Secular trends in cardiovascular disease mortality, incidence, and case fatality rates in adults in the United States

CLINICAL STUDIES

Secular Trends in Cardiovascular Disease Mortality, Incidence, and Case Fatality Rates in Adults in the United States Ahmet Ergin, MD, PhD, MPH, Paul Muntner, PhD, MHS, Roger Sherwin, MB, BChir, Jiang He, MD, PhD PURPOSE: To assess the effects of changes in cardiovascular disease incidence and case fatality rates on secular trends in mortality in the U.S. population between 1971–1982 and 1982– 1992. METHODS: Using data from the first National Health and Nutrition Examination Survey Epidemiologic Follow-up Study, two cohorts (10,869 subjects in the 1971–1982 cohort and 9774 in the 1982–1992 cohort) of participants aged 35 to 74 years were created. Baseline medical history questionnaires were administered in 1971–1975 and 1982–1984, with follow-up interviews, hospital record reviews, and death certificate searches conducted in 1982– 1984, 1986, 1987, and 1992. RESULTS: Between 1971–1982 and 1982–1992, age-standardized cardiovascular disease mortality declined from 79.1 (95% confidence interval [CI]: 75.2 to 83.0) to 53.0 (95% CI: 49.5 to 56.5) per 10,000 person-years, while cardiovascular

disease incidence rates decreased from 293.5 (95% CI: 284.5 to 302.4) to 225.1 (95% CI: 216.6 to 233.5) per 10,000 person-years. The 28-day case fatality rate for cardiovascular disease declined from 15.7% (95% CI: 14.5% to 16.8%) to 11.7% (95% CI: 10.3% to 13.0%). After adjustment for age, sex, and race, rates were 31% lower for cardiovascular disease mortality, 21% lower for incidence, and 28% lower for 28day case fatality in the 1982–1992 cohort than in the 1971– 1982 cohort (each P ⬍0.001). CONCLUSION: The decrease in cardiovascular disease mortality between 1971–1982 and 1982–1992 was due to declines in both the incidence and case fatality rates in this national sample. These findings suggest that both primary and secondary prevention and treatment contributed to the decline in cardiovascular disease mortality in the United States. Am J Med. 2004; 117:219 –227. ©2004 by Elsevier Inc.

A

dence or case fatality rate of cardiovascular disease. However, there has been a lack of national data on these two variables. Incidence and case fatality each holds very different implications; the former may reflect the effect of primary prevention, while the latter may represent the effect of secondary prevention and treatment. Further, data from several community surveillance programs and cohort studies not only have limited generalizability to the general population but also have reported conflicting findings on the incidence of cardiovascular disease from the early 1970s to the mid-1990s (4 –7,9,10,14 –20). Several of these studies reported a reduction in incidence (4,5,7,10,14,15), whereas others found no change or even an increase over time (6,9,16 –18). The National Hospital Discharge Survey indicates that the in-hospital case fatality rate of acute myocardial infarction and stroke has declined since the late 1960s (1). However, it does not provide information on out-of hospital cardiovascular disease mortality and long-term survival after a cardiovascular disease event (21). In this study, we took advantage of a large cohort of adults from the National Health and Nutrition Examination Survey (NHANES) I Epidemiologic Follow-up Study to determine secular trends in age-adjusted mortality, incidence,

lthough cardiovascular disease remains the leading cause of death in the United States, mortality from cardiovascular disease, including coronary heart disease and stroke, has declined markedly during the past several decades (1–3). National vital statistics data indicate that the age-adjusted mortality from coronary heart disease declined 58.3%, from 448.0 per 100,000 in 1970 to 186.6 per 100,000 in 2000, while ageadjusted mortality from stroke declined 61.5%, from 147.7 per 100,000 to 56.8 per 100,000 during the same period. These trends have been reported among men and women, and blacks and whites (1). It is unclear why cardiovascular disease mortality has declined since the early 1960s (3–13). One suggestion is that this decline might be due to a reduction in the inci-

From the Department of Epidemiology (AE, PM, RS, JH), Tulane University School of Public Health and Tropical Medicine, and Department of Medicine (PM, JH), Tulane University School of Medicine, New Orleans, Louisiana. Requests for reprints should be addressed to Jiang He, MD, PhD, Department of Epidemiology, Tulane University School of Public Health and Tropical Medicine, 1440 Canal Street Suite 2000, New Orleans, Louisiana 70112, or [email protected]. Manuscript submitted April 25, 2003, and accepted in revised form March 13, 2004. © 2004 by Elsevier Inc. All rights reserved.

