You are on page 1of 12

Investigación original / Original research

Prevalence of lifestyle-related
cardiovascular risk factors in Peru:
the PREVENCION study
Josefina Medina-Lezama,1 Oscar L. Morey-Vargas,1 Humberto Zea-Díaz,1
Juan F. Bolaños-Salazar,1 Fernando Corrales-Medina,1 Carolina Cuba-
Bustinza,1 Diana A. Chirinos-Medina,1 and Julio A. Chirinos 2

Suggested citation Medina-Lezama J, Morey-Vargas OL, Zea-Díaz H, Bolaños-Salazar JF, Corrales-Medina F, Cuba-
Bustinza C, et al. Prevalence of lifestyle-related cardiovascular risk factors in Peru: the PREVENCION
study. Rev Panam Salud Publica. 2008;24(3):169–79.

ABSTRACT Objectives. To estimate the prevalence of lifestyle-related cardiovascular risk factors in the
adult population of Arequipa, the second largest city in Peru.
Methods. The prevalence and patterns of smoking, alcohol drinking, lack of physical activ-
ity, high-fat diet, and low fruit and vegetable intake were evaluated among 1 878 subjects (867
men and 1 011 women) in a population-based study.
Results. The age-standardized prevalence of current smoking, former smoking, and never
smoking were 21.6%, 14.3%, and 64.1%, respectively. The prevalence of current smoking was
significantly higher in men than women (31.1% vs. 12.1%; P < 0.01). The prevalence of cur-
rent alcohol use was 37.7% and significantly higher in men than women (55.5% vs. 19.7%;
P < 0.01). Similarly, the prevalence of binge drinking was 21.2%, and the percentage of men
who binge drink (36.1%) was significantly higher than for women (6.4%; P < 0.01). The vast
majority of alcohol drinkers reported a pattern of alcohol consumption mainly on weekends and
holidays rather than regular drinking with meals during the week. The proportion of insuffi-
ciently active people was 57.6% and was significantly higher in women than men (63.3% vs.
51.9%; P < 0.01). Overall, 42.0% of adults reported consuming high-fat diets, 34.5% reported
low fruit intake, and 33.3% reported low vegetable intake.
Conclusions. The high prevalence of lifestyle-related cardiovascular risk factors found in
this Andean population is of concern. Preventive programs are urgently needed to deal with
this growing problem.

Key words Alcohol drinking, cardiovascular diseases, diet, physical activity, prevalence, risk fac-
tors, smoking, Peru.

1 Santa María Catholic University Schools of Medi- Cardiovascular diseases (CVD) are demiologic transition characterized
cine and Psychology and Santa María Research In- now the leading cause of death in most by a shift from infectious and defi-
stitute, Arequipa, Peru.
2 University of Pennsylvania School of Medicine, countries of Latin America (1). Al- ciency diseases to the preponderance
Philadelphia, Pennsylvania, United States of Amer- though CVD mortality has declined of non-communicable, chronic ill-
ica. Send correspondence to: Julio A. Chirinos, Di-
vision of Cardiology, 8B111, Philadelphia VA Med- in economically developed nations, nesses (2–8).
ical Center, 3900 Woodland Avenue, Philadelphia, growing rates of CVD morbidity and The growing rates of CVD seen in
PA 19104, U.S.A; telephone: (215) 823-5800, ext.
6791; Fax: (215) 823-4440; e-mail: Julio.chirinos@ mortality have been observed in de- recent decades in Latin America have veloping countries reflecting an epi- resulted, in large part, from the signif-

Rev Panam Salud Publica/Pan Am J Public Health 24(3), 2008 169

Original research Medina-Lezama et al. • Lifestyle-related cardiovascular risk factors in Peru

icant changes in lifestyle associated metropolitan area. The city was consid- Interview
with urbanization and economic de- ered to be ideal for a study of this na-
velopment (2–8). Important lifestyle- ture due to its compact geography, rel- After initial contact with the partici-
related cardiovascular risk factors in- atively large number of inhabitants, pants at the household by one of the
clude tobacco use, lack of physical and low number of reference hospi- investigators, a comprehensive evalu-
activity, harmful alcohol consumption, tals, facilitating enrollment and pro- ation was performed at the study
and poor dietary patterns. The impor- spective follow-up, which is planned headquarters, including an epidemio-
tance of these risk factors in Latin for the second phase of the study (10). logic questionnaire applied by previ-
America has been shown in recent in- The population of Arequipa is compa- ously trained interviewers (10). The
vestigations, including the INTER- rable to other urban populations in questionnaire assessed basic demo-
HEART study, where the presence of a Peru and resembles urban popula- graphic information, level of educa-
healthy lifestyle—avoidance of smok- tions from Andean countries such as tion, family health history, personal
ing, regular exercise, and regular fruit Bolivia and Ecuador. This population medical history, and lifestyle behav-
and vegetable consumption—was as- consists largely of mestizos, with the iors including tobacco smoking, alco-
sociated with a reduction in the risk of degree of admixture being predomi- hol drinking patterns, physical activity
acute myocardial infarction, suggest- nantly Andean Amerindian (i.e., auto- levels, high-fat diet, low fruit intake,
ing that an important proportion of chthonous Quechua and Aymara pop- and low vegetable intake. Educational
these coronary events could be ulations), with small contributions level was classified as low (no formal
avoided by lifestyle modifications in from Spanish whites and minimal con- education to completed secondary
the region (9). tributions from West African popula- school) or high (at least one year of
We believe that estimates of the prev- tions (11–13). post-secondary education). Ethnicity
alence, patterns, and population distri- was self-reported. The study was ap-
bution of cardiovascular risk factors in proved by the Santa María Catholic
Andean and other Latin American pop- Sampling design University Human Research Commit-
ulations are needed in order to design tee, and all participants gave informed
effective disease prevention programs. Details regarding the sampling consent (10).
In this paper, we report on the findings strategy have been published (10). The
regarding the prevalence of tobacco sampling frame for the study was
and alcohol consumption, lack of phys- based on Peru’s most recent popula- Cigarette smoking
ical activity, high-fat diet, and low fruit tion and household National Census
and vegetable intake in the PREVEN- and was provided by the Peruvian The questionnaire included ques-
CION study, a population-based study National Institute of Statistics and In- tions regarding smoking habit and the
of Peruvians (10). formatics (10). The sampling strategy frequency and quantity of cigarettes
was probabilistic, multistage, clus- consumed by the smokers (14). Sub-
tered, and stratified according to jects were also asked if they were ex-
MATERIALS AND METHODS geographic location and socioeco- posed to tobacco smoke at home or at
nomic status. We aimed to enroll a their workplace. Current smokers
Study population minimum total of 1 600 subjects with were defined as those reporting hav-
at least 200 subjects per gender in each ing smoked ≥ 100 cigarettes during
The general objectives, design, and pre-defined age group (20–34 years, their lifetime and smoking at the time
operation plan of the PREVENCION 35–49 years, 50–64 years, and 65–80 of the survey. Former smokers were
study have been previously published years) to increase statistical power defined as those who reported smok-
(10). PREVENCION, the Peruvian for age-group comparisons. This was ing ≥ 100 cigarettes during their life-
study of the prevalence of cardio- achieved through deliberate over- time but who did not smoke at the
vascular disease and coronary risk sampling of subjects aged ≥ 65 years time of the survey. Never-smokers
factors (Estudio peruano de prevalencia old to improve the precision of esti- were defined as those reporting not
de enfermedades cardiovasculares), is a mates in these groups. The overall having smoked ≥ 100 cigarettes dur-
population-based study undertaken in individual response rate was 85.3% ing their lifetime. Current smokers
Arequipa, the second largest city in and the final sample consisted of were further categorized as daily
Peru. The first phase of PREVENCION 1 878 individuals (867 men and 1 011 (those who smoked at least one ciga-
was designed to determine the preva- women) from 626 households en- rette every day) or occasional smokers.
lence of CVD and cardiovascular risk rolled between March 2004 and Janu- Heavy smokers were defined as sub-
factors in the adult study population ary 2006. Exclusion criteria included jects who smoke ≥ 20 cigarettes per
(10). Arequipa is located in the south- subjects living less than five years in day. Passive smokers were defined as
ern Peruvian Andes at an altitude of Arequipa, pregnant women, and per- those who reported exposure to to-
2 325 meters. Its population of over sons with the diagnosis of any malig- bacco smoke at home or at their work-
800 000 inhabitants live in the urban nant neoplasm. place at the time of the survey.

