1200
JACC Vol. 30, No. 5
November 1, 1997:1200 –5
ATHEROSCLEROSIS
Mortality From Coronary Heart Disease and Cardiovascular Disease
Among Adult U.S. Hispanics: Findings From the National Health
Interview Survey (1986 to 1994)
YOULIAN LIAO, MD, RICHARD S. COOPER, MD, FACC, GUICHAN CAO, MS,
JAY S. KAUFMAN, PHD, ANDREW E. LONG, PHD, DANIEL L. MCGEE, PHD
Maywood, Illinois
Objectives. We sought to estimate the coronary heart disease
(CHD) and cardiovascular disease (CVD) mortality experience of
U.S. Hispanics.
Background. Limited information is available concerning the
mortality from CHD among U.S. Hispanics, the nation’s second
largest minority group.
Methods. The study used data from the National Health
Interview Survey (1986 to 1994), including representative national
samples of 246,239 non-Hispanic whites, 38,042 blacks and 14,965
Hispanics who were >
245 years old at baseline. Mean follow-up of
mortality was 5 years (range 1 to 10).
Results. During the follow-up period, 27,702 whites (11%),
4,976 blacks (13%) and 1,061 Hispanics (7%) died. Among men,
the age-adjusted total mortality per 100,000 person-years was
3,089 in whites and 2,466 in Hispanics, and among women, it was
1,897 and 1,581 in whites and Hispanics, respectively. The His-
panic/white mortality rate ratio for CHD was 0.77 (95% confidence
interval [CI] 0.64 to 0.93) and 0.82 (95% CI 0.66 to 1.01) for men
and women, respectively. The rate ratio was 0.79 (95% CI 0.68 to
0.91) and 0.80 (95% CI 0.69 to 0.94), respectively, for mortality
from cardiovascular diseases. Given the lower all-cause mortality
in Hispanics, the proportion of total deaths due to CHD and CVD
was similar between the two populations for the same gender and
were, respectively, 29.7% and 44.7% in white men, 28.1% and
44.3% in Hispanic men, 24.9% and 43.2% in white women and
24.1% and 41% in Hispanic women.
Conclusions. These data from a cohort of a large national
sample are consistent with vital statistics that show that all-cause,
CHD and CVD mortality is ;20% lower among adult Hispanics
than among whites in the United States.
(J Am Coll Cardiol 1997;30:1200 –5)
©1997 by the American College of Cardiology
The Hispanic population, which numbers 22.4 million people
in the continental United States, is the second largest minority
group and is increasing at a rate five times that of the rest of
the United States (1). It has been estimated that Hispanics will
become the largest minority group early in the 21st century (2).
However, a comprehensive description of the mortality experience of Hispanics is still lacking. Vital statistics data show
that cardiovascular mortality has been lower in Hispanic men
(3–7) and in both men and women, (8 –10) than in nonHispanic whites. Follow-up data on coronary heart disease
(CHD) mortality from cohort studies are scarce (11,12). The
present report used multiple cause of death data from the
National Health Interview Survey (NHIS) (1986 to 1994),
including ;15,000 Hispanics, ;250,000 non-Hispanic whites
and ;38,000 blacks $45 years old. The primary purpose of the
present analysis was to estimate and compare the CHD and
cardiovascular disease (CVD) mortality experience of U.S.
Hispanics, blacks and whites.
From the Department of Preventive Medicine and Epidemiology, Loyola
University Stritch School of Medicine, Maywood, Illinois. This work was
supported by a cooperative agreement (U83/CCU512480) from the Centers for
Disease Control, Atlanta, Georgia.
Manuscript received March 7, 1997; revised manuscript received July 1, 1997,
accepted July 9, 1997.
Address for correspondence: Dr Youlian Liao, Department of Preventive
Medicine and Epidemiology, Loyola University Stritch School of Medicine, 2160
South First Avenue, Maywood, Illinois 60153. E-mail: [email protected].
©1997 by the American College of Cardiology
Published by Elsevier Science Inc.
