Abstract
Background
Case‐fatality rates in acute myocardial infarction (AMI) have significantly decreased; however, the prevalence of diabetes mellitus (DM), a risk factor for AMI, has increased. The purposes of the present study were to assess the prevalence and clinical impact of DM among patients hospitalized with AMI and to estimate the impact of important clinical characteristics associated with in‐hospital mortality in patients with AMI and DM.
Methods and Results
We used the National Inpatient Sample to estimate trends in DM prevalence and in‐hospital mortality among 1.5 million patients with AMI from 2000 to 2010, using survey data‐analysis methods. Clinical characteristics associated with in‐hospital mortality were identified using multivariable logistic regression. There was a significant increase in DM prevalence among AMI patients (year 2000, 22.2%; year 2010, 29.6%, Ptrend<0.0001). AMI patients with DM tended to be older and female and to have more cardiovascular risk factors. However, age‐standardized mortality decreased significantly from 2000 (8.48%) to 2010 (4.95%) (Ptrend<0.0001). DM remained independently associated with mortality (adjusted odds ratio 1.069, 95% CI 1.051 to 1.087; P<0.0001). The adverse impact of DM on in‐hospital mortality was unchanged over time. Decreased death risk over time was greatest among women and elderly patients. Among younger patients of both sexes, there was a leveling off of this decrease in more recent years.
Conclusions
Despite increasing DM prevalence and disease burden among AMI patients, in‐hospital mortality declined significantly from 2000 to 2010. The adverse impact of DM on mortality remained unchanged overall over time but was age and sex dependent.
Keywords: Diabetes mellitus, mortality, myocardial infarction
Introduction
Mortality following acute myocardial infarction (AMI) in the United States has steadily declined over many decades.1–2 Numerous factors have been proposed to explain this favorable trend and include better adherence to contemporary guideline‐based therapies, more efficient and effective in‐hospital and postdischarge processes of care, and changes in the cardiovascular (CV) risk profile of patients presenting with AMI.3–5
Despite this favorable trend in AMI‐related mortality, certain patients continue to carry disproportionate risk. The presence of diabetes mellitus (DM) among patients with CV disease has historically predicted worse outcomes compared with patients without DM.6–8 CV disease or, more specifically, coronary heart disease is the leading cause of death among patients with DM,9 and a history of DM has been considered equivalent in risk to a known history of coronary heart disease.10 Among patients with AMI and DM, female sex has been observed to confer increased risk of adverse CV outcomes compared with men,11–12 although more recent data suggest that this differential risk may be narrowing.13 A recent report pointed to a reversal of the above‐mentioned secular trend in CV‐related mortality in persons younger than 55 years along with an increase in risk factors for DM in this cohort.14
Given the increasing prevalence of DM in the US population15–16 and a continuing focus on the impact of age and sex on CV outcomes,16–17 we examined trends among patients hospitalized for AMI to assess trends in the prevalence of DM among patients hospitalized for AMI from 2000 to 2010, to assess trends in in‐hospital mortality among AMI patients with DM, and to describe the factors associated with in‐hospital mortality with a focus on the impact of age, sex, and time.
Methods
Data Source
The National Inpatient Sample (NIS) is the largest available all‐payer inpatient database in the public domain and is sponsored by the Agency for Healthcare Research and Quality (AHRQ) and the Healthcare Cost and Utilization Project. The NIS consists of discharge data from more than 1000 hospitals across a majority of states and is designed to approximate a 20% stratified sample of US community hospitals.18 The NIS provides patient discharge‐level demographic and clinical characteristics that are searchable using International Classification of Diseases, ninth revision, clinical modification (ICD‐9‐CM) or Clinical Classification System codes. Each release of the NIS includes patient‐level hospital discharge abstract data for 100% of discharges from the sample of hospitals in participating states. We used NIS severity files to extract clinician‐verified comorbid conditions of patients established by AHRQ. Statistical sampling weights provided by the NIS allow extrapolation to estimate hospital discharge rates for the nation.19 After weighting, this reflects ≈95% of hospital discharges within the United States. The study was considered exempt from formal review by the University of New Mexico institutional review board because the NIS is a public database without personal identifiers.
Data Quality
A summary data quality report is available for review for each year of the NIS.20 Individual reports for the years 2000–2010 were reviewed by one of us (W.K.L.). Edit check failure (missing data) rates were consistently <0.5% for key data elements (eg, age, diagnoses, procedures).
Study Samples
We analyzed data in NIS for patients aged 18 years or older from 2000 to 2010. All records with a primary discharge diagnosis of AMI using ICD‐9‐CM codes 410.0 to 410.8 were identified (sample 1). The total number of AMI hospitalizations was calculated as the sum over all AMI ICD‐9‐CM codes. We then obtained the proportion of AMI discharges that occurred over the same time interval with a diagnosis of type 2 DM (T2DM; sample 2), identified by ICD‐9‐CM code 250.0 to 250.9 with a fifth digit of 0 or 0 or 2 because the majority of diagnosed cases of DM in adults are of the type 2 variety.21 These ICD‐9‐CM codes allow for the concomitant use of insulin in persons with T2DM, and an additional diagnosis code (v.58.67) allows for the specific identification of insulin use.
Data Analysis
We excluded records from analysis if they were missing vital status at discharge, DM status, age, or sex. Weighted continuous variables are summarized as mean±SE, and weighted categorical variables are summarized as counts or percentages ±SE. For analysis purposes, age was categorized as <55, 55 to 64, 65 to 74, 75 to 84, and ≥85 years, with the “<55 years” category serving as the reference group. We used survey regression procedures designed to incorporate NIS‐specified weights for descriptive statistics and multivariable models. Trends in categorical variables were tested using the Wald chi‐square statistics (SAS PROC SURVEYFREQ; SAS Institute Inc.).
Multivariable Analysis
Demographic, clinical, and hospital characteristics in the NIS data sets from 2002 to 2010 were used to develop a model for in‐hospital death (data sets from 2000 and 2001 were missing information on obesity and tobacco and were excluded from this portion of the analysis). Our first model (compare with sample 1 under “Study Samples”) included all patients with AMI regardless of DM status, and a second model (compare with sample 2 under “Study Samples”) included only those records that included AMI and T2DM diagnoses. Covariates identified as AHRQ‐defined comorbidities likely present before admission were chosen for their clinical relevance, their presence in the NIS data sets, and their known association with in‐hospital mortality (the dependent variable). Additional covariates that may have been identified during hospitalization and are known to be associated with mortality (eg, shock, ventricular fibrillation) were also included in this explanatory model. An indicator variable encoding for any ICD‐9‐CM–identified coronary revascularization procedure (surgical or percutaneous) that was performed during the hospitalization was created and added to the list of covariates. Multivariable logistic regression models that accounted for survey methodology and hospital clustering (SAS PROC SURVEYLOGISTIC) were developed to estimate the magnitude of the association between T2DM status and in‐hospital mortality (using sample 1); the magnitude of association between clinical, temporal, and demographic covariates and in‐hospital mortality in patients with AMI and T2DM (sample 2); and whether survival at discharge had improved over time.
In the above‐noted models, the overall effects of sex, year, and age on the odds of mortality are represented by their respective β‐coefficients. We added the interaction terms “year×sex,” “year×age (category),” and “year×sex×age (category)” to the above‐described multivariable model to test for modification of the effect of time (year) by sex or age. Year was modeled as a continuous, linear function, whereas age and sex maintained their categorical status. Model fit was excellent (test for linear fit, P<0.001) and was not further improved with consideration of nonlinear relationships with time. Predicted probabilities were calculated from the inverse logit transformation and plotted and smoothed for display using a Hamming's window filter (MATLAB; MathWorks Inc.).
Rate Decomposition Analysis
In order to distinguish age versus age‐independent factors driving the observed decrease in mortality rate over time, the method of rate decomposition was used.22 Briefly, the difference in crude mortality rate (CMR) from 2000 to 2010 can be viewed as the sum of a “composition effect” (reflecting the difference in the age composition of the sample from 2000 to 2010) and a “rate effect” (reflecting the differences in the distribution of age stratum‐specific mortality rates from 2000 to 2010): Δ CMR2000–2010=composition effect+rate effect. Age standardization was performed using the average of the 2000 and 2010 NIS data sets as the standard population.
Sensitivity Analysis
Due to changing biomarker‐defined criteria for AMI (particularly for the non–ST‐segment elevation myocardial infarction [non‐STEMI] category) as well as dissemination of these criteria into routine coding practice over the time interval of this study, we performed a subgroup analysis confined to patients with STEMI—a more consistently defined group—to assess the impact of coding (for AMI) certainty on the conclusions. Additional sensitivity analyses were conducted in 2 important subgroups of patients with AMI and T2DM: patients receiving adjunctive insulin and patients with morbid obesity.
All analyses were conducted using SAS version 9.1 and higher (SAS Institute). Given the large sample size and the multiplicity of comparison testing, 2‐sided P values were considered statistically significant at ≤0.001. Estimated measures of association (logistic regression) are expressed as odds ratios (ORs) and 95% CIs.
Results
Data Quality and Data Quality Assurance
Differences in the number of states contributing data over time could result in biased estimates despite the sampling methodology used in the NIS. Over the 10 years from 2000 to 2010, the number of states contributing data to the NIS increased (Figure 1). Although there was “drop‐out” in the number of states contributing data in the first half of the decade, these same states “dropped in” in subsequent years. However, loss of states contributing data was infrequent, and the number of participating states increased steadily from 28 in 2000 to 45 states by 2010. Due to the sampling methodology used in the NIS, the number of hospitals and the number of discharge records in the sample remained relatively flat.
Figure 1.
National Inpatient Sample activity, 2000–2010. Participation by states increased over time. Sampling methodology maintained the number of hospitals and discharges within a narrow range from 2000 to 2010.
There were 86 622 872 records in the combined 2000–2010 data sets, with 0.2% missing one of the above‐mentioned key variables (ie, these records were excluded from further analysis). Of 86 593 459 primary diagnoses at discharge, full data were available for 99.97%. There were 1 547 859 unique principal discharge diagnoses of AMI (1.8%) in our sample and more than 7.5 million AMI records in the weighted sample (Table 1).
Table 1.
