The Prevalence of Hypertension in Older Mexicans and Mexican Americans

Abstract

Objective

To evaluate the prevalence of hypertension in older Mexicans in the United States and Mexico.

Design

Stratified by sex, logistic regression models to predict physician diagnosed hypertension were conducted using the Hispanic EPESE (wave 3) and the Mexican Health and Aging Study (MHAS- 70 years and older) datasets.

Setting

Five Southwestern States of Texas, Arizona, California, Colorado, and New Mexico in the United States.

Participants

Older Mexican and Mexican Americans ages 70 and over living in the United States and Mexico.

Main Outcome Measures

Physician diagnosed hypertension.

Results

Older Mexican and Mexican American women have a greater prevalence of hypertension than their male counterparts. Mexican women, who have migrated to the United States and returned to Mexico, have similarly high rates of hypertension as their female counterparts in the United States. After adjusting for demographic characteristics, obesity, and smoking, older Mexican and Mexican American women who have migrated or immigrated to the United States are at increased risk for hypertension.

Conclusions

Gender differences exist in hypertension risk for older Mexicans and Mexican Americans living in the United States and Mexico. Older women who migrate to the United States are at a particular risk for hypertension in both the United States and Mexico.

INTRODUCTION

Despite their greater likelihood of living in poverty, Mexican Americans have hypertension prevalence rates similar to non-Hispanic whites (non-Hispanic white males 23.2%, females 23.5%; Mexican American males 19.7%, females 22.1%) [1,2]. In Mexico, the prevalence of hypertension nationally is substantially higher than in the United States (34.2% males, 26.3% females) in Mexicans between the ages of 20 to 69 [3,4]. Nevertheless, no previous study has systematically compared the seventy and older populations from Mexico and the United States. Since the population is aging rapidly in both countries and chronic diseases are important to identify and treat in older age, it is important to understand the etiology of the disease in this subpopulation.

The effect of migration to the United States from Mexico on health has been debated in the recent literature [5]. At younger ages, migrants are said to benefit from a healthy immigrant effect [6–9]. However, with time this salubrious advantage declines with the process of assimilation and acculturation, as migrants adopt negative health behaviors [10] and bear the stress of trying to navigate an unwelcoming country thousands of miles away from their families and social networks [11]. Therefore, Mexicans who migrate to the United States may be vulnerable to physical conditions that are associated with stress or health behaviors, such as, hypertension.

We make use of the Hispanic EPESE and the Mexican Health and Aging Study (MHAS) datasets to evaluate the prevalence of hypertension in older Mexicans in the United States and Mexico. In addition we will determine what factors are associated with hypertension in similar cohorts of older Mexicans in the United States and Mexico. We pay particular attention to migration status, since this information is available for both the United States and Mexico and an important factor to consider with respect to the health of older Mexicans and Mexican Americans on both sides of the border.

METHODS

Data

The data that was used for this analysis are the Mexican Health and Aging Survey (MHAS-Mexico) and the Hispanic Established Population of Epidemiologic Studies of the Elderly (Hispanic EPESE-United States). The Mexican Health and Aging Survey (MHAS) is a nationally representative panel survey of Mexicans aged 50 or older in the year 2000 and their spouses in Mexico (N=15,186). Participants were identified in conjunction with the 2000 National Employment Survey/ Encuesta Nacional de Empleo (ENE). A follow-up wave of interviews was conducted in 2003 with the individuals or a proxy respondent (N = 14,277; Subjects that died before follow-up - 546; Subjects lost – 363; 93.32% response rate).

The Hispanic EPESE is a cohort longitudinal study of older Mexican Americans living in the Southwest United States. The sample was selected using probability design to represent older Mexicans living in Texas, Arizona, California, Colorado, and New Mexico. The original data was collected in 1993–1994 and has four subsequent waves (1995–1996, 1998–1999, 2000–2001, and 2003–2004). If a respondent was not located in person due to death or relocation, proxy information about him or her was collected from family or friends. The follow-up rate is nearly 86% of the original sample of 3,050.

For this analysis we use wave 3 from the Hispanic EPESE in order to equivocate the same period in time as the MHAS sample in 2001. Also, in the Hispanic EPESE we dropped 105 cases that did not self-identify as Mexican descent. Since the Hispanic EPESE dataset’s age distribution is quite old (i.e. 70 years and older in wave 3), cases will be limited to those 70 years and older in the MHAS dataset to optimize comparability of the two datasets. In addition, cases for the MHAS were limited to sampled subjects and therefore their spouses where dropped from the analysis. Also, because diagnosis of hypertension is contingent upon seeing a medical doctor, subjects who reported never seeing a physician were eliminated from this study. The final sample sizes for this analysis, therefore, were 2,295 (969 men; 1,326 women) for the MHAS and 1875 (762 men; 1,113 women) for the Hispanic EPESE. Individual level sample weights created by the principal investigators of the data sets will be used for the statistical analysis using this dataset.