170 Rev Panam Salud Publica/Pan Am J Public Health 24(3), 2008

Medina-Lezama et al. • Lifestyle-related cardiovascular risk factors in Peru Original research

Alcohol drinking patterns ticularly in developing countries, groups according to their total fat con-
where occupation and transportation tent: group 1 contains higher total fat
All participants were asked about activities represent a substantial pro- foods and group 2 contains lower fat
the presence, frequency, quantity, and portion of the total activity of individu- choices. Because the majority of high-
patterns of alcohol drinking, and the als. The IPAQ method also has the ad- fat foods listed in group 1 also con-
most frequently consumed alcoholic vantage of being culturally adaptable, tribute notable amounts of saturated
beverage. A drink of alcohol was de- allowing changes in examples of activ- fat and cholesterol, this criterion sim-
fined as one can or bottle (330 mL) of ities taking into account a population’s plifies the grouping process without
beer, one glass of wine, one cocktail, or customs. sacrificing accuracy (24).
a shot of liquor. Current drinkers were IPAQ assesses physical activity un- The questionnaire provides options
defined as those reporting having con- dertaken during the seven days pre- for weekly consumption and approxi-
sumed at least one drink of alcohol ceding the interview and defines mod- mate serving size of foods eaten
within three months prior to the in- erate activities as those requiring within each group and food category.
terview. We considered two alcohol modest physical effort and producing Each food group is given a numeric
drinking patterns: the Anglo-Saxon small increases in respiratory rate for score based on weekly consumption
pattern was defined as drinking alco- at least 10 minutes of duration, and serving size. These scores are to-
hol mainly on weekends and holidays, whereas vigorous activities are de- taled to obtain a final score, from 0 to
whereas the Mediterranean pattern fined as those requiring hard physical 216 points (23, 24). Lower MEDFICTS
was defined as regular consumption of effort and producing large increases in scores indicate diets containing less di-
alcohol during the week, particularly respiratory rate for at least 10 minutes. etary fat. We considered scores > 38
with meals (15). We also assessed for Following recommendations from the points for the identification of subjects
episodes of binge drinking, defined as authors of the IPAQ short form (19), consuming high-fat diets that do not
the consumption of five or more drinks we did not apply the questionnaire to meet the population-wide American
of alcohol on one occasion for males individuals older than 69 years old, Heart Association dietary guidelines
and as the consumption of four or due to lack of validation data. To as- for fat intake (24, 25).
more drinks of alcohol on one occasion sess physical activity levels, we ana- Finally, low fruit intake was defined
for females (16, 17), in the three months lyzed the answers collected from the as the consumption of fruits for less
prior to the interview. This gender dis- questionnaire and, based on interna- than three days per week (including
tinction takes into account differences tional recommendations (20–22), we fresh, canned, dried, or natural fruit
in both weight and metabolism. classified subjects as sufficiently active juices). Low vegetable intake was de-
people if they engaged in moderate- fined as the consumption of fresh veg-
intensity activities (including brisk etables for less than three days per
Physical activity levels walking) for at least 30 minutes on five week.
or more days per week, or 20 minutes
The short version of the Interna- of vigorous-intensity activities on
tional Physical Activity Questionnaire three or more days a week. Those not Statistical analysis
(IPAQ) was used to assess physical ac- engaging in these levels of physical ac-
tivity (18). The IPAQ short form is an tivity were classified as insufficiently Statistical analysis accounted for
instrument designed primarily for active. the complex sampling strategy of the
population surveillance of physical ac- study. Sampling weights were calcu-
tivity among adults. A key feature of lated for each subject according to the
this questionnaire is that it can be used High-fat diet and low fruit population distribution. This strategy
in all types of societies, from industri- and low vegetable intake allowed for accurate point estimates
alized to developing nations (18). The that account for the study design, ac-
IPAQ short form assesses physical ac- The MEDFICTS questionnaire was cording to which certain age groups
tivity undertaken in four major do- used to assess dietary fat intake (23). were oversampled to increase statis-
mains of day-to-day life: at work (espe- This instrument identifies individuals tical power for age group compar-
cially if the job involves manual labor), consuming a diet higher in total fat, isons. Variances were calculated tak-
for transport (for example, walking or saturated fat, and cholesterol than ing the complex sampling design into
cycling to work), in domestic duties currently recommended by the Amer- account, as failure to account for the
(e.g., housework or gathering fuel), ican Heart Association (24, 25). The design effects of a complex sample
and in leisure time (e.g., participating MEDFICTS questionnaire consists of design will likely result in underesti-
in sports or recreational activities) (19). eight food categories: meats, eggs, mation of variance (26). Data for con-
Assessing physical activity in these dairy, fried foods, fat in baked goods, tinuous variables are presented as
four domains avoids systematic under- convenience foods, fats added at the means, whereas proportions are pre-
estimation in women and individuals table, and snacks. Within each cate- sented as frequencies and percentages,
from lower socioeconomic status, par- gory, foods are assigned to one of two with 95% confidence intervals (CI). All

Rev Panam Salud Publica/Pan Am J Public Health 24(3), 2008 171

Original research Medina-Lezama et al. • Lifestyle-related cardiovascular risk factors in Peru

