Estimating the prevalence of risky behaviors by using

network scale-up method in Larestan City

Bashir Hashemi

and Abbas Yazdanpanah

and Parviz Aghaei

Department of Healthcare Management, Marvdasht Branch, Islamic Azad University, Marvdasht, Iran

Assistant professor, Department of Healthcare Management, Marvdasht Branch, Islamic Azad University,

Marvdasht, Iran

Assistant Professor, Departmentof Medical Education Management, Cellular and Molecular Research Center,

Yasuj University of Medical Sciences, Yasuj, Iran

ABSTRACT

Risky behavior is a series of behaviors that not only seriously damages the person engaged in this behavior and

important people in his life, but also causes the unintentional harm to other innocent people. This study aimed to esti-

mate the prevalence of risky behaviors by using Network Scale-Up method in Larestan in 2016. This cross-sectional

study was conducted on 800 people aged 18 to 30 in Larestan in 2014. It was used the data checklist made by the

researcher to collect data. In order to assess the demographic variables in the subjects, SPSS software (version 16)

was used. Also, to estimate the prevalence of risky behaviors, Stata version 11 was used.Chi- Square test was used

to compare the prevalence of risky behaviors in men and women. In all analyzes, the signi cance level was 0.05.

The  ndings of the study showed that the most common risky behavior in the age group of 18-30 years in Larestan

in both males and females is the tobacco products consumption in the amount of 16.82 percent and 11.06 percent,

respectively. In contrast, tattoo risk behavior on both males and females respectively in the amount of 1.18 percent

and 0.46 percent has the lowest rate. Despite the low prevalence of risky behaviors in Larestan, paying more atten-

tion to sexually active individuals can play an important role in reducing this phenomenon in the society. Therefore,

developing serious planning by the agencies that are responsible for the health, especially the health centers is essen-

tial to reduce these behaviors in the community.

KEY WORDS: LARESTAN, RISKY BEHAVIOR, NETWORK SCALE-UP METHOD

133

ARTICLE INFORMATION:

*Corresponding Author: abbas_yaz@miau.ac.ir

Received 27

Nov, 2016

Accepted after revision 18

March, 2017

BBRC Print ISSN: 0974-6455

Online ISSN: 2321-4007

Thomson Reuters ISI ESC and Crossref Indexed Journal

NAAS Journal Score 2017: 4.31 Cosmos IF : 4.006

reserved.

Online Contents Available at: http//www.bbrc.in/

Biosci. Biotech. Res. Comm. Special Issue No 1:133-138 (2017)

134 ESTIMATING THE PREVALENCE OF RISKY BEHAVIORS BY USING NETWORK SCALE-UP METHOD BIOSCIENCE BIOTECHNOLOGY RESEARCH COMMUNICATIONS

Hashemi and Yazdanpanah and Aghaei

INTRODUCTION

Risky behavior is a series of behaviors that not only

seriously hurts the person engaged in this behavior and

important people in his life, but also causes the unin-

tentional harm to other innocent people. Nowadays, the

prevalence of risky behaviors among young people is

one of the main concerns of the human societies and

despite the measures taken in the past three decades, the

risky behaviors have the exponential growth throughout

the world (Headquarters, 2015).

According to the report of UN Of ce on Drugs and

Crime (UNODC) in 2009, about 147 to 272 million peo-

ple in 15-64 year age group have used drugs at least

once during the past 12 month.The most common health

risky behaviors include the excessive consumption of

alcohol, drug abuse, unsafe sex, reckless driving, dan-

gerous sports, gambling and illegal acts (TW, 2006).

Researches has shown that most of these behaviors

occur on the campus (MT, 2005) and the risky behaviors

such as excessive alcohol consumption, illicit drug use

and unsafe sexual behavior can lead to high levels of

morbidity and mortality among them (Wilson , 1995).

Health risky behaviors usually begin from the child-

hood and adolescence, are established at a young age

and will continue into adulthood(Yach D, 2004). Due to

the fact that half of our population is below the age of

25 years, exploring these behaviors is important to plan

for the future. Also, since the group of society is very

vulnerable, the awareness of risky behaviors among

them is very important (Momen Nasab, 2006).Risky

behaviors have causes that must be understood and we

shall strive to eliminate the favorable social contexts for

the people’s suffering. Unfortunately, in our country, the

exact statistics on the prevalence of these behaviors in

different age groups is not available (Akbar, 2013).

