It has been of note that despite the increasing obesity prevalence in the USA observed in previous decades, a decline in CVD risk factors has occurred which has been attributed to the influence of lifestyle changes and therapies [
Many studies have documented race/ethnic differences in the prevalence of all major CVD risk factors [
Data from two NHANESs conducted from 1988 to 2004 (NHANES III: 1988–1994 and NHANES 1999–2004) were used. NHANESs are a set of large national cross-sectional surveys of the USA civilian, noninstitutionalized population based on interview, examination, and laboratory information. Detailed descriptions of the plan and operation of each survey have been published [
NHANES data for both survey periods were collected via standardized questionnaires administered by bilingual interviewers at participants’ homes, standardized examinations, and nonfasting laboratory tests conducted and collected by health examiners at NHANES mobile examination centers. NHANES III and NHANES 1999–2004 underwent institutional review board approval and included written informed consent. The sample for this analysis included women and men aged 25–84 years who completed both a home-based questionnaire and the medical examination. The overall response rates for both the questionnaire and examination and the sample populations for NHANES III and NHANES 1999–2004 were 73.4% (14,341 persons) and 72.2% (12,360 persons), respectively.
Race and ethnicity were self-reported using the categories of Black, White, or other (the latter category was created by combining three other race categories) for race, and non-Hispanic, Mexican American, or other Hispanic for ethnicity. Four race/ethnic groups were defined for both survey periods: non-Hispanic White (NHW), non-Hispanic Black (NHB), Mexican American (MA), and other race/ethnicity. The first three race/ethnic groups were used for analysis. The sample sizes for the 1988–1994 and 1999–2004 study periods for NHW were 6,028 and 6,253, NHB were 4,009 and 2,434, and MA were 3,734 and 2,766, respectively.
The five CVD risk factors in this study were measured using similar methods in the two survey periods, as defined in previous publications [
Obesity was defined as body mass index (BMI) ≥30 kg/m2. Height and weight were obtained using standard protocols [
Current smoking status was determined based on positive answers to the following questions asked during the household survey: “Have you smoked at least 100 cigarettes in your life?” and “Do you smoke cigarettes now?”
Diabetes status was defined as the presence of a fasting plasma glucose ≥126 mg/dL or self-reported current use of a glucose-lowering medication (insulin or hypoglycemic medications). Blood samples for fasting plasma glucose were assessed using previously described procedures for blood collection and processes [
Hypertension was defined as systolic blood pressure (SBP) ≥140 mm Hg or diastolic blood pressure (DBP) ≥90 mm Hg or self-reported current use of antihypertensive medication. Blood pressure was measured according to standard protocols [
Hypercholesterolemia was defined as total cholesterol levels ≥200 mg/dL or self-reported use of lipid-lowering medication. Total cholesterol concentrations were collected during the medical examination and processed according to criteria of the CDC-National Heart, Lung and Blood Institute Lipid Standardization Program.
Statistical analyses were performed using SAS version 9.1 (SAS Institute Inc., Cary, NC, USA) statistical software. The estimated means and standard errors were computed using the procedures SURVEYFREQ, SURVEYMEANS, and SURVEYREG [
The distribution of age groups, sex, and education categories were generally similar among the NHW, NHB, and MA populations over the two survey periods: period I (1988–1994) and period II (1999–2004). There were some exceptions: education levels across all race/ethnic groups increased over time, and the proportion of younger (age 25–34 years) individuals decreased, while the proportions of those aged 45–54 and 75–84 increased over time. MA had increased proportions of younger individuals; both NHB and MA populations had greater proportions of less educated participants, compared with the NHW population (Table
Demographic characteristics among NHW, NHB, and MA adults, by period I (1988–1994) and period II (1999–2004).