0002-9343/04/$–see front matter 219 doi:10.1016/j.amjmed.2004.03.017

Cardiovascular Disease Mortality, Incidence, and Case Fatality Rates/Ergin et al

and case fatality rates of cardiovascular disease between 1971–1982 and 1982–1992, overall and by sex and race, and to assess the effect of changes in incidence and case fatality rates on secular trends in mortality.

METHODS NHANES I Epidemiologic Follow-up Study The NHANES I comprised a representative sample of the civilian noninstitutionalized population of the United States. The NHANES I survey was conducted between 1971 and 1975, and data collection included medical history, standardized medical examination, dietary history, laboratory tests, and anthropometric measurements (22). The NHANES I Epidemiologic Follow-up Study cohort included 14,407 persons 25 to 74 years of age who had completed the NHANES I medical examination (23). Follow-up data were collected in 1982–1984, 1986, 1987, and 1992. Each follow-up wave included identical sequential procedures for identifying cardiovascular disease events and all deaths: tracking a participant or proxy to a current address; performing in-depth interviews with the participant or proxy; obtaining hospital and nursing home records, including electrocardiograms; and, for decedents, acquiring a death certificate. Tracing rates for each completed wave ranged from 90% to 94%, and interview rates ranged from 91% to 96% for those traced. Death certificates were obtained for 98.7% of the decedents in the follow-up study cohort (24).

Outcome Definitions The outcomes of interest included the mortality, incidence, 28-day case fatality, and long-term survival rates of cardiovascular disease, coronary heart disease, acute myocardial infarction, and stroke. Incident cardiovascular disease events were based on documentation of an event that met prespecified diagnostic criteria and occurred during the period between the baseline examination and last follow-up interview. Cause-specific incidence was identified by either death certificate reports in which the underlying cause of death was recorded using an International Classification of Diseases, Ninth Revision (ICD-9) code of 390 to 459 (cardiovascular disease), 410 to 414 (coronary heart disease), 410 (acute myocardial infarction), or 430 to 434 or 436 to 438 (stroke), or by one or more hospital or nursing home stays for which the participant had a discharge diagnosis with one of these codes. Cause-specific mortality was identified by underlying cause of death using the above-mentioned ICD-9 codes. Recurrent myocardial infarction was defined as a myocardial infarction among those who had a previous history of myocardial infarction. Twenty-eight– day case fatality was defined as deaths within 28 days after the first cardiovascular disease event. Long-term case fatality was defined as deaths among participants having an event and 220

August 15, 2004

THE AMERICAN JOURNAL OF MEDICINE威

surviving 28 days past the event date. The date of record for incident events was identified by the date of first hospital admission with an established study event or the date of death from a study event in the absence of hospital or nursing home documentation of such an event. Validity of study outcome data from both death certification and discharge diagnosis of hospital/nursing home stay has been documented (25).

Cohort Analysis Two national cohorts of adults aged 35 to 74 years were created from the NHANES I Epidemiologic Follow-up Study data. 1971–1982 cohort. Of the 14,407 participants who underwent the baseline examination during 1971–1975, 10,869 were 35 to 74 years of age and composed the 1971– 1982 cohort. The baseline medical history questionnaire included queries about selected health conditions and medications used for these conditions. We excluded participants who responded affirmatively to the question “Have you ever been told by a doctor or other health professional that you had a heart attack, heart failure, or stroke?” or who had used medication for heart disease. We excluded 1206 prevalent cardiovascular disease cases (637 men), 654 prevalent coronary heart disease cases (398 men), and 277 prevalent stroke cases (133 men). The follow-up period for this cohort was from the date of the 1971–1975 interview to the date of death, the admission date to a hospital due to the outcome of interest before December 31, 1984, or the date of the self- or proxy interview during 1982–1984, whichever occurred first. 1982–1992 cohort. Of the 12,220 participants who completed a self-interview or had a proxy interview during 1982–1984, 9774 were 35 to 74 years of age. The baseline medical history questionnaire included identical questions to the 1971–1975 survey regarding selected health conditions and medications for heart disease. We excluded 1381 prevalent cardiovascular disease cases (616 men), 571 prevalent coronary heart disease cases (309 men), and 185 prevalent stroke cases (77 men). The follow-up period for this cohort was from the date of the 1982–1984 interview to the date of death, the date of hospitalization due to the outcome of interest before July 19, 1993, or the date of the last follow-up interview in 1992, whichever occurred first.

Statistical Analysis Age-specific mortality and incidence rates per 10,000 person-years for cardiovascular disease, coronary heart disease, acute myocardial infarction, and stroke were calculated using the number of cause-specific deaths or new cases as the numerator and the total years of follow-up as the denominator for the following age groups: 35 to 44 years, 45 to 54 years, 55 to 64 years, and 65 to 74 years. The variances of these rates were estimated using the Chiang method (26).

