.27

2002 6.06 8.10 2.04 13.60 2.92 8.41

2003 5.58 7.62 2.04 13.54 2.95 8.80

2004 5.57 7.46 1.89 13.02 2.84 8.24

2005 5.42 7.49 2.08 12.89 3.00 8.26

(Continued)

Kostis et al Myocardial Infarction Mortality After Discharge 583



ownloaded from


from 26.6% to 20.2%; annual change, �0.24; 95% CI, �0.32to �0.17; R2�0.66). However, the decrease in mortality at 30days (8.0%) and at 1 year (6.4%) was less pronounced thanthat at discharge (9.4%). This was due to increasing mortalityafter discharge. Thirty-day postdischarge mortality increasedamong patients discharged alive by 0.6% (from 3.3% to3.9%; annual change, 0.05; 95% CI, 0.03 to 0.07; R2�0.56),as did mortality at 1 year after discharge (by 1.8%, from12.1% to 13.9%; annual change, 0.15; 95% CI, 0.10 to 0.20;R2�0.62; P�0.0001 for all trends).

Piecewise linear regression dividing the 22 years of thestudy into 3 time intervals confirmed these trends (OnlineTable B and Figure 2). In-hospital and 30-day mortalitydecreased in all 3 time intervals, whereas mortality at 1 yearfrom admission decreased in the first and in the last timeinterval. Mortality at 30 days and at 1 year from dischargeincreased significantly in the first 2 time intervals and showed

no significant change in the last (annual change, 0.22; 95%CI, 0.15 to 0.30; P�0.0001 for the 1986 to 1992 timeinterval; and 0.47; 95% CI, 0.37 to 0.57; P�0.001 for 2002 to2007 for 1-year postdischarge mortality).

Multivariate logistic regression adjusting for age, sex, race,AMI site, comorbidities, complications, and invasive cardiacprocedures indicated a significant decrease in in-hospital mor-tality per year (odds ratio [OR], 0.986; 95% CI, 0.983 to 0.988;P�0.0001) and a significant increase in 30-day postdischargemortality (OR, 1.028; 95% CI, 1.024 to 1.032; P�0.0001)(Table 2). Cox regression adjusting for the above confoundersalso indicated a significant increase in 1-year postdischargemortality (hazard ratio [HR], 1.008; 95% CI, 1.006 to 1.010;P�0.0001).

Effect of Length of StayLength of stay decreased by 5.8 days, from 11.5�9.5 in 1986to 5.7�6.7 in 2007. This could have displaced some deaths in

Table 1. Continued

In-Hospital30 Days From

Admission� (30-Day)(In-Hospital)

1 Year FromAdmission

30 Days FromDischarge

1 Year FromDischarge

2006 5.47 7.24 1.77 12.35 2.56 7.53

2007 4.72 6.45 1.73 10.62 2.38 6.44


�0.36(�0.39 to �0.33)

�0.33(�0.36 to �0.29)

0.03(0.02 to 0.07)

�0.38(�0.43 to �0.33)

�0.01(�0.03 to �0.01)

�0.07(�0.11 to �0.03)

R2 0.96 0.95 0.49 0.91 0.07 0.36

P value �0.0001 �0.0001 0.0003 �0.0001 0.23 0.003

Non-CVD death, %

1986 4.75 4.60 �0.16 6.29 0.29 2.08

1987 4.56 4.51 �0.05 6.24 0.44 2.34

1988 4.39 4.22 �0.16 6.04 0.42 2.32

1989 4.42 4.33 �0.08 6.53 0.51 2.71

1990 4.13 4.02 �0.12 6.40 0.50 2.96

1991 4.07 4.11 0.04 6.50 0.59 3.03

1992 3.43 3.50 0.07 5.83 0.63 2.94

1993 2.99 3.15 0.15 5.40 0.57 2.94

1994 3.43 3.46 0.02 6.11 0.60 3.26

1995 3.86 3.91 0.05 6.64 0.68 3.34

1996 3.79 4.12 0.33 6.74 0.80 3.53

1997 2.76 3.22 0.46 5.97 0.88 3.69

1998 2.88 3.32 0.43 6.62 0.91 4.29

1999 3.37 3.87 0.50 7.58 1.12 4.79

2000 3.40 4.05 0.65 7.87 1.19 5.03

2001 3.17 3.85 0.68 8.17 1.33 5.66

2002 3.24 3.94 0.70 8.29 1.28 5.62

2003 3.05 3.72 0.67 8.12 1.24 5.65

2004 2.90 3.61 0.71 8.08 1.35 5.73

2005 2.99 3.70 0.71 8.10 1.34 5.70

2006 2.67 3.37 0.70 8.00 1.26 5.87

2007 2.82 3.79 0.97 9.57 1.54 7.45


�0.08(�0.11 to �0.06)

�0.03(�0.05 to �0.01)

0.05(0.047 to 0.15)

0.14(0.10 to 0.18)

0.06(0.05 to 0.06)

0.22(0.20 to 0.25)

R2 0.73 0.26 0.93 0.71 0.95 0.94

P value �0.0001 0.016 �0.0001 �0.0001 �0.0001 �0.0001




ownloaded from


2007 from the inpatient to the outpatient setting. A markedincrease in the difference of mortality at 30 days fromdischarge minus in-hospital mortality occurred (from 1.30%in 1986 to 2.71% in 2007; average annual change, 0.09%;95% CI, 0.07 to 0.22; P�0.0001). Of patients who diedwithin 30 days of admission, 92.9% died in the hospital in1986 compared with 73.6% in 2007.

The shorter length of stay may explain part of the increase inpostdischarge mortality at 30 days but cannot account for theincrease in mortality from 30 days to 1 year after discharge from8.7% in 1986 to 10.0% in 2007 (a 1.2% increase; average annualchange, 0.10%; 95% CI, 0.06 to 0.23; Figure 2). A smallminority of patients (1.4%, n�4019) had length of stay longerthan 30 days. Of these, 1386 died within 1 year after discharge,representing 0.5% of the patients included in the study. Exclud-ing these deaths from the analysis did not alter the results (theincrease in mortality remained 1.2%).

Cardiovascular and Noncardiovascular MortalityDirectionally different trends were observed in CVD andNCVD mortality. The trends of increasing postdischargemortality described above for all-cause mortality were pri-marily for NCVD mortality (Table 1). Thirty-day postdis-charge NCVD mortality increased 5-fold, from 0.3% to 1.5%(annual change, 0.06; 95% CI, 0.05 to 0.06; R2�0.95;P�0.0001). Mortality at discharge to 1 year after dischargeincreased more than 3-fold, from 2.1% to 7.5% (annualchange, 0.22; 95% CI, 0.20 to 0.25; R2�0.94; P�0.0001). Onthe contrary, significant changes in CVD postdischarge mor-tality were not observed.

Piecewise linear regression showed an increase in 30-day and1-year postdischarge NCVD mortality in the first and secondtime intervals studied and no significant effects in the last timeinterval (online Table B). Thirty-day and 1-year postdischargeCVD mortality did not change in the first time interval andincreased in the second and decreased last time interval.

Effect of Age on Mortality TrendsThe increase in NCVD postdischarge mortality was morepronounced in the older age groups. In the youngest agegroup, postdischarge NCVD mortality at 30 days doubledfrom 0.18% to 0.36% (annual change, 0.01; 95% CI, 0.00 to0.01; R2�0.35; P�0.004; online Table C), whereas in theoldest age group, the rate increased 8-fold, from 0.39% to3.32% (annual change, 0.13; 95% CI, 0.11 to 0.15; R2�0.88;P�0.0001). Postdischarge NCVD mortality at 1 year in-creased from 0.84% to 1.64% (annual change, 0.04; 95% CI,0.03 to 0.05; R2�0.65; P�0.0001 in the youngest age group;and increased more than 4-fold from 3.61% to 16.23% in theoldest age group (annual change, 0.46; 95% CI, 0.40 to 0.52;R2�0.91; P�0.0001).

In the youngest age group, piecewise linear regressionshowed a significant increase in 30-day and 1-year postdis-charge NCVD mortality only in the second (1993 to 2001)time interval, whereas in the oldest age group an increase wasobserved in all time periods (online Table D). In the oldestgroup, 1-year postdischarge CVD mortality increased signif-icantly in all 3 time intervals (Table D of the online-only DataSupplement).

Figure 1. In hospital (lower line, circles), 30-day (middle line, squares), and 1-year (upper line, triangles) mortality of patients admittedwith a first AMI in New Jersey hospitals for the years 1986 to 2007.




ownloaded from


Effect of Type (Subendocardial VersusTransmural) of AMI on Mortality TrendsA pronounced increase in NCVD postdischarge mortality anda stable CVD postdischarge mortality were observed amongpatients with subendocardial infarction. In these patients,postdischarge NCVD mortality at 30 days increased morethan 5-fold, from 0.35% to 1.88% (annual change, 0.07; 95%CI, 0.07 to 0.08; R2�0.93; P�0.0001), and mortality at 1year after discharge more than tripled, from 2.48% to 9.42%(annual change, 0.27; 95% CI, 0.24 to 0.31; R2�0.92;P�0.0001, Table E of the online-only Data Supplement).One-year CVD postdischarge mortality remained relativelystable in these patients. On the contrary, in patients withtransmural infarction, a decrease in postdischarge CVDmortality was observed both at 30 days by 0.66% (annualchange, �0.04; 95% CI, �0.05 to �0.02; R2�0.47;P�0.0004) and at 1 year by 4.89% (�0.19; 95% CI, �0.22to �0.16; R2�0.88; P�0.0001). The effects of age andtype of infarction were interrelated. Younger individualswith transmural infarctions had the best long-term trends,especially for CVD mortality, whereas the opposite wastrue for older patients with subendocardial infarctions,especially for NCVD mortality.