Pre‐ and In‐Hospital Patient Characteristics Among All Patients With Acute Myocardial Infarction From 2000 to 2010
| 2000 | 2001 | 2002 | 2003 | 2004 | 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | P Value (Trend) | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| N (sample) | 157 263 | 154 693 | 158 029 | 156 672 | 143 222 | 135 141 | 138 374 | 126 231 | 131 380 | 125 777 | 121 077 | |
| N (weighted) | 768 407 | 773 858 | 764 133 | 760 718 | 695 063 | 662 345 | 675 121 | 624 936 | 644 657 | 633 356 | 604 784 | |
| Prehospital, % | ||||||||||||
| Sex | ||||||||||||
| Female | 41.9 | 41.1 | 41.0 | 40.9 | 40.8 | 40.7 | 39.9 | 40.4 | 40.3 | 39.5 | 39.6 | <0.0001 |
| Male | 59.1 | 58.9 | 59.0 | 59.1 | 59.2 | 59.3 | 60.1 | 59.6 | 59.7 | 60.5 | 60.4 | |
| Age group, y | ||||||||||||
| <55 | 19.6 | 19.5 | 20.1 | 20.1 | 19.8 | 20.3 | 21.4 | 21.1 | 20.6 | 21.2 | 20.9 | <0.0001 |
| 55 to 64 | 18.9 | 18.8 | 19.5 | 20.1 | 20.3 | 20.3 | 21.3 | 21.4 | 21.5 | 22.1 | 22.7 | <0.0001 |
| 65 to 74 | 24.1 | 23.4 | 22.8 | 22.0 | 22.1 | 21.3 | 21.1 | 21.2 | 21.3 | 21.8 | 21.7 | <0.0001 |
| 75 to 84 | 25.2 | 25.4 | 24.7 | 24.6 | 24.5 | 23.9 | 23.0 | 22.2 | 22.1 | 21.1 | 20.7 | <0.0001 |
| >84 | 12.1 | 12.9 | 13.0 | 13.3 | 13.4 | 14.2 | 13.3 | 14.0 | 14.5 | 13.8 | 14.0 | <0.0001 |
| Race | ||||||||||||
| Non‐Hispanic White | 83.1 | 82.2 | 79.7 | 78.1$ | 78.9 | 80.2 | 78.6 | 76.8 | 77.6 | 76.6 | 75.9 | <0.0001 |
| Black | 7.4 | 7.6 | 8.4 | 8.7 | 9.0 | 7.2 | 8.6 | 9.9 | 9.0 | 9.2 | 11.2 | <0.0001 |
| Hispanic | 5.5 | 5.9 | 6.7 | 8.3 | 6.9 | 7.4 | 7.6 | 7.2 | 6.5 | 7.2 | 7.0 | <0.0001 |
| Asian | 1.5 | 1.5 | 1.9 | 2.1 | 2.1 | 1.7 | 1.9 | 2.3 | 2.4 | 2.2 | 2.3 | <0.0001 |
| AI/NA | 0.3 | 0.4 | 0.2 | 0.2 | 0.3 | 0.3 | 0.5 | 0.7 | 0.9 | 0.6 | 0.9 | <0.0001 |
| Other | 2.2 | 2.4 | 3.1 | 2.6 | 2.8 | 3.1 | 2.9 | 3.2 | 3.6 | 4.3 | 2.7 | <0.0001 |
| Type of AMI | ||||||||||||
| STEMI | 41.2 | 38.7 | 37.5 | 35.0 | 32.5 | 30.6 | 31.4 | 29.5 | 28.6 | 27.2 | 27.2 | <0.0001 |
| Comorbidities | ||||||||||||
| T2DM | 22.2 | 22.3 | 24.1 | 24.1 | 25.0 | 26.0 | 26.3 | 26.6 | 27.9 | 29.2 | 29.6 | <0.0001 |
| TIA/stroke | 3.8 | 3.8 | 3.9 | 3.9 | 3.9 | 3.9 | 4.0 | 3.9 | 4.3 | 4.2 | 4.0 | <0.0001 |
| Heart failure | 32.2 | 31.6 | 31.9 | 34.1 | 35.0 | 34.6 | 33.3 | 33.9 | 33.6 | 33.9 | 33.9 | <0.0001 |
| Prior MI | 66.2 | 67.8 | 69.1 | 70.4 | 71.0 | 72.3 | 74.3 | 74.9 | 76.7 | 78.4 | 78.3 | <0.0001 |
| Hypertension | 50.7 | 52.7 | 55.2 | 57.1 | 59.1 | 60.8 | 62.9 | 64.2 | 66.1 | 67.8 | 68.8 | <0.0001 |
| Renal failure | 5.7 | 6.4 | 7.0 | 7.7 | 8.3 | 10.9 | 15.7 | 21.5 | 22.7 | 24.4 | 25.5 | <0.0001 |
| AFib | 16.1 | 16.5 | 16.6 | 16.5 | 16.9 | 17.1 | 17.0 | 16.8 | 15.8 | 15.9 | 16.0 | <0.0001 |
| Dyslipidemia | 28.4 | 31.5 | 35.6 | 37.3 | 40.8 | 44.0 | 47.1 | 50.7 | 52.3 | 54.4 | 56.7 | <0.0001 |
| PAD | 4.2 | 4.4 | 4.7 | 5.0 | 5.0 | 5.1 | 5.2 | 5.8 | 6.2 | 6.2 | 6.3 | <0.0001 |
| Cancer | 0.7 | 0.7 | 0.7 | 0.7 | 0.8 | 0.8 | 0.8 | 0.9 | 0.9 | 0.9 | 0.9 | <0.0001 |
| Dementia | 0.9 | 0.9 | 0.8 | 0.9 | 0.8 | 0.9 | 0.8 | 0.7 | 0.8 | 0.7 | 0.6 | <0.0001 |
| Tobacco use | — | — | 16.5 | 16.6 | 17.5 | 19.0 | 20.4 | 21.0 | 21.9 | 22.6 | 22.8 | <0.0001 |
| Obesity | — | — | 6.4 | 6.8 | 7.0 | 7.7 | 8.0 | 9.3 | 10.8 | 12.1 | 11.9 | <0.0001 |
| In‐hospital, % | ||||||||||||
| Shock | 4.1 | 4.1 | 4.3 | 4.5 | 4.5 | 4.6 | 4.7 | 5.1 | 5.4 | 5.8 | 5.7 | <0.0001 |
| VFib | 2.7 | 2.5 | 2.6 | 2.4 | 2.5 | 2.4 | 2.5 | 2.6 | 2.6 | 2.6 | 2.7 | 0.007 |
| Revasc | 56.9 | 59.1 | 59.8 | 61.2 | 61.8 | 63.8 | 66.5 | 65.9 | 66.4 | 69.5 | 69.7 | <0.0001 |
AFib indicates atrial fibrillation; AI/NA, American Indian or Native American; MI, myocardial infarction; PAD, peripheral arterial disease; Revasc, coronary revascularization procedure; STEMI, ST‐segment elevation myocardial infarction; T2DM, type 2 diabetes mellitus; TIA, transient ischemic attack; VFib, ventricular fibrillation.
AMI Sample (Sample 1)
In the weighted AMI sample, the mean age was 68.1±0.2 years in 2000 and 67.4±0.2 years in 2010 (P<0.0001). Women represented ≈40% of the sample (Table 1). There were statistically significant increases in the prevalence of T2DM, obesity, hypertension, dyslipidemia, prior AMI, and tobacco use over time (Table 1). There was a significant change in sex ratio over time as well as a change in the age distribution over time, with an increase in the proportion of the <55 years and 55 to 64 years age groups and a decrease in the proportion of the 65 to 84 years age group (P<0.0001). The type of AMI also changed over the study interval, with an increase in the proportion of non‐STEMI to STEMI.
AMI and T2DM Sample (Sample 2)
Within the AMI sample, there were 435 265 records (28.1%) with a coexistent T2DM diagnosis code (Table 2). As seen in Figure 2, there was a significant increase in the prevalence of T2DM over time (year 2000, 22.2%; year 2010, 29.6%, P for trend <0.0001). AMI patients with T2DM were younger in 2010 compared with 2000 (mean age in 2000, 68.9±0.4 years; mean age in 2010, 67.8±0.3 years; P for overall trend <0.0001) but were older than AMI patients without T2DM (P<0.0001) (Table 3). The majority of CV risk factors increased in prevalence over the study period (Table 2), and the proportions of these risk factors were significantly higher compared with patients without T2DM (Table 3). As observed in the overall AMI sample, the proportion of non‐STEMI to STEMI increased significantly over time. Coronary revascularization procedures, the vast majority (>75%) of which were percutaneous, steadily and significantly increased over the observed time (Figure 3). The use of these procedures in patients with AMI and coexistent T2DM was consistently greater than in AMI patients overall (Tables 1 and 2).
Table 2.
Pre‐ and In‐Hospital Patient Characteristics Among Patients With AMI and Coexistent Diabetes Mellitus
| 2000 | 2001 | 2002 | 2003 | 2004 | 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | P Value (Trend) | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| N (sample) | 36 753 | 37 752 | 40 673 | 41 633 | 39 107 | 38 635 | 40 745 | 38 993 | 41 235 | 40 238 | 39 501 | |
| N (weighted) | 179 646 | 188 898 | 196 987 | 200 244 | 189 870 | 189 348 | 198 903 | 193 004 | 202 011 | 202 774 | 197 492 | |
| Prehospital, % | ||||||||||||
| Sex | ||||||||||||
| Male | 55.3 | 55.5 | 55.6 | 55.4 | 56.2 | 55.9 | 56.6 | 56.9 | 56.9 | 57.8 | 57.5 | <0.0001 |
| Female | 44.7 | 44.5 | 44.4 | 44.6 | 43.8 | 44.1 | 43.4 | 43.1 | 43.1 | 42.2 | 42.5 | |
| Age group, y | ||||||||||||
| <55 | 14.5 | 14.4 | 15.1 | 14.9 | 15.2 | 15.6 | 16.9 | 16.6 | 16.3 | 16.9 | 16.9 | <0.0001 |
| 55 to 64 | 20.4 | 20.0 | 21.0 | 21.5 | 22.3 | 22.3 | 22.6 | 23.1 | 23.0 | 23.5 | 23.8 | <0.0001 |
| 65 to 74 | 28.5 | 28.2 | 27.3 | 26.7 | 26.2 | 25.5 | 25.4 | 25.3 | 25.7 | 25.9 | 25.6 | <0.0001 |
| 75 to 84 | 26.6 | 27.5 | 26.6 | 26.5 | 26.2 | 25.6 | 24.8 | 24.4 | 24.2 | 22.8 | 22.8 | <0.0001 |
| >84 | 9.9 | 10.0 | 10.1 | 10.4 | 10.2 | 11.0 | 10.2 | 10.7 | 10.9 | 10.8 | 10.9 | <0.0001 |
| Race | ||||||||||||