Volume 117

Cardiovascular Disease Mortality, Incidence, and Case Fatality Rates/Ergin et al

Rates were standardized by age using the direct method and the 1980 total U.S. population as the standard; the variances of these rates were calculated by Taylor series approximation (27). Percentage change in rates were calculated as the rate in the 1982–1992 cohort minus the rate in the 1971–1982 cohort, divided by the rate in the 1971–1982 cohort, and multiplied by 100. Poisson regression was used to examine differences in rates across periods after adjustment for age, sex, and race (28). Twenty-eight– day case fatality rates were also age-standardized using the direct method, but the age distribution of all incident cases from the two cohorts pooled together was applied as the standard. Logistic regression was used to examine the differences in 28-day case fatality rates across periods after adjustment for age, sex, and race. Case fatality rates were calculated only for the first incident event. Age-adjusted cumulative mortality curves among patients surviving to day 28 were generated by cohort from Cox regression models to estimate long-term case fatality. Analyses were performed using SAS software (SAS Institute Inc., Cary, NC). P values ⬍0.05 were considered significant.

RESULTS Mortality Overall, age-standardized mortality for cardiovascular disease, coronary heart disease, acute myocardial infarction, and stroke declined significantly between 1971–1982 and 1982–1992 (Figure 1, top panel). These declines were significant in all sex and race groups (Table 1), except for declines in myocardial infarction mortality among white women (⫺6.3%, P ⫽ 0.2). Age, race, and sex-adjusted mortality declined by 31% for cardiovascular disease, 33% for coronary heart disease, 37% for myocardial infarction, and 43% for stroke between 1971–1982 and 1982–1992 (all P ⬍0.001).

Incidence Overall, age-standardized incidence of cardiovascular disease, coronary heart disease, and stroke was significantly lower in the 1982–1992 cohort than in the 1971– 1982 cohort (Figure 1, middle panel). However, the incidence of acute myocardial infarction did not decline. The decline in cardiovascular disease incidence between the two time periods was observed across all race and sex groups, but the decline among black men was not significant (Table 2). The decline in age-standardized incidence of coronary heart disease was significant only among white men and women. The age-standardized incidence of acute myocardial infarction rate was nonsignificantly lower among white and black men during 1982–1992 as compared with 1971–1982 but increased by 21% (P ⬍0.001) among white women. The increase among black women was not significant (13%, P ⫽ 0.3). The age-standardized incidence of stroke declined significantly between 1971–1982 and 1982–1992 among white men and women as well as black women, but not black

Figure 1. Age-standardized mortality (top), incidence (center), and 28-day case fatality (bottom) rates of cardiovascular disease in the 1971–1982 and 1982–1992 cohorts. AMI ⫽ acute myocardial infarction; CHD ⫽ coronary heart disease; CVD ⫽ cardiovascular disease.

men. Between 1971–1982 and 1982–1992, age-, race-, and sex-adjusted incidence rates declined by 21% for cardiovascular disease, 13% for coronary heart disease, and 20% for stroke (each P ⬍0.001), whereas the incidence of myocardial infarction increased slightly (2%, P ⫽ 0.5). The age-standardized rate of recurrent myocardial infarction declined significantly among white men and black men and women but increased among white women.

28-Day Case Fatality Rate Overall, age-standardized 28-day case fatality rates for cardiovascular disease, coronary heart disease, acute

August 15, 2004

THE AMERICAN JOURNAL OF MEDICINE威

Volume 117 221

Cardiovascular Disease Mortality, Incidence, and Case Fatality Rates/Ergin et al

Table 1. Age-Standardized Mortality Rates of Cardiovascular Diseases in the 1971–1982 and 1982–1992 Cohorts, by Race and Sex 1971–1982 Cohort

No. of Events

Outcomes

Mortality Per 10,000 Person-Years

1982–1992 Cohort

(95% Confidence Interval)

No. of Events

Mortality Per 10,000 Person-Years

(95% Confidence Interval)

% Change

P Value*

White Men 35,773.2 Person-Years Cardiovascular disease Coronary heart disease Acute myocardial infarction Stroke

543 363 248

107.4 72.4 50.6

74

13.2

26,683.0 Person-Years 214 134 79

75.5 46.5 27.1

(68.3–82.4) (40.5–52.6) (22.2–32.1)

⫺29.7 ⫺35.8 ⫺46.4

⬍0.001 ⬍0.001 ⬍0.001

(11.1–15.4) 26 White Women

9.5

(7.0–11.9)