Noncardiovascular Causes of DeathThe most frequent causes of NCVD death were diabetes,respiratory and renal diseases, and cancer (online Table F).As expected, the cause-specific NCVD mortality was higherin the older age groups, especially for respiratory diseases andcancer. This age-related change was more pronounced forolder patients with subendocardial infarction.

A statistically significant increase in cause-specific NCVDmortality was observed in the later years of the study: for cancer(annual change, 0.06; 95% CI, 0.04 to 0.07; R2�0.86,P�0.0001), for respiratory diseases (0.04; 95% CI, 0.02 to 0.05;R2�0.78, P�0.0003), for renal failure (0.03; 95% CI, 0.02 to0.04; R2�0.82; P�0.0001), for septicemia (0.02; 95% CI, 0.02to 0.03; R2�0.86; P�0.0001), and for Alzheimer disease (0.01;95% CI, 0.01 to 0.01; R2�0.84, P�0.0001; online Table F).

Sensitivity analysis (limited to patients admitted between1994 and 2007, n�185 483), using logistic regression adjust-ing for insurance type in addition to the variables used in theprimary analysis, yielded similar trends (online Table G).

DiscussionThis statewide study shows that the natural history of AMIhas changed with marked decrease of in-hospital mortalitybut an increasing mortality after discharge. This increase in

Figure 2. One-year all-cause mortality of patients admitted with a first AMI in New Jersey hospitals for the years 1986 to 2007 compar-ing 3 time intervals: 1986 to 1992 (dotted lines, red); 1993 to 2001 (continuous lines, blue); and 2002 to 2007 (dashed lines, green). Leftupper panel: Starting from admission. Right upper panel: Starting from discharge. Lower panel: Starting from 30 days after discharge.




ownloaded from


postdischarge mortality was observed primarily among theolder age groups and is attributable to deaths from NCVD,especially cancer, respiratory and renal disease, and septice-mia. AMI management has improved with marked loweringof in-hospital mortality.24 This decrease is probably due tomany factors including process improvement initiatives, morefrequent use of pharmacological agents such as �-blockers,and, especially for ST-elevation–AMIs, the use ofthrombolytic therapy and primary percutaneous coronaryintervention.1,3,4,13–15 Also, primary prevention efforts withwider use of pharmacological therapy may have resulted insmaller AMIs with expected lower CVD mortality andpostponement of some AMIs to an older age. Yeh et al24 andothers25 also observed a reduction in case fatality in spite oftreating older patients with more comorbidities. They sug-gested that more frequent use of guideline-based pharmaco-logical therapy may be in part responsible for the betteroutcome. These investigators did not report specifically onpostdischarge mortality.

The decrease in CVD mortality combined with older ageand higher rates of comorbidities has resulted in higherNCVD mortality after discharge. We can speculate that theshorter length of stay in recent years may have contributed to

the increase in postdischarge mortality at 30 days by displac-ing some deaths from the inpatient to the outpatient setting. Asmall portion of the increase may be due to shortening of thelength of stay, with displacement of some deaths from theinpatient to the outpatient setting. However, this cannotexplain the increase in mortality occurring between 30 daysand 1 year after discharge. Previous investigators havepresented data compatible with either no change or implyingworsening of AMI mortality after discharge. For example,Roger et al1 reported lower mortality at 30 days afterdischarge but no change in mortality at 1 year after dischargeamong 2816 AMI patients hospitalized in Minnesota from1987 to 2006. Masoudi et al8 observed a 25% adjustedrelative risk reduction for in-hospital mortality but only a13% adjusted relative risk reduction at 1 year, implyingworsening prognosis after discharge. The higher postdis-charge NCVD mortality of subendocardial AMIs, some ofwhich were diagnosed by biomarkers only and were presum-ably smaller, may counterbalance the expected lower CVDmortality.1,2,12 Limitations of the study include the lack ofdetailed clinical information may have affected the accuracyof the diagnosis of AMI and the classification of the cause ofdeath (CVD versus NCVD). We have validated the informa-

Table 2. Multivariate Models of In-Hospital Death, 30-Day Mortality After Discharge, and 1-Year Mortality After Discharge, 1986 to 2007

In-Hospital Death 30 Days After Discharge 1 Year After Discharge

Variable OR Lower CL Upper CL OR Lower CL Upper CL Hazard Ratio Lower CL Upper CL

Discharge year 0.986 0.983 0.988 1.028 1.024 1.032 1.008 1.006 1.010

Age per year 1.05 1.05 1.05 1.04 1.04 1.04 1.05 1.05 1.05

Sex

Female 1.00 1.00 1.00

Male 0.88 0.85 0.90 1.00 0.96 1.05 1.03 1.01 1.06

Ethnic group

Hispanic 1.00 1.00

Non-Hispanic white 0.97 0.90 1.04 1.18 1.05 1.32 1.11 1.04 1.17

Non-Hispanic black 1.24 1.14 1.35 0.95 0.82 1.09 1.18 1.10 1.27

AMI site

Anterior 1.00 1.00 1.00

Interior/inferolateral 0.84 0.81 0.86 0.78 0.73 0.83 0.78 0.75 0.81

Other/unspecified 1.75 1.68 1.82 0.99 0.91 1.07 1.12 1.08 1.17

Subendocardial 0.34 0.33 0.36 0.64 0.61 0.68 0.94 0.91 0.97

Renal disease 1 vs 0 1.01 0.97 1.04 1.02 0.97 1.08 1.15 1.12 1.18

Anemia 1 vs 0 0.72 0.69 0.75 0.95 0.89 1.01 1.22 1.19 1.26

COPD 1 vs 0 0.95 0.91 0.98 1.18 1.11 1.25 1.35 1.31 1.39

Hypertension 1 vs 0 0.56 0.55 0.58 0.77 0.74 0.81 0.84 0.82 0.86

Diabetes 1 vs 0 1.00 0.97 1.03 1.17 1.11 1.22 1.26 1.23 1.29

Atrial fibrillation 1 vs 0 1.08 1.05 1.12 1.28 1.22 1.35 1.31 1.27 1.34

Heart failure 1 vs 0 2.62 2.55 2.69 2.31 2.20 2.42 2.14 2.08 2.19

Cancer 1 vs 0 1.85 1.74 1.97 2.13 1.94 2.33 2.59 2.48 2.70

Cerebrovasc Dx 1 vs 0 1.78 1.70 1.87 1.52 1.41 1.65 1.51 1.45 1.57

PCI 1 vs 0 0.50 0.47 0.53 0.27 0.24 0.31 0.41 0.38 0.43

CABG 1 vs 0 0.76 0.70 0.83 0.25 0.21 0.30 0.32 0.29 0.35

CL indicates 95% confidence limit; COPD, chronic obstructive pulmonary disease; cerebrovasc Dx, cerebrovascular disease; PCI, percutaneous coronaryintervention; CABG, coronary artery bypass graft; 1, yes; and 0, no.




ownloaded from


tion in MIDAS by comparing it with the hospital record in arandomly selected sample using predefined criteria for thediagnosis of AMI and found the information to be valid forthis in �90% of the cases and for in-hospital death in 99% ofthe cases.19 Also, other investigators have found acceptablesensitivity and positive predictive value of administrativedata bases for the diagnosis of AMI.26,27 However, cause-specific NCVD mortality rates may not be precise, andchanges in coding practices may have affected the results.Coady et al28 have reported that the death certificate overes-timated CHD mortality (by approximately 20%) in the ARICcommunities and that the overestimation was consistent overtime. This would detract from the precision of the estimatespresented in the present study. However, the bias is in adirection strengthening a conclusion of the study, for exam-ple, the increased NCVD mortality after discharge. Anotherlimitation is that MIDAS does not include AMIs or deathsoccurring outside New Jersey. Deaths occurring outside thestate within 1 year of discharge accounted for approximately1.4%, or 33 of 2351 in 2007 of the total deaths, a percentageunlikely to affect the conclusions of the study. Exclusion ofnon–New Jersey residents (n�12 381, 4.3%) did not affectthe overall trends reported in this study; for example, from1986 to 2007, 30-day mortality decreased by 8.00% (18.24%to 10.24%) when all patients are included versus 7.87%(18.32% to 10.45%) for New Jersey residents only. Therespective changes at 1 year were 6.43% (26.62% to 20.19%)versus 6.36% (27.01% to 20.74%). An additional limitation,which may have affected the precision of our findings as theyrelate to some subsets but not the overall conclusions of thereport, is that the terms “subendocardial” and “transmural” donot correspond well to the arteriographic or pathologicalfindings, and the administrative codes are not perfectlyreliable.26,27 However, the new definitions and diagnosticcategories of AMI were not used by New Jersey hospitalsduring the years under consideration.9 Steinberg et al29

examined 165 691 admissions entered in the National Regis-try of Myocardial Infarction to evaluate the adoption of the2005 classification by coders. They found little change incoding, possibly reflecting a lack of awareness of thissubstantial change in classification.29 It is also possible that inthe early years of the study, some recurrent AMIs wereincluded in the analysis as first AMIs. To examine thispossible limitation, we performed sensitivity analysis exclud-ing the first 8 years (1986 to 2003) of the study, whichyielded similar trends.

The inclusion of all hospitalizations for AMI in the state(other than to Veterans Affairs hospitals) strengthens theconclusions of the study and complements information fromclinical trials and prospective registries, which usually in-clude selected patients or hospitals.30 In addition, this studyincludes a large number of patients, encompasses all AMIs inthe state without an upper age limit, and provides long-termlongitudinal follow-up for up to 22 years.