tests were two-sided and an alpha Cigarette smoking The majority of current smokers
level of less than 0.05 was considered were occasional smokers (68.8%; 95%
to be statistically significant. The com- Overall, the age-standardized prev- CI = 62.5%–74.5%), the remaining mi-
plex sample module from SPSS for alence of current smokers, former nority being daily smokers (31.2%;
Windows, version 13 (Chicago, IL) smokers, and never-smokers were 95% CI = 25.5%–37.5%). A small mi-
was used for all analyses. Age-stan- 21.6%, 14.3%, and 64.1%, respectively. nority (2.4%; 95% CI = 1.2%–4.7%) of
dardized estimates presented were The prevalence of current smoking all current smokers were heavy smok-
calculated according to the standard was significantly higher in men than ers, which represented only 7.9% (95%
world population published by the women (31.1% vs. 12.1%; P < 0.01). CI = 4.2%–14.6%) of all daily smokers.
World Health Organization (27). In ad- This gender difference was present in The age-standardized prevalence of
dition to age-specific prevalence esti- all age groups. A decline in the prev- passive smoking was 31.1% (95% CI =
mates, age-standardized estimates are alence of current smoking in both 27.4%–35.0%), without significant gen-
presented for men and women sepa- genders was observed with increasing der differences.
rately. For standardized estimates in age (Table 2). The prevalence of cur-
the entire population, a male-to-female rent smoking was significantly higher
ratio of 1:1 was assumed. among those with high (26.3%; 95% Alcohol drinking patterns
CI = 22.9–30.0%) than in those with
low educational level (17.2%; 95% CI = Overall, the age-standardized prev-
RESULTS 14.1%–20.8%; P < 0.01). alence of current alcohol drinking was
The prevalence of former smok- 37.7%. The prevalence of current
The study sample consisted of 867 ing was 19.7% in men and 8.9% in drinking was significantly higher in
(46.2%) male and 1 011 (53.8%) female women (P < 0.01). There was an evi- men than women (55.5% vs. 19.7%; P <
subjects. The mean age of the study dent increase in the proportion of for- 0.01), which was true for all age
subjects was 48.5 years for women and mer smokers with increasing age in groups (Table 3). Younger males (20 to
49.6 years for men. Women consti- men but not in women (Table 2). In 34 years old) had the highest preva-
tuted the majority of the sample, men, the observed age-related in- lence of current drinking (61.5%). The
which was apparent for all age groups. crease in the prevalence of former proportion of current drinkers was
The sample consisted predominantly smoking was proportional to the de- significantly higher among those with
of self-defined mestizos or Amerindi- crease in current smoking. The prev- high (42.9%; 95% CI = 38.9%–46.9%)
ans (92.5%), with a minor proportion alence of never-smokers was signif- than in those with low educational
of white (7.3%) and black (0.2%) sub- icantly higher in women (79.1%) level (33.1%; 95% CI = 29.6%–36.9%;
jects. Some important demographic than men (49.2%; P < 0.01), which P < 0.01).
parameters of the study sample are was true for all age groups studied It is notable that the vast majority
shown in Table 1. (Table 2). (99.1%; 95% CI = 98.3%–99.6%) of cur-
rent drinkers reported an Anglo-Saxon
drinking pattern, whereas only 0.9%
TABLE 1. Selected characteristics of a sample of men and women in Are- (95% CI = 0.4%–1.7%) reported a
quipa, Peru, 2004–2006 Mediterranean drinking pattern.
Overall, 21.2% of adults reported
Men (%) Women (%) Overall (%)
binge drinking episodes. The percent-
(n = 867) (n = 1 011) (n = 1 878)
age of subjects who reported binge
Age groups drinking was significantly higher in
20–34 years 24.1 24.9 24.5 men (36.1%) than women (6.4%; P <
35–49 years 25.6 26.9 26.3 0.01), with the highest prevalence
50–64 years 26.5 26.5 26.5
65–80 years 23.8 21.7 22.6 observed among men aged 20 to 34
Race/ethnicitya years old (Table 3). The majority of
Mestizob/Amerindian 92.5 92.5 92.5 current drinkers reported binge drink-
White 7.4 7.2 7.3 ing (59.1%; 95% CI = 54.2%–63.8%).
Black 0.1 0.3 0.2
More than two-thirds of the male cur-
Marital status
Married 68.3 57.3 62.4 rent drinkers (67.4%; 95% CI = 61.9%–
Not married 31.7 42.7 37.6 72.5%), and more than one-third of the
Educational levelc female current drinkers (35.8%; 95%
High 52.0 44.5 48.0 CI = 27.8%–44.8%) reported binge
Low 48.0 55.5 52.0
drinking episodes.
a Ethnicity was self-reported. Among current drinkers, the alco-
b Mestizo: mixed Amerindian and white.
c High education level: at least one year of post-secondary education; low education level: no schooling to holic beverage most frequently con-
completed secondary school. sumed was beer (66%; 95% CI = 61.1%–

172 Rev Panam Salud Publica/Pan Am J Public Health 24(3), 2008

Medina-Lezama et al. • Lifestyle-related cardiovascular risk factors in Peru Original research

TABLE 2. Prevalence of current and former and of persons who never smoked among men and women of different age
groups in Arequipa, Peru, 2004–2006

Men Women Overall

Tobacco use / age strata % (95% CI) % (95% CI) % (95% CI)

All subjects (crude) 32.2 (28.4–36.3) 12.6 (10.3–15.4) 22.3 (19.8–25.0)
All subjects (age-standardized) 31.1 (27.5–34.9) 12.1 (9.9–14.6) 21.6 (19.3–24.1)
Age 20–34 years 38.2 (31.3–45.6) 15.1 (11.2–20.0) 26.5 (22.2–31.3)
Age 35–49 years 31.3 (25.2–38.2) 14.5 (10.4–19.7) 22.7 (18.8–27.1)
Age 50–64 years 27.4 (21.5–34.3) 8.4 (5.6–12.6) 17.9 (14.3–22.2)
Age 65–80 years 14.5 (10.0–20.4) 2.1 (0.9–5.0) 8.2 (5.8–11.4)
All subjects (crude) 18.4 (15.7–21.5) 8.4 (6.6–10.6) 13.4 (11.5–15.4)
All subjects (age-standardized) 19.7 (17.0–22.7) 8.9 (7.1–11.0) 14.3 (12.5–16.3)
Age 20–34 years 11.4 (7.6–16.8) 6.2 (3.7–10.1) 8.8 (6.1–12.4)
Age 35–49 years 17.6 (13.0–23.4) 10.4 (7.3–14.7) 13.9 (11.0–17.5)
Age 50–64 years 27.3 (21.7–33.7) 11.0 (7.6–15.6) 19.1 (15.7–23.0)
Age 65–80 years 39.3 (32.6–46.5) 7.6 (4.2–13.1) 23.2 (19.2–27.8)
All subjects (crude) 49.4 (45.2–53.6) 79.0 (75.7–81.9) 64.4 (61.4–67.2)
All subjects (age-standardized) 49.2 (45.2–53.1) 79.1 (76.0–81.9) 64.1 (61.3–66.9)
Age 20–34 years 50.4 (43.2–57.6) 78.7 (73.2–83.4) 64.7 (59.6–69.5)
Age 35–49 years 51.1 (44.1–58.2) 75.1 (69.4–80.0) 63.4 (58.6–67.9)
Age 50–64 years 45.2 (38.5–52.1) 80.6 (75.2–85.0) 63.1 (58.2–67.6)
Age 65–80 years 46.2 (39.0–53.6) 90.4 (84.8–94.1) 68.6 (63.7–73.2)
a Persons who reported smoking ≥ 100 cigarettes during their lifetime and at the time of the interview reported smoking daily or on some days.
b Persons who reported smoking ≥ 100 cigarettes during their lifetime but who did not smoke at the time of the interview.
c Persons reporting not having smoked ≥ 100 cigarettes during their lifetime.