⫺28.0

⬍0.001

(99.9–114.9) (65.8–78.9) (44.8–56.4)

50,278.7 Person-Years Cardiovascular disease Coronary heart disease Acute myocardial infarction Stroke

341 182 108

49.9 26.3 16.0

72

9.7

47,362.0 Person-Years 141 90 60

35.3 22.9 15.0

(31.5–39.1) (19.9–25.9) (12.4–17.5)

⫺29.3 ⫺12.9 ⫺6.3

⬍0.001 ⬍0.001 0.2

(8.2–11.3) 21 Black Men

5.3

(3.8–6.8)

⫺45.4

⬍0.001

(45.8–53.9) (23.2–29.3) (13.5–18.6)

6058.7 Person-Years Cardiovascular disease Coronary heart disease Acute myocardial infarction Stroke

177 57 32

133.4 61.8 35.5

22

21.7

3579.5 Person-Years 28 10 4

79.2 28.9 10.8

(56.6–101.8) (13.8–44.0) (2.9–18.6)

⫺40.6 ⫺53.2 ⫺70.0

⬍0.001 ⬍0.001 ⬍0.001

(13.3–30.2) 4 Black Women

8.6

(0.3–17.0)

⫺60.4

⬍0.001

(109.4–157.3) (46.7–76.9) (23.8–47.2)

9833.4 Person-Years Cardiovascular disease Coronary heart disease Acute myocardial infarction Stroke

8547.5 Person-Years

106 46 26

93.9 37.5 20.3

(77.8–101.1) (27.9–47.1) (13.6–27.0)

42 13 9

55.6 18.4 11.7

(43.0–68.1) (10.4–26.3) (5.8–17.6)

⫺40.8 ⫺50.9 ⫺42.4

⬍0.001 ⬍0.001 ⬍0.001

34

28.3

(19.9–36.8)

9

10.1

(4.3–16.0)

⫺64.3

⬍0.001

* P values are from the Poisson regression model, stratified by race and sex and adjusted for age.

myocardial infarction, and stroke declined between 1971–1982 and 1982–1992 (Figure 1, bottom panel), although the decrease for stroke was not statistically significant. The age-standardized 28-day case fatality rate for cardiovascular disease in the 1982–1992 cohort as compared with the 1971–1982 cohort was nonsignificantly lower among white men and women but was significantly lower in black men and women (Table 3). Although 28-day case fatality rates for coronary heart disease and myocardial infarction declined in all race and sex groups, this decline was not significant for white and black women. The decline in 28-day case fatality rate for stroke was nonsignificant across all race and sex groups. Age-, race-, and sex-adjusted 28-day case fatality rates declined by 28% for cardiovascular disease (P ⬍0.001), 36% for coronary heart disease (P 222

August 15, 2004

THE AMERICAN JOURNAL OF MEDICINE威

⬍0.001), 36% for myocardial infarction (P ⬍0.001), and 27% for stroke (P ⫽ 0.1) from 1971–1982 to 1982– 1992.

Trends in Mortality among Patients Surviving to Day 28 The age-adjusted cumulative mortality after 28 days of the first cardiovascular disease event was significantly lower in the 1982–1992 cohort than in the 1971–1982 cohort (P ⬍0.001; Figure 2).

DISCUSSION We found that cardiovascular disease mortality, incidence, and case fatality rates declined in a large national sample of adults from 1971–1982 to 1982–1992, and that

Volume 117

Table 2. Age-Standardized Incidence Rates of Cardiovascular Diseases in the 1971–1982 and 1982–1992 Cohorts, by Race and Sex 1971–1982 Cohort

Outcomes

Person-Years

No. of Events

Incidence, Per 10,000 Person-Years

1982–1992 Cohort (95% Confidence Interval)

Person-Years

No. of Events

Incidence Per 10,000 Person-Years

(95% Confidence Interval)

% Change

P Value*

White Men

THE AMERICAN JOURNAL OF MEDICINE威 Volume 117 223

27,843.2 30,865.2 31,998.6

1128 694 324

342.6 177.0 79.6

(326.6–358.6) (165.6–188.4) (71.5–87.7)

20,605.1 22,367.4 23,259.9

524 345 180

264.2 151.2 73.4

(248.4–280.1) (139.2–163.2) (64.8–82.1)

⫺22.9 ⫺14.6 ⫺7.8

⬍0.001 ⬍0.001 0.1

34,829.4

251

53.0

(46.9–59.1)

25,041.0

106

41.4

(36.3–46.4)