In summary, the postdischarge mortality of patients whowere hospitalized for a first AMI has increased over the last2 decades. This is due to increasing NCVD mortality,especially in the older age groups. Attention to comorbiditiesduring hospitalization or soon after discharge and preventive

and/or therapeutic strategies for diabetes, cancer, and respi-ratory and renal disease should become a routine part ofpost-AMI patient care and may result in better long-termoutcomes. We cannot make firm inferences for the causesof the observed trends, based on the administrative datapresented in the present study. Information from registriesor controlled trials would be useful in examining thereasons for the increased NCVD mortality and in evaluat-ing whether diagnostic, preventive, or therapeutic inter-ventions for NCVD may further improve long-term out-comes of AMI patients.

AcknowledgmentsThe authors thank Jeanne Dobrzynski, BA, for technical andeditorial assistance.

Sources of FundingThis study was supported by the Robert Wood Johnson Foundationand the Schering-Plough Foundation.

DisclosuresNone.

References1. Roger VL, Weston SA, Gerber Y, Killian JM, Dunlay SM, Jaffe AS, Bell

MR, Kors J, Yawn BP, Jacobsen SJ. Trends in incidence, severity, andoutcome of hospitalized myocardial infarction. Circulation. 2010;121:863–869.

2. Sanfilippo FM, Hobbs MST, Knuiman MW, Hung J. Impact of newbiomarkers of myocardial damage on trends in myocardial infarctionhospital admission rates from population-based administrative data. Am JEpidemiol. 2008;168:225–233.

3. Movahed M-R, John J, Hashemzadeh M, Jamal MM, Hashemzadeh M.Trends in the age adjusted mortality from acute ST segment elevationmyocardial infarction I the United States (1988–2004) based on race,gender, infarct location and comorbidities. Am J Cardiol. 2009;104:1030–1034.

4. Furman AMI, Dauerman HL, Goldberg RJ, Yarzebski J, Lessard D, GoreJM. Twenty-two year (1975 to 1997) trends in the incidence, in-hospitaland long-term case fatality rates from initial q-wave and non-q-wavemyocardial infarction: a multi-hospital, community-wide perspective.J Am Coll Cardiol. 2001;37:1571–1580.

5. Krumholz HM, Wang Y, Chen J, Drye EE, Spertus JA, Ross JS, CurtisJP, Nallamothu BK, Lichtman JH, Havranek EP, Masoudi FA, RadfordMJ, Han LF, Rapp MT, Straube BM, Normand SL. Reduction in acutemyocardial infarction mortality in the United States: risk-standardizedmortality rates from 1995–2006. JAMA. 2009;302:767–773.

6. Setoguchi S, Glynn RJ, Avorn J, Mittleman MA, Levin R, WindelmayerWC. Improvements in long-term mortality after myocardial infarction andincreased use of cardiovascular drugs after discharge: a 10-year trendanalysis. J Am Coll Cardiol. 2008;51:1247–1254.

7. Briffa T, Hickling S, Knuiman M, Hobbs M, Hung J, Sanfilippo FM,Jamrozik K, Thompson PL. Long term survival after evidence basedtreatment of acute myocardial infarction and revascularisation: follow-upof population based Perth MONICA cohort, 1984–2005. BMJ. 2009;338:b36.

8. Masoudi FA, Foody JM, Havranek EP, Wang Y, Radford MJ, AllmanRM, Gold J, Wiblin RT, Krumholz HM. Trends in acute myocardialinfarction in 4 US states between 1992 and 2001: clinical characteristics,quality of care, and outcomes. Circulation. 2006;114:2806–2814.

9. Alpert JS, Thygesen K, Antman E, Bassand JP. Myocardial infarctionredefined: a consensus document of the Joint European Society of Car-diology/American College of Cardiology Committee from the redefi-nition of myocardial infarction. J Am Coll Cardiol. 2000;36:959–969.

10. Cannon CP. Update to International Classification of Diseases, 9th

Revision: distinguishes STEMI from NSTEMI. Crit Pathways Cardiol.2005;4:185–186.

11. Parikh NI, Gona P, Larson MG, Fox CS, Benjamin EJ, Murabito JM,O’Donnell CJ, Vasan RS, Levy D. Long-term trends in myocardialinfarction incidence and case fatality in the National Heart, Lung, and




ownloaded from


Blood Institute’s Framingham Heart Study. Circulation. 2009;119:1203–1210.

12. Chan MY, Sun JL, Newby LK, Shaw LK, Lin M, Peterson ED, CaliffRM, Kong DF, Roe MT. Long-term mortality of patients undergoingcardiac catheterization for ST-elevation and non-ST-elevation myocardialinfarction. Circulation. 2009;119:3110–3117.

13. Heidenreich PA, Lewis WR, LaBresh KA, Schwamm LH, Fonarow GC.Hospital performance recognition with the get with the guidelinesprogram and mortality for acute myocardial infarction and heart failure.Am Heart J. 2009;158:546–553.

14. Bradley EH, Nallamothu BK, Herrin J, Ting HH, Stern AF, NembhardIM, Yuan CT, Green JC, Kline-Rogers E, Wang Y, Curtis JP, WebsterTR, Masoudi FA, Fonarow GC, Brush JE Jr, Krumholz HM. Nationalefforts to improve door-to-balloon time: results from the Door-to-BalloonAlliance. J Am Coll Cardiol. 2009;54:2423–2429.

15. Silvet H, Spencer F, Yarzebski J, Lessard D, Gore JM, GoldbergRJ. Communitywide trends in the use and outcomes associated withbeta-blockers in patients with acute myocardial infarction: the WorcesterHeart Attack Study. Arch Intern Med. 2003;163:2175–2183.

16. Ford ES, Ajani UA, Croft JB, Critchley JA, Labarthe DR, Kottke TE,Giles WH, Capewell S. Explaining the decrease in US deaths fromcoronary disease, 1980–2000. N Engl J Med. 2007;356:2388–2398.

17. Eagle KA, Nallamothu BK, Mehta RH, Granger CB, Steg PG, Van deWerf F, Lopez-Sendon J, Goodman SG, Quill A, Fox KA; GlobalRegistry of Acute Coronary Events (GRACE) Investigators. Trends inacute reperfusion therapy for ST-segment elevation myocardial infarctionfrom 1999 to 2006: we are getting better but we have got a long way togo. Eur Heart J. 2008;29:609–617.

18. Yarzebski J, Granillo E, Spencer FA, Lessard D, Gurwitz JH, Gore JM,Goldberg RJ. Changing trends (1986–2003) in the use of lipid loweringmedication in patients hospitalized with acute myocardial infarction:a community-based perspective. Int J Cardiol. 2009;132:66–74.

19. Kostis JB, Wilson AC, O’Dowd K, Gregory P, Chelton S, Cosgrove NM,Chirala A, Cui T. Sex differences in the management and long-termoutcome of acute myocardial infarction: a statewide study. MIDAS StudyGroup: Myocardial Infarction Data Acquisition System. Circulation.1994;90:1715–1730.

20. Kostis WJ, Demissie K, Marcella SW, Shao Y-H, Wilson AC, MoreyraAE, for the Myocardial Infarction Data Acquisition System (MIDAS 10)Study Group. Weekend versus weekday admission and mortality inpatients with myocardial infarction. N Engl J Med. 2007;356:1099–1109.

21. Campbell KM. Rule Your Data with The Link King (a SAS/AF appli-cation for record linkage and unduplication), in SUGI 30 Proceedings.2005: Philadelphia, Pa.

22. Campbell KM. Record linkage software in the public domain: a com-parison of Link Plus, The Link King, and a ‘basic’ deterministicalgorithm. Health Informatics J. 2008;14:5–15.

23. Elixhauser A, Steiner C, Palmer L. Clinical Classifications Software(CCS), 2008. US Agency for Healthcare Research and Quality. Availableat: http://www.hcup-us.ahrq.gov/toolssoftware/ccs/ccs.jsp.

24. Yeh RW, Sidney S, Chandra M, Sorel M, Selby JV, Go AS. Populationtrends in the incidence and outcomes of acute myocardial infarction.N Engl J Med. 2010;362:2155–2165.

25. van Domburg RT, Sonnenschein K, Nieuwlaat R, Kamp O, Storm CJ,Bax JJ, Simoons ML. Sustained benefit 20 years after reperfusion therapyin acute myocardial infarction. J Am Coll Cardiol. 2005;46:15–20.

26. Rosamond WD, Chambless LE, Sorlie PD, Bell EM, Weitzman S, SmithJC, Folsom AR. Trends in the sensitivity, positive predictive value,false-positive rate, and comparability ratio of hospital discharge diagnosiscodes for acute myocardial infarction in four US communities,1987–2000. Am J Epidemiol. 2004;160:1137–1146.

27. Petersen LA, Wright S, Normand SL, Daley J. Positive predictive valueof the diagnosis of acute myocardial infarction in an administrativedatabase. J Gen Intern Med. 1999;14:555–558.

28. Coady SA, Sorlie PD, Cooper LS, Folsom AR, Rosamond WD, Conwill DE.Validation of death certificate diagnosis for coronary heart disease: theAtherosclerosis Risk in Communities (ARIC) Study. J Clin Epidemiol. 2001;54:40–50.

29. Steinberg BA, French WJ, Peterson E, Frederick PD, Cannon CP, for theNational Registry of Myocardial Infarction Investigators. Is coding formyocardial infarction more accurate now that coding descriptions havebeen clarified to distinguish ST-elevation myocardial infarction fromnon-ST elevation myocardial infarction? Am J Cardiol. 2008;102:513–517.

30. Krumholz HM. Registries and selection bias: the need for accountability.Circ Cardiovasc Qual Outcomes. 2009;2:517–518.




ownloaded from


for the Myocardial Infarction Data Acquisition System (MIDAS14) Study GroupWilliam J. Kostis, Yingzi Deng, John S. Pantazopoulos, Abel E. Moreyra, John B. Kostis andTrends in Mortality of Acute Myocardial Infarction After Discharge From the Hospital

Print ISSN: 1941-7705. Online ISSN: 1941-7713 Copyright © 2010 American Heart Association, Inc. All rights reserved.