TABLE 3. Prevalence of current drinking and binge drinking in men and women of different age groups in Arequipa, Peru,

Men Women Overall

Alcohol use / age strata % (95% CI) % (95% CI) % (95% CI)

Current drinkinga
All subjects (crude) 56.7 (52.6–60.6) 20.7 (17.6–24.2) 38.6 (35.7–41.5)
All subjects (age-standardized) 55.5 (51.7–59.2) 19.7 (16.8–22.9) 37.7 (35.0–40.4)
Age 20–34 years 61.5 (54.5–68.1) 26.5 (20.9–32.9) 43.9 (39.0–49.0)
Age 35–49 years 59.4 (52.3–66.1) 22.2 (17.4–28.0) 40.6 (35.9–45.5)
Age 50–64 years 48.8 (42.3–55.5) 9.4 (6.4–13.7) 29.0 (24.8–33.5)
Age 65–80 years 38.2 (31.4–45.5) 9.5 (6.2–14.3) 23.7 (19.7–28.2)
Binge drinkingb
All subjects (crude) 37.2 (33.2–41.3) 6.9 (5.1–9.2) 21.8 (19.5–24.3)
All subjects (age-standardized) 36.1 (32.4–40.1) 6.4 (4.8–8.4) 21.2 (19.0–23.6)
Age 20–34 years 42.9 (36.0–50.1) 10.2 (6.8–14.9) 26.5 (22.4–31.0)
Age 35–49 years 37.5 (30.8–44.7) 5.8 (3.5–9.4) 21.2 (17.4–25.6)
Age 50–64 years 34.5 (28.3–41.3) 1.8 (0.7–4.6) 17.8 (14.4–21.9)
Age 65–80 years 13.2 (8.9–19.2) 5.1 (2.8–9.0) 9.1 (6.5–12.5)
a Consumption of at least one drink of alcohol (one can or bottle of beer, one glass of wine, one cocktail or shot of liquor) in the three months prior to the interview.
b Consumption of five or more drinks of alcohol on one occasion (men) or four or more drinks of alcohol (women) on one occasion in the three months prior to the interview.

70.7%), followed by liquor/spirits Physical activity 57.6%. The prevalence of insufficient

(29.8%; 95% CI = 25.3%–34.8%), wine activity was significantly higher in
(4%; 95% CI = 2.5%–6.2%), and chicha, The age-standardized prevalence of women than men (63.3% vs. 51.9%; P <
a fermented maize beverage (0.2%; insufficient physical activity in the 0.01). The percentage of insufficient
95% CI = 0.04%–1.0%). population aged 20 to 69 years old was physical activity by gender and age

Rev Panam Salud Publica/Pan Am J Public Health 24(3), 2008 173

Original research Medina-Lezama et al. • Lifestyle-related cardiovascular risk factors in Peru

group is shown in Table 4. Among (41.0% among men and 43.1% among without significant gender differences
those with high educational level the women). The percentage of subjects (35.6% among men and 33.4% among
proportion with insufficient physical consuming high-fat diets was highest women). There was an age-related de-
activity was 58.6% (95% CI = 54.7%– in men (46.8%) and women (48.7%) be- cline in low fruit consumption in both
62.4%), whereas it was 56.1% (95% tween 20 and 34 years old; it was low- genders (Table 5). Among those with
CI = 51.4–60.6%) in subjects with low est among men and women who were high educational level the proportion
educational level. This difference was 65 to 80 years old (31.7% and 31.8%, re- of low fruit intake was 33.6% (95%
not statistically significant. spectively) (Table 5). Analysis by level CI = 29.9%–37.4%), whereas in subjects
of education showed that high-fat diet with low educational level this pro-
consumption was significantly higher portion was 37.2% [95% CI = 33.1%–
High-fat diet and low fruit among those with high (46.4%; 95% 41.5%; P = NS (not significant)]. The
and low vegetable intake CI = 42.1%–50.9%) than in those with age-standardized prevalence of low
low educational level (38.1%; 95% CI = vegetable intake was 33.3%. The pro-
The age-standardized prevalence of 33.8%–42.5%; P < 0.05). portion of low vegetable intake was
high-fat diet consumption was 42.0%, Overall, the age-standardized prev- 35.7% in men, and 30.8% in women
without significant gender differences alence of low fruit intake was 34.5%, (P = NS). An age-related decline in low

TABLE 4. Prevalence of insufficient activity in men and women of different age groups in Arequipa, Peru, 2004–2006 a

Men Women Overall

Age group % (95% CI) % (95% CI) % (95% CI)

Age 20–69 years (crude) 51.9 (47.6–56.2) 63.0 (59.1–66.8) 57.6 (54.4–60.6)
Age 20–69 years (age-standardized) 51.9 (47.8–55.9) 63.3 (59.6–66.9) 57.6 (54.6–60.6)
Age 20–34 years 50.3 (43.0–57.5) 62.0 (55.5–68.1) 56.2 (51.1–61.2)
Age 35–49 years 56.0 (49.0–62.8) 61.7 (55.5–67.4) 58.9 (54.1–63.5)
Age 50–69 years 49.3 (43.2–55.3) 67.6 (62.1–72.6) 58.6 (54.3–62.8)
a Insufficient activity is defined as not engaging in at least 30 minutes of moderate-intensity activities including brisk walking ≥ 5 days per week, or 20 minutes of vigorous-
intensity activities ≥ 3 days per week.

TABLE 5. Prevalence of high-fat diet, low fruit and low vegetable intake in men and women of different age groups in
Arequipa, Peru, 2004–2006

Men Women Overall

% (95% CI) % (95% CI) % (95% CI)

High-fat dieta
All subjects (crude) 41.9 (37.8–46.2) 44.0 (39.3–48.7) 42.9 (39.5–46.3)
All subjects (age-standardized) 41.0 (37.1–45.1) 43.1 (38.7–47.5) 42.0 (38.8–45.3)
Age 20–34 years 46.8 (39.7–54.2) 48.7 (40.5–57.0) 47.7 (42.1–53.4)
Age 35–49 years 44.0 (36.6–51.6) 45.7 (38.7–52.9) 44.8 (39.3–50.4)
Age 50–64 years 31.8 (25.4–39.0) 36.5 (30.0–43.6) 34.0 (29.0–39.3)
Age 65–80 years 31.7 (24.7–39.6) 31.8 (25.2–39.2) 31.7 (26.4–37.5)
Low fruit intakeb
All subjects (crude) 36.3 (32.4–40.4) 34.1 (30.5–37.9) 35.2 (32.3–38.2)
All subjects (age-standardized) 35.6 (31.9–39.5) 33.4 (30.0–37.1) 34.5 (31.7–37.4)
Age 20–34 years 40.6 (33.6–48.0) 35.3 (29.2–41.8) 37.9 (33.1–43.0)
Age 35–49 years 36.1 (29.8–43.0) 36.9 (31.1–43.1) 36.5 (32.0–41.3)
Age 50–64 years 31.9 (25.9–38.6) 31.8 (26.2–37.9) 31.8 (27.4–36.6)
Age 65–80 years 24.3 (18.5–31.2) 22.7 (17.4–29.2) 23.5 (19.2–28.4)
Low vegetable intakec
All subjects (crude) 36.5 (32.6–40.5) 31.1 (27.7–34.8) 33.8 (31.0–36.7)
All subjects (age-standardized) 35.7 (32.1–39.6) 30.8 (27.5–34.2) 33.3 (30.6–36.1)
Age 20–34 years 40.5 (33.8–47.7) 32.9 (27.4–38.9) 36.7 (32.1–41.5)
Age 35–49 years 35.1 (29.0–41.8) 30.6 (25.1–36.8) 32.8 (28.4–37.6)
Age 50–64 years 32.1 (26.1–38.8) 28.8 (23.7–34.6) 30.5 (26.2–35.1)
Age 65–80 years 29.3 (23.0–36.5) 28.5 (22.6–35.3) 28.9 (24.2–34.1)
a MEDFICTS questionnaire scores > 38 points.
b Consumption of fruits on less than three days per week.
c Consumption of fresh vegetables on less than three days per week.

