Cardiovascular Risk Factors in the Mexican Population

Abstract

Chronic degenerative disorders have become a major health problem in Mexico. Cardiovascular diseases represent the first cause of death in our country. Diabetes mellitus (DM) has emerged as the main health problem in Mexico. Its prevalence doubled from < 3% in the 1960s to 6% in the 1980s. Between 1993 and 2000, diabetes mellitus increased from 6.7% to 8.2%, a 22% growth over a seven-year period. In 1995, the cost of the treatment of DM represented 15.48% of the health budget and 0.79% of the GDP. The prevalence of hypertension (HTN) increased from 10% in 1933 to 20% in 1990 and from 23.8% to 30.7% between 1993 and 2000. The expenditures from HTN in 1999 corresponded to 13.9% of the health budget, and 0.71% of GDP.

Dyslipidemias are very common. Close to 40% of the population has levels of HDL cholesterol < 35 mg/dL, 24.3% has fasting triglycerides > 200 mg/dL, and 10% has hypercholesterolemia. The prevalence of obesity increased from 21.4% in 1993, to 23.7% in the year 2000. Eight percent of the population has a glomerular filtration rate < 60 mL/min, and 9.1% has proteinuria. Twenty-four percent uses tobacco regularly, and 13% had the habit in the past. Smoking is more frequent among diabetics (34%).

In conclusion, cardiovascular risks factors are highly common among the Mexican population and increasing at alarming rates. Preventive programs targeted to decrease their prevalence are urgently needed in Mexico and should become a national priority.

Keywords:

• diabetes
• hypertension
• obesity
• proteinuria
• hyperlipidemia
• smoking

INTRODUCTION

As in other developing nations, diabetes mellitus (DM) and other non-communicable diseases are reaching epidemic levels. In 1960, communicable diseases like pneumonia, gastroenteritis, tuberculosis, and malaria were the leading causes of death. By 1980, five non-communicable diseases (cardiovascular, malignancy, hepatic cirrhosis, and DM) were reported among the ten main causes of death. By 1999, cardiovascular disease, malignancy, and DM represented the three leading causes of death.[1]

Factors such as the increase in life expectancy, a sedentary life, a diet rich in fat, as well as tobacco consumption have lead to an increase in the prevalence of obesity, DM, hypertension (HTN), chronic kidney disease, and dyslipidemias among the country's population.

This report reviews the epidemiology of cardiovascular risk factors and their impact in the health of the Mexican population.

DIABETES MELLITUS

Between 1922 and 1950, the mortality due to diabetes mellitus doubled, from 2 to 4 deaths per 100,000 inhabitants. However, in a similar 30-year period between 1950 and 1990, the mortality had an eight-fold increase, reaching 32 deaths per 100,000 inhabitants. The mortality rate has continued to grow at alarming rate. In only four years, between 1998 and 2002, the mortality rate increased from 43.3 to 53.2 deaths per 100,000 inhabitants, a 23% growth (see Figure 1). As a result, DM is now the main cause of death in women and second only to coronary heart disease in men. It is also the leading cause of premature retirement, blindness, and kidney failure.[2]

Figure 1 Diabetes Mortality Rate, 1922–2002.[2]

The impact of this disease on the Mexican health care system is enormous. In 2000, DM was the eleventh most frequent cause of hospitalization and second most common cause of hospital mortality. Since 1984, DM has been among the ten most frequent causes for seeking medical attention.[2]

Several cross-sectional studies have measured the prevalence of this disease in diverse populations.[3–6] These reports are difficult to compare due to different diagnostic criteria and selection bias. Nonetheless, the prevalence in the 1960s was estimated to be < 3%, 2–4% in the 1970s, and 5–8% in the 1980s.[2]

Three population-based studies have provided a more accurate estimate of the frequency of DM. The first one, the Mexico City Study,[7] was carried out in six low-income neighborhoods in the nation's capital, identifying 3,326 study-eligible men and non-pregnant women age 35–64 years. Of these, 84% completed a home interview and 68.5% completed a baseline medical examination from 1990–1992; the survey detected a baseline prevalence of 12.9%.

In 1993, the National Survey of Chronic Diseases (NSCD)[8] reported a prevalence of 6.7% in an adult, 20–69-year-old, urban population. The highest prevalence (9.0%) was observed in the north of the country, while the lowest (6.6%) was seen in the south. DM prevalence varied inversely with the level of education, being more frequent in the illiterate (11%) than in those with graduate studies (4.4%).

The prevalence increased with age, observing the highest frequency (25%) in the 60–69-year range.[8],[9] Of great concern was the fact that 8.4% of the cases were < 40 years old. As a result, the long-term adverse effects of DM make this group highly susceptible to chronic complications associated with this disease.