Due to the sensitive issue of risky behaviors among

sexually active population and due to legal and reli-

gious restrictions, there is not much information about

this sensitive issue. Therefore, there are very few reliable

local studies in the  eld that can be cited as a prec-

edent and because of the limitation of direct methods to

estimate the risky behaviors, the need to be done simi-

lar studies as soon as possible using indirect methods

will be determined to estimate such a critical behaviors

in the country (Hamdye M, 2008).Therefore, this study

aimed to estimate the prevalence of risky behaviors by

using Network Scale-Up method in Larestan in 2016.

MATERIALS AND METHODS

This is a cross-sectional study. The study population

consisted of sexually active people who were exposed

to health risky behaviors in 2016 in the city of Lar. The

instrument used is the self-made checklist. The check

list contains 2 sections. The  rst part of demographic

information includes gender, age, occupation, place of

education, place of residence and marital status, and the

second part of checklist is the questions related to the

risky behaviors (smoking, hookah, pipe, opium, trama-

dol, alcohol, ecstasy and unknown drugs, migration

history, self-mutilation, drug injection as well as the

prevalence of sexual behaviors outside of marriage). The

self-made checklists were completed by the clients after

providing the necessary explanations by the interviewer,

and in the case of illiterate persons, they answered the

questions with the help of a questioner.

Since the prevalence of risky behaviors in Network

Scale-Up method is indirectly estimated, unlike other

studies, there is no formula for determining the sam-

ple size. Thus, by classifying people across age groups,

a sample with an appropriate volume is considered to

make the accuracy of estimates desirable. According to

the number of people who are sexually active in 2016

(28192 women and 30390 men), 800 patients were

studied.

After extracting and summarizing, the data was

encoded and then entered into SPSS version 16 and

Microsoft excel and  nally was analyzed. The results

were reported as “number (percent)” for the qualita-

tive variables. Chi- Square test was used to compare the

prevalence of risky behaviors in men and women.

Moreover, Stata 11 software was used to estimate the

prevalence of risky behaviors in Wald method. The sig-

ni cance level was 0.05 in tests.

RESULTS

In total, 400 men and 400 women were studied. In table

1, it is shown the frequency distribution of demographic

variables in 18-30-year-old people who completed the

questionnaire in Larestan in 2016. As the results show

the people over 25 years old have the highest frequency

(45.42 percent). Also, most of the subjects were married

and high school graduates and were living in the city

and (Table 1).

Table 2 shows that the total of social networks in men

and women is respectively 6515 people with an average

16.29 and 5803 people with an average of 14.50 people.

In this study, the social network size (C) was found 296

people. The  ndings of the study showed that the most

common risky behaviors among the age group of 18- 30

years in both males and females in Larestan is the use of

tobacco products with a frequency of 16.82 percent and

11.06 percent, respectively. In contrast, the tattoo risk

behavior on both males and females respectively with a

frequency of 1.48 percent and 0.46 percent has the low-

est rate (Table 3).

BIOSCIENCE BIOTECHNOLOGY RESEARCH COMMUNICATIONS ESTIMATING THE PREVALENCE OF RISKY BEHAVIORS BY USING NETWORK SCALE-UP METHOD 135

Hashemi and Yazdanpanah and Aghaei

Table 1. The frequency distribution of

demographic variables in 18-30-year-old

people who completed the questionnaire in

Larestan in 2016.

Variable Number Percent

Age groups

8 2

11.05

Under 20 years old

20-25 years old 323 43.53

25-30 years old 337 45.42

Marital status

313 42.64

Single

Married 401 54.63

Widow 20 27.0

Address

598 78.58

City

Village 163 21.42

Education

205 27.82

Under diploma

High school Diploma 340 46.13

Collegiate 192 26.05

Table 3 shows the frequency of risky behaviors dif-

ferentiated by gender. The results of frequency table

estimated show the prevalence of tobacco consumption

among young people aged 18-30 years in Larestan. As

can be seen, the prevalence of tobacco consumption

among men is more than women. This difference was

statistically signi cant (P<0.001).

Also, the prevalence of drugs in men was signi -

cantly higher than women (P<0.001). The results of table

3 show the increased prevalence of psychotropic drug

use in men than in women that this difference was sta-

tistically signi cant (P=0.003). Also, in the prevalence

of dangerous actions and the estimated frequency, there

statistically was a signi cant difference between men

and women (P<0.001).

