University of Southern California Dissertations and Theses

Comparison of senior middle school students' smoking behavior between urban and rural area in China

(USC Thesis Other)

Comparison of senior middle school students' smoking behavior between urban and rural area in China

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content COMPARISON OF SENIOR MIDDLE SCHOOL STUDENTS’ SMOKING
BEHAVIOR BETWEEN URBAN AND RURAL AREA IN CHINA

by

Yan Zhou

A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(BIOSTATISTICS)

August 2007

Copyright 2007 Yan Zhou

ii

DEDICATION

To my parents, Rupei Zhou & Suhua Zhang,
my husband, Ning Xu,
my son, Jiejun Xu for their support and understanding

iii

ACKNOWLEDGEMENTS

This work was done under the direction and supervision of my guidance
committee chair, Dr. Stanley Paul Azen, and co-chair, Dr. Chih-Ping Chou. I would
like to express my gratitude and appreciation for their invaluable guidance and
support throughout the duration of my study. I would also like to extend my special
thanks to my guidance committee, Dr. Jennifer Unger for her guidance and support
while I was working on my Master’s thesis. Furthermore, I would also like to extend
my appreciation to Dr. Huiyan Ma for her advice and suggestions for the preparation
of this manuscript.

iv

TABLES OF CONTENTS

Dedication

ii
Acknowledgements

iii
List of Tables

v
Abstract

vi
Introduction
1
Methods
Data Sources and Sample Selection
Data Collection
Measures
Data analysis

4
6
7
8

Results
Demographic Characteristics
Prevalence of Lifetime Smoking
Prevalence of Past Month Smoking
Ever Smoked a Whole Cigarette
Smoking Onset Age

11
13
14
15
17

Discussion 18

Bibliography 21

v

LIST OF TABLES

Table 1: Demographic Characteristics of Urban & Rural 9
th
Graders by
Region

11
Table 2: Lifetime Smoking among Urban & Rural 9
th
Graders by Gender and
Region

13
Table 3: Past Month Smoking among Urban & Rural 9
th
Graders by Gender and
Region

15
Table 4: Ever Smoked a Whole Cigarette among Urban & Rural 9
th
Graders
by Gender and Region

16
Table 5: Smoking Onset Age of Urban & Rural 9
th
Graders by Gender and
Region
17

vi

ABSTRACT

This study examined the smoking behavior among 3818 middle school
students (9
th
graders’) in either urban or rural area across four geographic regions in
China. After adjusting for age, peer and parental smoking, rural boys demonstrate
more severe smoking behaviors compared to urban boys (odds ratio [OR] =1.89,
95% confidence interval [CI] =1.36-2.62 for lifetime smoking; OR =1.28, 95% CI
=0.93-1.77 for past month smoking; and OR =2.04, 95% CI =1.27-3.26 for a lifetime
whole cigarette use). Rural girls, on the other hand, were less likely to smoke than
urban girls. Smoking initiation age was not statistically significantly different
between rural and urban students. Across the four regions, the students in the
southwest region had a higher smoking prevalence than those in the other three
regions. Our findings suggest that rural boys and urban girls, especially those in the
southwest region, are important populations for tobacco control interventions.

1

INTRODUCTION

Over 320 million tobacco smokers, about one-third of all smokers worldwide,
live in China (Yang et al., 1999) where tobacco use poses a critical public health
problem. Many of the major causes of morbidity and mortality in China, including
respiratory illnesses, cerebrovascular and cardiovascular disease, and cancer (Gao et
al., 1994; Woodward et al., 2005; Yang et al., 2004), are related to tobacco smoking.
Epidemiologists project that by 2025, two million Chinese will die annually from
tobacco-related disease (WHO, 2001).
The prevalence of cigarette smoking (at least one cigarette daily at the time of
survey) among adult Chinese males is alarming - 61% in 1984 and 63% in 1996
(Weng et al., 1987; Yang et al., 1999; Yuan et al., 1996). The prevalence of tobacco
use has escalated according to recent data from China (Gu et al., 2004; Sung et al.,
2006). Historically, smoking prevalence was typically low in Chinese females.
However, recent studies observed a more rapid increase in smoking prevalence
among female Chinese adults than their male counterparts (Li et al., 1999).
Tobacco use in China is not limited to adults as it is also a “Youth Epidemic”.
Two thirds of the men began to smoke daily before reaching 25 years of age (Yang
et al., 1999). Based on the National Epidemiological Smoking Behavior Survey in
1996, about 9 million Chinese adolescents between the ages of 15–19 were ever
smokers in 1996, accounting for 18% of all male and 0.28% of all female (making a