174 Rev Panam Salud Publica/Pan Am J Public Health 24(3), 2008

Medina-Lezama et al. • Lifestyle-related cardiovascular risk factors in Peru Original research

vegetable intake was observed spe- prevalence of current smoking was populations (30). We also found that
cially among men (Table 5). Analysis seen with increasing age in both gen- the proportion of current smokers was
by level of education showed that ders. This decrease, also observed in significantly higher among those with
low vegetable intake was 31.9% (95% other Latin American populations and higher educational level, a group asso-
CI = 28.3%–35.7%) among those with in the U.S. (29, 32), may be a conse- ciated with higher income. This sug-
high education, and 36.0% (95% CI = quence of one or a combination of co- gests that cigarette prices and purchas-
32.1%–40.1%; P = NS) in those with hort effects (different generations may ing power are variables that have an
low education. smoke differently), an increase in the important impact on cigarette con-
number of persons who quit smoking sumption in this population.
as they get older, or a survival bias
DISCUSSION (the premature death of smokers com-
pared to never- and former smokers). Alcohol drinking patterns
Cigarette smoking We found that in men, the decline in
the prevalence of current smoking We found high levels of alcohol con-
Smoking is a well-established, major with increasing age was accounted for sumption in the study population;
risk factor for CVD as well as for can- by increases in the proportions of for- overall, more than half of men and
cer and other serious illness (14). In mer smokers in the older groups. In around one-fifth of women were cur-
this study, we found that 21.6% of contrast, among women, the decline in rent drinkers. We believe that this gen-
adults were current smokers, 14.3% the prevalence of current smoking der difference may be explained by so-
were former smokers, and 64.1% never with increasing age was mainly ac- cial and cultural elements present in
smoked. The prevalence of current counted for by increases in the propor- Latin American populations that make
smoking was almost three-fold higher tions of never-smokers in the older drinking alcohol generally more ac-
in men (31.1%) than women (12.1%), groups (Table 2). Although our cross- ceptable for men than for women (33).
findings that are in line with the sig- sectional results can not quantify sur- Although especially high percentages
nificantly higher prevalence of smok- vival bias in the study population, of current drinkers were observed
ing observed in men in many Latin they suggest that the decline of current among young adults, alcohol intake
American populations and among smoking with age in men is largely was by no means confined to the
Hispanics living in the United States due to smoking cessation, whereas in younger groups. Males between 20
(28–32). According to the 2005 Na- women the cohort effect may be more and 34 years old had the highest prev-
tional Health Interview Survey, the important (meaning that more recent alence of current drinking (61.5%), and
overall prevalence of current cigarette cohorts are adopting tobacco use at a rates remain high among men 35 to 49
smoking among U.S. adults was esti- higher rate than did their predeces- years old (59.4%), 50 to 64 years old
mated at 23.9% for men and 18.1% for sors). Further studies are required to (48.8%), and 65 to 80 years old (38.2%).
women, whereas the prevalence of examine health perceptions related to In women, the prevalence of current
current smoking among Hispanics liv- tobacco use and potentially important drinking among those 20 to 34 years
ing in the U.S. was 21.1% in men and determinants of smoking cessation old (26.5%) and 35 to 49 years old
11.1% in women (32). In Latin America (or lack thereof) in Peruvian men and (22.2%) was comparable, but was sub-
and the Caribbean, the prevalence of women, as these culturally related stantially lower among those 50 years
current smoking has been estimated to variables may provide important clues old and older (around 9%).
range from 24.1% (Paraguay) to 66.3% for the design of effective smoking ces- Alcohol consumption is a public
(Dominican Republic) for men and sation and education programs. health problem in Latin America and
from 5.5% (Paraguay) to 26.6% (Uru- In contrast to findings from other the Caribbean, where is it estimated
guay) for women (30). Although there populations, where the majority of that 4.5% of all deaths can be attrib-
are important differences in the smokers smoke daily (e.g., in the U.S. uted to alcohol use (33). In general, in
methodology and definitions of vari- more than 80% of current smokers are Latin American countries substantial
ous categories of smoking status used daily smokers) (32), only 31.2% of cur- alcohol consumption is observed in
by previous studies in the region, rent smokers were daily smokers in the form of irregular, heavy drinking
which makes comparisons of smoking Arequipa. Also, the percentage of bouts (34). The pattern of alcohol con-
prevalence problematic, the available heavy smokers was considerably sumption is an important factor to
data suggest that Latin America is het- lower (2.4% of all current smokers or consider when investigating the ef-
erogeneous regarding tobacco use and 7.9% of daily smokers). We speculate fects of alcohol on cardiovascular
that the variation in smoking preva- that this smoking pattern is related to health. An increasing amount of evi-
lence among countries is considerable economic factors, as it has been ob- dence suggests that the effects on car-
(28–31). served that the price of cigarettes and diovascular health of a pattern of
The highest prevalence of current the population income are potential steady, daily alcohol consumption are
smoking in our study was among factors in varying cigarette consump- different from those associated with a
young adults, and a decrease in the tion levels in other Latin American pattern of alternating heavy drinking

Rev Panam Salud Publica/Pan Am J Public Health 24(3), 2008 175

Original research Medina-Lezama et al. • Lifestyle-related cardiovascular risk factors in Peru

