Sei sulla pagina 1di 8

Original Paper

Ann Nutr Metab 2010;56:233–240 Received: January 4, 2009

Accepted after revision: February 10, 2010
DOI: 10.1159/000288313
Published online: March 30, 2010

Nutritional Knowledge, Attitude and Practice of

Tehranian Adults and Their Relation to Serum Lipid
and Lipoproteins: Tehran Lipid and Glucose Study
Parvin Mirmiran a, b Fatemeh Mohammadi-Nasrabadi a Nasrin Omidvar b

Firoozeh Hosseini-Esfahani a Homeira Hamayeli-Mehrabani a Yadollah Mehrabi c


Fereidoun Azizi a   

Obesity Research Center, Research Institute for Endocrine Sciences, b Faculty of Nutrition Sciences and Food

Technology, National Nutrition and Food Technology Institute c Faculty of Public Health, Shahid Beheshti

University of Medical Sciences, Tehran, Iran

Key Words sured, based on standard protocols, and LDL-cholesterol was

Knowledge, nutritional ⴢ Non-communicable disease ⴢ calculated. Associations between nutritional knowledge, at-
Risk factors ⴢ Tehran Lipid and Glucose Study titude, practice, sex, age and educational level, and between
knowledge, attitude and practice and BMI, waist to hip ratio
and levels of lipid and lipoproteins, were determined. Re-
Abstract sults: For knowledge 26.5, 52.7 and 20.8%, for attitude, 27.6,
Background/Aim: The high prevalence of non-communica- 48.9 and 23.5% and for practice, 27.4, 51.7 and 20.9% of
ble diseases and their risk factors among the general urban individuals had desirable, moderate and weak knowledge
population of Tehran necessitates an urgent implementa- scores, respectively. With increases in educational levels,
tion of nutritional interventional programs to curtail these knowledge scores increased (p ! 0.001). Attitudes of illiterate
risk factors. This study aimed to provide preliminary data on and poorly literate groups were weaker than those of other
the knowledge, attitude and practices of urban Tehranian groups (p ! 0.001). However, illiterate and poorly literate
adults regarding nutrition and their relation to major non- groups had higher practice scores as compared to middle
communicable disease risk factors and to assess the need for school and high school graduates (p ! 0.001). Mean practice
educational interventions. Methods: This cross-sectional scores of university graduates were higher than high school
study was performed within the framework of the second graduates (p ! 0.05). Those aged 30–50 years had higher
phase of the Tehran Lipid and Glucose Study (TLGS). Subjects knowledge scores than older individuals (p ! 0.001). LDL-
were 826 individuals, 369 men and 457 women, aged 20–70 and total cholesterol were correlated to practice scores (p !
years old, selected randomly from among 15,005 partici- 0.001, r = 0.23). Subjects with borderline triglyceride levels
pants in the TLGS. Knowledge, attitude and practices of sub- had better practice scores than normal individuals (p !
jects were assessed by a ‘knowledge, attitude and practice’ 0.001). Conclusions: The findings revealed that age, educa-
questionnaire; height, weight, waist and hip circumferences tional level and gender are factors that can influence knowl-
were measured, and BMI and waist to hip ratio were calcu- edge, attitude and practices, regarding nutrition. It is vital to
lated. Biochemical data including fasting blood glucose, develop approaches and implement programs to improve
cholesterol, triglycerides and HDL-cholesterol were mea- the nutritional practices of the population.
Copyright © 2010 S. Karger AG, Basel

© 2010 S. Karger AG, Basel Parvin Mirmiran, PhD

0250–6807/10/0563–0233$26.00/0 Obesity Research Center, Research Institute for Endocrine Sciences
Fax +41 61 306 12 34 Shahid Beheshti University of Medical Sciences
E-Mail Accessible online at: PO Box 19395-4763, Tehran (Iran) Tel. +98 21 2243 2484, Fax +98 21 2240 2463, E-Mail mirmiran @
Introduction Method