Methods
NHIS. As one of the major data collection programs of the
National Center for Health Statistics, NHIS is a principal
source of information on the health of the resident, civilian,
noninstitutionalized population of the United States (13). This
nationwide survey has been conducted continuously since 1957
and is based on a multistage area probability sample through
personal household interview. Most households chosen are
contacted by mail before the interviewers arrive. Interviewers
make repeated trips to households when respondents have not
been available. Data are collected for all family members.
Proxy responses are used for children and adults who are not
available or are unable to respond for themselves. The average
annual sample consists of ;45,000 households, including
120,000 persons (86,000 $18 years old). The response rate was
96% to 98% over the years analyzed. Race and Hispanic origin
were determined according to the participant’s choice of the
ethnic designation that best represented his or her race: Aleut,
Eskimo or American Indian; Asian or Pacific Islander; black,
0735-1097/97/$17.00
PII S0735-1097(97)00278-7
JACC Vol. 30, No. 5
November 1, 1997:1200 –5
Abbreviations and Acronyms
CHD 5 coronary heart disease
CI
5 confidence interval
CVD 5 cardiovascular disease
NHIS 5 National Health Interview Survey
white or other. Then he or she was asked whether his or her
national origin or ancestry was any of the following: Puerto
Rican, Cuban, Mexican/Mexicano, Mexican American, Chicano, other Latin American or other Spanish. Those who
responded as having any of the above national origins were
considered Hispanic. The present report focuses on persons
$45 years old only. Because the numbers were small for
persons other than the three primary groups of interest, they
were omitted from this report (n 5 7,568 [2.4%]). The term
“white” was used in this report to represent non-Hispanic
whites.
Family income was determined by asking which category
best represented the total combined family income during the
past 12 months from wages, salaries and all other sources. The
27 income categories ranged from ,$1,000 (including loss) to
$$50,000, with $1,000 increments for each category up to
$20,000 and $5,000 increments thereafter to $49,999. Data on
annual family income were available for 81% of the respondents in the study.
Mortality follow-up. Beginning with survey year 1986, vital
status of the NHIS respondents $18 years old is matched with
files in the National Death Index system (14). The system is a
computer database of all deaths in the United States since 1979
and has been shown (15) to provide a high level of death
ascertainment. To date, multiple cause of death data are
available for NHIS survey years of 1986 to 1994, with follow-up
to include all deaths that occurred through December 31, 1995.
The matching methodology used in linking to the National
Death Index is a modification of the probabilistic approaches
(16). We used the algorithm provided by the National Center
for Health Statistics (17) to determine which potential matches
should be classified as deaths. Sufficient information to permit
linkage with the National Death Index was unavailable in
10,524 persons (3.4%), and these persons were excluded from
the analysis.
Causes of death were coded using the 9th Revision of the
International Classification of Diseases. Codes 410 to 414 and
429.2 were defined as CHD; codes 430 to 438 as cerebrovascular disease; codes 390 to 448 as CVD; codes 140 to 208 as
cancers; and all other codes as other causes. Both underlying
cause and multiple cause (any mention) of death were considered in the analyses.
Data analysis. Death rates were calculated for each gender
and race group by dividing the number of deaths during 1986
to 1995 by the total number of person-years of follow-up. The
95% confidence intervals for the black/white and Hispanic/
white mortality ratios were estimated by the delta method (18).
LIAO ET AL.
CORONARY HEART DISEASE MORTALITY IN HISPANICS
1201
Age standardization was accomplished by the direct method
for 10-year intervals using the entire cohort as the reference.
The Cox proportional hazards model was used to estimate the
relative risk of mortality for blacks and Hispanics compared
with whites, adjusted for age (years) and family income. Family
income for different survey years was standardized to the
equivalent 1990 dollar value. The software program SUDAAN
(19) was used to account for the complex sampling design of
the study. Analyses were first performed for participants from
survey years 1986 to 1989 and 1990 to 1994 separately. The
results of racial comparisons were comparable for the two
analytic cohorts with lower statistical power. The final report
was presented with combined data of 1986 to 1994.