Body weight played a significant role in the prevalence of the disease. The prevalence and the risk of diabetes increased with body mass index (BMI); for those subjects with BMI < 25 kg/m², the prevalence was 4.1%, increased two-fold for those overweight (BMI between 25 and 29.9 kg/m²), and reached 12.5% for those with obesity (BMI ≥ 30 kg/m2).[8]

In 2000, the National Health Survey (NHS2000),[1] a cross-sectional study that included individuals from 400 cities, obtained information from 45,294 subjects with a response rate of 96.3%. It showed that the prevalence of DM had increased from 6.7% to 8.2%, a 22% growth over a seven-year period. Gender distribution had a 1:1 ratio. One-fifth of diabetics did not know they have the disease; the highest percentage (38.2%) of newly detected cases was found among the young. Similar to the NSCD study, prevalence increased with age in both sexes, but at an age ≥ 50 became more prevalent among women (see Figure 2). Of interest, in the group < 40 years old, the prevalence increased from 8.4% in 1993 to 13% in 2000, a 50% growth over a decade. This is of concern, because this is the group with the higher risk of developing chronic complications.[1],[8]

Figure 2 Prevalence distribution of diabetes mellitus by age and gender.

Geographically, the prevalence distribution was similar to that found in the NSCD survey. Also, the prevalence of diabetes varied inversely with the level of education. While the prevalence was < 5% among those with high school and graduate education, it was two-fold and three-fold higher among those with primary or none education at all, respectively. Interestingly, DM was more prevalent among the population with social security (10%) than among the uninsured (6.2%).[1]

In a separate analysis of the NHS2000 data, it was estimated that the risk of DM begins to increase significantly from 22 to 24 kg/m² BMI in both sexes, and from a 75 to 80 cm waist circumference values in men and 70 to 80 cm in women.[10] Additionally, other risks factors play a role in the development of diabetes. In a conjunctive consolidation analysis of the NHS2000 data,[11] the impact of the interaction between age, weight, gender, and HTN on the prevalence of DM was evaluated. For example, in women aged 20–34 with normal weight, the presence of HTN increased the prevalence of DM from 3.3% to 5.3%. However, if the patient was obese, the prevalence rose to 9.2%. Similarly, in men, the presence of HTN increased the prevalence of DM from 3.4% to 6.3%, and in those with HTN and obesity to 12.6%. These additive effects were magnified by age. In women aged 35–54 and 55–59, the prevalence of DM in non-obese, non-hypertensive subjects was 7.9% and 12.9%, respectively; however, the prevalence of DM increased to 19.8% and 28.6%, respectively, among obese, hypertensive subjects. Similar trends were observed in men. In a multivariate regression analysis, identifiable risks factors for DM were proteinuria (OR 1.48, CI 95% 1.35–1.62), male gender (OR 1.20, CI 95% 1.13–1.28), age (1.99, CI 95% 1.90–2.07), obesity (OR 1.91, CI 95% 1.79–2.04), and HTN (OR 1.43, CI 95% 1.33–1.52).[11]

Finally, 85% of the patients with known DM had received treatment; at the time of the survey, 69% were on an oral glucose-lowering agent and 5.8% received insulin. Only 24.5% appeared to be following a diet. Traditional medicine (including herbal products and homeopathy) was the chosen therapy in 11% of the cases. As a result, more than half of them were poorly controlled, as shown by blood glucose levels > 140 mg/dL.[12]

Few studies have addressed the prevalence of DM among indigenous populations. DM was found in 8.4% of Sonoran Pima Indians,[13] above the prevalence among the general population. However, it was lower among the Otomi Indians (4.4%) from central Mexico[14] and the Mazatecos (2.1%) from Oaxaca.[15] No case of DM was detected among the Huichol, Tepehuan, and Mexicanera Indians from Durango.[16]

Not surprisingly, mortality due to DM is also increasing. In the period between 1980 and 2000, the age-adjusted DM mortality rate increased 47%. Like DM prevalence, mortality follows a similar geographic pattern. Standardized mortality rates (SMR) of 127.6 were reported in northern states, compared with 97.0 in the south. However, the age-adjusted mortality rate increased 92% in the south, while only 9% in the north, shifting the difference from 18.5 to 5.3 deaths/100,000 inhabitants in that period. Additionally, the magnitude of prematurity of mortality, assessed by means of potential lost life years index (PLLYI), was low in the south (95.6) and higher in the north (106.3). Mexico City had the highest values, with a SMR 135.9 and a PLLYI of 126.6. The latter could be explained in part by aggregation bias because the nation's capital has large public hospitals, and patients from the central and southern regions migrate to Mexico City in search of better medical care.[17]

Treatment of DM is expensive. The average annual cost per diabetic patient, including those with ESRD on peritoneal dialysis, was estimated in USD $708.00, with a total annual cost of diabetics of $2.6 billion USD. These figures represented 15.48% of the country,s health expenditures and 0.79% of the GDP in the year 1995.[18] These costs are expected to rise. In the three main public institutions of the country,s health care system, the total expenditures diabetes in 2005 have been estimated at USD $317,631,206, including USD $140,410,816 in direct costs and USD $177,220,390 in indirect costs, a 26% increase from 2003.[19]