Greenville Avenue, Dallas, TX 75231is published by the American Heart Association, 7272Circulation: Cardiovascular Quality and Outcomes

doi: 10.1161/CIRCOUTCOMES.110.9578032010;3:581-589; originally published online October 5, 2010;Circ Cardiovasc Qual Outcomes.

http://circoutcomes.ahajournals.org/content/3/6/581World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circoutcomes.ahajournals.org/content/suppl/2010/10/12/CIRCOUTCOMES.110.957803.DC1Data Supplement (unedited) at:

http://circoutcomes.ahajournals.org//subscriptions/

at: is onlineCirculation: Cardiovascular Quality and Outcomes Information about subscribing to Subscriptions:

http://www.lww.com/reprints

Information about reprints can be found online at: Reprints:

document. Question and AnswerPermissions and Rightspage under Services. Further information about this process is available in the

which permission is being requested is located, click Request Permissions in the middle column of the WebCopyright Clearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of theCirculation: Cardiovascular Quality and Outcomesin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:



ownloaded from

http://circoutcomes.ahajournals.org/content/3/6/581

http://circoutcomes.ahajournals.org/content/suppl/2010/10/12/CIRCOUTCOMES.110.957803.DC1

http://www.ahajournals.org/site/rights/

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circoutcomes.ahajournals.org//subscriptions/


CIRCVOQ/2010/957803/R2

SUPPLEMENTAL MATERIALS

Table A. Patient Characteristics Over Time

Year 1986-1987 1988-1989 1990-1991 1992-1993 1994-1995 1996-1997 1998-1999 2000-2001 2002-2003 2004-2005 2006-2007 P for Trend

Number of Patients 26498 24238 23802 25376 26046 26286 27147 28544 28521 25474 23465

Age (Mean/Std) 66.8(12.7) 67.2(12.7) 67.2(13.0) 67.3(13.0) 67.6(13.3) 68.0(13.5) 69.0(13.8) 69.4(14.0) 69.1(14.4) 69.2(14.5) 68.7(14.7) <.0001

Gender (% female) 40.0% 40.4% 40.5% 40.8% 40.2% 41.4% 41.9% 42.7% 42.8% 42.9% 42.4% <.0001

Race (% white) 87.8% 86.0% 84.6% 85.8% 83.5% 80.7% 79.8% 77.2% 72.0% 73.8% 74.6% <.0001

Insurance Type <.0001---Commercial NA NA NA NA 25.2% 20.9% 19.0% 17.4% 17.5% 18.5% 20.3%---Governmental (Medicare/Medicaid)

NA NA NA NA 61.1% 57.6% 54.7% 56.1% 57.0% 56.6% 54.2%

---Self-Pay NA NA NA NA 4.7% 4.6% 4.3% 4.3% 5.3% 5.5% 5.2%

---HMO NA NA NA NA 9.0% 16.9% 22.1% 22.2% 20.2% 19.4% 20.3%

MI site <.0001

---Anterior 31.6% 31.7% 29.7% 27.4% 25.9% 24.7% 21.2% 17.5% 15.2% 12.9% 13.0%

---Inferior/Lateral/Post. 37.6% 37.6% 37.0% 36.3% 35.5% 33.2% 28.8% 24.0% 21.5% 18.1% 17.6%

---Other/Unspecified 14.3% 11.8% 9.9% 8.2% 7.4% 7.0% 7.4% 7.6% 8.0% 7.6% 6.6%

---Subendo 16.6% 18.9% 23.5% 28.1% 31.2% 35.2% 42.7% 50.9% 55.3% 61.4% 62.8%

---Q wave (non-subendo) 69.2% 69.3% 66.8% 63.8% 61.5% 57.9% 50.0% 41.6% 36.7% 31.0% 30.6%

Comorbidities

---% with diabetes 22.7% 23.8% 24.5% 25.1% 25.3% 26.2% 27.2% 27.3% 27.4% 27.2% 27.1% <.0001

---% with COPD 10.2% 10.0% 9.9% 10.8% 11.3% 11.5% 13.3% 13.5% 14.2% 14.8% 14.3% <.0001

---% with anemia 6.9% 7.6% 10.1% 10.7% 10.2% 10.7% 12.1% 12.9% 13.3% 12.7% 12.2% <.0001

---% with hypertension 34.6% 36.1% 38.5% 41.2% 44.3% 48.2% 50.5% 53.6% 56.4% 58.6% 59.7% <.0001

---% with liver disease 0.3% 0.3% 0.4% 0.4% 0.4% 0.3% 0.3% 0.5% 0.4% 0.5% 0.6% <.0001

---% with chronic renal disease 12.5% 14.3% 16.7% 18.6% 20.5% 21.5% 22.8% 24.1% 23.5% 24.8% 26.9% <.0001

---% with cerebrovascular disease 4.4% 5.0% 4.9% 5.0% 4.9% 5.1% 5.5% 5.5% 5.0% 4.6% 4.3% 0.80

---% with cancer 1.9% 1.9% 2.0% 2.3% 2.5% 2.6% 3.0% 3.2% 3.5% 3.3% 3.4% <.0001

---% with stroke 1.0% 1.1% 1.1% 1.2% 2.1% 2.3% 2.4% 2.4% 2.3% 2.1% 2.0% <.0001

---% with heart failure 28.5% 29.6% 29.9% 30.3% 29.9% 29.6% 30.4% 30.9% 30.5% 31.7% 28.3% <.0001

Length of stay(Mean/Std) 11.3(10.1) 10.7(11.3) 10.2(10.7) 9.8(10.7) 8.3(9.9) 7.0(7.8) 6.7(7.7) 6.5(7.5) 6.3(7.8) 6.0(6.9) 5.7(6.9) <.0001

Procedures received during index admission

---% Catheterization 9.4% 14.0% 19.8% 25.1% 26.7% 29.6% 34.7% 36.0% 42.6% 48.0% 52.9% <.0001

---% PCI 1.3% 2.0% 4.1% 6.2% 8.4% 10.7% 13.5% 16.0% 21.1% 27.9% 34.1% <.0001

---% CABG 0.7% 0.9% 2.3% 3.8% 4.2% 4.7% 4.9% 5.7% 6.0% 5.5% 5.6% <.0001

Discharge Destination <.0001

---Home 69.2% 61.2% 58.0% 56.4% 53.0% 49.3% 48.5% 44.8% 44.7% 46.2% 50.0%

---Nursing Home 1.8% 2.3% 2.9% 3.2% 3.8% 4.6% 6.8% 8.6% 9.0% 12.1% 12.7%

---Other 12.2% 20.3% 24.8% 28.1% 31.3% 35.1% 34.5% 36.7% 37.4% 33.5% 29.5%

HMO: Health Maintenance Organization COPD: Chronic obstructive pulmonary disease

PCI:Percutaneous coronary intervention CABG: Coronary artery bypass graft

Table A

- )o - )n )g

- )o - )n - )g )

)

)o )ng

Table B. Piecewise Linear regression1986-1992 1993-2001 2002-2007

All-cause death(%)In-hospital -0.33 (-0.37~-0.30) <.0001 0.30 (-0.40~-0.19 0.0009 -0.30 (-0.39~-0.20 0.00330 days from admissi -0.24 (-0.29~-0.20) <.0001 0.12 (-0.23~-0.01 0.07 -0.32 (-0.41~-0.22 0.0031-year from admissio -0.08 (-0.16~0.00) 0.08 0.18 (0.05~0.31) 0.03 -0.36 (-0.42~-0.30 0.000330 days from dischar 0.05 (0.02~0.08) 0.006 0.15 (0.11~0.19) 0.0002 -0.07 (-0.15~0.01) 0.21-year from discharge 0.22 (0.15~0.30) <.0001 0.47 (0.37~0.57) <.0001 -0.11 (-0.25~0.03) 0.2

CVD death(%)In-hospital -0.29 (-0.31~-0.27) <.0001 0.27 (-0.32~-0.22 <.0001 -0.21 (-0.31~-0.10 0.0230 days from admissi -0.26 (-0.29~-0.23) <.0001 0.18 (-0.25~-0.11 0.001 -0.27 (-0.38~-0.15 0.011-year from admissio -0.30 (-0.36~-0.24) <.0001 0.12 (-0.20~-0.05 0.01 -0.53 (-0.78~-0.28 0.0130 days from dischar -0.01 (-0.04~0.01) 0.3 0.05 (0.01~0.09) 0.03 -0.10 (-0.18~-0.03 0.061-year from discharge -0.04 (-0.11~0.02) 0.2 0.14 (0.09~0.19) 0.001 -0.39 (-0.61~-0.17 0.03

Non-CVD death(%)In-hospital -0.04 (-0.07~-0.01) 0.02 -0.02 (-0.12~0.08) 0.7 -0.09 (-0.14~-0.04 0.0330 days from admissi 0.01 (-0.02~0.05) 0.4 0.06 (-0.03~0.15 0.2 -0.05 (-0.14~0.04) 0.31-year from admissio 0.22 (0.17~0.27) <.0001 0.30 (0.19~0.41) 0.001 0.17 (-0.09~0.44) 0.330 days from dischar 0.06 (0.05~0.07) <.0001 0.10 (0.09~0.11) <.0001 0.04 (-0.01~0.08) 0.21-year from discharge 0.27 (0.23~0.31) <.0001 0.33 (0.27~0.39) <.0001 0.28 (0.03~0.54) 0.09

Table B

Table C. All-Cause, CVD and NCVD Death Rates at Different Time Intervals by Age Group and Year (1986-2007) (%)