Results
Source of participants. During the years 1986 to 1994, the
NHIS sampled ;300,000 persons $45 years old from the
U.S. population, of whom 246,239 were whites, 38,042 were
blacks, and 14,965 were of Hispanic origin (Table 1). Among
both men and women, Hispanics were younger than either
whites or blacks; thus, all analyses were performed with age
adjustment. The mortality follow-up period ranged from 1 to
10 years (mean 5), with 27,702 white (11%), 4,976 black (13%)
and 1,061 Hispanic deaths (7%).
All-cause and cause-specific mortality. All-cause and specific underlying cause of deaths, age-adjusted mortality rates
(per 100,000 person-years), rate ratios, using whites as the
reference, and their 95% confidence intervals are shown in
Table 2. Compared with whites, the total mortality was greater
in blacks and lower in Hispanics for both men and women. The
95% confidence intervals of the rate ratios were all significantly
different from 1.0. The specific causes with a significantly
greater rate ratio for blacks included CHD and cerebrovascular disease in women and CVD in both men and women. In
general, Hispanics had ;20% lower CHD and CVD mortality
than did whites. The rate ratios were significantly lower than
1.0, except for CHD among women. The confidence intervals
of the Hispanic/white ratio for cerebrovascular disease were
wide as the result of the small number of deaths among
Hispanics.
When mortality rates for CHD and CVD were estimated
using any mention of them on the death certificates, the
Hispanic/white rate ratios were essentially unchanged among
men (CHD: 0.77, 95% confidence interval [CI] 0.65 to 0.92,
CVD: 0.81, 95% CI 0.72 to 0.91). Among women, the ratio
increased to 0.93 (95% CI 0.78 to 1.10) for CHD and 0.93
(95% CI 0.83 to 1.04) for CVD, respectively.
Components of causes of death. The age-adjusted proportion of total deaths due to various causes in Hispanics and
whites are presented in Figure 1 for men and Figure 2 for
women. These data provide an indication of the relative
importance of various causes of death in each subpopulation.
There was no apparent racial difference in the proportion of
death due to CHD among both men (28.1% vs. 29.7%,
Hispanics vs. whites) and women (24.1% vs. 24.9%). Likewise,
1202
LIAO ET AL.
CORONARY HEART DISEASE MORTALITY IN HISPANICS
JACC Vol. 30, No. 5
November 1, 1997:1200 –5
Table 1. Number of Respondents at Baseline and Deaths During Follow-Up by Gender, Race and
Age Groups
Age
Group
(yr)
Whites
No. (%) at
Baseline
Blacks
No. of
Deaths
No. (%) at
Baseline
Hispanics
No. of
Deaths
No. (%) at
Baseline
No. of
Deaths
354
621
828
701
2,504
3,058 (46)
2,048 (31)
1,096 (16)
514 (8)
6,716 (100)
95
156
161
162
574
305
557
721
889
2,472
3,509 (43)
2,405 (29)
1,538 (19)
797 (10)
8,249 (100)
56
109
142
180
487
Men
45–54
55– 64
65–74
751
Total
39,159 (35)
32,032 (29)
26,715 (24)
14,357 (13)
112,263 (100)
1,275
2,886
5,112
5,367
14,640
5,541 (36)
4,676 (30)
3,444 (22)
1,765 (11)
15,426 (100)
Women
45–54
55– 64
65–74
751
Total
40,771 (30)
35,406 (26)
33,749 (25)
24,050 (18)
133,976 (100)
887
1,891
3,960
6,324
13,062
7,900 (35)
6,522 (29)
5,033 (22)
3,161 (14)
22,616 (100)
similar proportions due to CVD were found (44.3% vs. 44.7%
for men, 41.0% vs. 43.2% for women, respectively). The
proportion of deaths due to cancer was smaller in Hispanics
than in whites for both men (23.7% vs. 28.4%, Hispanics vs.
whites) and women (23.8% vs. 29.6%) but greater for other
causes (32.1% vs. 26.8% for men, 35.2% vs. 27.2% for women,
respectively).