The TLGS, initiated in 1999, is a large-scale, prospective study

Non-communicable diseases (NCDs) are a major of non-communicable diseases and their risk factors among Teh-
health burden in industrialized countries, and are on the ran’s urban population. Its aim is to develop population-based
rise in developing countries owing to demographic tran- measures to change the lifestyle of the population and to prevent
sitions and changing lifestyles among people. In the the rising trend of diabetes mellitus and dyslipidemia. The ethical
Global Burden of Disease Study 1990 [1], which estimated committee of the National Research Council of the Islamic Re-
public of Iran has approved this research. The sampling frame was
the distribution of deaths by region, non-communicable chosen from urban district 13 of Tehran, the capital city of the
diseases ranked first as cause of death in developed coun- Islamic Republic of Iran. The age distribution of the population
tries, in many developing countries and in the world as a of district 13 is representative of the urban population of Tehran
whole. In the developed countries, 3 out of 4 deaths are [5, 7]. The TLGS was conducted on 15,005 individuals, both males
due to cardio-vascular disease (CVDs), cancer, accidents and females, between the ages of 3 and 69 years, who were ran-
domly sampled from Tehran’s urban district 13 and assessed for
or other violent causes. In many developing countries, the presence of the following non-communicable risk factors: di-
NCDs, especially cardiovascular disease, are already a abetes, hyperlipidemia, smoking, hypertension, glucose intoler-
more common cause of death than infectious diseases [2]. ance, obesity, physical inactivity and dietary habits, as well as cor-
Thus, addressing the problems and issues connected with onary artery disease. A sub-sample consisting of 826 individuals
non-communicable diseases will lead to major health (369 males and 457 females) aged 20 years and over, was random-
ly selected from the TLGS population for knowledge, attitude and
benefits. practice study. Subjects were divided into 4 groups (illiterate and
Research has clearly shown that non-communicable poorly literate, middle school, diploma and university graduates),
diseases have their roots in unhealthy lifestyles or adverse based on their educational levels and, again into 2 groups, based
physical and social environments. Over a prolonged pe- on marital status (single and married).
riod, risk factors such as unhealthy nutrition, smoking,
Anthropometrical Assessment
physical inactivity, excessive use of alcohol, and psycho- Weight was recorded without shoes and socks using the Seca
social stress are among the major lifestyle issues. How- 707 weighing machine (range 0.1–150 kg) with an accuracy of up
ever, the incidence of these factors and CVD on a longi- to 100 g. The weighing machine was repeatedly checked by a stan-
tudinal basis has not been determined in most developing dard weight for accuracy after every 10th weighing. Height was
countries [3]. The most systematic finding throughout measured without shoes with a tape meter stadiometer with a least
count of 1 mm. BMI was calculated by dividing weight (kg) to
the limited number of countries studied was the rapid height squared (m2). Underweight, overweight and obesity were
increase of overweight and obesity, probably as a result of defined according to recommended BMI cut-off values for adults
shifts toward energy-dense diets, sedentary lifestyles and [8]. Waist and hip circumferences were measured according to
increased risk factors for NCDs [4]. standard protocols to the nearest 0.5 cm. Waist to hip ratios 61.0
Identification of the presence and magnitude of CVD in males, and 60.8 in females, were considered as abdominal obe-
sity [7].
risk factors is vital for the development of proper ap-
proaches to the prevention of NCDs. In response to this Biochemical Assessment
necessity, the Tehran Lipid and Glucose Study (TLGS) After 12–14 h fasting, a venous blood sample was taken from
has been designed to study the trend of NCDs and their each subject for measurement of fasting blood glucose, total cho-
risk factors in an urban population in Tehran, and also to lesterol and HDL-cholesterol, and triglycerides. FBG was mea-
sured by enzymatic glucose oxidase. The auto analyzer Selectra 2
consider changing lifestyles to prevent and decrease the performed all serum lipid and lipoprotein measurements on the
risk of major NCDs and their outcomes [5]. same day. Total cholesterol was measured by the enzymatic meth-
For promotion of effective healthy eating, it is neces- od, using cholesterol esterase and cholesterol oxidase. HDL-cho-
sary to understand the attitudes towards and beliefs lesterol was measured using the same method after precipitation
about nutrition among the general public [6]. The pres- of apolipoprotein B by phosphotungstic acid. Serum triglyceride
concentration was measured by the enzymatic method, using
ent study was conducted to provide baseline data about glycerol phosphate oxidase. LDL-cholesterol was calculated in se-
nutritional knowledge, attitude and practices of urban rum samples with TG !400 mg/dl using the Friedwald formula
Tehranians, and to identify the obstacles that prevent [9]. The classification criteria for serum lipid and lipoprotein lev-
implementation of healthy changes and to assess the els in adults as recommended by the National Cholesterol Educa-
tion Program were used [10, 11].
need for educational intervention within the framework
of the TLGS. Knowledge, Attitude, Practice Assessment
Knowledge, attitude and practice of subjects, regarding nutri-
tion, were assessed by questionnaires and face-to-face interviews