⫺21.9

⬍0.001

33,872.9

272

57.0

(51.3–62.6) 24,105.7 White Women

109

42.1

(36.4–47.9)

⫺26.1

⬍0.001

Cardiovascular disease Coronary heart disease Acute myocardial infarction Recurrent myocardial infarction Stroke

41,120.2 45,636.5 47,083.7

1116 573 178

246.8 101.7 30.0

(235.1–258.4) (94.8–108.5) (26.1–33.9)

38,391.7 41,628.6 42,664.6

636 325 136

190.8 90.1 36.3

(180.1–201.4) (83.1–97.2) (31.7–40.9)

⫺22.7 ⫺11.4 21.0

⬍0.001 0.008 ⬍0.001

49,110.3

97

14.5

(12.2–16.8)

43,697.2

72

19.6

(17.2–22.0)

35.2

47,644.4

253

39.2

(35.4–42.9) 42,849.2 Black Men

115

31.2

(27.1–35.3)

⫺20.4

⬍0.001

Cardiovascular disease Coronary heart disease Acute myocardial infarction Recurrent myocardial infarction Stroke

4450.0 5205.9 5392.2

206 114 44

368.6 162.7 62.4

(324.7–412.5) (162.7–131.9) (42.5–80.3)

2456.1 2768.9 2842.2

69 36 18

291.1 132.6 61.6

(243.0–339.6) (100.7–164.5) (41.3–87.9)

⫺21.0 ⫺18.5 ⫺1.3

0.08 0.09 0.9

5588.5

28

32.2

(21.2–44.4)

2886.1

6

21.0

(14.4–27.6)

⫺34.8

⬍0.001

5357.0

61

75.6

(56.8–94.3) 2798.9 Black Women

23

75.7

(53.2–98.3)

0.1

Cardiovascular disease Coronary heart disease Acute myocardial infarction Recurrent myocardial infarction Stroke

7643.7 8743.4 9016.8

247 120 37

303.1 122.7 33.1

(272.6–333.6) (103.7–141.7) (23.9–42.3)

6100.5 6722.9 6938.4

141 72 24

262.1 118.0 37.4

(230.5–293.8) (96.7–139.6) (26.1–49.2)

⫺13.5 ⫺3.8 13.0

0.007 0.9 0.3

9275.2

24

19.5

(13.1–26.0)

7062.1

8

14.6

(8.6–20.5)

⫺25.1

0.009

8959.0

70

69.5

(54.9–84.0)

6957.5

33

50.9

(36.8–65.0)

⫺26.8

0.01

* P values are from the Poisson regression model, stratified by race and sex and adjusted for age.

0.03

0.9

Cardiovascular Disease Mortality, Incidence, and Case Fatality Rates/Ergin et al

August 15, 2004

Cardiovascular disease Coronary heart disease Acute myocardial infarction Recurrent myocardial infarction Stroke

224 August 15, 2004

Table 3. Age-Standardized 28-Day Case Fatality Rates of Cardiovascular Diseases in the 1971–1982 and 1982–1992 Cohort, by Race and Sex 1971–1982 Cohort

No. of Deaths

(95% Confidence Interval)

No. of Cases

No. of Deaths

Case Fatality Rate (%)

(95% Confidence Interval)

% Change

P Value*

White Men

Volume 117

Cardiovascular disease Coronary heart disease Acute myocardial infarction Stroke

1128 694 324 272

202 190 162 50

17.0 27.2 50.2 17.9

(15.2–18.9) 524 (24.4–30.0) 345 (45.6–54.8) 180 (14.4–21.3) 109 White Women

65 59 56 17

13.5 19.0 33.8 16.1

(11.1–15.8) (15.7–22.4) (28.1–39.4) (11.3–20.8)

⫺20.6 ⫺30.1 ⫺32.7 ⫺10.1

0.06 ⬍0.001 ⬍0.001 0.07

Cardiovascular disease Coronary heart disease Acute myocardial infarction Stroke

1116 573 178 253

128 95 75 46

10.7 15.3 40.1 16.5

(9.2–12.2) 636 (12.8–17.9) 325 (33.1–47.0) 136 (12.8–20.2) 115 Black Men

50 42 50 16

9.4 14.2 36.7 16.0

(7.7–11.1) (11.5–16.9) (30.5–42.8) (11.8–20.1)

⫺12.1 ⫺7.2 ⫺8.5 ⫺3.0

0.3 0.5 0.6 0.5

Cardiovascular disease Coronary heart disease Acute myocardial infarction Stroke

206 114 44 61

61 44 29 17

27.8 39.2 64.8 28.9

(22.4–33.3) 69 (31.2–47.2) 36 (52.9–76.7) 18 (19.2–38.5) 23 Black Women

10 5 3 2

15.4 10.8 14.3 6.4

(8.5–22.2) (4.6–17.1) (5.8–22.7) (0.0–14.6)