Effect of income difference. The standardized mean annual
family income was, respectively, $35,800 and $30,400 among
white men and women, $25,200 and $20,500 among blacks and
$28,700 and $25,800 among Hispanics. Lower income was
significantly associated with increased all-cause and causespecific mortality in each group (data not shown). Adjustment
for income, in addition to age, narrowed the gap of mortality
from CHD and CVD between blacks and whites (Table 3). As
expected, the relative risk for Hispanics compared with whites
was reduced, to a smaller degree, however, after the differences in income distribution were considered.
Discussion
Vital statistics. Until recently our knowledge about CVD
mortality in Hispanic populations in the United States was
limited mostly to data from vital statistics. Before the early
1980s, vital statistics from New Mexico (3,8), Texas (4,6) and
California (5) indicated lower CHD and CVD among Hispanic
Table 2. Age-Adjusted All-Cause and Cause-Specific Mortality by Gender and Race: National Health
Interview Survey, 1986 to 1994
Men
Mortality
All causes
Whites
Blacks
Hispanics
CHD
Whites
Blacks
Hispanics
CVA
Whites
Blacks
Hispanics
CVD
Whites
Blacks
Hispanics
Women
No. of
Deaths
Rate*
Rate Ratio
(95% CI)
No. of
Deaths
Rate*
Rate Ratio
(95% CI)
14,640
2,504
574
3,089
3,895
2,466
1.00
1.26 (1.20 –1.33)
0.80 (0.73– 0.88)
13,062
2,472
487
1,897
2,339
1,581
1.00
1.23 (1.17–1.29)
0.83 (0.76 – 0.92)
4,326
604
153
928
951
717
1.00
1.02 (0.93–1.13)
0.77 (0.64 – 0.93)
3,390
616
113
490
605
401
1.00
1.23 (1.12–1.36)
0.82 (0.66 –1.01)
807
138
45
178
218
198
1.00
1.22 (0.99 –1.50)
1.11 (0.78 –1.58)
1,043
191
39
149
191
131
1.00
1.28 (1.07–1.53)
0.88 (0.63–1.22)
6,500
1,063
243
1,398
1,680
1,101
1.00
1.20 (1.12–1.29)
0.79 (0.68 – 0.91)
5,880
1,167
193
844
1,144
682
1.00
1.36 (1.26 –1.46)
0.80 (0.69 – 0.94)
*Age-adjusted rate/100,000 person-years. CI 5 confidence interval; CHD 5 coronary heart disease; CVA 5
cerebrovascular accident; CVD 5 cardiovascular disease.
JACC Vol. 30, No. 5
November 1, 1997:1200 –5
LIAO ET AL.
CORONARY HEART DISEASE MORTALITY IN HISPANICS
1203
Table 3. Adjusted Risk of Death From Coronary Heart and
Cardiovascular Disease: Blacks and Hispanics Versus Whites
Adjustment
CHD
[RR (95% CI)]
CVD
[RR (95% CI)]
Men
Black vs. white
Age
Age and income
Hispanic vs. white
Age
Age and income
1.07 (0.96 –1.20)
0.92 (0.82–1.03)
1.29 (1.18 –1.41)
1.12 (1.03–1.23)
0.75 (0.61– 0.93)
0.68 (0.55– 0.84)
0.79 (0.67– 0.94)
0.72 (0.62– 0.85)
Women
Figure 1. Proportional mortality in white and Hispanic men, $45
years old: NHIS, 1986 to 1994. CVA 5 cerebrovascular accident.
American men than non-Hispanic men, but less clearly so in
Hispanic American women. Recent reports (7,10) extended
the findings to the subsequent decade. There was no evidence
to indicate that the mortality gap between Mexican American
and non-Hispanic white men was closing. A specific item about
Hispanic origin was included in the death certificate in 15
states after 1980 and was gradually extended to 49 states and
the District of Columbia by 1993. At the national level,
Hispanics experienced about two-thirds the heart disease
mortality of non-Hispanic whites for both men and women (9).