YearAnnual change,

% (95% CI) R² P 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995AGE GROUP 35-60; N 4171 3987 3618 3485 3453 3640 3688 3852 3847 3911--CVD death(%)---Deaths at discharge -0.12 (-0.15~-0.10) 0.82 <0.0001 4.22 3.74 3.90 3.30 3.13 2.34 2.39 2.36 2.52 2.66

---Deaths at 30 days(from admission) -0.15 (-0.18~-0.12) 0.87 <0.0001 5.18 4.56 4.81 4.71 3.91 3.19 3.17 3.17 3.22 3.40---Deaths at 1 year(from admission) -0.22 (-0.26~-0.19) 0.90 <0.0001 8.03 6.80 6.88 6.69 5.56 4.97 5.21 5.06 4.99 4.78

---Deaths at 30 days(from discharge) -0.04 (-0.06~-0.03) 0.73 <0.0001 1.33 1.22 1.29 1.78 1.12 1.08 1.07 1.07 1.05 0.91---Deaths at 1 year(from discharge) -0.11 (-0.13~-0.09) 0.85 <0.0001 4.10 3.29 3.21 3.63 2.61 2.77 3.01 2.87 2.55 2.27

--Non-CVD death(%)---Deaths at discharge -0.04 (-0.06~-0.02) 0.56 <0.0001 1.73 1.86 1.52 1.75 1.45 1.29 1.17 0.86 0.65 1.46

---Deaths at 30 days(from admission) -0.03 (-0.04~-0.01) 0.34 0.005 1.77 1.93 1.35 1.81 1.25 1.35 1.19 0.99 0.73 1.48---Deaths at 1 year(from admission) 0.00 (-0.02~0.02) 0.01 0.69 2.47 2.53 2.18 2.78 2.26 2.31 1.76 1.61 1.74 2.63

---Deaths at 30 days(from discharge) 0.01(0.00~0.01) 0.35 0.004 0.18 0.24 0.09 0.21 0.15 0.23 0.25 0.19 0.24 0.24---Deaths at 1 year(from discharge) 0.04(0.03~0.05) 0.65 <0.0001 0.84 0.85 0.85 1.18 1.03 1.14 0.67 0.80 1.18 1.28

AGE GROUP 61-70; N 3838 3592 3521 3369 3270 3176 3300 3378 3311 3207--CVD death(%)---Deaths at discharge -0.36(-0.41~-0.31) 0.92 <0.0001 10.55 10.02 10.22 8.88 7.68 6.93 7.12 6.57 5.80 5.27

---Deaths at 30 days(from admission) -0.39(-0.44~-0.35) 0.93 <0.0001 12.43 11.83 12.01 10.21 9.66 8.22 8.85 8.02 7.40 6.74---Deaths at 1 year(from admission) -0.53(-0.59~-0.47) 0.94 <0.0001 17.95 16.73 17.32 15.14 13.73 12.66 12.94 12.40 11.27 10.17

---Deaths at 30 days(from discharge) -0.08(-0.10~-0.07) 0.80 <0.0001 3.03 2.94 3.47 2.93 2.88 2.34 2.82 2.47 2.55 1.94---Deaths at 1 year(from discharge) -0.22(-0.25~-0.20) 0.92 <0.0001 8.69 7.99 8.42 7.24 6.97 6.51 6.55 6.50 6.09 5.38

--Non-CVD death(%)---Deaths at discharge -0.08(-0.10~-0.05) 0.65 <0.0001 3.44 3.93 3.12 3.03 3.21 3.09 2.67 2.34 2.96 3.12

---Deaths at 30 days(from admission) -0.05(-0.07~-0.03) 0.50 0.0002 3.39 3.70 3.12 3.12 3.18 3.21 2.82 2.46 3.05 3.18---Deaths at 1 year(from admission) 0.05(0.01~0.08) 0.23 0.02 5.13 5.29 4.69 4.90 4.98 5.01 4.94 4.47 4.98 5.43

---Deaths at 30 days(from discharge) 0.03(0.02~0.03) 0.64 <0.0001 0.33 0.26 0.33 0.57 0.45 0.42 0.50 0.52 0.53 0.48---Deaths at 1 year(from discharge) 0.12(0.09~0.14) 0.76 <0.0001 2.21 1.91 2.00 2.26 2.23 2.20 2.62 2.57 2.28 2.69CVD: Cardiovascular Disease; NCVD: Non-cardiovascular Disease

Table C Page 1

Table C. All-Cause, CVD and NCVD Death Rates at Different Time Intervals by Age Group and Year (1986-2007) (%)(Cont'd)

Year 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007AGE GROUP 35-60; N 4014 3870 3905 3823 3992 4057 4373 4271 3976 3786 3849 3739--CVD death(%)---Deaths at discharge 2.22 2.17 1.66 2.09 2.28 1.40 1.44 1.64 1.16 1.37 1.69 1.58

---Deaths at 30 days(from admission) 2.82 3.07 2.15 2.59 2.68 2.02 2.04 1.99 1.76 2.03 2.00 1.98---Deaths at 1 year(from admission) 3.94 4.39 3.35 3.71 3.73 3.43 3.09 3.14 2.72 2.98 2.94 2.86

---Deaths at 30 days(from discharge) 0.72 0.99 0.58 0.51 0.60 0.78 0.66 0.53 0.69 0.76 0.48 0.55---Deaths at 1 year(from discharge) 1.78 2.35 1.74 1.70 1.50 2.12 1.71 1.54 1.64 1.68 1.28 1.31

--Non-CVD death(%)---Deaths at discharge 1.44 1.01 0.95 1.18 0.90 0.99 0.91 1.05 0.96 1.06 0.78 0.78



AGE GROUP 61-70; N 3183 2905 2802 2899 2865 2800 2860 2665 2535 2395 2304 2249--CVD death(%)---Deaths at discharge 5.00 5.78 4.71 4.48 4.19 4.75 3.88 3.45 3.31 3.13 2.86 2.71





---Deaths at 30 days(from discharge) 0.65 0.67 0.34 0.89 0.48 0.80 0.82 0.56 0.75 0.75 0.96 0.89---Deaths at 1 year(from discharge) 2.58 2.57 3.07 3.36 2.64 3.90 4.73 3.74 3.17 3.74 3.92 4.91CVD: Cardiovascular Disease; NCVD: Non-cardiovascular Disease

Table C Page 2


YearAnnual change,

% (95% CI) R² P 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995AGE GROUP 71-80; N 3556 3447 3270 3203 3198 3238 3416 3651 3551 3656--CVD death(%)---Deaths at discharge -0.61(-0.67~-0.55) 0.95 <0.0001 17.10 17.26 17.31 15.14 14.51 13.93 12.53 11.04 10.64 10.67


---Deaths at 30 days(from discharge) -0.07(-0.09~-0.04) 0.55 0.0001 4.23 3.99 4.51 4.38 3.78 3.75 3.91 3.74 3.94 4.48---Deaths at 1 year(from discharge) -0.26(-0.31~-0.20) 0.81 <0.0001 14.15 13.45 14.85 13.65 11.02 11.02 11.44 11.00 11.03 11.11

--Non-CVD death(%)---Deaths at discharge -0.14(-0.17~-0.10) 0.74 <0.0001 6.41 5.74 6.09 5.71 5.19 5.37 4.33 4.03 4.31 4.38

---Deaths at 30 days(from admission) -0.08(-0.11~-0.04) 0.47 0.0004 6.16 5.77 5.87 5.59 5.07 5.40 4.33 4.33 4.51 4.43---Deaths at 1 year(from admission) 0.13(0.08~0.18) 0.54 <0.0001 8.52 8.21 8.47 8.84 8.79 8.80 7.73 7.59 8.11 7.93

---Deaths at 30 days(from discharge) 0.06(0.05~0.07) 0.92 <0.0001 0.37 0.68 0.68 0.67 0.70 0.96 0.77 0.87 0.89 0.87---Deaths at 1 year(from discharge) 0.26(0.22~0.29) 0.91 <0.0001 2.98 3.58 3.51 4.18 4.87 4.48 4.30 4.35 4.74 4.41

AGE GROUP 81+; N 1941 1966 1923 1849 1884 1943 1942 2149 2227 2336--CVD death(%)---Deaths at discharge -0.72(-0.79~-0.65) 0.95 <0.0001 23.49 22.58 24.54 22.39 22.72 23.01 20.44 20.38 18.59 19.22


---Deaths at 30 days(from discharge) 0.09(0.04~0.13) 0.37 0.003 5.73 5.44 5.81 5.26 5.04 4.37 5.21 4.62 6.32 5.71---Deaths at 1 year(from discharge) -0.11(-0.20~-0.01) 0.20 0.04 22.59 21.97 21.74 21.20 21.26 19.74 19.06 18.80 19.88 20.85--Non-CVD death(%)---Deaths at discharge -0.23(-0.28~-0.18) 0.82 <0.0001 10.82 9.10 9.20 9.73 8.86 8.70 7.42 6.10 7.54 8.09

---Deaths at 30 days(from admission) -0.11(-0.16~-0.06) 0.46 0.0005 10.20 9.00 8.84 9.14 8.76 8.59 7.57 6.10 7.09 8.13---Deaths at 1 year(from admission) 0.22(0.14~0.30) 0.58 <0.0001 12.73 12.05 11.65 12.55 12.42 12.97 11.74 9.91 12.12 12.97

---Deaths at 30 days(from discharge) 0.13(0.11~0.15) 0.88 <0.0001 0.39 0.97 1.02 0.80 1.09 1.21 1.57 0.95 1.03 1.65---Deaths at 1 year(from discharge) 0.46(0.40~0.52) 0.91 <0.0001 3.61 5.06 4.71 4.86 5.74 6.93 6.64 5.95 7.05 7.07CVD: Cardiovascular Disease; NCVD: Non-cardiovascular Disease