and abstinence (35). Epidemiologic high proportion of 20 to 34-year-old (45). The few studies that have quanti-
studies have consistently associated men who reported binge drinking epi- fied physical activity in Latin Ameri-
regular, moderate alcohol drinking sodes (42.9%). Furthermore, our study can countries also document high rates
with decreased risk of cardiovascu- indicates that the Anglo-Saxon pattern of sedentary lifestyle (29, 46–48).
lar events (36, 37). At the same time, was by far the predominant pattern of Lack of comparability has been a
acute, heavy, episodic alcohol intake alcohol drinking (99.1%) and that most major limitation in studies on physical
(e.g., binge drinking) appears to be as- of the alcohol consumed comes from activity, mostly due to the use of dif-
sociated with a higher risk of coronary beer and spirits. The results presented ferent methodological instruments. In
events and mortality (38–40). here indicate that alcohol is consumed order to minimize this problem, a
Differences in the physiological ef- mainly during weekends and holi- standardized International Physical
fects of regular, moderate drinking days, usually in the form of binges Activity Questionnaire (IPAQ) has
and binge drinking may explain these when beer or spirits are consumed, been proposed as an instrument for
findings. A variety of mechanisms and that the pattern of daily, moderate cross-national monitoring of physical
have been described by which the drinking associated with protection activity and inactivity (18). A key fea-
moderate intake of alcohol may reduce against CVD is uncommon. Clearly, ture of this particular questionnaire is
cardiovascular risk; these include an educational campaigns are required to that it can be used in all types of soci-
increase in high-density lipoprotein address the harmful patterns of alco- eties, from industrialized nations to
cholesterol, a decrease in platelet ag- hol use in this population, and further developing countries.
gregation and coagulation effects, in- studies in Peru and the rest of Latin Our findings in the Arequipa popu-
creased fibrinolysis, and beneficial ef- America are needed to determine the lation indicate that the majority (57.6%)
fects on endothelial function (36). overall impact of alcohol consumption of adults aged 20 to 69 years do not
There are, however, a number of pos- as a risk factor of CVD. engage in the internationally recom-
sible ways by which heavy, episodic mended levels of physical activity. We
alcohol consumption might have neg- found that the prevalence of insuffi-
ative cardiovascular effects, including Physical activity cient activity was higher for women
induction of ventricular arrhythmias, than for men, probably reflecting gen-
coronary vasoconstriction, acute ele- There is compelling evidence that der differences mainly in activities re-
vation of blood pressure, and in- physical activity has many important lating to work and leisure time. Like
creased clotting activity (35, 41). Binge health benefits and that sedentary other studies evaluating the four major
drinking has also been associated with habits are associated with an increased domains of physical activity (46, 47),
an increased risk of all types of stroke risk of numerous chronic diseases and and unlike studies evaluating only
(particularly in younger men) (42), decreased longevity (21). Inactivity is leisure time activity, our data did not
atrial fibrillation, other forms of ar- recognized as a risk factor for coronary show an association between insuffi-
rhythmias, and sudden cardiac death artery disease and cardiovascular cient activity and age (Table 4).
(35). This may be linked to the Satur- mortality, whereas regular physical Comparable data from population-
day, Sunday, and Monday peaks in activity plays a role in both primary based studies that have used the IPAQ
cardiovascular mortality and morbid- and secondary prevention of CVD survey questions in Latin America are
ity documented in some countries (22). Physical activity also favorably limited. Matsudo et al. (46) reported
characterized by a high alcohol intake modifies several CVD risk factors, in- overall prevalence of insufficient activ-
on weekends (43). Many other conse- cluding obesity, high blood pressure, ity of 46.5% in São Paulo, Brazil. The
quences of binge drinking have espe- harmful lipid patterns, and insulin re- age-specific percentages of insufficient
cially high social and economic costs, sistance (22). The U.S. Department of activity in the São Paulo sample were
including unintentional injuries, inter- Health and Human Services recom- 43.7%, 48.6%, and 46.4% for subjects
personal violence (e.g., homicide, as- mends that adults should engage in at 15–29 years old, 30–49 years old, and
sault, domestic violence, rape, and least 30 minutes of moderate-intensity 50–69 years old, respectively. These
child abuse), unsafe sexual practices, activity on 5 or more days per week, or values are lower than the age-specific
unintended pregnancy, child neglect, 20 minutes of vigorous-intensity activ- percentages reported in our study
and lost productivity (44). ity on 3 or more days per week (20), (Table 4). In a population-based sur-
We found widespread binge drink- with additional health and fitness ben- vey of adult residents of Bogotá,
ing in the Arequipa, Peru, population, efits deriving from becoming more Colombia, Gómez et al. found overall
particularly in men. The percentage of physically active or including more prevalence of insufficient activity of
men who binge drink was 36.1% vigorous activities (21). However, de- 63.2% (48). The age-specific prevalence
(more than two-thirds of male current spite the known health benefits of reg- rates in Bogotá of insufficient activity
drinkers), whereas the percentage of ular physical activity and community- were 59.9%, 64.1%, and 69.0% for sub-
women who binge drink was 6.4% wide campaigns, more than half of jects 18–29 years old, 30–49 years old,
(more than a third of female current U.S. adults do not engage in the mini- and 50–65 years old, respectively (val-
drinkers). It is important to note the mum recommended activity levels ues higher than those found in our

176 Rev Panam Salud Publica/Pan Am J Public Health 24(3), 2008

Medina-Lezama et al. • Lifestyle-related cardiovascular risk factors in Peru Original research

study). These data indicate that low We found that an important per- tion studies, self-reported use of alco-
levels of physical activity are already a centage (42%) of adults in Arequipa hol is likely to underestimate the real
serious problem in urban areas of the reported consuming high-fat diets, consumption level. Over-reporting of
region and that some variations in the which tended to decrease with increas- physical activity using the IPAQ in-
prevalence of insufficient activity exist ing age. Many possible factors that strument has been previously reported
in different cities of Latin America. need further investigation may ex- and this could also be a problem in our
plain this last finding (e.g., prevalence study, particularly among the older age
of chronic diseases that are managed groups (52).
High-fat diet and low fruit with low-fat diets is higher in the older
and low vegetable intake age groups, different generational diet
preferences, survival bias, among oth- Conclusions
Epidemiologic studies have shown ers). Interestingly, the proportion of
that people that consume high amounts high-fat diet consumption was signifi- We report alarmingly high preva-
of saturated fatty acids and cholesterol cantly higher among those with high lence rates of tobacco smoking, harm-
have an increased risk of developing educational level, probably reflecting ful alcohol consumption patterns, in-
CVD, mostly by modulating plasma cultural and economic variables pre- sufficient physical activity, high-fat
lipoprotein concentrations (49). Total sent in different social strata of this diets, and low fruit and low vegetable
fat and saturated fat intake has in- population. Finally, more than one- intake in this Andean population,
creased dramatically in Latin America third of subjects reported consuming along with detailed data on the popu-
(50, 51). Populations in South America’s fruits or fresh vegetables less than lation distribution of these cardiovas-
Southern Cone, which have tradition- three days per week. This is a trou- cular risk factors. Control programs
ally obtained around 25% of the total bling finding considering that fruits for CVD risk factors that include sys-
energy supply from fats, approached a and vegetables are important compo- tematic surveillance and the adoption
level of 30% by the 1990s. Brazil had the nents of a healthy diet, and that suffi- of multiple intervention strategies are
most rapid increase, from just over 16% cient daily consumption could help urgently needed to deal with this
to about 28% in 30 years. In the Andean prevent major diseases such as certain growing problem. Data such as that
region there has also been an increase cancers and CVD (51). provided here will aid in the design of
in fat intake, from about 18% of total The findings of the study are subject programs to prevent CVD in Peru and
energy during the 1960s to about 24% to certain limitations. The results were the Andean region.
during the 1990s. These data are based based on self-report, and thus the prev-
on country data from the Food and alence observed may have been af- Acknowledgments. We are in-
Agriculture Organization of the United fected by recall bias. Estimates for cig- debted to the Peruvian National Insti-
Nations food balance sheets for the pe- arette smoking were not validated by tute of Statistics and Informatics for
riod 1964–1996 (51). biochemical tests. As in other popula- their support with this project.