234 Ann Nutr Metab 2010;56:233–240 Mirmiran et al.

with regard to NCD risk factors and body weight, dietary lipids, Table 1. Structure and scores of knowledge, attitude and practice
sugar, fiber, fruits and vegetables and salt (table  1). Knowledge questionnaire
questions included yes/no and correct/false questions (46 ques-
tions, each scoring 1 point). Attitude was assessed based on the Questionnaire category Number of questions
Likert criteria from 1 point, as the weakest, to 5 points, as a desir-
able score (9 questions). For assessing practice, the food frequency knowledge attitude practice
questionnaire including dairy and meat products (low and high
fat), vegetables and fruits was used (35 questions, each scoring 1 Body weight 16 2 9
point). Dietary lipids 16 2 10
Focus groups and experts authenticated face and content va- Sugar 6 2 7
lidity of the questionnaires and their reliability was determined Fiber, fruits and vegetables 6 2 7
by pre- and post-tests. Following finalization of the question- Salt 2 1 2
naire, 4 interviewers were educated about completion of the ques- Total 46 9 35
tionnaires in a sample of 42 individuals as pre- and post-tests, Score range 0–46 9–45 0–35
with a 1-month interval, following which the association between
the scores of the 2 tests was determined (r = 0.75, p ! 0.001).
After calculating the scores of each of the 3 sections (knowl-
edge, attitude and practice), the quartiles were determined and
first quartile was coded as weak, the second and third quartiles
were coded as medium and the last quartile was coded as desir- Table 2. Characteristics of the study population
able. Minimum scores of knowledge, attitude and practice were
considered to be 0, 9 and 0, respectively and maximum scores Variable Men Women Total
were considered to be 46, 45 and 35, respectively. Respective (n = 369) (n = 457) (n = 826)
thresholds and ranges for weak, medium and desirable scores
were set as follows: knowledge ! 29, 29–39 and 139; attitude !30, Age group, years
30–35 and 135, practice ! 20, 20–26 and 126. 20–29 99 (26.8) 144 (31.5) 243 (29.4)
30–39 74 (20.1) 122 (26.7) 196 (23.7)
Statistical Analysis 40–49 86 (23.3) 113 (24.7) 199 (24.1)
Statistical analysis was performed using the Statistical Pack- 50–59 73 (19.8) 57 (12.5) 130 (15.7)
age for Social Sciences (SPSS) version 10.0, and data were present- ≥60 37 (10.0) 21 (4.6) 58 (7.0)
ed as mean 8 SD. To determine the association between nutri- Educational level
tional knowledge, attitude, practice, sex, age and educational lev- Illiterate or poorly literate 90 (24.4) 118 (25.8) 208 (25.2)
els of subjects, the ␹2 test was used after age and sex adjustment; Middle school 66 (17.8) 92 (20.1) 158 (19.1)
for associations between knowledge, attitude and practice and High school diploma 121 (32.9) 178 (38.9) 299 (36.2)
BMI, waist to hip ratio and levels of lipid and lipoproteins, the University graduate 92 (24.9) 69 (15.2) 161 (19.5)
Spearman correlation was used instead of Pearson, as the scores Marital statusa
lacked normal distribution. For detecting significant differences Single 82 (22.2) 100 (21.9) 182 (22.0)
between the key variables in age, sex, educational levels and BMI Married 286 (77.5) 333 (72.9) 619 (74.9)
groups and the normal, borderline and high concentrations of Divorced/widowed 1 (0.3) 24 (5.2) 25 (3.1)
total cholesterol, triglycerides, HDL- and LDL-cholesterol, the
Kruskal-Wallis test was used. Linear regression was performed Numbers in parentheses are percentages.
for testing different potential correlations. R 2 was 1 0.04 for dif-
ferent models with total cholesterol, BMI, triglycerides and LDL-
cholesterol as dependent variables and total practice score as in-
dependent variable. P values !0.05 were considered to be signifi-
Knowledge, Attitude and Practice Scores
In figure 1, the frequency of persons based on the score
Results of knowledge, attitude and practices is shown. Most had
medium scores of knowledge, attitude and practice, with
General Characteristics desirable scores of knowledge, attitude and practice being
Subjects of this study were 369 men and 457 women observed in less than 30% of subjects. Table  3 shows
with mean ages of 40.8 8 13.7 and 37.4 8 12.2 years, re- scores of various age, sex, educational level, marital status
spectively. Table 2 shows the general characteristics of the and weight groups. Younger people (30–50 years old) had
subjects studied. More than three fourths of the subjects significantly higher knowledge scores than those 150
were 20–49 years of age; almost half of the population years of age (p ! 0.01). Regarding attitudes, there was no
were high school and university graduates and three significant difference between age groups. Scores for
fourths were married. practice increased with increasing age (p ! 0.001), and