The major weaknesses of using vital statistics to estimate
mortality in Hispanics are misclassification (i.e., incorrectly
assigning race to decedents) and underregistration (20). Most
misclassified decedents are falsely reported to be white, understating Hispanic mortality. Comparing the information on
death certificates and the National Mortality Followback Survey demonstrated that 19.6% more Hispanic decedents were
reported on the questionnaire than on the death certificate
(21). In contrast, the 1980 census undercounted Hispanics by
as much as 8%, blacks by 6% and white and others by ,1%
Figure 2. Proportional mortality in white and Hispanic women, $45
years old: NHIS, 1986 to 1994. CVA 5 cerebrovascular accident.
Black vs. white
Age
Age and income
Hispanic vs. white
Age
Age and income
1.28 (1.13–1.44)
1.14 (1.01–1.29)
1.42 (1.30 –1.54)
1.29 (1.19 –1.40)
0.83 (0.66 –1.04)
0.80 (0.63–1.00)
0.85 (0.72–1.01)
0.83 (0.70 – 0.98)
RR 5 relative risk; other abbreviations as Table 2.
(22). The aggregate effect of these errors is not precisely
known.
Cohort studies. To date, two cohort studies (11,12) have
reported mortality from CVD among the Hispanic population.
The National Longitudinal Mortality Study (11) followed up
;700,000 respondents, including 40,000 Hispanics, from 1979
to 1987. Hispanics ($25 years old) had significantly lower
CVD mortality than did whites (standardized rate ratios: 0.65
for men, 0.80 for women) (11). The recent 8-year follow-up of
the San Antonio Heart Study (12) reported that the death rate
from CVD was nonsignificantly higher (rate ratio 1.3) in
Mexican Americans than non-Hispanic whites. The number of
deaths for all cohorts was only 137; hence, the analysis was
performed with men and women combined.
Coronary heart disease in Hispanics. A less favorable
cardiovascular risk factor profile has been documented in
Hispanics than in non-Hispanic whites. On average, Hispanics
have higher body mass indexes (23), more central obesity
(23,24) and lower high density lipoprotein cholesterol and
higher triglycerides levels (23). Cigarette smoking (23,25) and
diabetes (23,26,27) are more prevalent in Hispanics, and the
prevalence of hypertension is similar to that in whites (28).
Overall, Hispanics have higher risk scores for CHD than
non-Hispanic whites (23), and the impact of the major risk
factors is similar to that of whites (12). Hispanics also have
lower levels of socioeconomic status and health insurance
coverage and utilize fewer preventive services than nonHispanic whites (29 –31). All these factors would imply higher
CHD morbidity and mortality in Hispanics. Morbidity data
were scarce, mainly from local areas and based on a small
number of events. The prevalence of CHD in Hispanics was
found to be lower than (32) or similar to (33) that in
non-Hispanic whites. CHD incidence has been reported (33)
to be similar among nondiabetic Hispanics and non-Hispanic
whites but lower in diabetic Hispanics. The Corpus Christi
1204
LIAO ET AL.
CORONARY HEART DISEASE MORTALITY IN HISPANICS
Heart Project (34) found that the rate of admission to coronary
care units was higher in Mexican Americans than in nonHispanic whites, and case fatality after myocardial infarction
was greater (35). Published data on sudden cardiac death and
out of hospital CHD deaths in Hispanics are unavailable.
Overall, a description of CHD in Hispanics is still incomplete.
Lower than expected CVD mortality in Hispanics has led
some investigators to formulate the concept of the “Hispanic
paradox” (36). Several hypotheses have been invoked, including the “healthy migrant” effect—a diet that includes more
fruits and vegetables, less chronic disease risk exposure before
leaving the country of birth and health-preserving cultural and
psychosocial effects (32,36). However, the findings of the
present study cannot be used to support the claim that
Hispanics are specifically immune to CHD. Although absolute
mortality rates from CHD were lower, the proportion of
deaths due to this cause was similar between Hispanics and
whites. Unlike cancer and other causes of death, the extent of
a lower Hispanic/white ratio of death from CHD was only
parallel to that of total mortality.