| Non‐Hispanic White | 77.7 | 76.4 | 73.1 | 71.0 | 72.3 | 73.9 | 71.8 | 70.2 | 71.3 | 70.1 | 69.3 | <0.0001 |
| Black | 9.3 | 9.5 | 10.8 | 10.9 | 11.0 | 9.2 | 10.9 | 11.9 | 11.1 | 11.5 | 13.4 | <0.0001 |
| Hispanic | 8.2 | 9.0 | 9.6 | 11.9 | 10.0 | 10.5 | 11.0 | 10.4 | 9.4 | 10.1 | 9.9 | <0.0001 |
| Asian | 2.0 | 2.1 | 2.6 | 2.9 | 3.1 | 2.3 | 2.4 | 3.0 | 3.2 | 2.8 | 3.1 | <0.0001 |
| AI/NA | 0.3 | 0.4 | 0.4 | 0.3 | 0.4 | 0.5 | 0.7 | 0.9 | 1.0 | 0.7 | 1.0 | <0.0001 |
| Other | 2.5 | 2.6 | 3.6 | 3.0 | 3.2 | 3.7 | 3.2 | 3.6 | 4.0 | 4.9 | 3.3 | <0.0001 |
| Type of AMI | ||||||||||||
| STEMI | 34.8 | 32.4 | 30.9 | 28.5 | 26.5 | 24.2 | 24.8 | 23.2 | 22.4 | 21.0 | 21.1 | <0.0001 |
| Comorbidities | ||||||||||||
| TIA/stroke | 4.0 | 4.0 | 4.1 | 4.2 | 4.2 | 4.3 | 4.3 | 4.2 | 4.6 | 4.5 | 4.3 | <0.0001 |
| Heart failure | 40.1 | 39.6 | 40.2 | 41.7 | 42.5 | 42.1 | 40.2 | 41.4 | 40.4 | 40.8 | 40.3 | 0.017 |
| Prior MI | 69.5 | 71.3 | 72.3 | 74.1 | 74.6 | 76.3 | 77.7 | 78.7 | 80.0 | 81.8 | 81.7 | <0.0001 |
| Hypertension | 62.5 | 64.0 | 67.1 | 69.0 | 71.2 | 72.5 | 74.6 | 75.6 | 76.6 | 78.4 | 79.4 | <0.0001 |
| Renal failure | 6.6 | 7.5 | 8.5 | 9.0 | 9.4 | 13.1 | 20.7 | 28.7 | 30.3 | 32.8 | 33.7 | <0.0001 |
| AFib | 15.5 | 15.5 | 16.1 | 15.6 | 15.7 | 15.9 | 16.3 | 16.4 | 15.7 | 15.9 | 15.9 | <0.0001 |
| Dyslipidemia | 30.7 | 34.6 | 39.6 | 42.0 | 46.1 | 49.2 | 52.7 | 56.8 | 58.2 | 59.9 | 62.5 | <0.0001 |
| PAD | 6.1 | 6.4 | 6.7 | 7.1 | 7.2 | 7.4 | 7.3 | 8.4 | 8.8 | 8.5 | 8.7 | <0.0001 |
| Cancer | 0.5 | 0.6 | 0.6 | 0.5 | 0.6 | 0.7 | 0.6 | 0.8 | 0.7 | 0.7 | 0.6 | <0.0001 |
| Dementia | 0.9 | 0.9 | 0.8 | 0.9 | 0.9 | 0.9 | 0.8 | 0.7 | 0.8 | 0.7 | 0.6 | <0.0001 |
| Tobacco use | — | — | 11.1 | 11.2 | 12.1 | 13.4 | 14.5 | 15.6 | 16.5 | 16.8 | 17.4 | <0.0001 |
| Obesity | — | — | 10.3 | 10.8 | 11.4 | 12.5 | 13.3 | 15.1 | 17.2 | 18.9 | 18.9 | <0.0001 |
| In‐hospital, % | ||||||||||||
| Shock | 3.4 | 3.4 | 3.7 | 3.5 | 3.6 | 3.6 | 3.7 | 4.1 | 4.6 | 5.0 | 5.0 | <0.0001 |
| VFib | 1.6 | 1.5 | 1.6 | 1.4 | 1.6 | 1.5 | 1.5 | 1.6 | 1.7 | 1.6 | 1.6 | <0.0001 |
| Revasc | 58.2 | 60.5 | 61.3 | 62.7 | 63.2 | 65.4 | 68.1 | 67.5 | 67.8 | 70.9 | 71.1 | <0.0001 |
AFib indicates atrial fibrillation; AI/NA, American Indian or Native American; AMI, acute myocardial infarction; MI, myocardial infarction; PAD, peripheral arterial disease; Revasc, coronary revascularization procedure; STEMI, ST‐segment elevation myocardial infarction; TIA, transient ischemic attack; VFib, ventricular fibrillation.
Figure 2.
Prevalence of T2DM over time in patients with AMI. Steady and significant increase in prevalence of T2DM among patients with AMI from 2000 to 2010. AMI indicates acute myocardial infarction; T2DM, type 2 diabetes mellitus.
Table 3.
Pre‐ and In‐Hospital Patient Characteristics Among Patients With AMI and Without Diabetes Mellitus
| 2000 | 2001 | 2002 | 2003 | 2004 | 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | P Value (Trend) | P Value (DM vs No DM) | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| N (sample) | 120 510 | 116 941 | 117 356 | 115 039 | 104 115 | 96 506 | 97 629 | 87 238 | 90 145 | 85 539 | 81 576 | ||
| N (weighted) | 588 826 | 584 900 | 567 133 | 550 154 | 504 964 | 472 811 | 476 086 | 43 175 | 442 524 | 430 546 | 248 774 | ||
| Prehospital, % | |||||||||||||
| Sex | |||||||||||||
| Male | 60.3 | 60.0 | 60.2 | 60.4 | 60.3 | 60.6 | 61.6 | 60.9 | 60.9 | 61.7 | 61.8 | <0.0001 | <0.0001 |
| Female | 39.7 | 40.0 | 39.8 | 39.6 | 39.7 | 39.4 | 38.4 | 39.1 | 39.1 | 38.3 | 38.2 | ||
| Age group, y | |||||||||||||
| <55 | 21.2 | 21.1 | 21.8 | 22.0 | 21.6 | 22.1 | 23.3 | 23.2 | 22.5 | 23.2 | 22.9 | <0.0001 | <0.0001 |
| 55 to 64 | 18.5 | 18.4 | 19.0 | 19.6 | 19.5 | 19.5 | 20.7 | 20.7 | 20.9 | 21.5 | 22.1 | <0.0001 | |
| 65 to 74 | 22.8 | 21.9 | 21.2 | 20.3 | 20.5 | 19.6 | 19.3 | 19.3 | 19.3 | 19.8 | 19.7 | <0.0001 | |
| 75 to 84 | 24.8 | 24.7 | 24.0 | 23.9 | 23.8 | 23.3 | 22.2 | 21.3 | 21.2 | 20.3 | 19.7 | <0.0001 | |
| >84 | 12.8 | 13.9 | 14.0 | 14.3 | 14.5 | 15.5 | 14.6 | 15.5 | 16.1 | 15.3 | 15.5 | <0.0001 | |
| Race | |||||||||||||
| Non‐Hispanic White | 84.8 | 84.1 | 82.0 | 80.7 | 81.5 | 82.8 | 81.5 | 79.8 | 80.6 | 79.7 | 79.1 | <0.0001 | <0.0001 |
| Black | 6.8 | 7.0 | 7.6 | 7.9 | 8.3 | 6.4 | 7.6 | 8.9 | 8.0 | 8.1 | 10.1 | <0.0001 | |
| Hispanic | 4.7 | 4.9 | 5.7 | 6.9 | 5.7 | 6.2 | 6.1 | 5.7 | 5.1 | 5.8 | 5.6 | <0.0001 | |
| Asian | 1.4 | 1.3 | 1.6 | 1.8 | 1.8 | 1.5 | 1.7 | 2.0 | 2.0 | 1.9 | 1.9 | <0.0001 | |
| AI/NA | 0.2 | 0.4 | 0.2 | 0.1 | 0.3 | 0.2 | 0.4 | 0.6 | 0.9 | 0.6 | 0.8 | <0.0001 | |
| Other | 2.1 | 2.4 | 2.9 | 2.5 | 2.6 | 2.9 | 2.7 | 3.0 | 3.5 | 4.0 | 2.4 | <0.0001 | |
| Type of AMI | |||||||||||||
| STEMI | 43.2 | 40.7 | 39.8 | 37.4 | 34.7 | 33.1 | 34.1 | 32.3 | 31.4 | 30.1 | 30.1 | <0.0001 | <0.0001 |
| Comorbidities | |||||||||||||
| TIA | 3.8 | 3.7 | 3.8 | 3.7 | 3.8 | 3.8 | 3.8 | 3.8 | 4.2 | 4.0 | 3.9 | <0.0001 | <0.0001 |
| Heart failure | 29.8 | 29.1 | 29.0 | 31.3 | 32.2 | 31.6 | 30.5 | 30.6 | 30.4 | 30.6 | 30.8 | <0.0001 | <0.0001 |
| Prior MI | 65.2 | 66.7 | 67.9 | 69.1 | 69.6 | 70.7 | 72.9 | 73.2 | 75.1 | 76.8 | 76.6 | <0.0001 | <0.0001 |
| Hypertension | 47.0 | 49.0 | 51.1 | 52.7 | 54.5 | 56.1 | 58.0 | 59.1 | 61.3 | 62.9 | 63.7 | <0.0001 | <0.0001 |
| Renal failure | 5.5 | 6.1 | 6.5 | 7.3 | 7.9 | 10.0 | 13.6 | 18.2 | 19.2 | 20.5 | 21.4 | <0.0001 | <0.0001 |
| AFib | 16.3 | 16.8 | 16.8 | 16.9 | 17.3 | 17.6 | 17.3 | 17.0 | 15.9 | 15.9 | 16.0 | <0.0001 | <0.001 |
| Dyslipidemia | 27.7 | 30.5 | 34.3 | 35.6 | 38.8 | 41.9 | 44.7 | 47.9 | 49.6 | 51.9 | 53.9 | <0.0001 | <0.0001 |
| PAD | 3.6 | 3.7 | 4.0 | 4.2 | 4.2 | 4.1 | 4.3 | 4.7 | 5.1 | 5.1 | 5.1 | <0.0001 | <0.0001 |
| Cancer | 0.7 | 0.7 | 0.8 | 0.7 | 0.9 | 0.9 | 0.9 | 1.0 | 1.0 | 1.0 | 1.0 | <0.0001 | <0.0001 |
| Dementia | 1.0 | 0.9 | 0.8 | 0.9 | 0.7 | 0.8 | 0.8 | 0.7 | 0.8 | 0.7 | 0.6 | <0.0001 | 0.9 |
| Tobacco use | — | — | 18.4 | 18.6 | 19.5 | 21.3 | 22.9 | 23.4 | 24.3 | 25.4 | 25.4 | <0.0001 | <0.0001 |
| Obesity | — | — | 5.0 | 5.3 | 5.3 | 5.7 | 5.9 | 6.7 | 7.9 | 8.8 | 8.5 | <0.0001 | <0.0001 |
| In‐hospital, % | |||||||||||||
| Shock | 4.3 | 4.3 | 4.6 | 4.8 | 4.8 | 5.0 | 5.1 | 5.5 | 5.8 | 6.2 | 6.1 | <0.0001 | <0.0001 |
| VFib | 3.0 | 2.8 | 2.9 | 2.8 | 2.9 | 2.7 | 2.9 | 3.0 | 3.0 | 3.1 | 3.2 | <0.0001 | <0.0001 |
| Revasc | 58.2 | 60.5 | 61.3 | 62.7 | 63.2 | 65.4 | 68.0 | 67.5 | 67.8 | 70.9 | 71.1 | <0.0001 | <0.0001 |
AFib indicates atrial fibrillation; AI/NA, American Indian or Native American; AMI, acute myocardial infarction; MI, myocardial infarction; PAD, peripheral arterial disease; Revasc, coronary revascularization procedure; STEMI, ST‐segment elevation myocardial infarction; TIA, transient ischemic attack; VFib, ventricular fibrillation.
Figure 3.
Frequency of coronary revascularization during hospitalization. From 2000 to 2010, there was a 23% increase in the use of coronary revascularization procedures in patients with acute myocardial infarction and type 2 diabetes mellitus. The majority (>75%) of these procedures were percutaneous coronary interventions.