Table C Page 3


Year 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007AGE GROUP 71-80; N 3646 3583 3742 3862 3930 3904 3664 3441 3197 2860 2726 2545--CVD death(%)---Deaths at discharge 10.45 9.07 8.34 7.33 7.89 7.86 6.58 5.84 6.41 6.01 5.76 4.68






AGE GROUP 81+; N 2502 2583 2943 3171 3504 3492 3626 3621 3383 3342 3052 3001--CVD death(%)---Deaths at discharge 16.03 17.38 15.60 15.48 12.64 13.23 12.82 11.54 11.65 11.13 11.96 10.16


---Deaths at 30 days(from discharge) 6.50 5.91 6.92 6.90 7.37 7.42 7.38 7.47 6.33 6.66 6.07 5.69---Deaths at 1 year(from discharge) 21.84 19.95 21.66 21.42 21.42 21.93 20.53 22.14 20.09 19.54 18.39 16.00--Non-CVD death(%)---Deaths at discharge 7.67 5.96 5.81 6.34 6.76 6.07 6.43 5.63 5.68 5.27 4.85 5.46


---Deaths at 30 days(from discharge) 1.83 1.87 2.38 2.46 2.87 3.09 2.66 3.00 3.04 3.15 2.36 3.32---Deaths at 1 year(from discharge) 7.02 7.58 8.26 9.44 10.30 11.28 10.38 11.20 11.37 11.70 12.05 16.23CVD: Cardiovascular Disease; NCVD: Non-cardiovascular Disease

Table C Page 4

Table D. Piecewise Linear Regression for All-Cause Death, CVD Death and NCVD Death at Three Time Intervals 1986-1992 1993-2001 2002-2007

Annual change % (95%CI) p Annual change % (95%CI) p Annual change % (95%CI) pAGE GROUP 35-60 --CVD death(%)---Deaths at discharge -0.32 (-0.41 ~ -0.24) 0.0006 -0.10 (-0.18 ~ -0.03) 0.03 0.03 (-0.07 ~ 0.13) 0.6

---Deaths at 30 days(from admission) -0.35 (-0.47 ~ -0.22) 0.003 -0.14 (-0.22 ~ -0.07) 0.008 0.00 (-0.05 ~ 0.05) 1.0---Deaths at 1 year(from admission) -0.48 (-0.63 ~ -0.33) 0.001 -0.22 (-0.30 ~ -0.14) 0.001 -0.04 (-0.11 ~ 0.03) 0.3

---Deaths at 30 days(from discharge) -0.04 (-0.14 ~ 0.05) 0.4 -0.06 (-0.10 ~ -0.02) 0.02 -0.02 (-0.07 ~ 0.04) 0.6---Deaths at 1 year(from discharge) -0.18 (-0.32 ~ -0.04) 0.05 -0.12 (-0.20 ~ -0.04) 0.02 -0.08 (-0.14 ~ -0.02) 0.07

--Non-CVD death(%)---Deaths at discharge -0.10 (-0.15 ~ -0.05) 0.01 0.00 (-0.07 ~ 0.08) 0.9 -0.04 (-0.09 ~ 0.01) 0.2

---Deaths at 30 days(from admission) -0.11 (-0.19 ~ -0.03) 0.04 0.01 (-0.07 ~ 0.09) 0.8 -0.01 (-0.04 ~ 0.01) 0.3---Deaths at 1 year(from admission) -0.09 (-0.19 ~ 0.01) 0.2 0.08 (0.00 ~ 0.16) 0.10 0.01 (-0.05 ~ 0.07) 0.8

---Deaths at 30 days(from discharge) 0.01 (-0.01 ~ 0.03) 0.4 0.02 (0.00 ~ 0.03) 0.05 0.02 (-0.04 ~ 0.08) 0.6---Deaths at 1 year(from discharge) 0.01 (-0.06 ~ 0.08) 0.8 0.07 (0.04 ~ 0.10) 0.00 0.05 (-0.06 ~ 0.17) 0.4

AGE GROUP 61-70 --CVD death(%)---Deaths at discharge -0.68 (-0.86 ~ -0.49) 0.001 -0.23 (-0.34 ~ -0.12) 0.0045 -0.22 (-0.26 ~ -0.19) 0.0003

---Deaths at 30 days(from admission) -0.73 (-0.94 ~ -0.51) 0.001 -0.27 (-0.39 ~ -0.15) 0.003 -0.22 (-0.32 ~ -0.11) 0.02---Deaths at 1 year(from admission) -0.96 (-1.22 ~ -0.69) 0.001 -0.37 (-0.54 ~ -0.20) 0.004 -0.37 (-0.55 ~ -0.19) 0.02

---Deaths at 30 days(from discharge) -0.09 (-0.20 ~ 0.03) 0.2 -0.07 (-0.14 ~ -0.01) 0.07 -0.01 (-0.17 ~ 0.15) 0.9---Deaths at 1 year(from discharge) -0.39 (-0.51 ~ -0.27) 0.002 -0.17 (-0.27 ~ -0.06) 0.02 -0.17 (-0.36 ~ 0.03) 0.2

--Non-CVD death(%)---Deaths at discharge -0.14 (-0.24 ~ -0.04) 0.04 -0.13 (-0.27 ~ 0.00) 0.09 -0.04 (-0.10 ~ 0.02) 0.3

---Deaths at 30 days(from admission) -0.09 (-0.17 ~ -0.02) 0.05 -0.09 (-0.21 ~ 0.03) 0.2 0.03 (-0.05 ~ 0.11) 0.5---Deaths at 1 year(from admission) -0.03 (-0.10 ~ 0.04) 0.4 0.02 (-0.14 ~ 0.17) 0.8 0.03 (-0.33 ~ 0.38) 0.9

---Deaths at 30 days(from discharge) 0.03 (0.00 ~ 0.07) 0.10 0.03 (-0.02 ~ 0.07) 0.3 0.04 (-0.01 ~ 0.10) 0.2---Deaths at 1 year(from discharge) 0.07 (0.01 ~ 0.14) 0.08 0.14 (0.05 ~ 0.23) 0.02 0.06 (-0.28 ~ 0.40) 0.8CVD: Cardiovascular Disease; NCVD: Non-cardiovascular Disease

Table D Page 1

Table D. Piecewise Linear Regression for All-Cause Death, CVD Death and NCVD Death at Three Time Intervals (Cont'd) 1986-1992 1993-2001 2002-2007

Annual change % (95%CI) p Annual change % (95%CI p Annual change % (95% pAGE GROUP 71-80 --CVD death(%)---Deaths at discharge -0.83 (-1.07 ~ -0.59) 0.001 -0.50 (-0.64 ~ -0.36) 0.0002 -0.29 (-0.50 ~ -0.08) 0.05


---Deaths at 30 days(from discharge) -0.08 (-0.18 ~ 0.02) 0.2 -0.07 (-0.17 ~ 0.03) 0.2 -0.06 (-0.27 ~ 0.15) 0.6---Deaths at 1 year(from discharge) -0.60 (-0.98 ~ -0.22) 0.03 -0.10 (-0.21 ~ 0.01) 0.1 -0.41 (-0.77 ~ -0.05) 0.09

--Non-CVD death(%)---Deaths at discharge -0.28 (-0.40 ~ -0.16) 0.006 -0.06 (-0.18 ~ 0.06) 0.4 -0.03 (-0.16 ~ 0.11) 0.7

---Deaths at 30 days(from admission) -0.25 (-0.36 ~ -0.14) 0.007 -0.01 (-0.11 ~ 0.10) 0.9 0.03 (-0.11 ~ 0.16) 0.7---Deaths at 1 year(from admission) -0.03 (-0.19 ~ 0.13) 0.7 0.28 (0.14 ~ 0.43) 0.006 0.20 (-0.07 ~ 0.47) 0.2

---Deaths at 30 days(from discharge) 0.06 (0.02 ~ 0.11) 0.03 0.07 (0.05 ~ 0.10) 0.0002 0.07 (-0.02 ~ 0.16) 0.2---Deaths at 1 year(from discharge) 0.25 (0.11 ~ 0.40) 0.02 0.35 (0.28 ~ 0.41) <.0001 0.24 (-0.10 ~ 0.58) 0.2

AGE GROUP 81+ --CVD death(%)---Deaths at discharge -0.36 (-0.76 ~ 0.03) 0.1 -0.91 (-1.14 ~ -0.67) 0.0001 -0.36 (-0.66 ~ -0.06) 0.08


---Deaths at 30 days(from discharge) -0.16 (-0.30 ~ -0.02) 0.07 0.29 (0.17 ~ 0.40) 0.002 -0.35 (-0.50 ~ -0.21) 0.009---Deaths at 1 year(from discharge) -0.55 (-0.70 ~ -0.40) 0.0008 0.30 (0.11 ~ 0.49) 0.02 -0.98 (-1.50 ~ -0.46) 0.02

--Non-CVD death(%)---Deaths at discharge -0.41 (-0.63 ~ -0.18) 0.02 -0.13 (-0.34 ~ 0.08) 0.3 -0.22 (-0.39 ~ -0.04) 0.07

---Deaths at 30 days(from admission) -0.31 (-0.47 ~ -0.16) 0.01 0.12 (-0.07 ~ 0.31) 0.3 -0.18 (-0.45 ~ 0.10) 0.3---Deaths at 1 year(from admission) -0.01 (-0.22 ~ 0.19) 0.9 0.49 (0.25 ~ 0.74) 0.006 0.59 (-0.05 ~ 1.23) 0.1