Further, the results in Table 3 show that the preva-

lence of alcohol consumption is higher in men than in

women, and this difference was statistically signi cant

in alcohol consumption (P<0.001). Moreover, the preva-

lence of unprotected sex outside the family was different

in men and women, so that the prevalence was reported

in men more than in women (P<0.001). In addition, the

prevalence of migration, particularly migration abroad

was higher among men than women (P=0.163). Finally,

the prevalence of tattooing was signi cantly higher

among men (18-30 years) than women in Larestan

(P<0.001) (Table 3).

DISCUSSION

The  ndings of the study showed that the most com-

mon risky behaviors among the age group of 18- 30

years in both males and females in Larestan is the use

of tobacco products with a frequency of 16.82 percent

and 11.06 percent, respectively. In contrast, the tattoo

risk behavior on both males and females respectively

with a frequency of 1.48 percent and 0.46 percent has

the lowest rate.

The results showed that the prevalence of tobacco

products (Cigarette and hookah) has the highest fre-

quency among the other high-risk behaviors (14.05 per-

cent). According to the results of this study, it was found

that the prevalence of (cigarette) smoking is more than

hookah smoking among the subjects. In addition, the

use of tobacco products (Cigarette and hookah) in all

subjects was more common in boys than girls. Atai et

al reported the prevalence of smoking is 31.3 in Isfa-

han (Ataei B, 2011).While Ismail Zadeh in his study

showed that the hookah smoking (59.2) is more prev-

alent compared with the experience of smoking (32.7)

among students (Ismail-Zadeh, 1393).Also, in the study

of Taremian et al conducted on 2997 students in six

Tehran universities in the academic year 2005-2006, the

prevalence of smoking and hookah was estimated 15.7

and 22.1 respectively, in the past year. Contrary to the

results of this study, the prevalence of hookah smoking

was higher that this could be due to changing patterns

of tobacco use among young people and the dif culty

of hookah smoking in the student dormitories as well as

ease of smoking.

The results of this study showed that the prevalence

of drug abuse in the 18-30 year-old young people is

2.5% in Larestan. While Ismail Zadeh reported in his

study that the illicit drug abuse is 7.3%(Ismail-Zadeh,

1393). The results of another study in Kerman and Raf-

sanjan showed that the prevalence of injecting drug

use has a frequency of 61.5 percent(Torkashvand F,

2015).

Table 2. Sample size and social network total of friends of respondents

Variable Sample size Social network total of

respondents’ friends

Social network total of

respondents’ friends

Male 400 6515 16.29

Female 400 5803 14.50

136 ESTIMATING THE PREVALENCE OF RISKY BEHAVIORS BY USING NETWORK SCALE-UP METHOD BIOSCIENCE BIOTECHNOLOGY RESEARCH COMMUNICATIONS

Hashemi and Yazdanpanah and Aghaei

Another study in Isfahan showed that the prevalence

of drug use among prisoners is 30.1%(Ataei 2011). Gar-

maroudi et al hjave also reported 2.7 percent of drug

use in the past month(Garmarodi, 2009). In the study

of Shokoohi et al, the estimate derived from the indi-

rect method showed that 13.1 percent of people has the

experience of drug use(B. M. Shokoohi M, Haghdoost AA

2012). The Sheikh Zadeh’s study  ndings showed that

Intravenous drug use has the lowest frequency among

students(Sheikhzadeh 2016). In explaining this relation-

ship, it can be concluded that men are more exposed

to the social and economic problems and work-related

stress compared to women, which may put them at risk

for drug use.

Also, the study results showed 11.61 percent preva-

lence of alcohol and alcoholic beverages in the region.

Ismail Zadeh reported 16.7 percent of drinking alcohol

in his study(Ismail-Zadeh, 1393).The results of a study

reported an experience 17.8% in Isfahan(Ataei B, 2011).

Garmaroudi et al also reported the alcohol consumption

to 7.4 percent in the past month(Garmarodi GhR, 2009).

In the study of Shokoohi et al, the estimate derived

from the probabilistic method was higher than the fre-

quency in the direct method. According to the proba-

bilistic method, 13.7% of men had used alcohol at least

once a year( Shokoohi et al., 2012). In this study, similar

to previous studies, the prevalence of risk behaviors is

higher among boys than girls that this can be justi ed

according to boys’ more freedom in the family and soci-

ety, more courage and ease of access to drugs and alco-

hol, psychological stress caused by unemployment on

the boys and on the other hand, family’s more precise

control over the behavior of girls.