In‐Hospital Mortality in AMI With T2DM
The CMR decreased by 3.1% in all AMI patients (year 2000, 8.4%; year 2010, 5.3%; P<0.0001). Among AMI patients with T2DM, there was a 3.2% absolute reduction in CMR (year 2000, 8.0%; year 2010, 4.8%; P<0.0001). Figure 4 depicts the significant downward trend in CMR and age‐standardized mortality rate among AMI patients with T2DM stratified on sex. Although women with T2DM had a higher initial age‐standardized mortality rate than men with T2DM, there was a greater absolute decline in mortality over the study period among women (women, −3.4%; men, −2.3%; P<0.0001).
Figure 4.
Decrease in crude and age‐standardized mortality rates by sex. Crude and age‐standardized mortality rates in both men and women decreased significantly from 2000 to 2010.
Rate Decomposition Analysis
Using the method of rate decomposition as described under “Methods,” the rate effect for men was 0.025 and the composition effect was 0.001. The total, 0.025+0.001, or 0.026, matches the difference in CMR for men from 2000 to 2010 and suggests that a change in stratum‐specific risk for mortality is the main driver of the observed decrease in mortality in men and is not due to differences in age structure of the populations. For women, the rate effect was 0.035 and the composition effect was 0.001. The sum of these 2 components, 0.034, matches the difference in CMR for women from 2000 to 2010 and suggests that, as with men, the main driver for the decrease in mortality is a change in risk structure rather than a change in age structure of the populations.
Characteristics Associated With In‐Hospital Mortality in Patients With AMI
Table 4 reports adjusted ORs and their respective 95% CIs for the associations between relevant clinical, demographic, year, and hospital‐level characteristics and in‐hospital mortality. Most notable is the significant overall adverse impact of DM on mortality (adjusted OR 1.069, 95% CI 1.051 to 1.087; P<0.001). The magnitude of this association, however, did not change significantly over time, as indicated by the adjusted OR for the interaction term DM×year (OR 1.04, 95% CI 0.989 to 1.090; P=0.132). There was a significant increase in the odds of mortality for each age group compared with the reference age group of <55 years. The use of coronary revascularization procedures was strongly and inversely associated with the odds of in‐hospital death.
Table 4.
Characteristics Associated With In‐Hospital Mortality in All Patients With Acute Myocardial Infarction, 2002–2010
| Effect | Odds Ratio | 95% CI | P Value |
|---|---|---|---|
| T2DM | 1.069 | 1.051 to 1.087 | <0.0001 |
| Year | 0.751 | 0.719 to 0.785 | <0.0001 |
| Female | 1.071 | 1.05 to 1.093 | <0.0001 |
| Age (vs <55), y | |||
| 55 to 64 | 1.651 | 1.578 to 1.727 | <0.0001 |
| 65 to 74 | 2.477 | 2.371 to 2.588 | <0.0001 |
| 75 to 84 | 3.93 | 3.758 to 4.109 | <0.0001 |
| >84 | 5.996 | 5.71 to 6.297 | <0.0001 |
| Race (vs white) | |||
| Black | 1.019 | 0.972 to 1.069 | 0.4267 |
| Hispanic | 1.072 | 1.016 to 1.132 | 0.0114 |
| Asian | 1.035 | 0.964 to 1.111 | 0.3471 |
| AI/NA | 0.871 | 0.738 to 1.028 | 0.1013 |
| Other | 0.979 | 0.916 to 1.046 | 0.5301 |
| STEMI | 1.308 | 1.276 to 1.34 | <0.0001 |
| TIA/stroke | 1.827 | 1.758 to 1.898 | <0.0001 |
| Heart failure | 1.228 | 1.199 to 1.257 | <0.0001 |
| Prior MI | 0.536 | 0.522 to 0.551 | <0.0001 |
| Hypertension | 0.773 | 0.756 to 0.791 | <0.0001 |
| Renal failure | 2.24 | 2.187 to 2.294 | <0.0001 |
| AFib | 1.072 | 1.047 to 1.096 | <0.0001 |
| Dyslipidemia | 0.467 | 0.456 to 0.479 | <0.0001 |
| PAD | 1.165 | 1.119 to 1.213 | <0.0001 |
| Cancer | 2.427 | 2.257 to 2.611 | <0.0001 |
| Dementia | 0.989 | 0.914 to 1.069 | 0.7736 |
| Smoke | 0.658 | 0.632 to 0.686 | <0.0001 |
| Obesity | 0.786 | 0.75 to 0.824 | <0.0001 |
| VFib | 4.986 | 4.751 to 5.233 | <0.0001 |
| Shock | 9.425 | 9.104 to 9.757 | <0.0001 |
| Revasc | 0.395 | 0.381 to 0.409 | <0.0001 |
| Hospital size (medium vs small) | 1.021 | 0.964 to 1.081 | 0.4781 |
| Hospital size (large vs small) | 0.993 | 0.941 to 1.049 | 0.8112 |
| Hospital location (urban vs rural) | 0.873 | 0.828 to 0.921 | <0.0001 |
AFib indicates atrial fibrillation; AI/NA, American Indian or Native American; MI, myocardial infarction; PAD, peripheral arterial disease; Revasc, coronary revascularization procedure; STEMI, ST‐segment elevation myocardial infarction; T2DM, type 2 diabetes mellitus; TIA, transient ischemic attack; VFib, ventricular fibrillation.
Characteristics Associated With In‐Hospital Mortality in Patients With AMI and T2DM
A primary interest in the present study centers on the group with AMI and coexistent T2DM. Table 5 reports adjusted ORs and their respective 95% CIs for the association between the same covariates listed in Table 4 and in‐hospital mortality. In the fully adjusted model, the nominal increase in the odds of mortality among women was not significant (OR 1.033, 95% CI 0.997 to 1.072; P=0.0765). Older age (compared with the reference age group of <55 years) was significantly associated with mortality. There was a significant decrease in the odds of death for each successive year when year was modeled as a continuous variable (OR 0.774, 95% CI 0.722 to 0.830; P<0.0001). Noted again is a strong and inverse association between the use of coronary revascularization procedures and in‐hospital death (OR 0.292, 95% CI 0.276 to 0.308; P<0.0001).
Table 5.
Characteristics Associated With In‐Hospital Mortality in Patients With Type 2 Diabetes Mellitus, 2002–2010
| Effect | Odds Ratio | 95% CI | P Value |
|---|---|---|---|
| Year | 0.774 | 0.722 to 0.83 | <0.0001 |
| Female | 1.033 | 0.997 to 1.072 | 0.0765 |
| Age (vs <55), y | |||
| 55 to 64 | 1.473 | 1.349 to 1.607 | <0.0001 |
| 65 to 74 | 2.001 | 1.835 to 2.182 | <0.0001 |
| 75 to 84 | 2.762 | 2.536 to 3.007 | <0.0001 |
| >84 | 3.475 | 3.165 to 3.816 | <0.0001 |
| Race (vs white) | |||
| Black | 0.926 | 0.869 to 0.986 | 0.0168 |
| Hispanic | 1.016 | 0.95 to 1.088 | 0.6398 |
| Asian | 0.873 | 0.781 to 0.976 | 0.0173 |
| AI/NA | 1.095 | 0.84 to 1.427 | 0.5016 |
| Other | 0.902 | 0.809 to 1.006 | 0.0636 |
| STEMI | 1.76 | 1.681 to 1.843 | <0.0001 |
| TIA/stroke | 1.753 | 1.635 to 1.879 | <0.0001 |
| Heart failure | 1.085 | 1.042 to 1.13 | <0.0001 |
| Prior MI | 0.754 | 0.724 to 0.785 | <0.0001 |
| Hypertension | 0.8 | 0.768 to 0.832 | <0.0001 |
| Renal failure | 1.991 | 1.91 to 2.075 | <0.0001 |
| AFib | 1.092 | 1.045 to 1.141 | <0.0001 |
| Dyslipidemia | 0.537 | 0.515 to 0.56 | <0.0001 |
| PAD | 1.127 | 1.06 to 1.199 | 0.0001 |
| Cancer | 1.751 | 1.491 to 2.056 | <0.0001 |
| Dementia | 0.929 | 0.803 to 1.075 | 0.3232 |
| Smoke | 0.747 | 0.69 to 0.809 | <0.0001 |
| Obesity | 0.825 | 0.771 to 0.883 | <0.0001 |
| VFib | 9.508 | 8.653 to 10.447 | <0.0001 |
| Shock | 12.086 | 11.409 to 12.804 | <0.0001 |
| Revasc | 0.292 | 0.276 to 0.308 | <0.0001 |
| Hospital size (medium vs small) | 1.107 | 1.025 to 1.195 | 0.0095 |
| Hospital size (large vs small) | 1.225 | 1.142 to 1.314 | <0.0001 |
| Hospital location (urban vs rural) | 1.137 | 1.055 to 1.224 | 0.0007 |
AFib indicates atrial fibrillation; AI/NA, American Indian or Native American; MI, myocardial infarction; PAD, peripheral arterial disease; Revasc, coronary revascularization procedure; STEMI, ST‐segment elevation myocardial infarction; TIA, transient ischemic attack; VFib, ventricular fibrillation.
Modification of the Effect of Time on Mortality by Age and Sex
As described under “Methods,” interaction terms were added to the final multivariable regression model with year modeled as a continuous linear function. The triple interaction term year×sex×age was statistically significant (OR 0.774, 95% CI 0.726 to 0.824; P<0.0001), as were the interaction terms year×age (OR 0.94, 95% CI 0.915 to 0.965; P<0.0001) and year×sex (OR 1.157, 95% CI 1.088 to 1.230; P<0.0001). Figure 5 attempts to graphically depict this complex set of interactions another way by showing the probability of death (y‐axis) as a function of time (x‐axis) by sex (taking into account the interaction of age and sex), whereas Figure 6 shows the probability of death as a function of time for each age category (taking into account the interaction of age category and sex). The presence of effect modification is reflected in the nonparallel relationship between group‐specific plots. Notably, the annual change in the probability of (lower) mortality was less in men than in women and less in younger patients than in older patients.
Figure 5.
Effect modification of time (year) by sex. Probability of death plotted against time stratified by sex. The nonparallel nature of the plots is consistent with a statistically significant interaction between time and sex. The probability of death for each sex takes into account the variation in probability of death with age.
Figure 6.
Effect modification of time (year) by age. Probability of death plotted against time stratified by age. The nonparallel nature of the plots is consistent with a statistically significant interaction between time and age. The probability of death within each age category takes into account the variation in probability of death with sex.
Sensitivity Analyses
A sensitivity analysis was undertaken to assess the effect of potential information bias in the coding of type of AMI by limiting the analysis to only patients with STEMI (Tables 6,6 through 8). In STEMI patients with T2DM (Table 7), in‐hospital crude mortality was higher compared with the overall STEMI group (Table 6), although mortality rates in all patients with STEMI and those with T2DM significantly declined over the observation period. In patients with STEMI, the frequency of CV risk factors increased over time, although the impact of diabetes was similar to the effect in AMI patients overall (T2DM: OR 1.11, 95% CI 1.066 to 1.156; P<0.0001) (Table 9).