Nutritional Knowledge, Attitude and Ann Nutr Metab 2010;56:233–240 235

Behavior of Tehranian Adults
Table 3. Comparison of the scores of knowledge, attitude and
60 Desirable Medium Weak practice between various groups
50 * Variable Knowledge Attitude Practice

40 Age groups, years

20–29 33.786.4 32.284.1 20.983.9b

30–39 34.986.5a 32.484.4 22.283.9b
40–49 34.785.9a 31.984.0 23.483.9b
50–59 32.985.5 32.283.5 24.783.6b
≥60 32.382.5 31.583.9 26.383.3b
Educational level
10 Illiterate or poorly literate 30.385.4b 31.383.6 24.084.0
Middle school 32.985.9b 32.284.1 22.484.0
0 High school diploma 35.685.8b 32.584.3 22.384.3
Men Women Men Women Men Women University graduate 36.885.4b 32.584.2 22.284.0
Knowledge Attitude Practice Marital status
Single 33.786.4 32.284.1 20.483.7b
Married 32.286.1 32.184.1 23.484.0
Fig. 1. Relative frequency of people with desirable, medium and Sex
poor nutritional knowledge, attitude and practice. *  p ! 0.001, Male 33.586.1c 31.683.8b 21.684.2
men vs. women. Female 34.586.2 32.684.2 22.184.1
BMI status
Normal 34.486.4 31.584.5c 21.884.4
Overweight 34.985.8 32.984.0 23.983.5
with increasing educational levels knowledge of subjects Obese 33.886.1 32.383.7 23.683.7
increased significantly (p ! 0.001). a
p < 0.01 compared to 50–59 and ≥60 years age group.
In the middle school group, more than half of the peo- b
p < 0.001 compared to other groups.
ple had good knowledge and medium practice scores. In c
p < 0.05 compared to other groups.
the university graduate group, all individuals had good
scores of knowledge (data not shown in table 3). Females,
compared to males, had higher scores in their knowledge
and attitudes. Married individuals had higher scores in
practices than did singles (p ! 0.001). Overweight and is not an important source of cholesterol; 40.1% of these
obese people had higher practice scores than those with reported animal fat to be more delicious than vegetable
normal weight (23.9 8 3.5 and 23.6 8 3.7 vs. 21.8 8 4.4, oil, and most (76%) knew that non-hydrogenated oils are
respectively; p ! 0.01). Attitudes of overweight people better than hydrogenated ones. However, the consump-
were better than their normal weight peers. Knowledge tion of hydrogenated oils, was much higher than vegeta-
scores of normal weight, overweight and obese groups ble oils (75 vs. 25%). Only about 30% of subjects had a
were not significantly different. general idea of the meaning of ‘fiber’, whereas more than
There were significant positive, but weak associations 90% of them were conscious of the necessity of vegetable
between knowledge and attitudes (r = 0.26, p ! 0.001), and fruit consumption for healthy living. Fruits were pre-
knowledge and practice (r = 0.18, p ! 0.001) and between ferred as snacks by 87% of people and only 6% of the sub-
attitude and practice (r = 0.30, p ! 0.001) of the partici- jects preferred confectionery products as snacks.
pants. Women thought themselves to be overweight or obese
and had tried out low calorie diets more than men (67 vs.
Components of Knowledge, Attitude and Practice 39%; p ! 0.001); over 90% of them were aware of their
Fields weight and weighed themselves regularly. Overall regular
Table 4 demonstrates the results of responses to key physical activity (95.8%), lower fat intake (94.3%), lower
questions of the study in both genders. Of the studied sugar intake (88.4%), low calorie diets (86.2%), lower con-
population, 64.8% claimed to know the meaning of ‘cho- sumption of fried foods (81.6%) and higher consumption
lesterol’, but 43.6% of them could not identify its food of fruits and vegetables were seen to be the best ways to
sources correctly and believed that food items such as lose and maintain an appropriate weight, respectively.
bread, cereals and nuts have cholesterol or that egg yolk Most of the people reported inactivity as the most impor-

236 Ann Nutr Metab 2010;56:233–240 Mirmiran et al.

Table 4. Correct answers to key components of knowledge, attitude and practice in study group