In addition, the study results showed that the preva-

lence of dangerous acts is 5.77 percent. The study of

Torkashvand et al showed that men compared with

women signi cantly had the behaviors such as self-

mutilation (p = 0/001), after the diagnosis and aware-

ness of their disease. The results of a study in Tehran

showed that most patients with self-mutilation suffer

from the borderline personality disorder and antiso-

cial personality disorder. More patients had attempted

self-mutilation in their upper limbs. Furthermore, most

patients with self-mutilation had the previous self-muti-

lation symptoms, and the average number of previous

self-mutilations was approximately 9 times and patients

with a history of 2 self-mutilations had the highest fre-

quency. In explaining the risky behavior prevalence, it

can be mentioned the people’s easy access to sharp tools

to harm themselves and attract others.

Also, the Immigration prevalence as another risky

behavior was estimated 5.75 percent in the whole sam-

ple. No studies have been conducted in this area. In

explaining the prevalence of this risky behavior (5.75%),

it can be noted that due to the city’s proximity to the

Persian Gulf states, most residents of Larestan city

migrate to the Persian Gulf states, especially Abu Dhabi,

Kuwait and Qatar.

Table 3. Risky behaviors prevalence among youths (18-30 years) differentiated by gender in

Larestan

Risky behavior

Male Female

Estimated

frequency

Prevalence 95%

con dence

interval

Estimated

frequency

Prevalence 95%

con dence

interval

Consumption of

tobacco products

5112 16.82 -17.75

15.92

3119 11.06 10.26

-11.89

drug use 1129 3.71 3.26

-4.20

364 1.29 1.01

-1.61

Psychotropic

drug use

708 2.33 1.98

-2.72

446 1.58 1.27

-1.94

Dangerous Acts 2020 6.64 6.05

-7.27

1360 4.82 4.28

-5.40

alcohol

consumption

4373 14.39 -15.27

13.55

2433 8.63 7.92

-9.38

Sex outside the

family

2631 8.65 7.98

-9.36

1756 6.23 5.62

-6.89

Emigration 1833 6.032 5.48

-6.63

1536 5.44 4.87

-6.06

Tattoo 359 1.18 0.93

-1.47

132 0.46 0.30

-0.67

BIOSCIENCE BIOTECHNOLOGY RESEARCH COMMUNICATIONS ESTIMATING THE PREVALENCE OF RISKY BEHAVIORS BY USING NETWORK SCALE-UP METHOD 137

Hashemi and Yazdanpanah and Aghaei

Another high-risk behavior examined in this study

was a history of sex before marriage that it was observed

nearly 7.5 percent of 15-30 year-olds have experienced

this behavior. The results of a study in Isfahan showed

that the risky behaviors, including partner’s illicit sex is

22.1%, illicit sexual relationship and a history of tem-

porary marriage are 17.8%(Ataei B, 2011). The study

results of Torkashvand et al in the city of Kerman and

Rafsanjan indicated that the unprotected sexual behav-

ior with a frequency of 40% is a risky behavior before

HIV diagnosis(Torkashvand 2015)Kolahi et al showed in

their study that the female sex workers (prostitutes) who

have non-commercial intercourse use condom 1.8 times

less than those that have commercial intercourse(Kolahi

AA). Garmaroudi et al reported that the prevalence of

health risk behaviors such as sexual activities is 20.2%

in Tehran(Garmarodi 2009).

In the study of Shokoohi et al in the indirect method,

sex outside the family was estimated 12% over the last

year, while sex with female sex workers was reported

7%(Shokoohi et al., 2012). Moreover, in the study of

Sheikh Zadeh et al in the indirect method, the alco-

hol consumption was the most common risky behavior

among men and sex outside of marriage was the most

common behavior in women(Sheikhzadeh 2016).The

history of premarital sex among boys was more than

girls that due to the cultural and religious structure of

Iranian society is a signi cant outbreak.

The high prevalence of premarital sex can be a sign of

fading interest in religion among young people, lack of

suitable conditions for marriage and sometimes a sign of

modernity.Although tattooing is often done with care, it

is not totally risk-free, and can transmit the blood-borne

infections. Limited serologic studies on people who have

a history of tattooing have shown that tattooing could

be a way to transfer the viral infections, particularly

hepatitis B virus.