To cover the service demand by diabetic patients, the Mexican Institute of Social Security's (IMSS) direct costs are expected to increase from USD $79,137,288 in 2004 to USD $88,966,408 in 2006; for the Institute for Social Security of Federal Employees (ISSSTE), they will increase from USD $18,048,804 to USD $20,282,976; and for the Health Secretariat (SSA), they will rise from USD $34,018,405 to USD $40,533,202.[20]

It has been pointed out that the prevalence of DM in Mexico will increase drastically over the next decades for a number of reasons.[21] First, obesity, which is a risk factor for type 2 DM, is increasing rapidly in Mexico. Second, pregestational obesity and gestational diabetes, which are associated with high birth weight and are both risk factors for later obesity, type 2 DM, hypertension, and the metabolic syndrome, are also increasing in Mexico. The fact that 58% of women of reproductive age are either overweight or obese is therefore a cause of great concern.[22] Finally, although the rate of undernourished children < 3 years old has decreased from 22.8% in 1988 to 17.7% in 1999, it remains a serious problem.[23] Undernutrition early in life, when followed by catch-up growth during childhood, is a high risk factor for the development of obesity, DM, and cardiovascular diseases in adulthood.[24],[25]

ARTERIAL HYPERTENSION

Several cross-sectional studies have attempted to determine the prevalence of HTN in the Mexican population. An excellent review is provided by Rodríguez-Saldaña et al.[26] Between 1933 and 1995, 30 studies were reported in the literature: 26 among the adult and 4 among the pediatric populations. The first of those surveys was conducted in 1933 among the Mayans in the state of Yucatan: 370 subjects at least 30 years old were included, and the reported prevalence was 10%. In the 1960s, the prevalence reported among physicians belonging the Mexican Institute for Social Security was 11.20% and 19% among cardiovascular patients from the same institution; in the 1970s, six studies found a mean prevalence of 16.5% in adults. In the 1980s, seven studies reported a mean prevalence of HTN of 10.0% (range 0.7–18.5%); in the 1990s, eight studies reported a prevalence of 20% (range 4.7–29.1%).

As has been pointed out, these results are difficult to interpret due to methodological differences, mainly the different cutoff values used to diagnose hypertension, the lack of standardization in the measurement of blood pressure, and the variability of the population studied.[26,][27] Nevertheless, they show an increasing trend in the prevalence of this disease (see Figure 3). Also, the prevalence of HTN among the rural population increased from 9.2% in 1981 to 21.9% in 1997.[28],[29]

Figure 3 Prevalence of arterial hypertension in Mexico, 1933–2000.

Two population-based studies have provided a more accurate view of the frequency of HTN in Mexico.[1],[8] Similar to that of DM, the prevalence of HTN—defined as the population previously diagnosed with HTN and on therapy regardless of the level of blood pressure, as well as those subjects with a systolic blood pressure ≥ 140 mmHg and/or diastolic blood pressure ≥ 90 mmHg—has increased from 23.8% in 1993 to 30.7% in the year 2000. This means that close to 15 million of Mexican adults have high blood pressure. Sixty-one percent of the subjects were unaware that they had high blood pressure.

A separate analysis of the NHS2000[11],[30] provided a more accurate view of the epidemiology of this disease. In the general population, the prevalence of HTN was higher among men (34.2%) than women (26.3%). However, as age increases, HTN becomes more prevalent, especially among women. Before the age of 50, the prevalence is higher among men, reaching 40.7% in the group 40–49 years old; thereafter, the prevalence is higher among women, reaching 54.6% in the age group 55–59 and 60.3% in those women ≥ 65 years old. Similar results in the elderly had been reported earlier.[31] Of concern is the fact that a significant number (17%) of subjects under 30 are hypertensive, increasing the time exposure to this disease and the risk of developing complications associated with high blood pressure.[4],[11],[30]

In hypertensive subjects, 34.2% had elevations of both diastolic and systolic blood pressures, 46.1% had pure diastolic HTN, and only 5% had isolated systolic HTN. The majority (70.9%) had stage I, and only 7.2% had stage III of the JNC-VI classification. The low prevalence of isolated systolic HTN compared to the rates reported in the United States and Canada could be explained in part by the age distribution of the hypertensive population. Although the prevalence of hypertension is 29.3% in the group 20–34 years old, this age group represents half of the population between 20–69 years old. This means than 75% of the hypertensive population in Mexico is < 54 years old, and therefore diastolic hypertension prevails.[11]