⫺44.6 ⫺72.4 ⫺77.9 ⫺77.9

0.04 0.01 0.004 0.1

Cardiovascular disease Coronary heart disease Acute myocardial infarction Stroke

247 120 37 70

51 32 23 24

20.5 26.5 58.2 34.8

(15.8–25.1) (19.5–33.4) (43.4–73.1) (25.0–44.6)

16 11 8 8

13.7 16.6 30.2 20.0

(8.8–18.7) (10.0–23.2) (15.2–45.3) (8.2–31.8)

⫺33.2 ⫺37.4 ⫺48.1 ⫺42.5

0.05 0.2 0.09 0.3

* P values were from a logistic regression model, stratified by race and sex and adjusted for age.

141 72 24 33

Cardiovascular Disease Mortality, Incidence, and Case Fatality Rates/Ergin et al

THE AMERICAN JOURNAL OF MEDICINE威

Outcomes

No. of Cases

Case Fatality Rate (%)

1982–1992 Cohort

Cardiovascular Disease Mortality, Incidence, and Case Fatality Rates/Ergin et al

Figure 2. Cumulative mortality for cardiovascular disease among patients surviving to day 28.

age, race, and sex-adjusted cardiovascular disease mortality was 31% lower in the 1982–1992 cohort. This downward trend was similar among all race and sex groups. Further, that the decrease in cardiovascular disease mortality was accompanied by a decline in incidence and improvements in 28-day case fatality and long-term survival suggests that the decrease in mortality from cardiovascular disease was likely due to a combination of primary prevention, which reduced cardiovascular disease incidence, and of secondary prevention and treatment, which reduced short- and long-term case fatality rates. The trends in cardiovascular disease mortality from 1971 to 1992 that we observed are consistent with U.S. vital statistics and data from community-based surveillance programs and cohort studies (1,3,14). However, our study also provides additional information on the secular trends of cardiovascular disease incidence and case fatality rates at the national level. Our analysis also found a substantial decline in overall age-adjusted mortality from coronary heart disease and myocardial infarction. The decline in coronary heart disease mortality was accompanied by considerable declines in incidence and case fatality rates in all race and sex groups. In contrast, the age-adjusted incidence of myocardial infarction was stable among white men and black men and women but significantly increased among white women. Recurrent myocardial infarction also declined in all race and sex groups, but it was significantly increased among white women. Although a number of studies reported declines in the incidence of myocardial infarction from the 1970s to mid-1990s (4,5,14), our finding of stable incidence rates are consistent with other studies covering the same period (6). In contrast to the stable rates of myocardial infarction incidence, we found that overall age-adjusted 28-day case fatality rates and long-term survival for myocardial infarction improved significantly in the 1982–1992 cohort. Twenty-eight– day case fatality rates declined in all race and sex groups, albeit insignificantly in some subgroups with a small number of outcomes. This reduction may

have been due to secondary prevention and treatment, which may have improved patient management, and to primary prevention, which may have resulted in a decrease in disease severity (19). A similar decline in the severity of myocardial infarction was found in Olmsted County, Minnesota, between 1983 and 1994 (29). Our findings of reduced stroke mortality and case fatality mirror national vital statistics and several community surveillance studies (1–3,7,10). However, our data showing a decline in overall stroke incidence differ from the National Hospital Discharge Survey data covering the same period. There are two possible explanations for this difference. First, our study did not include transient ischemic attack discharges, unlike in the National Hospital Discharge Survey. Second, we calculated stroke incidence using the first event, whereas the National Hospital Discharge Survey does not separate incident and recurrent hospitalizations due to stroke. Even at the community level, conflicting findings have been observed. For example, an increase in age-adjusted stroke incidence during the same time frame as that in our study was reported in Rochester, Minnesota (9), whereas another Minnesota study showed a persistent decline in acute stroke incidence between 1970 and 1985 (7). Likewise, populationbased data from Olmsted County have showed a substantial decline in stroke incidence (30). A large reduction in mean blood pressure levels and improvement in detection, treatment, and control of hypertension in the U.S. population may have contributed to the reduction in stroke incidence and mortality (31). In our study, the difference in mortality due to cardiovascular disease between whites and blacks was smaller in the 1982–1992 cohort. It seems that improvements in the 28-day case fatality rate among black men were the main contributor to this decrease. Data from the Atherosclerosis Risk in Community study showed an increased rate of hospitalization for myocardial infarction among blacks despite declining myocardial infarction mortality (32). However, it should be noted that our study comprised a small number of blacks, limiting inferences regarding race-specific results. Another limitation of the NHANES I Epidemiologic Follow-up Study is that participants were passively followed for clinical outcomes and periodic clinical examinations were not employed. Therefore, subclinical events or even clinical events may have been missed. However, the same diagnostic procedure and criteria were used in both periods, and previous validation studies have reported little misclassification in coronary heart disease (25). Still, the validity for heart failure, peripheral arterial disease, and cardiovascular disease as a whole might be more questionable. One other limitation is that three follow-up interviews were conducted for the 1982–1992 cohort, whereas only one interview was conducted the 1971–1982 cohort. However, any differential reporting of