Questions Correct answers, n (%)

men women

Do you know the meaning of ‘cholesterol’? 231 (62.6) 304 (66.5)
Egg yolk is an important source of cholesterol 233 (63.1) 303 (66.3)
Fish is a source of cholesterol 87 (23.6) 95 (20.8)
Bread, cereals and pasta are not sources of cholesterol 95 (25.7) 109 (23.9)
Nuts and seeds are not sources of cholesterol 57 (15.4) 65 (14.2)
Non-hydrogenated oils are better for health than hydrogenated ones 292 (79.1) 386 (84.5)
Fiber is a substance in vegetables that cannot be digested and absorbed 120 (32.5) 158 (34.6)
For healthy nutrition, consumption of fruits and vegetables should be increased 348 (97.0) 444 (97.1)
The best ways to lose and maintain an appropriate weight
Lower fat intake 189 (93.1) 234 (93.2)
Lower sugar intake 170 (83.7) 225 (89.6)
Low calorie diets 170 (83.7) 223 (88.8)
Lower consumption of fried foods 157 (77.3) 217 (86.5)
Higher consumption of fruits and vegetables 44 (22.7) 46 (18.3)
Inactivity is the most important factor contributing overweight 195 (96.1) 231 (92.0)
People with normal weights run less risk of coronary heart disease 185 (91.1) 233 (92.8)
Overweight persons probably have higher blood pressure than do others 182 (89.7) 200 (79.7)

Answered ‘yes’, n (%)

men women

Animal fat is not more delicious than vegetable oil 124 (17.9) 66 (27.1)
I prefer fruits as snacks 312 (84.6) 415 (90.8)
I prefer sweets as snacks 28 (7.6) 23 (5.0)
I think myself to be overweight or obese and try out low calorie diets 144 (39.0) 306 (67.0)
I always use salt on my foods 155 (42.0) 284 (62.1)
I always use vegetable oil for cooking 109 (23.9) 108 (29.3)
I often weigh myself regularly 187 (92.1) 232 (92.4)

tant factor contributing to overweight and were aware ing the influence of fat consumption on the incidence of
of the relation between overweight, hypertension and non-communicable disease and its risk factors (6.0 8 1.5
CVDs. vs. 6.8 8 1.7; p ! 0.001) and their practice in adding salt
Regarding fat and risk factors of CVD, the practices of to foods (1.1 8 0.8 vs. 1.5 8 0.7; p ! 0.001) were all better
illiterate and poorly literate groups were better than mid- than men.
dle school or diploma graduate groups (p ! 0.001). The
practices of university graduates in the field of fat and Association of Lipids with Knowledge, Attitude,
confectionary consumption scored higher than diploma Practice
graduates (p ! 0.05) as the university graduates reported Subjects with high serum total cholesterol and LDL-
high consumption of low fat dairy products and low con- cholesterol concentrations had better practices than oth-
sumption of sweets (data not shown). ers (p ! 0.001). Also, subjects who had borderline serum
In addition to significant differences in knowledge triglyceride levels had more desirable practices than those
and attitudes between men and women, knowledge of with normal triglyceride levels (mean score of practice
women regarding disadvantages of sweet consumption 24.3 8 3.7 vs. 22.6 8 4.2; p 1 0.01; fig. 2); however, no
(4.6 8 1.2 vs. 4.8 8 1.0; p ! 0.05), their attitude regard- significant difference between serum lipids and nutri-

Nutritional Knowledge, Attitude and Ann Nutr Metab 2010;56:233–240 237

Behavior of Tehranian Adults
tice [14–16]. Some who can benefit from making dietary
24.5 Normal Borderline High changes may not perceive a need to do so [14, 17], or some
* * individuals may know what changes to make but not
know how to make them [18, 19]. Broader social and cul-
23.5 tural factors also play an important role in shaping food
choices. Another explanation for the inconsistent asso-
Practice score