Limitations and unanswered questions. The overall response rate is very high in NHIS (96% to 98%), but a racial
difference in the response may exist. The response rate was
somewhat lower in the Hispanic Health and Nutrition Examination Survey (1982 to 1984) than that in the other National
Health and Nutrition Examination Surveys (37). However,
nonresponse to the present survey has been much smaller than
nonresponse to the Health and Nutrition Examination Surveys. This result is due primarily to the NHIS practice of
allowing proxy response to medical history questions. Hence,
ill and elderly persons are more likely to be included in this
study.
The present study used the National Death Index as the
source for mortality follow-up data. In a validation study (17),
the recommended cutoff matching scores to determine vital
status correctly classified .97% of the true matches and .97%
of the false matches for all race/ethnic groups combined. The
correct classification rates for known decedents who are nonwhites dropped to 86%, whereas the rate for living persons
remained high at ;97%. Misclassification of vital status could
be introduced by the reliance on matching by the social
security number. Hispanics may be less likely to have a social
security number or one that is accurate (38). In the 1990 NHIS,
66.6% of non-Hispanic whites, 63.2% of blacks and 59.3% of
Hispanics provided their social security number (unpublished
data, National Center for Health Statistics). Other potential
difficulties associated with using computer matching of personal identifiers for Hispanics include the Spanish surname.
For hyphenated last names, the present study submitted both
portions, as well as the hyphenated form.
It has been suggested (38,39) that some Hispanics may
return to their country of birth when they retire or have a
potentially fatal illness, especially those with a lower education,
lower wages and fewer employment experiences. Selective
outmigration would obviously lead to underestimates of mortality with the National Death Index. The follow-up period of
JACC Vol. 30, No. 5
November 1, 1997:1200 –5
this cohort is relatively short, which may result in less bias
because fewer persons would be expected to return to their
home countries.
There are limitations in using the death certificate to
classify cause of death. It has been shown (40,41) that blacks
have a higher out of hospital death rate than whites, and the
coding of CHD on death certificates is less specific for blacks
than for whites. The cause of death for out of hospital deaths
is often based on minimal to nonexistent information and is
often signed as due to CHD because of a lack of better
diagnoses. In contrast, death from CHD may be attributed to
other causes because many physicians may believe that this
disease is rare in blacks. Whether a similar ethnic bias in the
coding of death certificates also exists in Hispanics deserves
further study. In the present study, when any mentioned cause
instead of underlying cause of death on the death certificate
was used, the gap of CHD mortality between Hispanic women
and white women reduced sizably. Although drawn from a
large survey, there is a relatively small number of events for
Hispanic women and insufficient statistical power to detect the
racial difference.
Conclusions. These data from a large national random
sample demonstrate that mortality from CHD and CVD is
lower in adult Hispanics than that observed among whites in
the United States. However, the proportion of total deaths due
to these two diseases is similar between the two racial groups.
More data on the prevalence, incidence and case fatality of
CHD as well as descriptive information from vital statistics and
cohort studies of the Hispanic population will need to be
accumulated so as to have a complete picture of CHD in this
increasing portion of our population.
References
1. U.S. Department of Commerce News. Census Bureau releases refined
estimates from postenumeration survey of 1990 census coverage. Washington (D.C.): Bureau of the Census, June 13, 1991:CB91–221.
2. Council on Scientific Affairs, American Medical Association. Hispanic
health in the United States. JAMA 1991;265:248 –52.
3. Buechley RW, Key CR, Morris DL, Morton WE, Morgan MV. Altitude and
ischemic heart disease in tricultural New Mexico: an example of confounding. Am J Epidemiol 1979;109:663– 6.
4. Bradshaw BS, Fonner E Jr. The mortality of Spanish-surname persons in
Texas, 1969 –71. In: Bean FD, Frisbie WP, editors. The Demography of
Racial and Ethnic Groups. New York: Academic Press, 1978:261– 82.