DM increases the risk of HTN. Sixteen percent of the hypertensive population had DM, compared with 8.2% among the non-diabetics. However, nearly half of the diabetics subjects in the NHS2000 survey were hypertensive, and 30% were diagnosed during the study. The odds ratio (age- and sex-adjusted) for HTN in diabetics was 1.54 (CI 95% 1.44–1.63).The coexistence of both diastolic and systolic HTN was the most frequent finding (50%). Isolated HTN was less commonly found (17%).[11],[12],[30]

Additionally, other risk factors play a role in the development of HTN. In a conjunctive consolidation analysis of the NHS2000 data, the impact of the interaction between age, weight, gender, and DM on the prevalence of HTN was evaluated. For example, in women aged 20–34 with normal weight, the presence of DM increased the prevalence from 8.0% to 12.5%. However, if additionally the patient was obese, the prevalence rose to 27.2%. Similarly, in men, the presence of DM increased the prevalence of HTN from 15.0 to 25.0%, and in those with DM and obesity to 57.8%. These additive effects were magnified by age. In women aged 35–54 and 55–59, the prevalence of HTN in non-obese, non-diabetic subjects was 19.2% and 50.2%, respectively; however, the prevalence of HTN rose to 50.2% and 74.%, respectively, among the obese, diabetic subjects. Similar trends were reported for men. In a multivariate regression analysis, identifiable risks factors for HTN were DM (OR 1.42, CI 95% 1.33–1.51), proteinuria (OR 1.33, CI 95% 1.24–1.43), gender male (OR 1.57, CI 95% 1.50–1.65), age (OR 2.23, CI 95% 2.16–2.30), obesity (OR 2.51, CI 95% 2.39–2.63), and smoking (OR 1.08, CI 95% 1.03–1.13).[11]

Like DM, HTN was more prevalent in the north (34%) than the south (27%) of the country. Mexico City, with one-fourth of the country,s population, had a prevalence of 29%. Similar to DM, HTN was more frequent among those with social security (34%) than the uninsured (28%). Also, it had an inverse correlation with the level of education, being more prevalent among the illiterate (44%) than those with graduate studies (23%).[1]

Among those previously diagnosed with HTN, 46% were receiving antihypertensive therapy, but only 20% had blood pressure readings < 140/90 mmHg.[30],[31] Among hypertensive diabetics receiving therapy, only 30% had a blood pressure < 140/90 mmHg. Fewer were able to achieve the goals proposed by the year 2000 recommendations of the American Diabetes Association, those being < 130 mmHg for systolic (11.3%) and < 85 mmHg for diastolic (26.6%).[12]

A follow-up of the hypertensive population identified in the 2000NHS study is now available[32]: 14,567 individuals were followed between the years 2000–2004, and 1,165 (8%) subjects had blood pressure readings < 135/85 mmHg and were considered false positives or non-hypertensive. Of the remaining 13,402 subjects, 335 died in the period of follow-up for an annual 1.5% mortality rate. The prevalence of well-controlled hypertensive subjects increased from 14.6% in the year 2000 to 19.2% in 2004. However, more than half of the subjects were seen in the emergency room, with an average annual number of hospitalizations of 4.1± 3. Twenty-five percent developed renal failure; 14.7%, peripheral vascular insufficiency; 12.3%, heart failure; and 1.2%, a cerebrovascular accident. Thirty-six percent had blood cholesterol levels > 200 mg/dL, and 43% had triglycerides levels > 200 mg/dL. The prevalence of diabetes increased from 16% in 2000 to 30% in 2004, for an annual absolute incidence rate of 5%; obesity increased from 38% to 49% in the same period of time.[32]

Like DM, the treatment of HTN is expensive. In 1999, it was estimated that the average annual expenditures from hypertension was $1.8 billion USD, corresponding to 13.9% of the budget allocated to health care and 0.71% of the country's GDP.[33]

In a separate study, the expected changes in the cost of the treatment of hypertension in the public sector between 2004 and 2006 were determined. To cover the service demand by hypertensive patients, the Mexican Institute of Social Security (IMSS) requirements will increase from USD $28,398,544 to $32,055,782; for the Institute for Social Security of Federal Employees (ISSSTE), they will rise from USD $10,960,600 to USD $12,216,198; and for the Health Secretariat (SSA), they will increase from USD $17,102,804 to USD $18,928,907. It is expected that the combined financial requirements for the treatment of DM and HTN in 2006 will amount 9.5% of the total budget for the uninsured (SSA) population and 13.5% for the insured (IMSS, ISSSTE) population.[20]

DYSLIPIDEMIAS

The mean national levels for cholesterol and triglycerides in the NSCD survey[8] were 183 mg/dL and 158 mg/dL, respectively. Twenty-nine percent of the population had a cholesterol level ≥ 200 mg/dL, and the prevalence rose to 45% among diabetic subjects. Hypercholesterolemia, defined as a level of blood cholesterol ≥ 240 mg/dL, was found in 8.9% of the population. The prevalence increased with age, reaching 17.4% in subjects in the 60–69 age range. As was the case with DM and HTN, a large proportion of subjects younger than 30 (7.4%) had blood cholesterol levels ≥ 240 mg/dL, increasing the probability of developing cardiovascular complications associated with this abnormality.