ciations between knowledge and dietary behavior is that
* knowledge may be poorly assessed. Psychometrics, has
defined a set of criteria such as face and content validity,
21.5 internal reliability and test-retest reliability for a valid test
21.0 [20]. For example, whereas more than 60% of the popula-
tion studied claimed to know the meaning of ‘cholester-
Total cholesterol LDL-cholesterol Triglyceride ol’, less than 50% of them were able to identify its food
sources correctly (marked weakness of applied knowl-
edge). Also, whereas more than 80% of them knew that
Fig. 2. Comparison of practice scores of subjects with normal, vegetable oils are better than hydrogenated ones for
borderline and high concentrations of total cholesterol, LDL-cho- health, only one third of the people used the former for
lesterol and triglycerides. * p ! 0.001, vs. other groups. cooking. These findings confirmed the Pirouznia opin-
ions that nutrition knowledge is not the only factor that
can influence eating behavior [21]. Lower consumption
of vegetable oil in the subjects studied can be due to the
tional knowledge and attitude was found. LDL- and total higher price of this oil as compared to hydrogenated ones.
cholesterol were correlated to practice scores (r = 0.23, Other factors and variables include physiological needs,
p  ! 0.001). BMI and triglyceride levels were also corre- body image, peer pressure, media social norms, and
lated to practice scores (r = 0.27 and r = 0.14, p ! 0.01). availability of foods, personal experiences and food pref-
Literature available on eating behavior and its relation
Discussion to nutrition knowledge is often contradictory [22, 23].
Some researchers have shown that nutrition knowledge
In recent years, with increasing urbanization and in- was highly and positively related to behaviors toward nu-
dustrialization, and with better access for populations to trition, while others found little correlation between nu-
public health care, the prevalence of communicable dis- trition knowledge and actual choices of healthy food. In
eases has reduced. In contrast, due to reduced physical the Saegert and Young [24] study, a strong positive rela-
activity and changes in dietary patterns, more fat con- tion between nutritional knowledge and practices of peo-
sumption in particular, non-communicable diseases ple regarding dietary guidelines was found. Wardle et al.
have increased [12, 13]. This phenomenon, referred to as [25] also found that general practitioners with higher
‘nutritional transition’, has emphasized the vital need for knowledge scores meet the fruit and vegetable recom-
new strategies to improve the health and nutritional sta- mendations more than their peers with lower knowledge.
tus of the population. To reach this goal, it is imperative However, Story showed a weak association between nu-
to promptly identify the factors that influence dietary tritional knowledge and food choices [26]. Other studies
practice and health, and educate the population regard- also reported that despite most people being aware of the
ing these risk factors. This study showed that, regarding relation between diet and health, their nutritional prac-
NCD-nutritional risk factors, more than half of adults tices are not desirable; for example, their fat and sugar
had moderate knowledge, attitude and practices, whereas consumption is high [27, 28].
desirable knowledge, attitude and practice were only ob- In this study, knowledge and attitude scores of women
served in a fourth of them. in some fields such as influence of consumption of fats,
A weak but significant association was found between sweets and salt on non-communicable diseases and their
knowledge, attitude and practice, a finding that indicated risk factors were significantly higher than men; however,
the sizeable gap that exists between what individuals there were no significant differences in their practices.
know and how much they put their knowledge into prac- As a result, in spite of higher knowledge and attitude of