---Deaths at 30 days(from discharge) 0.15 (0.07 ~ 0.22) 0.01 0.27 (0.24 ~ 0.30) <.0001 0.04 (-0.14 ~ 0.22) 0.7---Deaths at 1 year(from discharge) 0.50 (0.31 ~ 0.68) 0.003 0.62 (0.48 ~ 0.75) <.0001 0.92 (0.31 ~ 1.53) 0.04CVD: Cardiovascular Disease; NCVD: Non-cardiovascular Disease

Table D Page 2

Table E. All-Cause, CVD and NCVD Death Rates at Different Time Intervals by Infarction Type and Year (1986-2007) (%)

YearAnnual change,%

(95%CI) R² p 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996Transmural, N 9298 9036 8689 8100 8005 7863 7967 8201 8011 7981 7811---All-Cause Death (%)---Deaths at discharge -0.40 (-0.45~-0.36) 0.94 <0.0001 16.05 16.06 16.79 15.22 14.78 14.41 12.75 12.05 11.96 12.77 11.97

---Deaths at 30 days(from admission) -0.36 (-0.41~-0.32) 0.93 <0.0001 17.32 17.54 18.28 16.54 16.08 15.87 14.32 13.64 13.64 14.46 14.21---Deaths at 1 year(from admission) -0.45 (-0.51~-0.40) 0.94 <0.0001 24.75 24.34 25.43 23.74 22.31 22.01 20.79 20.03 19.99 20.65 19.88

---Deaths at 30 days(from discharge) -0.02 (-0.04~0.00) 0.13 0.11 3.25 3.32 3.68 3.66 3.14 3.09 3.51 3.34 3.33 3.22 3.56---Deaths at 1 year(from discharge) -0.12 (-0.16~-0.08) 0.67 <0.0001 10.72 10.35 10.89 10.50 9.37 9.18 9.48 9.44 9.43 9.34 9.25--CVD death(%)---Deaths at discharge -0.31 (-0.34~-0.27) 0.95 <0.0001 11.57 11.85 12.51 10.80 10.92 10.33 9.53 9.29 8.88 8.82 8.17

---Deaths at 30 days(from admission) -0.29 (-0.33~-0.26) 0.94 <0.0001 12.97 13.39 14.16 12.28 12.40 11.74 11.05 10.71 10.55 10.49 10.09---Deaths at 1 year(from admission) -0.44 (-0.48~-0.40) 0.96 <0.0001 18.94 18.78 19.68 17.67 16.88 16.10 15.69 15.42 14.83 14.68 13.87

---Deaths at 30 days(from discharge) -0.04 (-0.05~-0.02) 0.47 0.0004 2.95 2.94 3.28 3.22 2.70 2.60 2.89 2.79 2.92 2.59 2.87---Deaths at 1 year(from discharge) -0.19 (-0.22~-0.16) 0.88 <0.0001 8.92 8.44 8.80 8.30 7.15 6.88 7.19 7.14 6.85 6.85 6.54--Non-CVD death(%)---Deaths at discharge -0.10 (-0.12~-0.07) 0.77 <0.0001 4.47 4.21 4.28 4.42 3.86 4.08 3.23 2.76 3.08 3.95 3.80

---Deaths at 30 days(from admission) -0.07 (-0.09~-0.04) 0.62 <0.0001 4.35 4.15 4.12 4.26 3.67 4.13 3.28 2.94 3.10 3.97 4.12---Deaths at 1 year(from admission) -0.01 (-0.04~0.01) 0.06 0.27 5.81 5.56 5.75 6.07 5.43 5.91 5.10 4.61 5.16 5.96 6.02

---Deaths at 30 days(from discharge) 0.02 (0.01~0.03) 0.52 0.0001 0.31 0.38 0.40 0.44 0.44 0.49 0.62 0.55 0.41 0.63 0.70---Deaths at 1 year(from discharge) 0.07 (0.06~0.08) 0.83 <0.0001 1.81 1.91 2.09 2.20 2.21 2.30 2.29 2.30 2.58 2.48 2.71CVD: Cardiovascular Disease; NCVD: Non-cardiovascular Disease

Table E Page 1

Table E. All-Cause, CVD and NCVD Death Rates at Different Time Intervals by Infarction Type and Year (1986-2007) (%)(Cont'd)Year 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007Transmural, N 7392 7025 6532 6125 5721 5397 5066 4228 3660 3751 3424---All-Cause Death (%)---Deaths at discharge 11.23 10.41 11.24 10.63 10.40 9.56 8.47 8.94 8.61 8.96 8.03

---Deaths at 30 days(from admission) 13.69 12.58 13.41 13.00 12.92 11.45 10.40 11.14 10.79 10.53 10.11---Deaths at 1 year(from admission) 19.17 18.73 19.81 19.25 19.07 16.84 16.76 16.53 15.68 15.62 14.89

---Deaths at 30 days(from discharge) 3.73 3.30 3.35 3.60 3.90 3.07 3.13 3.35 3.14 2.52 2.99---Deaths at 1 year(from discharge) 9.11 9.45 9.81 9.68 9.89 8.07 9.23 8.39 7.86 7.44 7.53--CVD death(%)---Deaths at discharge 8.47 7.49 8.07 7.54 7.76 6.97 6.10 6.53 5.85 6.42 5.61


---Deaths at 30 days(from discharge) 3.00 2.56 2.47 2.83 3.04 2.17 2.52 2.65 2.54 1.87 2.29---Deaths at 1 year(from discharge) 6.58 6.50 6.80 6.49 6.55 5.04 5.74 5.35 5.17 4.33 4.03--Non-CVD death(%)---Deaths at discharge 2.76 2.92 3.17 3.09 2.64 2.59 2.37 2.41 2.76 2.53 2.42


---Deaths at 30 days(from discharge) 0.73 0.75 0.88 0.77 0.86 0.90 0.60 0.70 0.60 0.64 0.70---Deaths at 1 year(from discharge) 2.53 2.96 3.02 3.20 3.34 3.03 3.49 3.04 2.69 3.10 3.49CVD: Cardiovascular Disease; NCVD: Non-cardiovascular Disease

Table E Page 2

Table E. All-Cause, CVD and NCVD Death Rates at Different Time Intervals by Infarction Type and Year (1986-2007) (%)(Cont'd)

YearAnnual change,%

(95%CI) R² p 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996Subendocardial, N 2155 2231 2187 2393 2580 3004 3316 3810 3948 4189 4589---All-Cause Death (%)---Deaths at discharge 0.04 (0.01~0.07) 0.25 0.02 6.40 5.69 6.77 6.14 5.66 6.16 6.36 6.30 6.64 6.68 7.19

---Deaths at 30 days(from admission) 0.16 (0.11~0.20) 0.70 <0.0001 7.61 6.81 8.14 7.15 7.17 6.92 7.24 7.32 7.98 8.67 9.24---Deaths at 1 year(from admission) 0.26 (0.19~0.33) 0.73 <0.0001 18.28 17.53 18.98 19.39 18.60 17.81 18.64 18.06 19.15 19.69 20.33

---Deaths at 30 days(from discharge) 0.11 (0.09~0.14) 0.79 <0.0001 2.23 2.76 3.04 2.89 3.08 2.59 3.00 2.66 3.34 3.76 3.73---Deaths at 1 year(from discharge) 0.22 (0.16~0.28) 0.73 <0.0001 13.24 13.17 13.68 14.34 14.13 12.84 13.69 13.03 13.65 14.33 14.46--CVD death(%)---Deaths at discharge -0.01(-0.03~0.01) 0.06 0.27 4.64 3.85 4.62 4.35 3.88 4.09 4.31 4.09 4.28 4.42 4.77

---Deaths at 30 days(from admission) 0.04 (0.01~0.08) 0.22 0.03 5.85 5.11 6.13 5.39 5.39 4.83 5.04 5.04 5.67 6.30 6.56---Deaths at 1 year(from admission) -0.05(-0.11~0.01) 0.13 0.11 14.34 12.95 14.72 14.33 12.98 12.18 12.76 12.52 13.25 13.34 13.82

---Deaths at 30 days(from discharge) 0.04 (0.01~0.06) 0.32 0.01 1.88 2.23 2.70 2.40 2.59 1.84 2.35 2.10 2.66 3.10 2.91---Deaths at 1 year(from discharge) -0.05 (-0.10~0.00) 0.14 0.08 10.76 9.89 11.18 10.82 9.90 8.87 9.34 9.19 9.63 9.80 9.88--Non-CVD death(%)---Deaths at discharge 0.05 (0.03~0.07) 0.57 <0.0001 1.76 1.84 2.15 1.80 1.78 2.06 2.05 2.20 2.36 2.27 2.42

---Deaths at 30 days(from admission) 0.12 (0.09~0.14) 0.84 <0.0001 1.76 1.70 2.01 1.76 1.78 2.10 2.20 2.28 2.30 2.36 2.68---Deaths at 1 year(from admission) 0.31 (0.27~0.35) 0.93 <0.0001 3.94 4.57 4.25 5.06 5.62 5.63 5.88 5.54 5.90 6.35 6.52

---Deaths at 30 days(from discharge) 0.07 (0.07~0.08) 0.93 <0.0001 0.35 0.52 0.34 0.49 0.49 0.74 0.64 0.56 0.68 0.67 0.82---Deaths at 1 year(from discharge) 0.27 (0.24~0.31) 0.92 <0.0001 2.48 3.28 2.50 3.52 4.23 3.97 4.35 3.84 4.02 4.53 4.58CVD: Cardiovascular Disease; NCVD: Non-cardiovascular Disease

Table E Page 3

Table E. All-Cause, CVD and NCVD Death Rates at Different Time Intervals by Infarction Type and Year (1986-2007) (%)(Cont'd)Year 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007Subendocardial, N 4658 5405 6185 7060 7481 7963 7816 7878 7772 7375 7357---All-Cause Death (%)---Deaths at discharge 6.08 5.81 6.26 6.80 7.45 7.38 7.31 6.84 7.10 6.59 6.23