1. Barceló A. Cardiovascular diseases in Latin 7. Popkin BM. The nutrition transition in low- population from Peru and regional migration
America and the Caribbean. Lancet. income countries: an emerging crisis. Nutr patterns of Amerindians in South America:
2006;368(9536):625–26. Rev. 1994;52(9):285–98. data from Y chromosome and mitochondrial
2. Reddy KS, Yusuf S. Emerging epidemic of 8. Yusuf S, Reddy S, Ounpuu S, Anand S. DNA. Hum Hered. 2001;51(1–2):97–106.
cardiovascular disease in developing coun- Global burden of cardiovascular diseases: 12. Modiano G, Bernini L, Carter ND, San-
tries. Circulation. 1998;97(6):596–601. part I: general considerations, the epidemio- tachiara-Benerecetti AS, Setter JC, Baur EW,
3. Chavez DR. Issues and challenges for CVD logic transition, risk factors, and impact of ur- et al. A survey of several red cell and serum
prevention in Ibero-America: the challenge of banization. Circulation. 2001;104(22):2746–53. genetic markers in a Peruvian population.
human resource development. Can J Cardiol. 9. Lanas F, Avezum A, Bautista LE, Diaz R, Am J Hum Genet. 1972;24(2):111–23.
1993;9(Suppl D):195D–196D. Luna M, Islam S, et al. Risk factors for acute 13. Brutsaert TD, Parra E, Shriver M, Gamboa A,
4. Medina-Lezama J, Chirinos JA, Chirinos- myocardial infarction in Latin America: the Palacios J, Rivera M, et al. Effects of birthplace
Pacheco J. Cardiovascular disease in Latin INTERHEART Latin American study. Circu- and individual genetic admixture on lung
America. Am Heart J. 2005;149(2):E13. lation. 2007;115(9):1067–74. volume and exercise phenotypes of Peruvian
5. Omran AR. The epidemiologic transition: a 10. Medina-Lezama J, Chirinos JA, Zea Díaz H, Quechua. Am J Phys Anthropol. 2004;123(4):
theory of the epidemiology of population Morey O, Bolanos JF, Munoz-Atahualpa E, et 390–8.
change. Milibank Mem Fund Q. 1971;49(4): al. Design of PREVENCION: a population- 14. World Health Organization (WHO). Guide-
509–38. based study of cardiovascular disease in lines for controlling and monitoring the to-
6. Frenk J, Lozano R, Bobadilla JL. The epidemi- Peru. Int J Cardiol. 2005;105(2):198–202. bacco epidemic. Geneva: WHO; 1998.
ological transition in Latin America. Notas 11. Rodriguez-Delfin LA, Rubin-de-Celis VE, 15. Rehm J, Rehn N, Room R, Monteiro M, Gmel
Población. 1994;22(60):79–101. Zago MA. Genetic diversity in an Andean G, Jernigan D, et al. The global distribution of

Rev Panam Salud Publica/Pan Am J Public Health 24(3), 2008 177

Original research Medina-Lezama et al. • Lifestyle-related cardiovascular risk factors in Peru

average volume of alcohol consumption and ization of rates: a new WHO standard. GPE Eastern Europe: potential physiological
patterns of drinking. Eur Addict Res. 2003; Discussion Paper Series: No. 31. Geneva: mechanism. J R Soc Med. 1998;91(8):402–7.
9(4):147–56. World Health Organization; 2006. 42. Mazzaglia G, Britton AR, Altmann DR, Che-
16. Wechsler H, Isaac N. Binge drinkers at Mas- 28. Bautista LE, Orostegui M, Vera LM, Prada net L. Exploring the relationship between
sachusetts colleges; prevalence, drinking GE, Orozco LC, Herran OF. Prevalence and alcohol consumption and non-fatal or fatal
style, time trends, and associated problems. impact of cardiovascular risk factors in Bu- stroke: a systematic review. Addiction. 2001;
JAMA. 1992;267(21):2929–31. caramanga, Colombia: results from the Coun- 96(12):1743–56.
17. Wechsler H, Nelson TF. Binge drinking and trywide Integrated Noncommunicable Dis- 43. Evans C, Chalmers J, Capewell S, Redpath
the American college student: what’s five ease Intervention Programme (CINDI/ A, Finlayson A, Boyd J, et al. “I don’t like
drinks? Psychol Addict Behav. 2001;15(4): CARMEN) baseline survey. Eur J Cardiovasc Mondays”—day of the week of coronary
287–91. Prev Rehabil. 2006;13(5)769–75. heart disease deaths in Scotland: study of
18. Craig CL, Marshall AL, Sjöström M, Bauman 29. Jadue L, Vega J, Escobar MC, Delgado I, Gar- routinely collected data. BMJ. 2000;320(7229):
AE, Booth ML, Ainsworth BE, et al. Interna- rido C, Lastra P, et al. Risk factors for non- 218–9.
tional Physical Activity Questionnaire communicable diseases: methods and global 44. Naimi TS, Brewer RD, Mokdad A, Denny C,
(IPAQ): 12-country reliability and validity. results of the CARMEN program basal sur- Serdula MK, Marks JS. Binge drinking among
Med Sci Sports Exerc. 2003;35(8):1381–95. vey. Rev Med Chil. 1999;127(8):1004–13. US adults. JAMA. 2003;289(1):70–5.
19. Guidelines for data processing and analysis 30. da Costa e Silva VL, Koifman S. Smoking in 45. Centers for Disease Control and Prevention
of the International Physical Activity Ques- Latin America: a major public health problem. (CDC). Adult participation in recommended
tionnaire (IPAQ)–short and long forms (No- Cad Saude Publica. 1998;14(Suppl 3):99S–108S. levels of physical activity—United States,
vember 2005) [Internet site]. Available from: 31. United States Department of Health and 2001 and 2003. MMWR Morb Mortal Wkly Accessed 4 No- Human Services (U.S. DHHS). Smoking and Rep. 2005;54(47):1208–12.
vember 2006. health in the Americas. Atlanta, Georgia: U.S. 46. Matsudo S, Matsudo V, Araújo T, Andrade D,
20. U.S. Department of Health and Human Ser- DHHS; 1992. Andrade E, Oliveira L, et al. Nível de ativi-
vices. Healthy people 2010. 2nd ed. Washing- 32. Centers for Disease Control and Prevention dade física da população do Estado de São
ton, D.C.: U. S. Government Printing Office; (CDC). Tobacco use among adults—United Paulo: análise de acordo com o gênero, idade,
2000. States, 2005. MMWR Morb Mortal Wkly Rep. nível sócio-econômico, distribuição geográ-
21. Pate RR, Pratt M, Blair SN, Haskell WL, Mac- 2006;55(42):1145–8. fica e de conhecimento. Rev Bras Ciênc Mov.
era CA, Bouchard C, et al. Physical activity 33. Pyne HH, Claeson M, Correia M. Gender di- 2002;10(4):41–50.
and public health. A recommendation from mensions of alcohol consumption and alco- 47. Hallal PC, Victora CG, Wells JC, Lima RC.
the Centers for Disease Control and Preven- hol-related problems in Latin America and Physical inactivity: prevalence and associated
tion and the American College of Sports Med- the Caribbean. Washington, D.C.: Interna- variables in Brazilian adults. Med Sci Sports
icine. JAMA. 1995;273(5)402–7. tional Bank for Reconstruction and Develop- Exerc. 2003;35(11):1894–1900.
22. NIH Consensus Development Panel on Phys- ment/The World Bank; 2002. (World Bank 48. Gómez LF, Duperly J, Lucumí DI, Gámez R,
ical Activity and Cardiovascular Health. Discussion Paper No. 433). Venegas AS. Physical activity levels in adults
Physical activity and cardiovascular health. 34. Rehm J, Monteiro M. Alcohol consumption living in Bogotá (Colombia): prevalence and
JAMA. 1996; 276(3): 241–246. and burden of disease in the Americas: impli- factors associated. Gac Sanit. 2005;19(3):
23. National Cholesterol Education Program cations for alcohol policy. Rev Panam Salud 206–13.
(NCEP) Expert Panel on Detection, Evalua- Publica. 2005;18(4–5):241–8. 49. Shekelle RB, Stamler J. Dietary cholesterol
tion, and Treatment of High Blood Choles- 35. Puddey IB, Rakic V, Dimmitt SB, Beilin LJ. In- and ischaemic heart disease. Lancet. 1989;
terol in Adults (Adult Treatment Panel III). fluence of pattern of drinking on cardiovas- 1(8648):1177–9.
Third Report of the National Cholesterol Ed- cular disease and cardiovascular risk fac- 50. Bermudez OI, Tucker KL. Trends in dietary
ucation Program (NCEP) Expert Panel on De- tors—a review. Addiction. 1999;94(5):649–63. patterns of Latin American populations. Cad
tection, Evaluation, and Treatment of High 36. Pearson TA. Alcohol and heart disease. Cir- Saude Publica. 2003;19(Suppl 1):S87–S99.
Blood Cholesterol in Adults (Adult Treat- culation. 1996;94(11):3023–5. 51. World Health Organization (WHO) and Food
ment Panel III), final report. Circulation. 2002; 37. Corrao G, Rubbiati L, Bagnardi V, Zambon A, and Agriculture Organization of the United
106(25):3143–421. Poikolainen K. Alcohol and coronary heart dis- Nations (FAO). Diet, nutrition and the pre-
24. Taylor AJ, Wong H, Wish K, Carrow J, Bell D, ease: a meta-analysis. Addiction. 2000;95(10): vention of chronic diseases. Report of a Joint
Bindeman J, et al. Validation of the MED- 1505–23. WHO/FAO Expert Consultation. Geneva:
FICTS dietary questionnaire: a clinical tool to 38. McElduff P, Dobson AJ. How much alcohol WHO; 2003. (WHO Technical Report Series
assess adherence to American Heart Associa- and how often? Population based case-con- 916).
tion dietary fat intake guidelines. Nutr J. trol study of alcohol consumption and risk of 52. Rzewnicki R, Vanden Auweele Y, De Bour-
2003;2(1):4. a major coronary event. BMJ. 1997;314(7088): deaudhuij I. Addressing overreporting on
25. Krauss RM, Eckel RH, Howard B, Appel LJ, 1159–64. the International Physical Activity Question-
Daniels SR, Deckelbaum RJ, et al. AHA di- 39. Kauhanen J, Kaplan GA, Goldberg DE, Salo- naire (IPAQ) telephone survey with a popu-
etary guidelines: revision 2000: A statement nen JT. Beer binging and mortality: results lation sample. Public Health Nutr. 2003;6(3):
for healthcare professionals from the Nutri- from the Kuopio ischaemic heart disease risk 299–305.
tion Committee of the American Heart Asso- factor study, a prospective population based
ciation. Circulation. 2000;102(18):2284–99. study. BMJ. 1997;315(7112):846–51.
26. Krewski D, Rao JNK. Inference from stratified 40. Mukamal KJ, Maclure M, Muller JE, Mittle-
samples: properties of the linearization, jack- man MA. Binge drinking and mortality after
knife, and balanced repeated replication acute myocardial infarction. Circulation.
methods. Ann Stat. 1981;9(5):1010–19. 2005;112(25):3839–45.
27. Ahmad OB, Boschi-Pinto C, Lopez AD, Mur- 41. McKee M, Britton A. The positive relation- Manuscript received on 6 June 2007. Revised version ac-
ray CJL, Lozano R, Inoue M. Age standard- ship between alcohol and heart disease in cepted for publication on 21 April 2008.