Table 6.
Pre‐ and In‐Hospital Characteristics Among Patients With ST‐Segment Elevation Myocardial Infarction
| 2000 | 2001 | 2002 | 2003 | 2004 | 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | P Value | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| N (sample) | 64 918 | 316 871 | 59 433 | 54 960 | 46 332 | 41 405 | 43 559 | 37 249 | 37 556 | 34 200 | 32 900 | |
| N (weighted) | 316 871 | 10 744 | 286 615 | 262 937 | 225 706 | 202 291 | 211 620 | 184 115 | 184 442 | 172 006 | 164 297 | |
| Prehospital | ||||||||||||
| Sex, % | ||||||||||||
| Male | 63.23 | 63.76 | 64.03 | 64.72 | 65.02 | 65.50 | 66.95 | 66.85 | 67.17 | 68.39 | 68.74 | <0.0001 |
| Female | 36.77 | 36.24 | 35.97 | 35.28 | 34.98 | 34.50 | 33.06 | 33.15 | 32.83 | 31.61 | 31.26 | <0.0001 |
| Age, y, % | ||||||||||||
| <55 | 25.81 | 26.13 | 27.19 | 27.84 | 27.31 | 28.11 | 30.00 | 29.99 | 29.40 | 30.46 | 29.96 | <0.0001 |
| 55 to 64 | 22.08 | 22.25 | 22.68 | 24.08 | 24.38 | 25.01 | 25.70 | 25.90 | 26.68 | 26.90 | 28.15 | <0.0001 |
| 65 to 74 | 23.24 | 22.72 | 22.03 | 21.11 | 20.99 | 20.15 | 19.62 | 19.27 | 20.18 | 20.09 | 20.04 | <0.0001 |
| 75 to 84 | 20.47 | 20.12 | 19.31 | 18.58 | 18.88 | 17.83 | 16.53 | 16.29 | 15.51 | 14.75 | 14.33 | <0.0001 |
| >84 | 8.40 | 8.78 | 8.79 | 8.39 | 8.44 | 8.90 | 8.15 | 8.55 | 8.23 | 7.80 | 7.51 | <0.0001 |
| Race, % | ||||||||||||
| White | 83.80 | 82.85 | 80.53 | 79.52 | 80.77 | 81.80 | 81.19 | 78.52 | 79.37 | 78.55 | 78.16 | <0.0001 |
| Black | 6.16 | 6.32 | 6.87 | 7.28 | 6.99 | 5.56 | 6.51 | 7.78 | 7.06 | 7.21 | 8.07 | <0.0001 |
| Hispanic | 5.55 | 6.05 | 6.91 | 7.85 | 6.64 | 7.42 | 7.04 | 7.37 | 6.28 | 6.85 | 7.36 | <0.0001 |
| Asian | 1.69 | 1.53 | 2.06 | 2.20 | 2.22 | 1.78 | 1.95 | 2.08 | 2.23 | 2.20 | 2.37 | <0.0001 |
| AI/AN | 0.26 | 0.37 | 0.22 | 0.18 | 0.30 | 0.27 | 0.43 | 0.57 | 0.99 | 0.64 | 0.93 | <0.0001 |
| Other | 2.53 | 2.88 | 3.41 | 2.98 | 3.10 | 3.18 | 2.88 | 3.68 | 4.09 | 4.56 | 3.11 | <0.0001 |
| Comorbidities, % | ||||||||||||
| T2DM | 19.73 | 20.42 | 21.26 | 21.71 | 22.29 | 22.68 | 23.34 | 24.29 | 24.56 | 24.74 | 25.42 | <0.0001 |
| TIA/stroke | 3.05 | 3.00 | 3.02 | 2.84 | 2.91 | 2.69 | 2.75 | 2.69 | 2.90 | 2.52 | 2.80 | <0.0001 |
| Heart failure | 25.39 | 24.16 | 24.42 | 25.20 | 25.43 | 24.24 | 23.45 | 24.04 | 23.23 | 23.39 | 22.81 | <0.0001 |
| Prior MI | 66.56 | 68.68 | 70.59 | 72.63 | 74.39 | 76.10 | 78.77 | 80.01 | 82.87 | 84.08 | 84.13 | <0.0001 |
| Hypertension | 46.32 | 48.07 | 50.07 | 51.46 | 53.05 | 54.43 | 56.36 | 57.99 | 59.42 | 60.81 | 61.32 | <0.0001 |
| Renal failure | 4.63 | 5.04 | 5.59 | 5.77 | 6.12 | 7.52 | 9.93 | 12.97 | 13.66 | 14.38 | 14.84 | <0.0001 |
| AFib | 12.73 | 12.85 | 12.99 | 12.26 | 12.80 | 12.53 | 12.16 | 12.00 | 11.25 | 11.19 | 10.95 | <0.0001 |
| Dyslipidemia | 30.38 | 33.50 | 37.68 | 39.52 | 43.26 | 46.92 | 49.91 | 53.47 | 55.06 | 57.06 | 58.82 | <0.0001 |
| PAD | 2.94 | 2.98 | 3.27 | 3.29 | 3.31 | 3.25 | 3.41 | 3.67 | 3.87 | 3.65 | 3.66 | <0.0001 |
| Cancer | 0.57 | 0.61 | 0.58 | 0.57 | 0.60 | 0.71 | 0.60 | 0.73 | 0.71 | 0.65 | 0.69 | <0.0001 |
| Dementia | 0.64 | 0.63 | 0.52 | 0.51 | 0.44 | 0.51 | 0.48 | 0.40 | 0.38 | 0.39 | 0.29 | <0.0001 |
| Tobacco use | — | — | 24.00 | 25.39 | 26.93 | 29.08 | 31.12 | 32.07 | 33.86 | 34.73 | 34.69 | <0.0001 |
| Obesity | — | — | 11.4075 | 12.4402 | 12.5151 | 14.701 | 14.9526 | 15.7799 | 17.9707 | 19.4745 | 19.9242 | <0.0001 |
| In‐hospital, % | ||||||||||||
| Shock | 6.19 | 6.51 | 7.01 | 7.23 | 7.52 | 8.25 | 8.39 | 9.28 | 9.96 | 10.71 | 10.75 | <0.001 |
| VFib | 4.27 | 4.23 | 4.61 | 4.59 | 4.85 | 4.88 | 5.11 | 5.46 | 5.72 | 6.16 | 6.19 | <0.0001 |
| Revasc | 70.52 | 72.84 | 74.60 | 76.56 | 77.72 | 81.19 | 83.38 | 84.68 | 86.53 | 89.02 | 89.65 | <0.0001 |
| Mortality | 9.09 | 8.57 | 8.36 | 7.93 | 7.77 | 7.65 | 6.85 | 7.14 | 6.93 | 6.63 | 6.32 | <0.0001 |
AFib indicates atrial fibrillation; AI/NA, American Indian or Native American; MI, myocardial infarction; PAD, peripheral arterial disease; Revasc, coronary revascularization procedure; T2DM, type 2 diabetes mellitus; TIA, transient ischemic attack; VFib, ventricular fibrillation.
Table 8.
Pre‐ and In‐Hospital Characteristics Among Patients With ST‐Segment Elevation Myocardial Infarction Without Type 2 Diabetes Mellitus
| 2000 | 2001 | 2002 | 2003 | 2004 | 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | P Value | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| N (sample) | 12 812 | 12 228 | 12 604 | 11 901 | 10 320 | 9393 | 10 161 | 9052 | 9231 | 8462 | 8355 | |
| N (weighted) | 62 539 | 61 147 | 60 933 | 57 118 | 50 340 | 45 901 | 49 407 | 44 722 | 45 303 | 42 561 | 41 762 | |
| Prehospital, % | ||||||||||||
| Sex, % | ||||||||||||
| Male | 59.27 | 59.49 | 59.69 | 59.57 | 61.00 | 61.73 | 62.70 | 63.73 | 63.35 | 64.36 | 63.95 | <0.0001 |
| Female | 40.73 | 40.51 | 40.31 | 40.43 | 39.00 | 38.27 | 37.30 | 36.27 | 36.65 | 35.64 | 36.05 | <0.0001 |
| Age, y, % | ||||||||||||
| <55 | 19.84 | 20.23 | 21.95 | 21.70 | 21.55 | 23.46 | 25.61 | 25.75 | 24.67 | 26.23 | 25.93 | <0.0001 |
| 55 to 64 | 23.81 | 24.05 | 24.12 | 25.55 | 26.26 | 27.42 | 27.16 | 27.05 | 27.73 | 27.16 | 29.43 | <0.0001 |
| 65 to 74 | 27.35 | 26.63 | 25.46 | 25.14 | 24.77 | 22.53 | 23.12 | 22.28 | 23.98 | 23.07 | 22.90 | <0.0001 |
| 75 to 84 | 21.71 | 21.99 | 20.93 | 20.09 | 20.48 | 19.29 | 17.40 | 18.02 | 16.80 | 16.46 | 15.46 | <0.0001 |
| >84 | 7.29 | 7.09 | 7.54 | 7.52 | 6.95 | 7.30 | 6.70 | 6.89 | 6.82 | 7.08 | 6.29 | <0.0001 |
| Race, % | ||||||||||||
| White | 77.22 | 75.90 | 72.99 | 71.11 | 73.23 | 74.26 | 73.74 | 70.58 | 72.38 | 70.77 | 69.78 | <0.0001 |
| Black | 8.35 | 8.45 | 9.19 | 9.53 | 9.27 | 7.28 | 8.66 | 10.02 | 9.64 | 9.69 | 10.40 | <0.0001 |
| Hispanic | 8.85 | 9.93 | 10.28 | 12.40 | 10.35 | 11.57 | 10.93 | 11.42 | 9.52 | 10.42 | 11.22 | <0.0001 |
| Asian | 2.23 | 2.00 | 3.04 | 3.05 | 3.11 | 2.53 | 2.60 | 2.99 | 2.87 | 2.96 | 3.39 | <0.0001 |
| AI/NA | 0.43 | 0.34 | 0.43 | 0.28 | 0.45 | 0.42 | 0.68 | 0.88 | 1.12 | 0.68 | 1.26 | <0.0001 |
| Other | 2.93 | 3.39 | 4.06 | 3.64 | 3.60 | 3.94 | 3.39 | 4.11 | 4.48 | 5.49 | 3.95 | <0.0001 |
| CV risk factors, % | ||||||||||||
| TIA | 3.23 | 3.56 | 3.23 | 3.22 | 3.27 | 3.25 | 3.04 | 3.13 | 3.26 | 2.69 | 3.12 | <0.0001 |
| Heart failure | 31.51 | 29.93 | 30.67 | 31.37 | 31.57 | 29.91 | 28.15 | 30.14 | 27.86 | 28.51 | 27.27 | <0.0001 |
| Prior MI | 68.05 | 71.35 | 71.98 | 74.73 | 76.40 | 78.75 | 80.52 | 82.42 | 84.91 | 85.72 | 85.84 | <0.0001 |
| Hypertension | 60.00 | 61.62 | 64.15 | 66.09 | 68.27 | 68.39 | 72.04 | 72.71 | 72.90 | 75.88 | 75.24 | <0.0001 |
| Renal failure | 5.38 | 5.71 | 6.89 | 7.30 | 7.37 | 9.68 | 13.41 | 18.51 | 19.51 | 20.53 | 20.33 | <0.0001 |
| AFib | 12.73 | 12.73 | 12.73 | 12.73 | 12.73 | 12.73 | 12.73 | 12.73 | 12.73 | 12.73 | 12.73 | <0.0001 |
| Dyslipidemia | 33.37 | 36.28 | 42.36 | 45.33 | 49.20 | 53.11 | 57.06 | 60.55 | 62.59 | 63.49 | 65.06 | <0.0001 |
| PAD | 4.29 | 4.62 | 4.47 | 4.78 | 4.74 | 4.70 | 4.58 | 5.32 | 5.65 | 5.17 | 5.01 | <0.0001 |
| Cancer | 4.29 | 4.62 | 4.47 | 4.78 | 4.74 | 4.70 | 4.58 | 5.32 | 5.65 | 5.17 | 5.01 | <0.0001 |
| Dementia | 0.71 | 0.66 | 0.56 | 0.60 | 0.62 | 0.53 | 0.52 | 0.48 | 0.48 | 0.54 | 0.28 | <0.0001 |
| Tobacco use | — | — | 15.83 | 16.10 | 18.09 | 20.41 | 22.22 | 23.99 | 25.35 | 24.90 | 26.38 | <0.0001 |
| Obesity | — | — | 11.41 | 12.44 | 12.52 | 14.70 | 14.95 | 15.78 | 17.97 | 19.47 | 19.92 | <0.0001 |
| In‐hospital, % | ||||||||||||
| Shock | 6.30 | 6.66 | 7.14 | 7.50 | 7.74 | 8.43 | 8.66 | 9.45 | 10.11 | 10.72 | 10.72 | <0.001 |
| VFib | 4.67 | 4.64 | 5.09 | 5.08 | 5.43 | 5.37 | 5.60 | 6.00 | 6.29 | 6.81 | 6.93 | <0.0001 |
| Re‐Vasc | 66.72 | 68.47 | 70.35 | 72.12 | 73.82 | 78.31 | 80.75 | 82.56 | 84.56 | 86.89 | 87.84 | <0.0001 |
AFib indicates atrial fibrillation; AI/NA, American Indian or Native American; CV, cardiovascular; MI, myocardial infarction; PAD, peripheral arterial disease; Revasc, coronary revascularization procedure; TIA, transient ischemic attack; VFib, ventricular fibrillation.