---Deaths at 30 days(from discharge) 3.95 4.05 4.36 4.83 4.65 4.65 4.67 4.43 4.78 4.24 4.15---Deaths at 1 year(from discharge) 15.73 15.64 16.57 17.02 17.95 17.78 17.31 16.72 16.27 16.08 16.81--CVD death(%)---Deaths at discharge 4.36 3.98 3.69 4.31 4.61 4.13 4.21 4.11 4.30 4.18 3.60


---Deaths at 30 days(from discharge) 2.99 3.02 3.04 3.39 3.15 3.13 3.06 2.75 3.14 2.73 2.26---Deaths at 1 year(from discharge) 10.56 9.96 10.31 10.65 10.85 10.43 10.30 9.58 9.34 8.93 7.39--Non-CVD death(%)---Deaths at discharge 1.72 1.83 2.57 2.49 2.83 3.25 3.10 2.73 2.80 2.41 2.62


---Deaths at 30 days(from discharge) 0.96 1.02 1.33 1.44 1.50 1.52 1.60 1.68 1.63 1.51 1.88---Deaths at 1 year(from discharge) 5.17 5.68 6.26 6.37 7.11 7.35 7.01 7.14 6.94 7.16 9.42CVD: Cardiovascular Disease; NCVD: Non-cardiovascular Disease

Table E Page 4

7 9 3 5 3 5

Table F. One-Year Post-Discharge Non-CVD Cause Specific Mortality by Two-Year Time Intervals

YearAnnual change,%

(95%CI) R² p 1986-198 1988-198 1990-19911992-199 1994-199 1996-1997 1998-1999 2000-20012002-200 2004-200 2006-2007NCVD death as % of all deaths 0.14 (0.09~0.19) 0.78 0.0003 6.27 6.28 6.45 5.61 6.37 6.36 7.11 8.02 8.20 8.09 8.77--Transmural 0.00 (-0.04~0.04) 0.00 0.91 6.67 6.65 6.71 5.57 6.48 6.28 6.63 7.16 6.29 6.32 6.41--Subendocardial 0.31 (0.27~0.35) 0.96 <.0001 4.26 4.67 5.62 5.70 6.13 6.51 7.75 8.85 9.75 9.20 10.17Most Common NCVD deaths:Cancer 0.06 (0.04~0.07) 0.86 <.0001 0.75 0.79 0.93 0.88 1.07 0.89 1.26 1.51 1.84 1.64 1.73--Transmural 0.03 (0.02~0.04) 0.84 0.0001 0.73 0.74 0.86 0.81 1.00 0.83 1.16 1.19 1.37 1.29 1.23--Subendocardial 0.06 (0.04~0.08) 0.80 0.0002 0.89 1.00 1.15 1.07 1.22 0.99 1.40 1.83 2.22 1.85 2.02Diabetes 0.01 (0.00~0.03) 0.24 0.13 0.54 0.81 1.13 0.85 1.03 1.05 1.11 1.12 1.12 1.08 0.83--Transmural 0.00 (-0.01~0.02) 0.02 0.67 0.52 0.84 1.11 0.85 1.04 1.00 1.01 0.91 1.00 0.82 0.74--Subendocardial 0.02 (0.00~0.04) 0.32 0.07 0.64 0.70 1.22 0.84 1.00 1.16 1.24 1.32 1.22 1.23 0.88Respiratory 0.04 (0.02~0.05) 0.78 0.0003 0.58 0.64 0.70 0.66 0.80 0.99 1.00 1.31 1.29 1.32 1.04--Transmural 0.01 (0.00~0.02) 0.19 0.19 0.55 0.66 0.56 0.59 0.68 0.77 0.70 0.86 0.67 0.86 0.54--Subendocardial 0.05 (0.03~0.07) 0.69 0.002 0.73 0.57 1.16 0.86 1.07 1.38 1.40 1.74 1.78 1.60 1.33Renal failure 0.03 (0.02~0.04) 0.82 0.0001 0.22 0.26 0.30 0.33 0.30 0.40 0.57 0.78 0.71 0.85 0.64--Transmural 0.02 (0.01~0.03) 0.76 0.0005 0.19 0.26 0.32 0.27 0.25 0.34 0.45 0.69 0.59 0.62 0.54--Subendocardial 0.03 (0.02~0.05) 0.78 0.0003 0.34 0.22 0.25 0.51 0.41 0.51 0.72 0.87 0.80 0.99 0.71Septicemia 0.02 (0.02~0.03) 0.82 0.0001 0.11 0.16 0.16 0.25 0.23 0.24 0.53 0.49 0.52 0.60 0.47--Transmural 0.01 (0.01~0.02) 0.63 0.004 0.12 0.14 0.16 0.20 0.19 0.22 0.44 0.38 0.42 0.40 0.27--Subendocardial 0.03 (0.02~0.04) 0.80 0.0002 0.07 0.24 0.16 0.38 0.32 0.28 0.66 0.61 0.60 0.73 0.58Alzheimer's 0.01 (0.01~0.01) 0.84 0.0001 0.01 0.02 0.01 0.02 0.04 0.05 0.08 0.12 0.19 0.16 0.14--Transmural 0.01 (0.00~0.01) 0.91 <.0001 0.01 0.02 0.01 0.02 0.04 0.05 0.06 0.08 0.12 0.10 0.10--Subendocardial 0.01 (0.01~0.01) 0.76 0.0004 0.00 0.04 0.02 0.03 0.02 0.04 0.12 0.15 0.24 0.20 0.16CVD: Cardiovascular Disease; NCVD: Non-cardiovascular Disease

Table F

e

Table G. Multivariate Models of In-Hospital Death, 30 Day Mortality After Discharge and One Year Mortality After Discharge, 1994-2007

In-Hospital Death1 30 Day After Discharge1 1 Year After Discharge2

Variable Odds Ratio (95% CL) Odds Ratio (95% CL) Hazard Ratio (95% CL)

Discharge Year 1.00 (0.99 - 1.00) 1.03 (1.02 - 1.03) 1.02 (1.01 - 1.02)Age per year 1.05 (1.05 - 1.05) 1.05 (1.04 - 1.05) 1.05 (1.05 - 1.05)Gender ---Female 1.00 1.00 1.00 ---Male 0.93 (0.89 - 0.96) 0.98 (0.93 - 1.03) 1.02 (0.99 - 1.05)Ethnic group ---Hispanic 1.00 1.00 1.00 ---Non-Hispanic White 0.99 (0.92 - 1.06) 1.15 (1.03 - 1.30) 1.10 (1.04 - 1.17)---Non-Hispanic Black 1.30 (1.19 - 1.43) 1.04 (0.90 - 1.21) 1.23 (1.14 - 1.33)Insurance Groups ---Medicare 1.00 1.00 1.00 ---Commercial Insurance 0.76 (0.71 - 0.81) 0.86 (0.78 - 0.96) 0.73 (0.69 - 0.78)---Medicaid 1.35 (1.17 - 1.56) 1.33 (1.05 - 1.67) 1.36 (1.21 - 1.53)---Self pay 1.37 (1.23 - 1.53) 0.93 (0.75 - 1.15) 0.79 (0.70 - 0.89)---HMO 0.76 (0.72 - 0.81) 0.86 (0.79 - 0.95) 0.79 (0.75 - 0.83)AMI site ---Anterior 1.00 1.00 1.00 ---Inferior / Inferolateral 0.87 (0.82 - 0.91) 0.75 (0.69 - 0.82) 0.79 (0.75 - 0.83)---Other / Unspecified 1.73 (1.63 - 1.83) 0.97 (0.88 - 1.08) 1.08 (1.02 - 1.15)---Subendocardial 0.37 (0.35 - 0.39) 0.65 (0.60 - 0.69) 0.93 (0.90 - 0.97)Renal disease Yes vs No 1.07 (1.03 - 1.11) 1.08 (1.02 - 1.15) 1.26 (1.23 - 1.30)Anemia Yes vs No 0.71 (0.68 - 0.75) 0.93 (0.86 - 1.00) 1.18 (1.14 - 1.23)COPD Yes vs No 1.01 (0.96 - 1.05) 1.20 (1.12 - 1.28) 1.34 (1.30 - 1.39)Hypertension Yes vs No 0.55 (0.53 - 0.57) 0.74 (0.70 - 0.78) 0.81 (0.79 - 0.84)Diabetes Yes vs No 0.91 (0.87 - 0.95) 1.10 (1.04 - 1.17) 1.19 (1.15 - 1.23)Atrial Fibrillation Yes vs No 1.15 (1.11 - 1.20) 1.28 (1.20 - 1.36) 1.29 (1.25 - 1.33)Heart Failure Yes vs No 2.49 (2.40 - 2.59) 2.03 (1.92 - 2.15) 1.98 (1.92 - 2.04)Cancer Yes vs No 1.82 (1.68 - 1.96) 2.15 (1.94 - 2.38) 2.49 (2.37 - 2.62)Cerebrovascular Diseas Yes vs No 1.83 (1.72 - 1.94) 1.66 (1.51 - 1.81) 1.52 (1.45 - 1.60)PCI Yes vs No 0.49 (0.46 - 0.53) 0.28 (0.25 - 0.32) 0.41 (0.39 - 0.44)CABG Yes vs No 0.68 (0.62 - 0.75) 0.27 (0.22 - 0.33) 0.33 (0.30 - 0.37)

1--Logistic regression 2-Cox Propotional Hazard ModelCL: Confidence limit; AMI: Acute myocardial infarction; COPD: Chronic obstructive pulmonary disease; PCI: Percutaneous coronary intervention; CABG: Coronary Artery Bypass Graft;

Table G

trends in mortality of acute myocardial infarction after...

Documents