Hypercholesterolemia was more frequent in men (10%) than women (8.1%). Unlike DM and HTN, the prevalence increased with the level of education, being of 9.1% among the illiterate and 11.7% in those subjects with graduate studies.[8] However, it was also common among the poor: in a survey of the prevalence of hypercholesterolemia among low-income urban population of Mexico City, 12.3% of men and 12.5% of women where found with cholesterol levels > 240 mg/dL, similar to the levels found in the more affluent population; severe hypercholesterolemia, defined by cholesterol levels ≥ 260 mg/dL, was found in 6.6% of men and 6.5% of women. Mean triglycerides, HDL, and LDL in men were 387 mg/dL, 31.5 mg/dL, and 186 mg/dL, respectively; in women, the levels reported were 253 mg/dL, 36.5 mg/dL, and 193 mg/dL, respectively. A disturbing finding in this study was the fact that only 15.9% of men and 7.8% of women were aware that they had hypercholesterolemia, and none were receiving treatment.[34]

Finally, 12% percent of the population had low-density lipoproteins (LDL) levels > 160 mg/dL; they increased with age, and high levels were more prevalent among men (13%) than women (10.7%). The most common abnormality was a HDL cholesterol concentration < 35 mg/dL, and 16.3% had triglycerides levels > 200 mg/dL.[8]

A more precise analysis of the lipid-related cardiovascular risk of the population in the ENEC survey was assessed by grouping the lipid abnormalities as lipid phenotypes.[35] The hypertriglyceridemia/hypoalphalipoproteinemia profile, usually seen in the insulin resistance syndrome, was observed in 12.9% of the general population. The prevalence was significantly higher in men compared with women (20.9% vs. 7.2%, p < 0.01). In men, the prevalence at ages 20–29 was almost as high as that observed in women ages 50–59 (13.1 vs. 15%). The association of normotriglyceridemia/hypoalphalipoproteinemia was among the most common forms of dyslipidemias and was present in 18.6% of the population. The prevalence was higher in men than in women (22% vs. 16%, p < 0.05). Several risk factors for this abnormality were found. Tobacco was used in 31.2% of the subjects, and insulin resistance was found in 59% of this group. A BMI between 25 and 30 Kg/m² was reported in 36% of the subjects, and 20.9% had a value > 30 Kg/m².[35]

The simultaneous elevation of cholesterol/triglycerides concentrations was observed in 12.6% of the population. Close to 20% of the subjects older than 50 had this phenotype. It was more frequently seen in men then in women (16.8% vs. 9.6%, p < 0.01). This abnormality was also common in men younger than 29 years (8.1%).

Severe triglyceridemia (> 500 mg/dL) was observed in 2.9% of the general population. On the other hand, severe hypercholesterolemia (> 300 mg/dL) with normal triglycerides levels were found in only 0.29% of the subjects. The prevalence of severe hypertriglyceridemia was higher in men than in women (5.5% vs. 1.4%), and remarkably, 3.1% of men 20–29 years old had this defect.[35]

Isolated hypertriglyceridemia was found in 15.8% of the population. It was more common in men than in women (21.3% vs. 12.1%), and more than 30% of the men older than 50 had a fasting triglyceride concentration > 200 mg/dL. Isolated hypercholesterolemia (> 240 mg/dL) with triglycerides levels < 200 mg/dL was found in 18.7% of the population. The vast majority of subjects had a cholesterol concentration of 200 and 240 mg/dL. Close to 10% of those aged 20–29 years had cholesterol levels within this range.[35]

DM increased the likelihood of having almost every class of dyslipidemia. The risk was statistically significant for mixed dyslipidemias (OR 3.1), severe dyslipidemia (OR 4.7), isolated hypertriglyceridemia (OR 6.4), and isolated hypercholesterolemia (OR 5.7). Also, hypertriglyceridemia (35%) and mixed dyslipidemias (18%) were more commonly seen in obese subjects; arterial hypertension was more frequently associated with hypertriglyceridemia (36.5%) and mixed dyslipidemia (21.2%) than in the general population. Also, the prevalence of the abnormal lipid profiles was higher in hypertensive subjects compared with the general population; however, none of the odds ratios was statistically significant. Hypertriglyceridemia (36.5%) and mixed dyslipidemias (21%) were the most common abnormalities seen in subjects with HTN.