Table 7.
Pre‐ and In‐Hospital Characteristics Among Patients With ST‐Segment Elevation Myocardial Infarction and Type 2 Diabetes Mellitus
| 2000 | 2001 | 2002 | 2003 | 2004 | 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | P Value | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| N (sample) | 12 812 | 12 228 | 12 604 | 11 901 | 10 320 | 9393 | 10 161 | 9052 | 9231 | 8462 | 8355 | |
| N (weighted) | 62 539 | 61 147 | 60 933 | 57 118 | 50 340 | 45 901 | 49 407 | 44 722 | 45 303 | 42 561 | 41 762 | |
| Prehospital, % | ||||||||||||
| Sex, % | ||||||||||||
| Male | 59.27 | 59.49 | 59.69 | 59.57 | 61.00 | 61.73 | 62.70 | 63.73 | 63.35 | 64.36 | 63.95 | <0.0001 |
| Female | 40.73 | 40.51 | 40.31 | 40.43 | 39.00 | 38.27 | 37.30 | 36.27 | 36.65 | 35.64 | 36.05 | <0.0001 |
| Age, y, % | ||||||||||||
| <55 | 19.84 | 20.23 | 21.95 | 21.70 | 21.55 | 23.46 | 25.61 | 25.75 | 24.67 | 26.23 | 25.93 | <0.0001 |
| 55 to 64 | 23.81 | 24.05 | 24.12 | 25.55 | 26.26 | 27.42 | 27.16 | 27.05 | 27.73 | 27.16 | 29.43 | <0.0001 |
| 65 to 74 | 27.35 | 26.63 | 25.46 | 25.14 | 24.77 | 22.53 | 23.12 | 22.28 | 23.98 | 23.07 | 22.90 | <0.0001 |
| 75 to 84 | 21.71 | 21.99 | 20.93 | 20.09 | 20.48 | 19.29 | 17.40 | 18.02 | 16.80 | 16.46 | 15.46 | <0.0001 |
| >84 | 7.29 | 7.09 | 7.54 | 7.52 | 6.95 | 7.30 | 6.70 | 6.89 | 6.82 | 7.08 | 6.29 | <0.0001 |
| Race, % | ||||||||||||
| White | 77.22 | 75.90 | 72.99 | 71.11 | 73.23 | 74.26 | 73.74 | 70.58 | 72.38 | 70.77 | 69.78 | <0.0001 |
| Black | 8.35 | 8.45 | 9.19 | 9.53 | 9.27 | 7.28 | 8.66 | 10.02 | 9.64 | 9.69 | 10.40 | <0.0001 |
| Hispanic | 8.85 | 9.93 | 10.28 | 12.40 | 10.35 | 11.57 | 10.93 | 11.42 | 9.52 | 10.42 | 11.22 | <0.0001 |
| Asian | 2.23 | 2.00 | 3.04 | 3.05 | 3.11 | 2.53 | 2.60 | 2.99 | 2.87 | 2.96 | 3.39 | <0.0001 |
| AI/AN | 0.43 | 0.34 | 0.43 | 0.28 | 0.45 | 0.42 | 0.68 | 0.88 | 1.12 | 0.68 | 1.26 | <0.0001 |
| Other | 2.93 | 3.39 | 4.06 | 3.64 | 3.60 | 3.94 | 3.39 | 4.11 | 4.48 | 5.49 | 3.95 | <0.0001 |
| CV risk factors, % | ||||||||||||
| TIA/stroke | 3.23 | 3.56 | 3.23 | 3.22 | 3.27 | 3.25 | 3.04 | 3.13 | 3.26 | 2.69 | 3.12 | <0.0001 |
| Heart failure | 31.51 | 29.93 | 30.67 | 31.37 | 31.57 | 29.91 | 28.15 | 30.14 | 27.86 | 28.51 | 27.27 | <0.0001 |
| Prior MI | 68.05 | 71.35 | 71.98 | 74.73 | 76.40 | 78.75 | 80.52 | 82.42 | 84.91 | 85.72 | 85.84 | <0.0001 |
| Hypertension | 60.00 | 61.62 | 64.15 | 66.09 | 68.27 | 68.39 | 72.04 | 72.71 | 72.90 | 75.88 | 75.24 | <0.0001 |
| Renal failure | 5.38 | 5.71 | 6.89 | 7.30 | 7.37 | 9.68 | 13.41 | 18.51 | 19.51 | 20.53 | 20.33 | <0.0001 |
| AFib | 12.69 | 12.59 | 12.64 | 11.89 | 11.75 | 12.08 | 11.95 | 12.28 | 11.44 | 10.92 | 11.20 | <0.0001 |
| Dyslipidemia | 33.37 | 36.28 | 42.36 | 45.33 | 49.20 | 53.11 | 57.06 | 60.55 | 62.59 | 63.49 | 65.06 | <0.0001 |
| PAD | 4.29 | 4.62 | 4.47 | 4.78 | 4.74 | 4.70 | 4.58 | 5.32 | 5.65 | 5.17 | 5.01 | <0.0001 |
| Cancer | 4.29 | 4.62 | 4.47 | 4.78 | 4.74 | 4.70 | 4.58 | 5.32 | 5.65 | 5.17 | 5.01 | <0.0001 |
| Dementia | 0.71 | 0.66 | 0.56 | 0.60 | 0.62 | 0.53 | 0.52 | 0.48 | 0.48 | 0.54 | 0.28 | <0.0001 |
| Tobacco use | — | — | 15.83 | 16.10 | 18.09 | 20.41 | 22.22 | 23.99 | 25.35 | 24.90 | 26.38 | <0.0001 |
| Obesity | — | — | 11.41 | 12.44 | 12.52 | 14.70 | 14.95 | 15.78 | 17.97 | 19.47 | 19.92 | <0.0001 |
| In‐hospital, % | ||||||||||||
| Shock | 5.77 | 5.95 | 6.54 | 6.28 | 6.78 | 7.64 | 7.50 | 8.76 | 9.52 | 10.65 | 10.82 | <0.001 |
| VFib | 2.66 | 2.65 | 2.82 | 2.82 | 2.80 | 3.22 | 3.47 | 3.75 | 3.96 | 4.20 | 4.02 | <0.0001 |
| Revasc | 66.72 | 68.47 | 70.35 | 72.12 | 73.82 | 78.31 | 80.75 | 82.56 | 84.56 | 86.89 | 87.84 | <0.0001 |
| Mortality | 9.6 | 8.48 | 8.65 | 8.27 | 8.16 | 8.02 | 6.78 | 7.41 | 6.99 | 7.24 | 6.43 | <0.0001 |
AFib indicates atrial fibrillation; AI/NA, American Indian or Native American; CV, cardiovascular; MI, myocardial infarction; PAD, peripheral arterial disease; Revasc, coronary revascularization procedure; TIA, transient ischemic attack; VFib, ventricular fibrillation.
Table 9.
Characteristics Associated With In‐Hospital Mortality in Patients With ST‐Segment Elevation Myocardial Infarction, 2002–2010
| Effect | OR | 95% CI | P Value |
|---|---|---|---|
| T2DM | 1.11 | 1.066 to 1.156 | <0.0001 |
| Year | 0.869 | 0.819 to 0.922 | <0.0001 |
| Female | 1.226 | 1.182 to 1.27 | <0.0001 |
| Age (vs <55), y | |||
| 55 to 64 | 1.519 | 1.425 to 1.62 | <0.0001 |
| 65 to 74 | 2.243 | 2.106 to 2.39 | <0.0001 |
| 75 to 84 | 3.445 | 3.23 to 3.674 | <0.0001 |
| >84 | 4.662 | 4.333 to 5.017 | <0.0001 |
| Race (vs white) | |||
| Black | 1.026 | 0.944 to 1.115 | 0.5413 |
| Hispanic | 1.069 | 0.992 to 1.151 | 0.0787 |
| Asian | 0.947 | 0.839 to 1.068 | 0.3749 |
| AI/NA | 1.037 | 0.799 to 1.346 | 0.783 |
| Other | 1.037 | 0.94 to 1.144 | 0.4689 |
| TIA | 1.982 | 1.833 to 2.143 | <0.0001 |
| Heart failure | 0.971 | 0.934 to 1.01 | 0.146 |
| Prior MI | 0.755 | 0.724 to 0.787 | <0.0001 |
| Hypertension | 0.864 | 0.833 to 0.896 | <0.0001 |
| Renal failure | 2.528 | 2.418 to 2.642 | <0.0001 |
| AFib | 1.098 | 1.052 to 1.145 | <0.0001 |
| VFib | 3.603 | 3.386 to 3.834 | <0.0001 |
| Shock | 8.417 | 8.016 to 8.839 | <0.0001 |
| Dyslipidemia | 0.468 | 0.448 to 0.489 | <0.0001 |
| PAD | 1.189 | 1.094 to 1.293 | <0.0001 |
| Cancer | 2.177 | 1.891 to 2.508 | <0.0001 |
| Dementia | 0.896 | 0.765 to 1.049 | 0.1708 |
| Smoke | 0.691 | 0.653 to 0.73 | <0.0001 |
| Obesity | 0.881 | 0.814 to 0.953 | 0.0015 |
| Revasc | 0.292 | 0.277 to 0.309 | <0.0001 |
| Hospital size (medium vs small) | 1.156 | 1.068 to 1.251 | 0.0003 |
| Hospital size (large vs small) | 1.275 | 1.184 to 1.373 | <0.0001 |
| Hospital location (urban vs rural) | 1.261 | 1.172 to 1.358 | <0.0001 |
AFib indicates atrial fibrillation; AI/NA, American Indian or Native American; MI, myocardial infarction; OR, odds ratio; PAD, peripheral arterial disease; Revasc, coronary revascularization procedure; T2DM, type 2 diabetes mellitus; TIA, transient ischemic attack; VFib, ventricular fibrillation.