Statistics

The outcome variable for this project is self-reported diagnosed hypertension. There are limitations to using self-report rather than actual measured conditions, however, the primary drawback is an underestimate of the actual number of cases that exist because many may unknowingly have the condition. Since, this is a potential problem that could occur in both the United States and Mexican data, there is no concern for inflation of cases, and the outcomes of these measures are expected to be consistent, yet, conservative estimates.

Our explanatory variables are measured as sex, level education (continuous), age (continuous), migration to the United States, obesity, and currently a smoker. In the United States, subjects were coded as US born or immigrant. In Mexico, subjects were classified as never migrated or previously migrated to the United States and have returned. Body Mass Index was measure as the respondents’ self-reported weight divided by their self-reported height squared. Subjects with a BMI of 30 or greater were then categorized as obese based on National Institutes of Health recommendations.

Prevalence of hypertension was calculated by sex for each dataset. Since the analysis was limited with respect to number of waves collected for the MHAS data, logistic regression will be used rather than hazard or growth curve models. Statistical models, therefore, for this analysis baseline hypertension is predicted (MHAS 2003, Hispanic EPESE 2001) by the explanatory variables at baseline. All analysis was conducted using STATA SE 9. We use the Jackknife Method of standard error estimation, which systematically replicates statistical analysis by selecting random sub-samples of the full sample.

RESULTS

Table 1 presents the prevalence of hypertension at follow-up by sex. Older Mexican men living in Mexico have the lowest prevalence of hypertension (29.6 percent), followed by women in Mexico (50.4 percent), Mexican American men (54.7 percent), and then Mexican American women (63.3 percent). Level of education among those with hypertension was lowest among Mexican women (2.60 years) and highest among Mexican American men (4.88 years). Unmarried Mexican American women have the greatest prevalence of hypertension at 63.7 percent. In addition, women who migrated to the United States who live in Mexico had strikingly similar incidence of hypertension (62.3 percent) to their Mexican American female counterparts (approximately 63 percent).

Table 1.

Prevalence of Hypertension in Older Mexicans in Mexico and the United States

	Mexico		United States
	Men	Women	Men	Women
Explanatory Variables
Age (mean ± se)	76.7 (± .32)	77.1 (± .23)	77.4 (±.27)	77.6 (± .22)
Years of Education (mean ± se)	3.45 (± .22)	2.60 (± .12)	4.88 (± .20)	4.59 (± .15)
Marital Status (%)
Married	140 (29.7)	113 (49.1)	297 (54.7)	234 (62.4)
Unmarried	147 (29.6)	555 (50.6)	120 (54.8)	471 (63.7)
Migration Status (%)
Former Migrant (Mexico)	63 (31.3)	38 (62.3)
Non-Migrant (Mexico)	224 (29.2)	630 (49.8)
US Born (United States)			192 (54.1)	396 (63.2)
Immigrant (United States)			225 (55.3)	309 (63.4)
Obese (%)	42 (35.3)	115 (55.6)	116 (66.3)	258 (70.9)
Smoker (%)	39 (18.1)	40 (40.8)	58 (51.8)	40 (60.6)
Total	287 (29.6)	668 (50.4)	417 (54.7)	705 (63.3)

Table 2 presents logistic regression results for hypertension at follow-up by sex. Beginning with women, in model 1, in Mexico, the odds of hypertension are increased by 3.026 for those that had migrated to the United States. Only about six percent of the effect of migration to the United States is explained in model 2 which includes obesity and current smoker. Obesity, in model 2, also increases the odds of hypertension by 1.813 for older Mexican women.

Table 2.

Odds Ratios from Logistic Regression at Baseline for Hypertension in Older Mexicans in Mexico and the United States

	Females				Males
	Mexico		United States		Mexico		United States
Explanatory Variables	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2
Demographics
Age	.9903	.9931	.9518***	.9541***	1.011	1.004	.9571*	.9652*
Years of Education	1.058†	1.052	.9888	.9893	1.042	1.039	.9732	.9753
Married	.8998	.9244	.6548*	.6398*	1.061	1.019	.8664	.8285
Former Migrant = 1 (Mexico)	3.026**	2.967***			1.132	1.088
Immigrant =1 (United States)			1.306	1.313†			.9934	.9824
Health Indicators
Diabetes
Obese		1.813*		1.357†		1.504		1.675*
Current Smoker		.5679		.7181		.3656**		1.331
Currently Uses Alcohol
n	1326	1326	1113	1113	969	969	762	762

Results are reported in Odds Ratios.

^†p<.10,

^*p<.05,

^**p<.01,

^***p<.001

In the United States, in table 2 (model 1), for women, age reduces the odds of hypertension by about five percent per year increase. Immigrant status, in model 1, increases the odds of hypertension by 31.3 percent, but is insignificant in model 1. The effect of immigrant status is increased by an addition 7 percent in model 2, causing the odds ratio to become marginally significant. Obesity in model 2 also increases the odds of hypertension by 1.357 for older Mexican American women, but, it is only marginally significant.