In conclusion, this analysis demonstrated that some particular forms of dyslipidemias are very common among Mexican adults. Close to 40% had low levels of HDL (< 35 mg/dL) and 24.3% had fasting triglycerides > 200 mg/dL. Hypercholesterolemia was less common (10%) (see Figure 4). The prevalence of dyslipidemia was higher in men than in women, especially in younger subjects. In young subjects, close to 7% had hypertriglyceridemia and low HDL, 10% had hypercholesterolemia, and 20% had isolated HDL cholesterol levels.[35]

Figure 4 Prevalence of lipids abnormalities in the general population. Abbreviations: HCH = hypercholesterolemia; HTG = hypertriglyceridemia; HDL = high density lipoproteins; LDL = low density lipoproteins.

In a more recent survey, the prevalence of hypercholesterolemia among the urban population in Mexico was evaluated. It was conducted in six large urban conglomerations and included 120,005 subjects. The prevalence reported was 43.%, among the highest in the world and significantly different from that reported in 1993. Overweight and obesity increased the risk of hypercholesterolemia; among subjects with BMI ≥ 25 to 29 Kg/m², the prevalence was 45.9%, and among those with BMI ≥ 30 Kg/m², the prevalence rose to 47.3%. Fifty-two percent of the hypertensive population had hypercholesterolemia. A similar prevalence was found among diabetics. In a multivariate variable analysis, age (OR 1.56, 95% CI, 1.53–1.59), HTN (OR 1.35, CI 95% 1.32–1.39), DM (OR 1.24, 95% CI 1.19–1.28), and smoking (OR 1.03, 95% CI 1.01–1.06) increased the risk of hypercholesterolemia in this population.[36]

Little is known about the prevalence of dyslipidemia among indigenous population in Mexico. In a cross-sectional study of Otomi Indians in central Mexico, the prevalence of hypercholesterolemia and hypertriglyceridemia was found in 7% and 26% of this population, respectively, similar to that reported in the general population in 1993.[14]

RENAL DISEASE

The prevalence of renal disease, defined by the knowledge of a previous diagnosis of any type of kidney disease, increased from 0.5% in 1993 to 10.9% in the 2000 National Health Survey.[1],[8] This means that currently close to 5.6 million subjects know they have some kind of kidney disease. As expected, the geographic distribution follows that of diabetes mellitus and hypertension. Also, it is more prevalent among overweight and obese subjects, and follows an inverse correlation with the level of education.[8] More recently, Amato et al.[37] evaluated the prevalence of chronic kidney disease in an urban population of central Mexico; this was a cross-sectional survey that included 3564 subjects older than 18. The prevalence rate reported for an estimated GFR < 60 mL/min was 80,788 per million population (pmp), and that of GFR < 15 mL/min was 1142 pmp. By KDOQI guidelines, the percentage distribution of CKD among the Mexican population was close to the one reported in the American population. The prevalence of DM and HTN was 10.9% and 20.6%, respectively. Proteinuria was present in 8.7% of subjects with Ccr ≥ 60 mL/min and 12.3% of those with DM and HTN. All diabetics with Ccr < 60 mL/min had proteinuria.

DM (OR 1.96, 95% CI 1.34–2.90), age > 65 years (OR 1.11, 95% CI 1.10–1.13,), educational level < primary (OR 1.96, 95% CI, 1.37–2.83), and income > USD $4.00/day (OR 0.53, 95% CI 0.36–0.76) were independent risk factors for CKD.

OBESITY

In the NHS2000 study, close to two-thirds of the population had a BMI ≥ 25Kg/m², compared to 55% in 1993.[1],[8] Obesity was present in 24% of the population, and 38% were overweight. This means that 18.5 million Mexicans were overweight, and 11.4 million were obese. In addition, in a recent nationwide nutrition survey, a 27% prevalence of overweight and obesity was found among children.[22]

The prevalence of obesity increased from 21.4% in 1993 to 23.7% in the year 2000.[1],[8] A higher prevalence was reported among the urban population,[4] where it reached 40%. However, the high prevalence of overweight and obesity is now also evident in poor rural communities in Mexico. In a cross-sectional study conducted in four relatively isolated communities, 42% of adult men and 40% of adult women were overweight; additionally, 9% of men and 33% of women were obese.[38]

Also, women were more obese than men (28% vs. 19%, respectively), especially in middle-age women, where the prevalence rose to 41%. However, men are more frequently overweight than women (41% vs. 36%). The prevalence of both problems increases with age. However, it is worrisome to see that of subjects younger than 29 years, 33.3% of them were overweight, and 14.4% were obese.[1]

Obesity is more prevalent in DM and HTN (see Figure 5). Of the population with DM, 75% had a BMI ≥ 25 kg/m², and 30% had a BMI ≥ 30 kg/m². Similarly, abdominal obesity, defined as waist size bigger than 102 cm in men and 88 cm in women, was more frequent in women than in men (59% vs. 21%) and more common in diabetics than non-diabetic control subjects. The latter was more evident in women (OR 4.2) than men (OR 2.2).[1],[12]

Figure 5 Prevalence of obesity in hypertension, diabetes mellitus, and the general population.