5. Schoen R, Nelson VE. Mortality by cause among Spanish surnamed
Californians, 1969 –71. Soc Sci Q 1981;62:259 –73.
6. Stern MP, Bradshaw BS, Eifler CW, Fong DS, Hazuda HP, Rosenthal M.
Secular decline in death rates due to ischemic heart disease in Mexican
Americans and non-Hispanic whites in Texas, 1970 –1980. Circulation 1987;
76:1245–50.
7. Goff DC Jr, Ramsey DJ, Labarthe DR, Nichaman MZ. Acute myocardial
infarction and coronary heart disease mortality among Mexican Americans
and non-Hispanic whites in Texas, 1980 through 1989. Ethnicity Dis 1993;
3:64 –9.
8. Becker TM, Wiggins C, Key CR, Samet JM. Ischemic heart disease mortality
in Hispanics, American Indians, and non-Hispanic whites in New Mexico,
1958 –1982. Circulation 1988;78:302–9.
9. National Center for Health Statistics. Health, United States, 1992 and
healthy people 2000 review. Hyattsville (MD): Public Health Service, 1993.
10. Wild SH, Laws A, Fortmann SP, Varady AN, Byrne CD. Mortality from
JACC Vol. 30, No. 5
November 1, 1997:1200 –5
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
coronary heart disease and stroke for six ethnic groups in California, 1985 to
1990. Ann Epidemiol 1995;5:432–9.
Sorlie PD, Backlund E, Johnson NJ, Rogot E. Mortality by Hispanic status
in the United States. JAMA 1993;270:2464 – 8.
Wei M, Mitchell BD, Haffner SM, Stern MP. Effects of cigarette smoking,
diabetes, high cholesterol, and hypertension on all-cause mortality and
cardiovascular disease mortality in Mexican Americans: the San Antonio
Heart Study. Am J Epidemiol 1996;144:1058 – 65.
Massey JT, Moore TF, Parsons VL, Tadros W. Design and estimation for the
National Health Interview Survey, 1985–94: National Center for Health
Statistics. Vital Health Stat [2] 1989;110:1–33.
Patterson JE, Bilgrad R. The National Death Index experience: 1981–1985.
In: Proceedings of the Workshop on Exact Matching Methodologies, May
9 –10, 1985. Washington (D.C.):Statistics of Income Division, Internal Revenue Service;1985:245–254. Publication No. 1299 (2– 86).
Stampfer MJ, Willett WC, Speizer FE, et al. Test of the National Death
Index. Am J Epidemiol 1984;119:837–9.
Rogot E, Sorlie P, Johnson NJ. Probabilistic methods in matching census
samples to the National Death Index. J Chronic Dis 1986;39:719 –34.
National Center for Health Statistics. Public Use Data File Documentation.
National Health Interview Survey Multiple Cause of Death, 1986 –1994
survey years: 1– 40.
Miller RG Jr. Survival Analysis. New York: Wiley, 1981:25–7.
Shah BV, Barnwell BG, Bieler GS. SUDAAN User’s Manual: Software for
Analysis of Correlated Data, Release 6.40. Research Triangle Park (NC):
Research Triangle Institute, 1995.
Hahn RA. The state of federal health statistics on racial and ethnic groups.
JAMA 1992;267:268 –71.
Poe GS, Powell-Griner E, McLaughlin JK, Placek PJ, Thompson GB,
Robinson K. Comparability of the death certificate and the 1986 National
Mortality Followback Survey. Vital Health Stat [2] 1993;118:1–53.
Fay RE, Passel JS, Robinson GJ, Gowan CD. The coverage of the
population in the 1980 census. Washington (D.C.): Bureau of the Census,
1988. U.S. Dept. of Commerce Publication No. PHC80-E4.
Mitchell BD, Stern MP, Haffner SM, Hazuda HP, Patterson JK. Risk factors
for cardiovascular mortality in Mexican-Americans and non-Hispanic whites:
the San Antonio Heart Study. Am J Epidemiol 1990;131:423–33.