For older men in Mexico, no odds ratios are significant in model 1. In model 2, however, being a current smoker reduces the odds of hypertension by .3656. In the United States, for Mexican American men, age significantly reduces the odds of hypertension by about 4 percent per year increase. In model 2, age remains significant and obesity increases the odds of hypertension in older Mexican American men by 1.675. In neither case of Mexican or Mexican American men is immigration status significant.

DISCUSSION

In this study, we set out to determine the prevalence of hypertension in two older Mexican populations living in Mexico and the United States. The major findings of this research are that prevalence differs by gender and migration to the United States has a significant negative effect for women who live in both countries. First, older Mexican women have a higher risk of hypertension then men in both the United States and Mexico. Our findings are similar to results from the 2000 Encuesta Nacional de Salud in Mexico in that women had a higher prevalence of hypertension than men over the age of 50 years [3]. Studies in the United States of Mexican Americans, have also consistently documented higher prevalence of hypertension in women [1, 12]. Therefore, this disadvantage continues into older age for both Mexican and Mexican American women.

Men in Mexico, on the other hand, have an unusually low rate of hypertension in this study. In the Mexican National Health Survey 2000, men 60 to 69 had a hypertension prevalence of 52.4 percent [4], which is significantly higher than what was found in this study and we would expect it be about the same or even higher for the 70 and older age group. We suspect that since hypertension was self-reported, the low rate of hypertension among Mexican men may be explained by under-reporting in this study.

Second, migration experience to the United States had a consistent negative effect for women so that, women who migrated in both countries were the most likely to have hypertension. This effect for Mexican and Mexican American women remained even after taking into account obesity and smoking. Findings from other countries have yielded similar results for other cardiovascular conditions, also after controlling for health behaviors [13]. There may be factors associated with migration that may increase the propensity to have hypertension, such as psychological or physiological stressors from being separated from loved ones or difficulties in navigating a foreign country that may have a greater effect on women than men [11]. Migrating men and women to the United States enter into very different labor markets [14]. In addition, women and men have different types of social networks or opportunities for social support, leading to very different pathways of health. Thus, as a result of these differences, women may be put at a greater risk of health problems than men.

In a previous study, Mexican immigrants and US born Mexican Americans on the border region of Texas were observed to be no different with respect to depressive symptoms, life-satisfaction, or self-esteem, but immigrants exhibited a significantly greater amount of stress than their native born counterparts [15]. For women, using the Hispanic EPESE, Black, Markides and Miller [16] found that older Mexican immigrant women were also more at risk for depressive symptoms than their US born Mexican American counterparts. In addition, other studies have shown that an immigrant advantage in mental health declines with longer duration of stay and earlier age at entry [17]. Therefore it may be that, in this study, although women who immigrate to the United States may have a mental health advantage earlier in the life course, stress associated with the experience may exhibit itself in poor mental health and physiological damage to the body over the years that results in chronic conditions such as hypertension in old age. This “weathering” likely continues across the border into Mexico, hence explaining the significant effects there. Still, little is known as to the physiological effects of migration to the United States on women in specific and future research is indicated.

Obesity and smoking did not explain the of effect of migration status for women in either Mexico or the United States, which is surprising given the well-documented increased propensity to adopt of negative health behaviors with longer time spent in the United States for Mexican immigrants. [10,11,18]. The more Mexican immigrants in the United States adopt these deleterious behaviors, the more likely they are to have associated physical conditions such as hypertension [19–21]. However, this did not prove to be an explanatory factor for older Mexican women in the United States or Mexico.

An unusual finding from this study was the protective effect of smoking on the propensity to have hypertension for older men in Mexico. Although insignificant, Mexican and Mexican American older women were also less likely to have hypertension if they were current smokers. These findings may be due to the fact that people who are diagnosed with hypertension, are likely to subsequently quit smoking after diagnosis and, therefore, those who are still smoking have yet to be diagnosed. Further research is indicated to better determine what is driving this relationship.

This study is the first to compare the risk of hypertension in older Mexican Americans in the United States to Mexicans in Mexico. As a result we have revealed similarities and differences that exist between these two populations. Still, limitations to this study exist. For example, relying on self-reported hypertension may not fully capture to the full extent the prevalence of hypertension in these two populations. Nonetheless, we believe that this study makes an important contribution to the literature on aging and health in older Hispanic populations.

In conclusion our findings suggest that hypertension is more prevalent in older age than in previous studies of younger cohorts and affects women to a greater extent than men. The results from this study also suggest that progression into disease occur at different points in the life course and at different time points for men and women. This may also influence variations between Mexican Americans in the United States and Mexicans living in Mexico. Paying attention to these differences may hold important clues in understanding areas in which interventions can be successfully implemented and may contribute a better grasp of the unusual mortality advantage that Mexican Americans have in the United States [22].