August 15, 2004

THE AMERICAN JOURNAL OF MEDICINE威

Volume 117 225

Cardiovascular Disease Mortality, Incidence, and Case Fatality Rates/Ergin et al

events resulting from the additional waves of follow-up for the later cohort would underestimate the decline in cardiovascular disease incidence. Finally, because of the small numbers of events in some strata, subgroups results need to be interpreted with caution. In summary, the current study provides data to fill the gap in our understanding of trends in U.S. national cardiovascular disease during the past several decades. The validity of our findings is supported by several strengths of the study, namely, the study population was derived from a large sample of the U.S. general population; an average of 8 to 10 years of follow-up was available for the two cohorts, during which a large number of outcome events occurred; cardiovascular disease events were ascertained by standardized methods; and high follow-up rates were achieved. Our results suggest that the decline in national mortality due to cardiovascular disease from the early 1970s to the mid-1990s can be explained by a decline in both cardiovascular disease incidence and case fatality rates. Although cardiovascular disease remains the leading cause of death in the United States, these results highlight improvements in the prevention and treatment of cardiovascular disease that have occurred during the past 25 years. Vigorous implementation of primary and secondary prevention efforts, and improved treatment of cardiovascular disease, should be emphasized to further reduce the societal burden of cardiovascular disease in the general population.

REFERENCES 1. National Heart, Lung, and Blood Institute. Morbidity and Mortality: 2002 Chart Book on Cardiovascular, Lung, and Blood Diseases. Rockville Maryland: U.S. Department of Health and Human Services, National Institutes of Health; 2002. 2. National Center for Chronic Disease Prevention and Health Promotion, CDC. Decline in deaths from heart disease and stroke— United States. 1900-1991. MMWR Morb Mortal Wkly Rep. 1999;48: 649 – 656. 3. Cooper R, Cutler J, Desvigne-Nickens P, et al. Trends and disparities in coronary heart disease, stroke, and other cardiovascular disease in the United States: findings of the national conference on cardiovascular disease prevention. Circulation. 2000;102:3137– 3147. 4. Goldberg RJ, Yarzebski J, Lessard D, Gore JM. A two-decades (1975 to 1995) long experience in the incidence, in-hospital and longterm case-fatality rates of acute myocardial infarction: a community-wide perspective. J Am Coll Cardiol. 1999;33:1533–1539. 5. McGovern PG, Pankow JS, Shahar E, et al. Recent trends in acute coronary heart disease. Mortality, morbidity, medical care and risk factors. N Engl J Med. 1996;334:884 – 890. 6. Rosamond WD, Chambless LE, Folsom AR, et al. Trends in the incidence of myocardial infarction and in mortality due to coronary heart disease, 1987 to 1994. N Engl J Med. 1998;339: 861– 867. 7. Shahar E, McGovern PC, Pankow JS, et al. Stroke rates during the 1980s. The Minnesota Stroke Survey. Stroke. 1997;28:275–279. 226