178 Rev Panam Salud Publica/Pan Am J Public Health 24(3), 2008

Medina-Lezama et al. • Lifestyle-related cardiovascular risk factors in Peru Original research

RESUMEN Objetivos. Estimar la prevalencia de factores de riesgo cardiovascular relacionados

con el estilo de vida de adultos de Arequipa, la segunda mayor ciudad de Perú.
Métodos. Se realizó un estudio de base poblacional para evaluar la prevalencia y los
Prevalencia de factores patrones de consumo de tabaco y bebidas alcohólicas, la falta de actividad física, la
de riesgo cardiovascular dieta rica en grasas y el bajo consumo de frutas y vegetales en 1 878 personas (867
relacionados con el estilo hombres y 1 011 mujeres).
Resultados. Las prevalencias estandarizadas por la edad de los fumadores actuales,
de vida en Perú: pasados y de los que nunca fumaron fueron 21,6%, 14,3% y 64,1%, respectivamente.
el estudio PREVENCIÓN La prevalencia de tabaquismo fue significativamente mayor en los hombres que en las
mujeres (31,1% frente a 12,1%; P < 0,01). La prevalencia del consumo de bebidas al-
cohólicas fue de 37,7%, significativamente mayor en los hombres que en las mujeres
(55,5% frente a 19,7%; P < 0,01). La prevalencia del consumo excesivo de alcohol fue
de 21,1%, mayor en los hombres que en las mujeres (36,1% frente a 6,4%; P < 0,01). La
gran mayoría de los bebedores presentó un patrón de consumo concentrado funda-
mentalmente en los fines de semana y los días feriados, más que el consumo habitual
con las comidas en los días laborables. La proporción de personas con insuficiente ac-
tividad fue de 57,6%, significativamente mayor en las mujeres que en los hombres
(63,3% frente a 51,9%; P < 0,01). En general, 42,0% de los adultos informaron consu-
mir dietas ricas en grasas, 34,5% dijo tener un bajo consumo de frutas y 33,3% un bajo
consumo de vegetales.
Conclusiones. La alta prevalencia de factores de riesgo cardiovascular relacionados
con el estilo de vida encontrada en esta población de los Andes es preocupante. Se
deben implementar urgentemente programas preventivos para resolver este creciente

Palabras clave Consumo de bebidas alcohólicas, enfermedades cardiovasculares, dieta, activi-

dad física, prevalencia, factores de riesgo, tabaquismo, Perú.

Ataque cardíaco y accidente cerebrovascular. Prevención

Esta publicación presenta datos de la eficacia de las inter-
venciones para la prevención secundaria de la cardiopatía
coronaria y la enfermedad cerebrovascular, así como para
la prevención de la vasculopatía periférica y la diabetes.
También se incluyen en ella recomendaciones clínicas y
se identifican las áreas en las cuales se necesita mayor

El libro está destinado a los profesionales médicos con la

responsabilidad de atención de pacientes con ECV en pobla-
ciones y países de ingresos bajos y medianos. Su objetivo es
proporcionar orientación general acerca de la eficacia de 2005, pp., 107
ISBN: 92 75 31610 4
intervenciones específicas, tanto farmacológicas como sobre US$ 15.00 en América
comportamientos y hábitos de vida. Latina y el Caribe/
US$ 22.00 en el resto del
Adquiera esta publicación por medio de la librería en línea de la OPS:; correo electrónico:;
Fax: (301) 209-9789; Oficina de país de la OPS/OMS Código: PC 610

Rev Panam Salud Publica/Pan Am J Public Health 24(3), 2008 179