Burchfiel CM, Hamman RF, Marshall JA, Baxter J, Kahn LB, Amirani
JJ. Cardiovascular risk factors and impaired glucose tolerance: the San Luis
Valley Diabetes Study. Am J Epidemiol 1990;131:57–70.
Haynes SG, Harvey C, Montes H, Nickens H, Cohen BH. Patterns of
cigarette smoking among Hispanics in the United States: results from
HHANES 1982–1984. Am J Public Health 1990;80 Suppl:47–54.
Flegal KM, Ezzati TM, Harris MI, et al. Prevalence of diabetes in Mexican
LIAO ET AL.
CORONARY HEART DISEASE MORTALITY IN HISPANICS
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
1205
Americans, Cubans, and Puerto Ricans for the Hispanic Health and
Nutrition Examination Survey, 1982–1984. Diabetes Care 1991;14 Suppl
3:628 –38.
Baxter J, Hamman RF, Lopez TK, Marshall JA, Hoag S, Swenson CJ. Excess
incidence of known non-insulin-dependent diabetes mellitus (NIDDM) in
Hispanics compared with non-Hispanic whites in the San Luis Valley,
Colorado. Ethnicity Dis 1993;3:11–21.
Caralis PV. Hypertension in the Hispanic-American population. Am J Med
1990;88 Suppl 3B:9 –16.
Andersen RM, Lewis SZ, Giachello AL, Aday LA, Chiu G. Access to
medical care among the Hispanic population of the southwestern United
States. J Health Social Behav 1981;22:78 – 89.
Trevino FM, Moyer ME, Valdez RB, Stroup-Benham CA. Health insurance
coverage and utilization of health services by Mexican Americans, mainland
Puerto Ricans, and Cuban Americans. JAMA 1991;265:233–37.
Ginzberg E. Access to health care for Hispanics. JAMA 1991;265:238 – 41.
Mitchell BD, Hazuda HP, Haffner SM, Patterson JK, Stern M. Myocardial
infarction in Mexican-Americans and non-Hispanic whites. Circulation
1991;83:45–51.
Rewers M, Shetterly SM, Hoag S, Baxter J, Marshall J, Hamman RF. Is the
risk of coronary heart disease lower in Hispanics than in non-Hispanic
whites? The San Luis Valley Diabetes Study. Ethnicity Dis 1993;3:44 –54.
Nichaman MZ, Wear ML, Goff DC Jr, Labarthe DR. Hospitalization rates
for myocardial infarction among Mexican-Americans and Non-Hispanic
whites: the Corpus Christi Heart Project. Ann Epidemiol 1993;3:42– 8.
Goff DC Jr, Ramsey DJ, Labarthe DR, Nichaman MZ. Greater case-fatality
after myocardial infarction among Mexican Americans and women than
among non-Hispanic whites and men: the Corpus Christi Heart Project.
Am J Epidemiol 1994;139:474 – 83.
Markides KS, Coreil J. The health of Hispanics in the Southwestern United
States: an epidemiologic paradox. Public Health Rep 1986;101:253– 65.
Rowland ML, Forthofer RN. Investigation of nonresponse bias: Hispanic
Health and Nutrition Examination Survey (National Center for Health
Statistics). Vital Health Stat [2] 1993;119:1–75.
Herrera CR, Stern MP, Goff D, Villagomez E. Mortality among Hispanics
[letter]. JAMA 1994;271:1237.
Reyes B. Dynamics of Immigration: Return Migration to Western Mexico.
San Francisco (CA): Public Policy Institute of California, 1997:1–98.
Lee MH, Borhani NO, Kuller LH. Validation of reported myocardial
infarction mortality in blacks and whites: a report from the Community
Cardiovascular Surveillance Program. Ann Epidemiol 1990;1:1–12.
Kuller LH. Commentary on coronary heart disease in blacks. Public Health
Rep 1995;110:570 –2.