August 15, 2004

THE AMERICAN JOURNAL OF MEDICINE威

8. Shahar E, McGovern PG, Sprafka JM, et al. Improved survival of stroke patients during the l980s. The Minnesota Stroke Survey. Stroke. 1995;26:1– 6. 9. Brown RD, Whisnant J, Sicks JD, O’Fallon WM, Wieber DO. Stroke incidence prevalence and survival: secular trends in Rochester, Minnesota, through 1989. Stroke. 1996;27:373–380. 10. McGovern PG, Burke GL, Sprafka JM, Xue S, Folsom AR, Blackburn H. Trends in mortality, morbidity, and risk factor levels for stroke from 1960 through 1990: the Minnesota Heart Survey. JAMA. 1992:268;753–759. 11. TunstalI-Pedoe H, Kuulasmaa K, Mahonen M, et al. Contribution of trends in survival and coronary-event rates to changes in coronary heart disease mortality: 10-year results from 37 WHO MONICA project populations. Monitoring trends and determinants in cardiovascular disease. Lancet. 1999;353:1547–1557. 12. Thotvaldsen P, Kuulasmaa K, Rajakangas AM, Rastenyte D, Sarti C, Wilhelmsen L. Stroke trends in the WHO MONICA project. Stroke. 1997;28:500 –506. 13. Tunstall-Pedoe H, Vanuzzo D, Hoobs M, et al. Estimation of contribution of changes in coronary care to improving survival, event rates, and coronary heart disease mortality across the WHO MONICA Project populations. Lancet. 2000;355:688 –700. 14. Sytkowski PA, Kannel WB, D’Agostino RB. Changes in risk factors and the decline in mortality from cardiovascular disease. The Framingham Heart Study. N Engl J Med. 1990;322:1635–1641. 15. Ives DG, Fitzpatrick AL, Bild DE, et al. Surveillance and ascertainment of cardiovascular events: the Cardiovascular Health Study. Ann Epidemiol. 1995;5:278 –285. 16. Fang J, Alderman MH. Trend of stroke hospitalization, United States, 1988-1997. Stroke. 2002;32:2221–2226. 17. Elveback L, Connolly DC, Melton LJ III. Coronary heart disease in residence of Rochester, Minnesota. VII. Incidence, 1950 through 1982. Mayo Clin Proc. 1986;61:896 –900. 18. Goldberg RJ, Gore JM, Alpert JS, et al. Recent changes in attack and survival rates of acute myocardial infarction (1975 through 1982): the Worcester Heart Attack Study. JAMA. 1986;255:2774 –2779. 19. Levy DL, Thom TJ. Death rates from coronary disease-progress and a puzzling paradox. N Engl J Med. 1998;339:915–917. 20. Heidenreich PA, McClellan M. Trends in treatment and outcomes for acute myocardial infarction: 1975-1995. Am J Med. 2001;110: 165–174. 21. Russell LB, Milan E, Jagannathan R. Comparison of two surveys of hospitalization: the National Hospital Discharge Survey and the NHANES I Epidemiologic Followup Study. Vital Health Stat 2. 1997;123:I-iii,1–16. 22. Miller HW. Plan and operation of the Health and Nutrition Examination survey, United States, 1971-1973. National Center for Health Statistics. Vital Health Stat 1. 1973;10a:1– 46. 23. Cohen BB, Barbano HE, Cox CS, et al. Plan and operation of the NHANES l Epidemiologic Follow-up Study, 1982-1984. National Center for Health Statistics. Vital Stat 1. 1987;22:1–142. 24. Cox CS, Mussolino ME, Rothwell ST, et al. Plan and operation of the NHANES I Epidemiologic Follow up Study, 1992. National Center for Health Statistics. Vital Health Stat 1. 1997;35:1–231. 25. Madans JH, Reuben CA, Rothwell ST, Eberhardt MS. Differences in morbidity measures and risk factor identification using multiple data sources: the case of coronary heart disease. Stat Med. 1995;14: 643– 653. 26. Kahn HA, Sempos CT. Statistical Methods in Epidemiology. New York, New York: Oxford University Press; 1989:217. 27. Cochran WG. Sampling Techniques. 3rd ed. New York, New York: John Wiley & Sons Inc.; 1977:155. 28. Frome EK. The analysis of rates using Poisson regression models. Biometrics. 1983;39:665– 674.

Volume 117

Cardiovascular Disease Mortality, Incidence, and Case Fatality Rates/Ergin et al 29. Hellermann JP, Reeder GS, Jacobsen SJ, Weston SA, Killian JM, Roger VL. Longitudinal trends in the severity of acute myocardial infarction: a population study in Olmsted County, Minnesota. Am J Epidemiol. 2002;156:246 –253. 30. Garraway WM, Whisnant JP, Furlan AJ, Phillips LH, Kurland LT, O’Fallon WM. The declining incidence of stroke. N Engl J Med. 1979;300:449 – 452.

31. Klag MJ, Whelton PK, Seidler AJ. Decline in US stroke mortality. Demographic trends and antihypertensive treatment. Stroke. 1989; 20:14 –21. 32. Goff Jr, DC Howard G, Wang CH, et al. Trends in severity of hospitalized myocardial infarction: the Atherosclerosis Risk in Communities (ARIC) Study, 1987–1994. Am Heart J. 2000;139: 874 – 880.

August 15, 2004

THE AMERICAN JOURNAL OF MEDICINE威

Volume 117 227