The prevalence of HTN among the obese was 46.8%, compared with 24.6% in the non-obese population. This represents a 2.6 increase risk of having hypertension among the obese. Additionally, 38% of the hypertensive population was obese, compared with 18.6% among the non-obese. It was more frequent in men (46.1%) than women (36.0%). Being obese increases the prevalence of HTN across all age groups. The odd ratio (age/sex adjusted) for having hypertension was 2.3 for obesity. Of interest, among half of the population that knew they were hypertensive but did not take antihypertensive medications, 71% were obese.[11],[30]

Additional information on the prevalence of overweight and obesity in women and children less than five years old was provided by the analysis information from the National Nutrition Surveys conducted in 1988 and 1999.[39] The combined prevalence of overweight and obesity in women 18–49 years of age were 33.4% in 1988 and 59.6% in 1999, an increase of 26.2 percentage points or 78% of the baseline prevalence. The prevalence of obesity in children under five years of age increased from 4.2% to 5.3% in the same period of observation, a 26% increase.

Also, the analysis of the changes in dietary habits between 1984–1998, using the quantities of food purchased per adult equivalent, showed a general decline in purchased food quantities of fruits, vegetable, legumes, meat, poultry, and eggs, but not sugars and refined carbohydrates, which increased 19%, and soda, which rose 25%. Fat intake represented 23.5% of energy intake in 1988 and increased to 30.3% in 1999.[39]

Parallel to the increase in the prevalence of obesity and the changes in dietary habits has been the rise of mortality rates for acute myocardial infarction, DM, and HTN. In an analysis of age-adjusted standardized mortality rates (SMRs) of these three diseases from 1980–1998, the SMR increased 53% for acute myocardial infarction, 62% for DM, and 55% for HTN.[39]

The authors concluded that although food expenditure data do not explain the increase in obesity prevalence, the increasing trends in the amount of sugars and refined carbohydrates purchased, and, more particularly, the purchased quantities of soda, could be associated with the increased SMR of acute myocardial infarction, DM, and HTN.

PROTEINURIA

In the NSCD study,[8] 1.6% of the population had proteinuria (urine protein > 100 mg/dL) and 8.2% had microalbuminuria (urinary albumin between 50–100 mg/dL).

Geographically, proteinuria was more frequent in the north (2.0%) than in the south (1.6%) of the country. The lower rates were observed in the central states (1.2%) and in Mexico City (1.5%). Also, microalbuminuria was equally frequent in the north (10.3%) as the south of Mexico (11.2%); like proteinuria, the lower prevalence rates were reported in Mexico City (5.4%) and the central states (5.8%). Similar to DM and HTN, microalbuminuria and proteinuria were more frequently present in subjects with no education (12% and 1.8%, respectively) than in those with university studies (6.1% and 1.7%, respectively). Likewise, they were highly prevalent in patients with past history of myocardial infarction (11.7% and 5.8%), cerebrovascular disease (13.8% and 6.3%), and gout (8.2% and 1.6%, respectively).[8]

In the NHS2000 study,[1] the prevalence of proteinuria was 9.1%. Obesity, HTN, and DM increased the risk of proteinuria (see Figure 6). Of those with proteinuria, 40% had HTN, compared to 30% among those without proteinuria. On the other hand, 10.4% percent of the hypertensive population had proteinuria, compared with 7.9% among the non-hypertensive. This means that hypertension increases by 30% the probability of developing proteinuria. Also, the prevalence of proteinuria is higher among those with uncontrolled HTN (12.6%) than with controlled blood pressure (7.9%). Finally, while the prevalence of proteinuria in non-hypertensive diabetics was 9.3%, it increased to 19.6% in those with hypertension.[11],[30],[40]

Figure 6 Prevalence of proteinuria in obesity, hypertension, diabetes mellitus, and the general population. Abbreviations: GP = general population, HTN = arterial hypertension, DM = diabetes mellitus.

Additionally, 56% of diabetic subjects had microalbuminuria. Subjects diagnosed with diabetes for at least ten years had an increased likelihood of a positive test (OR 1.25). The same was found for arterial hypertension (OR 1.50).[12] A recent study has identified microalbuminuria as a predictor of myocardial infarction in the Mexican population.[41]

SMOKING

In all, 21–24% of the population used tobacco regularly, and 13% had the habit in the past. This means than close to 18 million of the population had been exposed to the deleterious effects of tobacco. Smoking is more prevalent among men, representing 80% of active smokers.[1] Forty-percent of the urban population lives or works close to an active smoker.[4] Of great concern, smoking was more frequent among diabetics (34%).[12]