Non-Hispanic White | Non-Hispanic Black | Mexican American | |||||
---|---|---|---|---|---|---|---|
NH 88–94 |
NH 99–04 |
NH 88–94 |
NH 99–04 |
NH 88–94 |
NH 99–04 |
||
Sex | Male | 48.3 | 48.4 | 44.5 | 43.9 | 51.6 | 53.1 |
Female | 51.7 | 51.6 | 55.5 | 56.1 | 48.4 | 46.9 | |
| |||||||
Age, years | 25–34 | 25.6 | 19.2 | 32.1 | 24.6 | 39.8 | 37.3 |
35–44 | 25.0 | 23.7 | 29.0 | 27.5 | 28.5 | 29.1 | |
45–54 | 16.7 | 23.0 | 13.8 | 22.2 | 15.2 | 17.5 | |
55–64 | 14.0 | 14.5 | 12.2 | 13.3 | 9.0 | 8.6 | |
65–74 | 12.4 | 11.7 | 8.8 | 8.2 | 5.7 | 5.4 | |
75–84 | 6.2 | 7.9 | 4.1 | 4.3 | 1.7 | 2.1 | |
| |||||||
Education | <HS | 20.1 | 13.8 | 32.9 | 33.4 | 59.5 | 55.0 |
HS | 34.8 | 27.6 | 36.9 | 24.4 | 22.2 | 19.6 | |
>HS | 45.1 | 58.6 | 30.2 | 42.2 | 18.3 | 25.4 |
Values are percentages.
NH: NHANES, HS: high school.
The prevalence of obesity significantly increased from period I to period II: NHW (22.8% to 30.8%,
Changes over time in age-adjusted prevalence of CVD risk factors among NHW, NHB, and MA adults, by sex and period I (1988–1994) and period II (1999–2004).
Current Smoking | Diabetes | Hypertension | Hypercholesterolemia | Obesity | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
I | II | Δ | I | II | Δ | I | II | Δ | I | II | Δ | I | II | Δ | |
Non-Hispanic White | |||||||||||||||
| |||||||||||||||
Men | 29.5 | 25.6 |
|
8.4 | 10.5 | 2.1 | 27.5 | 30.6 |
|
58.3 | 59.5 |
|
21.3 | 29.7 |
|
Women | 25.9 | 23.3 |
|
5.4 | 5.2 | −0.2 | 24.5 | 29.3 |
|
56.2 | 58.4 |
|
24.2 | 31.9 |
|
| |||||||||||||||
Total |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |||||||||||||||
Non-Hispanic Black | |||||||||||||||
| |||||||||||||||
Men | 42.3 | 33.9 |
|
10.3 | 11.5 | 1.3 | 38.8 | 42.5 |
|
51.6 | 51.1 |
|
21.2 | 30.1 |
|
Women | 28.8 | 22.1 |
|
12.3 | 13.9 | 1.6 | 40.3 | 45.6 |
|
56.0 | 50.0 |
|
40.3 | 52.0 |
|
| |||||||||||||||
Total |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |||||||||||||||
Mexican-American | |||||||||||||||
| |||||||||||||||
Men | 27.6 | 27.4 |
|
13.7 | 15.4 | 1.7 | 28.1 | 28.2 |
|
58.5 | 56.9 |
|
25.7 | 30.0 |
|
Women | 14.0 | 13.8 |
|
15.3 | 15.0 | −0.3 | 26.3 | 30.9 |
|
55.6 | 54.3 |
|
37.5 | 40.6 |
|
| |||||||||||||||
Total |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Values are percentages, except that the change column values are percentage points.
I = NHANES 1988–1994; II= NHANES 1999–2004.
The asterisks (* and **) indicate significant changes overtime,
Δ = change.
Prevalence of current smoking decreased significantly from period I to period II for NHW (27.6% to 24.4%,
Current smoking prevalence and race/ethnic disparity in each period and over time; NHANES periods I (1988–1994) and II (1999–2004).
Prevalence in each period and change over time (%)
Race/ethnic disparity in each period and over time (percentage points)
Prevalence of diabetes nonsignificantly increased over time for all race/ethnic groups from period I to period II: NHW (6.8% to 7.8%); NHB (11.5% to 13%); MA (14.6% to 15.3%). (Table
Diabetes prevalence and race/ethnic disparity in each period and over time; NHANES periods I (1988–1994) and II (1999–2004).
Prevalence in each period and change over time (% and percentage points)
Change in race/ethnic disparity in each period and over time (percentage points)
Hypertension prevalence and change in race/ethnic disparity in each period and over time; NHANES periods I (1988–1994) and II (1999–2004).