238 Ann Nutr Metab 2010;56:233–240 Mirmiran et al.

women, nutritional practices in the family followed men’s ‘stages of change’ theory [34], those individuals with a
preferences. Thus, there is a need to increase funding for health condition or disease may progress through the
the education of men. Population-based food and nutri- stages to the action stage, where nutrition knowledge, be-
tion studies consistently demonstrate that women are sig- liefs and dietary practices would be more appropriate in
nificantly more likely to comply with dietary guidelines order to reduce risks. Overweight and obese people also
than men. At least 4 possible explanations for this finding had better reported practices than their normal weight
are suggested in the literature: (1) women are more knowl- peers [35].
edgeable regarding food, nutrition, health and related as- The findings of the present paper revealed that age,
sociations; (2) women exhibit more positive health and educational level, gender and marital status are factors
food-related beliefs and attitudes; (3) women manifest a that can influence knowledge, attitude and practice, re-
more heightened concern about their personal and phys- garding nutrition. However, our previous data showed
ical appearance than do men, and (4) women are more that dietary behavior may not accord with the nutrition-
likely than men to report that they prefer the taste of al knowledge of individuals [36]. Health education strate-
healthy foods and meals [29]. gies to encourage individuals to choose healthy diets
In the present study, educational level, age and marital must be developed and enhanced to achieve health pro-
status were considered as factors that can influence motion in Tehran. Programs designed to provide nu-
knowledge, attitude and practice. More highly educated trition education can benefit from the present data on
people scored higher in knowledge than did those with lifestyle and nutrition behavior. The ability to look at
lower levels of education. Practices of illiterate or poorly interrelationships among all these measures could
literate subjects regarding fat intake and risk factors of significantly improve our ability to appropriately design
non-communicable diseases scored significantly higher and target interventions, such as nutrition education
than others. Since measurement of family income levels programs [37].
is complicated, in some studies the educational level has
been used as an indicator of family income and socioeco-
nomic status [30]. It is concluded that more highly edu- Acknowledgments
cated people with higher socio-economic status, in spite
This research project was supported by the combined support
of their higher levels of knowledge, consumed more fast
of the National Research Council of the Islamic Republic of Iran
and fatty foods than others. and Research Institute for Endocrine Sciences (grant No. 121).
Nutritional knowledge of the 30–50 year group was We acknowledge and appreciate the efforts of the personnel of
better than older age groups, but with increasing age, the Glucose and Lipid Unit of the Research Institute for Endo-
practice scores increased. Higher knowledge scores of crine Sciences and the participants of this study.
younger groups are probably due to higher educational
levels, with a positive significant correlation being seen
between educational levels and knowledge [15, 17, 18].
However, better practice scores in older adults could be References 1 Nissinen A, Berrios X, Puska P: Community-
based noncommunicable disease interven-
due to more motivation in meeting the nutritional rec- tions: lessons from developed countries for
ommendations of their physician or other health care developing ones. Bull World Health Organ
personnel, since 30–50-year-old subjects usually have 2001;79:963–970.
2 Murray GYL, Lopez AD: Mortality by cause
one or more NCDs or at least one of their risk factors. for eight regions of the world: Global Burden
Another finding of the current study was the better of Disease Study. Lancet 1997; 349: 1269–
practices of persons with abnormal total cholesterol lev- 1276.
3 Fuentes R: Hypertension in developing eco-
els, LDL-cholesterol and triglycerides compared with nomics: a review of population-based studies
those who had normal or borderline levels; this is similar carried out from 1980 to 1998. J Hyperten-
to the Dian et al. [31] study, which found that those with sion 2000;18:521–529.
4 WHO Multi-Country Study: The Nutrition
heart disease and hypercholesterolemia had better knowl- Transition Program. 2003.
edge and better dietary practices about fat than those (accessed February 25,
without this condition. Compared with the total popula- 2010).
5 Azizi F, Rahmani M, Emami H, Madjid M:
tion, these people benefit from lifestyle changes in the Tehran Lipid and Glucose Study: rationale
best way and it appears that they are highly motivated to and design. CVD Prevention 2000; 3: 242–
comply with changes in lifestyle [32, 33]. Based on the 247.