Tattoo prevalence was estimated 0.84% in the total

sample. A study in 2002 showed that people who

have been tattooed become infected with the Hepati-

tis C virus 9 times more than those who have not been

tattooed. Also, among the 454 students, 50 percent

of them pierced parts of the body such as the lip, ear

and navel, and 25 percent had a history of tattooing.

Almost 20 percent of those who had the tattoo effects

on their body were suffering from the complications

such as bacterial infection, bleeding and skin and tis-

sue damage in the area of tattooing. The results of a

study showed that the rate of positive HBsAg in the tat-

tooed women was 7.9% in Zahedan(Shari moud and

Metanat 2007). Torkashvand et al stated in their study

that a third of people with HIV had a history of tat-

tooing before the diagnosis of the disease (Torkashvand

2015).

The prevalence of psychotropic drug use was 0.84

percent in the total sample. In the study of Hamdieh

et al, the prevalence of psychotropic drugs was reported

3.8% in youth and adolescents in Tehran(Hamdieh et al.,

2008). In the study conducted by Ahmadi et al on the

students of Shiraz city, a frequency of smoking mari-

juana, heroin, morphine and cocaine was 0.8, 1, 0.8 and

0.5 percent, respectively(Ghaderi, 2015). Pourasl et al

stated in their study that 20 percent of high school stu-

dents have the experience of psychotropic drug use in

Tabriz. Perhaps, a possible justi cation for the increase

in other studies is that people in those areas have easy

access to the psychoactive drugs.

In the results of this study, social network size in

males and females was estimated 16.29 and 14.5, respec-

tively. This suggests that on average, men and women

in the age group 18-30 years are familiar with 16 males

and 14 females in this age group. This implies that on

average, each man communicates with 1.12 people more

than women. While a study in Kerman showed that the

social network size in men and women is 25.8 and 29.5,

respectively. On average, each man communicates with

3.8 people more than women(Sheikhzadeh 2016).

Moreover, in the study of Shokoohi et al in 2010, it

was suggested that the social network size estimated in

Iranian population is 303 people( Shokoohi et al., 2012).

The value has been estimated much more in this study

and a study done in Kerman. This difference may be due

to the different de nitions of social network in the stud-

ies. In a study of Kerman, it was related only to the social

network of students on campus. In this study, it also

covered the sexual age groups 18-30 years(Sheikhzadeh

2016).

According to the results, the following suggestions

are recommended:

In addition, to reduce the prevalence of risky behav-

iors among young and active population of our country,

the following are recommended:

Improving the knowledge and attitude of people, especially

sexually active people aged 18-30 years in conjunction

with risky behaviors

Improving the knowledge and attitude of mental health care

workers and psychiatrists in relation to the risky behaviors

through mandatory training sessions for all employees

Development of life skills training as a primary prevention

program of alcohol, tobacco and other drugs in the youth

in order to reduce drug use among young people by Educa-

tion’s of cials and planners

A comprehensive review and evaluation on a large scale

in the  eld of implementing the program by counseling

center for the prevention of behavioral health and identify-

ing the weaknesses and providing the strategies to improve

the program

Evaluation of the knowledge and attitude of mental health

care workers in relation to the risky behaviors

138 ESTIMATING THE PREVALENCE OF RISKY BEHAVIORS BY USING NETWORK SCALE-UP METHOD BIOSCIENCE BIOTECHNOLOGY RESEARCH COMMUNICATIONS

Hashemi and Yazdanpanah and Aghaei

Evaluation of the knowledge and attitude of target groups

in relation to the risky behaviors

Further studies on the causes and motivations of people for

the tendency toward the risky behaviors

CONCLUSION

Despite the low prevalence of risky behaviors in Lar-

estan, paying more attention to sexually active individ-

uals can play a critical role in reducing this phenome-

non in the community. Therefore, developing the serious

planning by the agencies that are responsible for the

health, especially the health centers is essential to reduce

these behaviors in the community.

STUDY LIMITATIONS

Entry criteria for the study includes all women and men

aged 18-30 years living in Larestan.

Exclusion criteria do not include other people outside

the age group 18-30 years are not included.

Exclusion criteria do not include other non-native

people.

ACKNOWLEDGMENTS

The authors thank and appreciate all those who have

cooperated in implementing this plan as well as health

personnel who have cooperated and assisted in complet-

ing the checklists.