Smoking increases the risk of hypertension. In the NSCD study, more than half of hypertensive subjects had been exposed to tobacco; 26% of active smokers and 33% of former smokers had high blood pressure.[8] The NHS2000 study reported that the prevalence of hypertension among smokers was 34.1%. The increment was higher among the group aged 20–40 years, which is the group where the largest percentage of active smokers was registered (24%).[1] The odds ratio (age- and sex-adjusted) for having hypertension was 1.26 for smoking.[11],[30]

DISCUSSION

Cardiovascular risks factors are highly prevalent among the Mexican population and increasing at alarming rates. The prevalence of obesity underwent a rapid increase in the last decade. The rates of overweight and obesity found in the NHS2000 are now similar to those reported among non-Hispanic whites (NHANES III) in the United States and exceeded those of the 1993 survey.[42] This high prevalence of obesity has been associated with shifts in food consumption to greater amounts of fat and refined carbohydrates in Mexico. At the national level, from 1988 to 1999, the percentage of total energy from fat went from 23.5% to 30.3%. The increase in fat intake occurred more in the north of the country (32%) than in the south (22%). Meanwhile, food purchases of refined carbohydrates and soda increased by 6.3% and 37.2%, respectively, in the same period of observation.[39]

Obesity has been established as a major risk factor for non-communicable chronic diseases. Parallel with the increased prevalence of overweight and obesity has been the increase in the frequencies of DM, HTN, and dyslipidemia. The prevalence of DM increased from 6.7% in 1993 to 8.2% in 2000; HTN increased from 23.8% to 30.7% over the same period of time.[1] Dyslipidemia is now very common among the Mexican population. Close to 50% have low levels of HDL cholesterol, 24% have hypertriglyceridemia, and 10% have hypercholesterolemia.[35]

As has been pointed out, a widespread misconception in developing countries is that obesity and non-communicable chronic diseases are problems associated with wealthy populations. The epidemiological data clearly show that this is not the case. Using information about house characteristics and possession of goods, the NHS2000 data was analyzed.[23] The results showed that the sum of overweight and obesity declines as socioeconomic conditions increased. Similar findings have been reported from Curacao and Chile.[43],[44] Perhaps under certain conditions, behavioral changes toward healthy lifestyles are associated with higher incomes. Others risk factors for chronic diseases are also more frequent among the poor. For example, the prevalence of DM and HTN varies inversely with the level of education. While it was low (<5% and 23%, respectively) among those with high school and university education, it was high (15% and 44%, respectively) in those with no education at all. The highest prevalence of obesity was reported in the developed north (30%) and central (22%) Mexico and lowest in the underdeveloped south (17%).[1],[8] Also, hyperlipidemia is commonly found among the poor.[34]

Another important issue is the aging of the population. Practically, the prevalence of all cardiovascular risks increased with age. Also, the deleterious combination of the presence of DM, HTN, and obesity is significantly amplified by age.[11] However, of great concern is the fact that an important percentage of the population < 40 years old is already at risk. Additionally, parallel to the rise in the prevalence of cardiovascular risk factors has been the increase in the mortality due to chronic diseases. The increases in the SMR from 1980–1998 were 53% for acute myocardial infarction, 62% for diabetes, and 55% for hypertension (see Figure 7).[39]

Figure 7 Age-adjusted death rates, relative to 1980, from hypertension, diabetes mellitus, and acute myocardial infarction.[39]

Additionally, chronic renal disease, a risk factor for cardiovascular disease,[45] is widely prevalent among the Mexican population, similar to the prevalence reported in the American population.[46] Another risk factor, proteinuria, is present in 9.1% of the population, and as expected, it is highly prevalent among subjects with DM and HTN.[11],[30],[40] Finally, 24% of the population uses tobacco regularly,[1] and the habit is more prevalent among diabetics.[12] Cigarette smoking has been identified as a strong predictor of cardiovascular disease, both in diabetic and non-diabetic populations.[47]

Last but not least, the costs of treatment of chronic diseases are enormous. The average annual cost for DM has been estimated in USD $2.6 billion, representing 15.48% of the country,s health expenditures and 0.79% of the GDP in the year 1995,[18] and they are expected to rise.[19] The average annual expenditures from hypertension in 1999 were $1.8 billion USD, corresponding to 13.9% of the budget allocated to health care and 0.71% of the country,s GDP.[33] Similar to DM, they are expected to rise. It is projected that the combined financial requirements for the treatment of DM and HTN in 2006 will amount to 9.5% of the total budget for the uninsured and 13.5% for the insured populations.[20]

In conclusion, cardiovascular risks factors are highly common among the Mexican population and increasing at alarming rates. They represent the main leading causes of morbidity and mortality in our country. Preventive programs targeted to decrease their prevalence are urgently needed in Mexico and should become a national priority.

Notes

8. Secretaria de Salud. Encuesta Nacional de Enfermedades Cronicas. Direccion General de Epidemiologia, Secretaria de Salud, Mexico, D.F., 1993.