Nutritional Knowledge, Attitude and Ann Nutr Metab 2010;56:233–240 239

Behavior of Tehranian Adults
6 Kearney JM, Gibney MJ, Livingstone BE, 15 Glanz K, Brug J, Van Assema P: Are aware- 27 Birgit R, Peter N, Jurgen B, et al: Eating hab-
Robson PJ, Kiely M, Tlarrington K: Attitudes ness of dietary fat intake and actual fat con- its, health status, and concern about health:
towards and beliefs about nutrition and sumption associated? A Dutch American a study among 1,641 employees in the Ger-
health among a random sample of adults in comparison. Eur J Clin Nutr 1997; 51: 542– man metal industry. Prev Med 2000, 30;
the Republic of Ireland and Northern Ire- 547. 295–301.
land. Public Health Nutr 2001;4:1117–1126. 16 Johansson L, Anderson LF: Who eats 5 a day? 28 Luthy J: The development of a nutrition
7 Azizi F, Ghanbarian A, Momenan AA, Ha- Intake of fruits and vegetables among Nor- policy in Switzerland. Nutr Rev 1997; 55
daegh F, Mirmiran P, Hedayati M, et al: Pre- wegians in relation to gender and lifestyle. J (suppl 1):S29–S31.
vention of non-communicable disease in a Am Diet Assoc 1998;98:689–691. 29 Turrell G: Determinants of gender differenc-
population in nutrition transition: Tehran 17 DeGraaf C, Van der Gaag M, Kafatos A, es in dietary behavior. Nutr Res 1997; 17:
Lipid and Glucose Study phase II. Trials Lennernas M, Kearney J: Stages of dietary 1105–1120.
2009;10:1–15. change among nationally representative 30 Kaplan GA, Keil JE: Socioeconomic factors
8 NIH/NHLBI (National Institute of Health/ samples of adults in the European Union. and cardiovascular disease: a review of the
National Heart, Lung and Blood Institute): Eur J Clin Nutr 1997;51(suppl 2):535–340. literature. Circulation 1993;88:1973–1998.
Clinical guidelines on the identification, 18 Buttriss JL: Food and nutrition: attitudes, be- 31 Dian K, Sonja L, Gary S, et al: Knowledge of
evaluation and treatment of overweight and liefs and knowledge in the United Kingdom: and attitudes toward coronary heart disease
obesity in adults: the evidence report. Obes Am J Clin Nutr 1997; 65(suppl 16):19855– and nutrition in Oregon families. Prev Med
Res 6(suppl 2):1998;51S. 19955. 1984;13:390–395.
9 Friedwald WT, Levy RI, Fredrickson DS: Es- 19 Harnack L, Block G, Subar A, Lane S: Cancer 32 Adhikari P, Dhungel S, Shrestha R, Khanal
timating of low-density lipoprotein choles- prevention related nutrition knowledge, be- S: Knowledge attitude and practice (KAP)
terol in plasma, without use of preparative liefs and attitudes of US adults: 1992 NHIS study regarding facts for life. Nepal Med Coll
ultracentrifuge. Clin Chem 1972; 18: 499– cancer epidemiology supplement. J Nutr J 2006;8:93–96.
502. Educ 1998;30:131–138. 33 Pon LW, Noor-Aini MY, Ong FB, Adeeb N,
10 Mirmiran P, Ramezankhani A, Hekmat- 20 Kline P: The Handbook of Psychological Seri SS, Shamsuddin K, Mohamed AL, Hap-
doost A, Azizi F: Effect of nutrition interven- Testing. London, Roultedge, 1993. izah N, Mokhtar A, Wan HW: Diet, nutri-
tion on non-communicable disease risk fac- 21 Pirouznia M: The correlation between nutri- tional knowledge and health status of urban
tors among Tehranian adults: Tehran Lipid tion knowledge and eating behavior in an middle-aged Malaysian women. Asia Pac J
and Glucose Study. Ann Nutr Metab 2008; American school: the role of ethnicity. Nutr Clin Nutr 2006;5:388–399.
52:91–95. Health 2000;14:89–107. 34 Prochaska JO, DiClemente CC: Transtheo-
11 He J, Bazzano LA: Effects of lifestyle modifi- 22 Cotunga N, Subar AF, Heimendinger J, Kah- retical therapy: towards a more integrative
cation on treatment and prevention of hy- le L: Nutrition and cancer prevention knowl- model of change. Psychother Ther Pract
pertension. Curr Opin Nephrol Hypertens edge, beliefs attitudes and practices: the 1987 1982;19:276–288.
2000;9:267–271. National Health Interview Survey. J Am Diet 35 Carlton DJ, Kicklighter JR, Jonnalagadda SS,
12 Kimiagar SM, Ghaffarpour M, Houshiar- Assoc 1992;92:963–968. Shoffner MB: Design, development, and for-
Rad A, Hormozdyari H, Zellipour L: Food 23 Dallongeville J, Marecaux N, Cottle D, et al: mative evaluation of ‘put nutrition into prac-
consumption pattern in the Islamic Republic Association between nutrition knowledge tice’, a multimedia nutrition education pro-
of Iran and its relation to coronary heart dis- and nutritional intake in middle-age men gram for adults. J Am Diet Assoc 2000; 100:
ease. East Mediatr Health J 1998;4:539–547. from Northern France. Public Health Nutr 555–563.
13 Ghassemi H, Harrison G, Mohammad K: An 2001;4:27–33. 36 Mirmiran P, Azadbakht L, Azizi F: Dietary
accelerated nutrition transition in Iran. Pub- 24 Saegert J, Young EA: Nutrition Knowledge behaviour of Tehranian adolescents does not
lic Health Nutr 2002;5:149–155. and health food consumption. Nutr Behav accord with their nutritional knowledge.
14 Poplein BM, Siega-Riz AM, Haines PS: A 1983;1:103–113. Public Health Nutr 2007;10:897–901.
comparison of dietary trends between racial 25 Wardle J, Parmenter K, Waller J: Nutrition 37 Murphy SP: Collection and analysis of intake
and socioeconomic groups in the United knowledge and food intake. Appetite 2000; data from the integrated survey. J Nutr 2003;
States. N Engl J Med 1996;335:216–220. 34:269–275. 133:585S–589S.
26 Story M: Adolescents’ view of food and nutri-
tion. J Nutr Educ 1989;18:188–192.

240 Ann Nutr Metab 2010;56:233–240 Mirmiran et al.