Additional sensitivity analyses were undertaken in 2 important subgroups: those coded for adjunctive insulin use (ICD‐9‐CM V58.67) and those coded for morbid obesity (ICD‐9‐CM 278.01). Adjunctive insulin use was identified in only 1.6% of the sample of patients with AMI and coexistent T2DM (Table 10) and is accompanied by higher risk of in‐hospital death for this subgroup compared with the group with AMI and coexistent T2DM overall. Similarly, morbid obesity was identified in 1.8% of the entire sample and is accompanied by higher risk of in‐hospital death (Table 10). There was no significant change in death rate in either subgroup over the observation time.
Table 10.
In‐Hospital Mortality in Patients With Acute Myocardial Infarction and Coexistent Type 2 Diabetes Mellitus in Additional Subgroups
| 2000 | 2001 | 2002 | 2003 | 2004 | 2005 | 2006 | 2007 | 2008 | 2009 | 2010 |
|---|---|---|---|---|---|---|---|---|---|---|
| Insulin use | ||||||||||
| — | — | — | – | 10.71 | 11.52 | 11.69 | 11.25 | 11.62 | 11.67 | 10.89 |
| Morbid obesity | ||||||||||
| 10.83 | 9.53 | 11.62 | 11.76 | 11.40 | 11.67 | 11.89 | 12.01 | 11.17 | 11.60 | 11.96 |
Data are shown as percentages.
Discussion
Using a nationally representative sample of more than 1.5 million patients hospitalized with AMI from 2000 to 2010, our findings support the following conclusions. First, over the past decade, there have been significant increases in the prevalence of T2DM and the prevalences of CV risk factors in AMI patients. Second, despite this increased disease burden, there has been a 40% reduction in in‐hospital mortality over time. Third, the reduction in mortality risk varied by age and sex. Fourth, the adverse impact of T2DM on in‐hospital mortality has not significantly changed over this time period.
Possible explanations for these observations include earlier diagnosis and aggressive management of this traditionally high‐risk group, inclusion of relatively lower risk subjects compared with earlier time periods, secular trends in AMI‐related incidence and mortality, or a combination of all of these. We explored the likelihood of inclusion of such potentially lower risk subjects as well as the likelihood of more aggressive ascertainment and treatment, either or both of which might contribute to an observed increase in prevalence of T2DM or improvement in in‐hospital mortality over the time of this study. The serum glucose threshold for a diagnosis of DM was modified in 1997 and 199923 and antedate the time period of the current study. HbA1c was not recommended for use as a diagnostic test for DM until 2009–2010.24 The present data from 2000 to 2010 would not have been affected by the changes in the serum glucose threshold for diagnosis of DM, and the widespread use of HbA1c as a diagnostic tool is not relevant to the time frame of this study. Consequently, the present observations are less subject to potential misclassification and/or spectrum bias. The present data demonstrate a decline in in‐hospital mortality beginning well before 2009–2010, minimizing potential spectrum bias from the use of HbA1c criteria.
Potentially influencing the observed decrease in in‐hospital mortality is the increasing prevalence of patients with non‐STEMI‐type AMI, a group felt to be at lower risk. Similar to other reports,1,6,25 we observed an increase in the absolute and relative prevalence of non‐STEMI from 2000 to 2010. Decreasing overall mortality rates, although potentially influenced by changing biomarker criteria for AMI, antedate the current universal use of troponin as the preferred biomarker for the diagnosis of AMI. In addition, ICD‐9‐CM codes for AMI changed in 2005.26 Despite higher in‐hospital mortality for patients with STEMI, both before and after 2005, the difference in mortality rates between STEMI and non‐STEMI was not statistically significant (2000–2005 difference, 2.17%; 2006–2010 difference, 2.26%; P=0.6), and mortality decreased equally over time in both STEMI and non‐STEMI groups. The results of our sensitivity analysis of the STEMI‐only patients (thereby obviating much of the uncertainty in diagnosis related to the increasing reliance on biomarker criteria) confirm the observed decrease in mortality in the AMI group overall and support the thesis that the diabetic condition itself confers an increased risk of mortality. Particularly high‐risk diabetic patients, such as those requiring supplemental insulin or morbidly obese patients, demonstrated in‐hospital mortality rates that were significantly higher than overall mortality and that did not decline. However, these latter observations must be qualified due to the small sample sizes and the possibility of undercoding or undercounting.
Secular improvements in primary and secondary prevention of CV disease and AMI treatment strategies over the study period may well have affected the continuing reduction in in‐hospital mortality rates in all patients.1,4,27–28 A decrease in mortality rate among DM patients is suggestive of a decrease in disease incidence, a decrease in case fatality rate, or a change in disease‐severity spectrum. The latter hypothesis is consistent with improved population‐based risk factor management of DM28–30 and/or sharing in overall favorable secular changes in the incidence of AMI and AMI‐related mortality. We observed lower odds over time for mortality among DM patients despite increasing frequencies of a history of AMI, a history of hypertension, and a history of dyslipidemia, an observation that may reflect the extent and effectiveness of evidence‐based medical therapy at the time of presentation. Similar conclusions have been reported from other large population‐based registries and studies.26–33 These latter studies and their conclusions are also consistent with the results of our rate decomposition analysis and suggest an overall decrease in age‐independent risk in patients either as a result of receiving appropriate care for risk factors identified prior to the time of AMI or a true population‐based shift in the spectrum of disease severity. Consequently, the countervailing effects of an increase in disease burden, namely, prevalence of risk factors, and effective treatment of these risk factors must be kept in mind when forecasting future trends.
As shown in the current analysis, both sex and age continue to affect the risk of in‐hospital mortality in AMI patients with DM. Older age has been known to be a risk factor for CV disease‐related mortality. In the present analysis, the impact of sex on death in AMI patients with DM was dependent not only on the specific age category but also on time. When taking time into account, the odds for mortality for each subsequent year lessened, to a smaller extent among younger patients, and was dependent on sex. The impact of cohort‐specific changes in mortality risk in these younger subjects cannot be determined from the present analysis; however, the increasing prevalence of obesity—a strong risk factor for T2DM—in these younger cohorts34 may contribute to the increase in prevalence of T2DM as well as the “leveling off” of the decreasing mortality risk observed in older cohorts.14
Limitations
The NIS database was used for the present retrospective analysis using ICD‐9‐CM and Clinical Classification System codes. Miscoding cannot be completely ruled out, although the large number of patients in the database would strongly mitigate significant misclassification bias. Prior analyses have shown excellent positive and negative predictive capability of ICD‐9‐CM codes for CV risk factors in general35 and, specifically, for AMI.36–37 The analysis could be biased by “upcoding” or “Diagnosis‐related group (DRG) creep,” which may have resulted in overreporting of comorbidities38; however, the impact of such would likely have been uniform across the groups, would be unlikely to bias CMRs, and would bias the results of a comparison toward the null.
Data accuracy and comparisons to other national data sets
Data quality assessment of the NIS is performed annually and ensures the internal validity of the data.20 Comparisons against other nationwide data sources (eg, the National Hospital Discharge Survey from the National Center for Health Statistics) provide external validation for the NIS.39–40
We were only able to assess in‐hospital mortality and do not have data on longer term outcomes that may be more relevant, particularly for younger patients. Observational studies may not be able to fully adjust for residual or unmeasured confounding that might affect our estimates for the reported associations between in‐hospital mortality and observed covariates. Finally, the absence of specific data on in‐hospital medical therapy in the NIS database precludes further analysis regarding the impact of prevalent treatment on outcomes. Notwithstanding the above caveats, the NIS represents the largest publicly available database with a statistically sound sampling design allowing for accurate identification of trends in specific diseases.
Conclusions and Clinical Relevance
Over the past decade, despite an increasing prevalence of T2DM in the general population and an increasing prevalence of CV risk factors in AMI patients with T2DM, AMI patients with T2DM exhibited a steady and significant decline in in‐hospital mortality. The impact of diabetes, expressed as the increased risk of in‐hospital death for diabetic patients compared with nondiabetic patients, remained unchanged over time. Secular changes in the diagnosis and management of diabetics in the general population may have contributed to an alteration in the spectrum of disease severity in AMI patients with T2DM. Although women remain at a slightly increased risk of mortality, there was a greater reduction in mortality among women, compared with men, over time. The decline in mortality over time was nearly flat among younger patients, whereas the biggest gains in survival were observed in the elderly. Continued efforts toward recognizing and treating a growing burden of risk factors, particularly DM and obesity, in younger patients41 should be used to improve the differential risk in in‐hospital mortality.
Sources of Funding
This study was supported, in part, by the Robert S. Flinn Foundation (University of New Mexico School of Medicine, Albuquerque, NM) for cardiovascular research.
Disclosures
No conflicts of interest on the part of any author exist with respect to the material presented herein.
Acknowledgments
The authors would like to express their appreciation to Ed Bedrick, PhD, and Faith Selzer, PhD, who provided critical reviews of this manuscript.