Prevalence in each period and change over time (%)
Change in race/ethnic disparity over time (percentage points)
Obesity prevalence and race/ethnic disparity in each period and over time; NHANES period I (1988–1994) and period II (1999–2004).
Prevalence in each period and change over time (% and percentage points)
Change in race/ethnic disparity in each period and over time (percentage points)
Prevalence of hypertension significantly increased from period I to period II, NHW (26.2% to 30.2%,
The prevalence of hypercholesterolemia decreased significantly from period I to period II in NHB (54.2% to 50.5%,
This study documents a significant increase in the prevalence of obesity and hypertension between 1988 and 2004 in NHW and NHB, with an increasing trend observed in MA. Also, an increasing trend (nonsignificant) in the prevalence of diabetes over time in all race/ethnic groups was found. This is consistent with the increasing prevalence of obesity and suggests that, in the future, unless effective methods to halt the obesity epidemic are implemented, the prevalence of diabetes will likely augment. Increases in diabetes prevalence have been reported in the last decade, which has disproportionately affected MA and NHB populations [
The finding that the prevalence of current smoking for most race/ethnic groups significantly decreased is encouraging, as is the narrowing of race/ethnic disparities in the prevalence of smoking over time. However, NHB still has the highest prevalence of current smoking and race/ethnic disparities persisted in 1999–2004; therefore, continuedpublic health efforts are needed to target this population, especially NHB men.
In this study, the prevalence of hypercholesterolemia significantly decreased for NHB and trended downward for MA and upward for NHW despite an increase in obesity prevalence. While this may appear paradoxical, previous studies have suggested that prevalence of high cholesterol and an elevated BMI are not necessarily related [
The increase in obesity prevalence among USA adults over time has been reported previously [
Growing trends in obesity prevalence may lead to increases in diabetes and hypertension as previously reported in other studies [
Although obesity prevention should be the primary public health focus in order to address future worsening trends in diabetes and hypertension, there are multiple challenging factors involved in a population-oriented strategy to address these trends. Observational studies and a randomized long-term social experiment recently published have shown the impact of neighborhood characteristics and built environment in the prevalence of obesity and diabetes [
The relationship between lifestyle-related behaviors, such as increased physical activity and healthful dietary habits, with obesity prevention and lowering of CVD risk is well known [
Our study has several limitations. First, race and ethnicity disparities research is a relatively burgeoning field, and controversy exists as to the interpretation of differences based on race and ethnicity data alone [
Nevertheless, our study is based on comprehensive national surveys for USA adults, including oversampling of NHB and MA individuals. Only since the release of 2002–2004 NHANES survey could accurate and nationally representative estimates among race/ethnic groups be used to examine changes over time for 10-year trend analysis.
Important temporal trends in the prevalence of major CVD risk factors in a representative sample of USA NHW, NHB, and MA adults were documented: a worrisome increase in the prevalence of obesity and hypertension in most groups and a decrease in the prevalence of current smoking. To what extent these unfavorable CVD risk factors trends may be offset by the decline in smoking is unknown. This study also found persistent race/ethnic differences for all CVD risk factors and time periods examined, with NHB and MA generally having worse profiles than NHW. Importantly, race/ethnic disparities did not improve over time (with the exception of smoking); in fact, they widened for obesity and hypercholesterolemia. Disproportionate increases in the prevalence of obesity may be implicated in persistent disparities observed for hypertension and diabetes currently and in the future. Furthermore, recent data suggest that the decline in CVD mortality observed in the last 50 years may be leveling off due to alarming increases in the prevalence of modifiable CVD risk factors [
None of the authors has any financial interest or gains from the Health Resources and Services Administration or from Fundacion Araucaria Foundation.
Dr. C. X. Romero was supported by Grant no. D33HP02610 for Preventive Medicine Residencies, from the Health Resources and Services Administration (HRSA). The contents are solely the responsibility of the authors and do not necessarily represent the official views of HRSA. Fundacion Araucaria Foundation assisted in the presentations of some of these data to the World Congress of Cardiology (Beijing, 2010) and the American College of Cardiology Scientific Sessions (New Orleans, 2011) and in the preparation and publication of the paper.