REFERENCES

Akbar, A. (2013). Past Human Development Quarterly Journal

of Social Deveopment, 7(3), 123-154.

Ataei B, K. F., Azadeh S, Nokhodian Z, Kassaian N, Anahita

Babak. (2011). The Prevalence of High Risk Behaviors among

Women Prisoners in Isfahan. Journal of Isfahan Medical School

2011, 29(150), 170-177.

Garmarodi GhR, M. J., Abasi Z. (2009). Health risky habits in

students of Tehran Monitoring 9(1), 9-13.

Ghaderi, M. A., Zeinab; Darabzadeh, Fatemeh; Nasiri, Morteza;

Fakouri, Elham. (2015). Correlation between Emotional Intel-

ligence and Risky Sexual Behaviors in Nursing Students of

Khozestan Province Universities in 2013. Journal of Clinical

Research in Paramedical Sciences, 4(1).

Hamdye M, M. N., Hasheri H, Berojerdi A. (2008). Prevalence

of stimulant drugs, alcohol and psychoactive drugs Youth and

adolescents 15-35 years in Tehran. Original Article, 32(4), 315.

Headquarters, I. D. C. (2015). Exponential Growth of Danger-

ous Behaviour Among Youngsters Etemad Newspaper, 16().

Ismail-Zadeh, H., Asadi, M., Miri, Mir Nadir, Kramtkar, Mary.

(1393). MaryThe prevalence of risky behaviors among Journal

of Epidemiology Specialist. adolescents in iran, 10, 75-82.

Kolahi AA, R. A., Abadi AR, Nabavi M, Sayyarifard A, Sohrabi

M. The knowledge and attitudes of a female at risk popula-

tion towards the prevention of AIDS and sexually transmitted

infections in Tehran. J Res Med Sci 16(11), 1452-1455.

Momen Nasab, M. N., Seyed Saeid; Kaveh, Mohammad Hos-

sein; Ahmadpour, Farnaz. (2006). Evaluation of Spreading

Amount of Risky Sanitary Behaviour Among the Students of

Higher Education in the City of Khorram Abad in 2005-06.

Hafteh, 8(2), 29-32.

Buelow MT (2005). The In uence of Cognitive, Personality and

Social Variables: Predicting Changes in Risky Behaviors over a

Two-Year Interval. College of Arts and Science.

Gestv R, B. (2013). Comparison of indicators related with

injecting drug users(IDUs) in Iran before and after the harm

reduction programme. Kerman University of Medical Sciences;.

Shari moud B. Metanat M, G. H. (2007). Prevalence Of Trans-

fusion-Transmitted Viral Infections Among Women With His-

tory Of Tattoo In The City Of Zahidan During 2004-06. Iranian

Journal of Infectious Diseases and Tropical Medicine, 12(37),

66-69.

Sheikhzadeh KH, B. M., Afshari M, Haghdoost AA (2016).

Comparing direct, network scale-up, and proxy respondent

methods in estimating risky behaviors among collegians. Jour-

nal of Substance Use, 21(1), 9-13.

Shokoohi M, B. M., Haghdoost A. (2012). Size Estimation of

Groups at High Risk of HIV/AIDS using Network Scale Up in

Kerman, Iran. International journal of preventive medicine,

3(7), 471.

Shokoohi M, B. M., Haghdoost AA (2012). Size estimation of

groups at high risk of HIV/AIDS using network scale up in Ker-

man, Iran. Int J Prev Med, 3, 471-476.

Torkashvand F, A. M., Sheikh Fathollahi M, Sheikhi E, Salehi

Shahrbabaki M.H, Hoseini OR, Bakhtar M, Bidaki R (2015).

Frequency of High Risk Behaviour in HIV Positive Patients

Referred to Centers for Behavioural Disorders of Rafsanjan

and Kerman. J RafsanjanUniv Med Sci in 2012, 14(7), 587-

598.

Boyer TW ( 2006). The development of risk-taking. multiper-

spective review. Dev Rev, 26, 291-345.

Wilson MD, J. A. (1995). Adolescent medicine. JAMA, 273(21),

1657-1659.

Yach D, H. C., Gould CL, Hoffman KJ. (2004). The Global Bur-

den of Chronic Disease: Overcoming impediments to preven-

tion and control. JAMA, 291

, 2616-2622.