References
- 1.Fox CS, Evans JC, Larson MG, Kannel WB, Levy D. Temporal trends in coronary heart disease mortality and sudden cardiac death from 1950 to 1999: the Framingham Heart Study. Circulation. 2004; 110:522-527. [DOI] [PubMed] [Google Scholar]
- 2.Yeh RW, Sidney S, Chandra M, Sorel M, Selby JV, Go AS. Population trends in the incidence and outcomes of acute myocardial infarction. N Engl J Med. 2010; 362:2155-2165. [DOI] [PubMed] [Google Scholar]
- 3.Ford ES, Ajani UA, Croft JB, Critchley JA, Labarthe DR, Kottke TE, Giles WH, Capewell S. Explaining the decrease in U.S. deaths from coronary disease. N Engl J Med. 2007; 356:2388-2398. [DOI] [PubMed] [Google Scholar]
- 4.Lewis WR, Peterson ED, Cannon CP. An organized approach to improvement in guideline adherence for acute myocardial infarction: results with the Get‐With‐The‐Guidelines quality improvement program. Arch Intern Med. 2008; 168:1813-1819. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Boyer N, Laskey WK, Cox M, Hernandez AF, Peterson ED, Bhatt DL, Cannon CP, Fonarow GC. Trends in clinical, demographic and biochemical characteristics of patients with acute myocardial infarction from 2003–2008: a report from the American Heart Association Get With The Guidelines Coronary Artery Disease Program. J Am Heart Assoc. 2012; 1:e00120610.1161/JAHA.112.001206 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Pries SR, Hwang SJ, Coady S, Pencina MJ, D'Agostino RB, Sr, Savage PJ, Levy D, Fox CS. Trends in all‐cause and cardiovascular disease mortality among women and men with and without diabetes mellitus in the Framingham Heart Study, 1950 to 2005. Circulation. 2009; 119:1728-1735. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Zuanetti G, Latini R, Maggioni AP, Santoro L, Maria GF. Influence of diabetes on mortality in acute myocardial infarction: data from the GISSI‐2 Study. J Am Coll Cardiol. 1993; 22:1788-1794. [DOI] [PubMed] [Google Scholar]
- 8.Malmberg K, Yusuf S, Gerstein HC, Brown J, Zhao F, Hunt D, Piegas L, Calvin J, Keltai M, Budaj A. Impact of diabetes on long‐term prognosis in patients with unstable angina and non‐Q wave myocardial infarction: results of the OASIS (Organization to Assess Strategies for Ischemic Syndromes) Registry. Circulation. 2000; 102:1014-1019. [DOI] [PubMed] [Google Scholar]
- 9.Center for Disease Control and Prevention. National Diabetes Fact Sheet: National Estimates and General Information on Diabetes and Prediabetes in the US 2011. 2011Washington, DC: US Department of Health and Human Services [Google Scholar]
- 10.Grundy SM, Benjamin IJ, Burke GL, Chait A, Eckel RH, Howard BV, Mitch W, Smith SC, Jr, Sowers JR. Diabetes and cardiovascular disease: a statement for healthcare professionals from the American Heart Association. Circulation. 1999; 100:1134-1146. [DOI] [PubMed] [Google Scholar]
- 11.Barrett‐Connor EL, Cohn BA, Wingard DL, Edelstein SL. Why is diabetes mellitus a stronger risk factor for fatal ischemic heart disease in women than in men? The Rancho Bernardo Study. JAMA. 1991; 265:627-631. [PubMed] [Google Scholar]
- 12.Hu FB, Stampfer MJ, Solomon CJ, Liu S, Willett WC, Speizer FE, Nathan DM, Manson JE. The impact of diabetes mellitus on mortality from all causes and coronary heart disease in women: 20 years of follow‐up. Arch Intern Med. 2001; 161:1717-1723. [DOI] [PubMed] [Google Scholar]
- 13.Gore MO, Patel MJ, Kosiborod M, Parsons LS, Khera A, de Lemos JA, Rogers WJ, Peterson ED, Canto JC, McGuire DKfor the National Registry of Myocardial Infarction Investigators. Diabetes mellitus and trends in hospital survival after myocardial infarction, 1994–2006: data from the National Registry of Myocardial Infarction. Circ Cardiovasc Qual Outcomes. 2012; 5:791-797. [DOI] [PubMed] [Google Scholar]
- 14.Ford ES, Capewell S. Coronary heart disease mortality among young adults in the U.S. from 1980 through 2002. Concealed leveling of mortality rates. J Am Coll Cardiol. 2007; 50:2128-2132. [DOI] [PubMed] [Google Scholar]
- 15.Boyle JP, Honeycutt AA, Narayan KM, Hoerger TJ, Geiss LS, Chen H, Thompson TJ. Projection of diabetes burden through 2050: impact of changing demography and disease prevalence in the U.S. Diabetes Care. 2001; 24:1936-1940. [DOI] [PubMed] [Google Scholar]
- 16.Heidenreich PA, Trogdon JG, Khavjou OA, Butler J, Dracup K, Ezekowitz MD, Finkelstein EA, Hong Y, Johnston SC, Khera A, Lloyd‐Jones DM, Nelson SA, Nichol G, Orenstein D, Wilson PWF, Woo YJ. Forecasting the future of cardiovascular disease in the United States: a policy statement from the American Heart Association. Circulation. 2011; 123:933-944. [DOI] [PubMed] [Google Scholar]
- 17.Huffman MD, Lloyd‐Jones DM, Ning H, Labarthe DR, Castillo MG, O'Flaherty M, Ford ES, Capewell S. Quantifying options for reducing coronary heart disease mortality by 2020. Circulation. 2013; 127:2477-2484. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Healthcare Cost and Utilization Project. Overview of the Nationwide Inpatient Sample. 2013. Available at: http://www.hcup-us.ahrq.gov/nisoverview.jsp. Accessed November 1, 2013.
- 19.Healthcare Cost and Utilization Project. HCUP methods series. 2010. Available at: http://www.hcup-us.ahrq.gov/reports/methods/methods_topic.jsp. Accessed November 1, 2013.
- 20.HCUP Quality Control Procedures. Healthcare Cost and Utilization Project (HCUP). Agency for Healthcare Research and Quality. Rockville, MD; September 2013. Available at: www.hcup-us.ahrq.gov/db/quality.jsp. Accessed January 1, 2014. [Google Scholar]
- 21.Cheung BM, Ong KL, Cherny SS, Sham PC, Tso AW, Lam KS. Diabetes prevalence and therapeutic target achievement in the United States, 1999 to 2006. Am J Med. 2009; 122:443-453. [DOI] [PubMed] [Google Scholar]
- 22.Kitagawa EM. Components of a difference between two rates. J Am Stat Assoc. 1955; 50:1168-1194. [Google Scholar]
- 23.Gabir MM, Hanson RL, Dabela D, Imperatore G, Roumain J, Bennett PH, Knowler WC. The 1997 American Diabetes Association and 1999 World Health Organization criteria for hyperglycemia in the diagnosis and prediction of diabetes. Diabetes Care. 2000; 23:1108-1112. [DOI] [PubMed] [Google Scholar]
- 24.Gillett MJ. International Expert Committee report on the role of A1C assay in the diagnosis of diabetes. Diabetes Care. 2009; 32:1327-1334. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Rogers WJ, Frederick PD, Stoehr E, Canto JG, Ornato JP, Gibson CM, Pollack CV, Jr, Gore JM, Chandra‐Strobos N, Peterson ED, French WJ. Trends in presenting characteristics and hospital mortality among patients with ST‐elevation and non‐ST‐elevation myocardial infarction in the National Registry of Myocardial Infarction from 1990 to 2006. Am Heart J. 2008; 156:1026-1034. [DOI] [PubMed] [Google Scholar]
- 26.Cannon CP. Update to International Classification of Diseases, 9th Revision: distinguishes STEMI from NSTEMI. Crit Pathw Cardiol. 2005; 4:185-186. [DOI] [PubMed] [Google Scholar]
- 27.Chen J, Normand SL, Wang Y, Drye EE, Schreiner GC, Krumholz HM. Recent declines in hospitalizations for acute myocardial infarction for Medicare fee‐for‐service beneficiaries: progress and continuing challenges. Circulation. 2010; 121:1322-1328. [DOI] [PubMed] [Google Scholar]
- 28.Yang Q, Cogswell ME, Flanders WD, Hong Y, Zhang Z, Loustalot F, Gillespie C, Merritt R, Hu FB. Trends in cardiovascular health metrics and associations with all‐cause and CVD mortality among US adults. JAMA. 2012; 307:1273-1283. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Hoerger TJ, Zhang P, Segel JE, Gregg EW, Narayan KMV, Hicks KA. Improvements in risk factor control among persons with diabetes in the United States: evidence and implications for remaining life expectancy. Diabetes Res Clin Pract. 2009; 86:225-232. [DOI] [PubMed] [Google Scholar]
- 30.Gregg EW, Li Y, Wang J, Burrows NR, Ali MK, Rolka D, Williams DE, Geiss L. Changes in diabetes‐related complications in the United States, 1990–2010. N Engl J Med. 2014; 370:1514-1523. [DOI] [PubMed] [Google Scholar]
- 31.Selvin E, Parrinello CM, Sacks DB, Coresh J. Trends in prevalence and control of diabetes in the United States, 1988–1994 and 1999–2010. Ann Intern Med. 2014; 160:517-525. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Gregg WE, Cheng YJ, Saydah S, Cowie C, Garfield S, Geiss L, Barker L. Trends in death rates among U.S. adults with and without diabetes between 1997 and 2006: findings from the National Health Interview Survey. Diabetes Care. 2012; 35:1252-1257. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Gaede P, Lund‐Andersen H, Parving HH, Pederson O. Effect of a multi‐factorial intervention on mortality in type 2 diabetes mellitus. N Engl J Med. 2008; 358:580-591. [DOI] [PubMed] [Google Scholar]
- 34.Ogden CL, Flegal KM, Carroll MD, Johnson CL. Prevalence and trends in overweight among US children and adolescents, 1999–2000. JAMA. 2002; 288:1728-1732. [DOI] [PubMed] [Google Scholar]
- 35.Birman‐Deych E, Waterman AD, Yan Y, Nilasena DS, Radford MJ, Gage BF. Accuracy of ICD‐9‐CM codes for identifying cardiovascular and stroke risk factors. Med Care. 2005; 43:480-485. [DOI] [PubMed] [Google Scholar]
- 36.Saydah SH, Geiss LS, Tierney ED, Benjamin SM, Engelgau M, Brancati F. Review of the performance of methods to identify diabetes cases among vital statistics, administrative data and survey data. Ann Epidemiol. 2004; 14:507-516. [DOI] [PubMed] [Google Scholar]
- 37.Quan H, Li B, Saunders LD, Parsons GA, Nilsson CI, Alibhai A, Ghali WAIMECCHI Investigators. Assessing validity of ICD‐9‐CM and ICD‐10 administrative data in recording clinical conditions in a unique dually coded database. Health Serv Res. 2008; 43:1424-1441. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Steinwald B, Dummit LA. Hospital case‐mix change: sicker patients or DRG creep? Health Aff. 1989; 8:35-47. [DOI] [PubMed] [Google Scholar]
- 39.Whalen D, Houchens R, Elixhauser A. 2003 HCUP Nationwide Inpatient Sample (NIS) comparison report. HCUP Methods Series Report x0023 2006–09. U.S. Agency for Healthcare Research and Quality; Available at: http://www.hcup-us.ahrq.gov/reports/methods.jsp. Accessed December 1, 2006. [Google Scholar]
- 40.Barrett M, Wilson E, Whalen D. 2007 HCUP Nationwide Inpatient Sample NIS) comparison report. HCUP Methods Series Report x0023 2010x201003. U.S. Agency for Healthcare Research and Quality; Available at: http://www.hcup-us.ahrq.gov/reports/methods.jsp. Accessed September 9, 2010. [Google Scholar]
- 41.Cheng YJ, Imperatore G, Geiss LS, Wang J, Saydah SH, Cowie CC, Gregg EW. Secular changes in the age‐specific prevalence of diabetes among U.S. adults: 1988–2010. Diabetes Care. 2013; 36:2690-2696. [DOI] [PMC free article] [PubMed] [Google Scholar]