2

total of 9 million teenagers); Furthermore, Chinese adolescents’ current regular
smoking rates reached 9.7% (Zhang and Cai, 2003). This youth epidemic has taken
a turn for the worse as the adolescent smoking rate has increased in recent years (Li
et al., 1999). Based on the China Seven Cities Study conducted in urban area of
Harbin, Shenyang, Wuhan, Chengdu, Kunming, Hangzhou, and Qingdao, Johnson et
al in 2002 found that 27% of middle school students, 38% of academic high school
students, 54% of vocational high school students had ever tried smoking cigarettes.
And smoking prevalence in the past month are 9% , 8% and 26% for these three
populations, respectively (Johnson et al., 2006).
More than two thirds of China 's population live in rural areas. However, there
are limited data describing the status of smoking consumption for Chinese citizens in
the rural areas, especially for rural adolescents. One study reported people who were
at least 15 years of age and lived in one of three rural areas neighboring three
Chinese provinces: Henan, Fujian, and Shanxi during 1996 and 1998, consumed
fewer cigarettes and were less likely to smoke than those in the neighboring urban
areas(Hu and Tsai, 2000). However, another study found that rural respondents
between the ages of 15-65 in the Huaihua District of Hunan Province of China were
more likely to be heavy smokers than urban populations, especially among rural
males, whereas urban females were more likely to smoke than rural counterparts
(Zhou et al., 2006). A study for adolescents in Wuhan, China, found that the
prevalence of adolescent smoking was more prevalent among rural boys than among

3

urban boys, but it was more prevalent among urban girls than among rural girls
(Unger et al., 2001). For the current study, we analyzed and compared the smoking
behavior using the cross-sectional data of 3818 urban and rural senior middle school
students in China. The cross-sectional data were obtained from the China Seven-City
study (CSCS) involving seven cities (Harbin, Shenyang, Wuhan, Chengdu, Kunming,
Hangzhou, and Qingdao) in four different geographic regions (Northeast, Central,
Southwest, Coastal region) of China. The measures of smoking behaviors compared
between rural and urban area and across four geographic areas in this article include
lifetime smoking, past month smoking, a whole cigarette use, and the initiation age
of a whole cigarette.

4

METHODS

Data Sources and Sample Selection
The senior middle school students’ (9
th
graders’) included in this analysis
were selected from the survey of China Seven Cities Study (CSCS) (Johnson et al.,
2006). The primary purpose of the CSCS is to gain a better understanding of the role
of rapid social, economic, and cultural change on tobacco use and related health
practices and outcomes in China, with the ultimate goal of developing and
implementing effective community-based approaches to tobacco use prevention and
control. The seven participating cities are located in four regions in China: Northeast
(Harbin, Shenyang), Center (Wuhan), Southwest (Chengdu, Kunming), and Coastal
(Hangzhou, Qingdao). Started in 2002, three waves of longitudinal data were
collected annually from youth and their parents in the urban area. At the third wave,
survey was expanded to the youth in the rural area.
Participants in urban area were originally recruited in 2002 from selected
schools in each of the seven cities. The schools were selected using a stratification
process considering (a) administrative district median income and (b) school
academic performance. Each city first identified the administrative districts in the
urban area with the highest, middle, and lowest residential incomes by consulting the
appropriate city agencies. The local Education Committees were then asked to
classify every middle and high school in the identified districts into one of three

5

levels of academic performance. This process resulted in a total of nine clusters (3
academic levels × 3 districts) of schools. Three districts representing three levels of
district income randomly selected one middle school and one traditional high school
from each of the 9 clusters were invited to participate in the study. One class each
from grades 7, 8 (middle school), 10, and 11 (high school) from each school was
recruited for the study. In addition, one professional high school was selected from
each district, and the three professional high schools selected from three districts in
each city matched on enrollment, type of vocational training, and ratio of male to
female enrollment. Major courses of study within each professional school were
randomly selected, and students in these majors were recruited from the 10th and
11th grades to participate in the study. In summary, 9 middle schools, 9 high schools,
and 3 professional schools were selected from each city, and a total of 147 schools
were selected across all seven cities at baseline (2002). In China, middle school
includes 7, 8 and 9th grades while high school includes 10, 11, and 12th grades. The
8
th
and 11
th
graders had been followed-up for 2 years while 7
th
and 10
th
graders had
been followed-up for 3 years. In other words, at the third-year survey, we only
obtained the follow-up data from urban youth who were 7th and 10th graders.
As described above, the third-year of the CSCS survey also covered the rural
areas governed by the seven cities. From the lowest income county town within the
county jurisdiction of each of our participating urban areas, local education

6

committees supplied the names of middle schools. From each middle school, two
7
th
and two 9
th
grade classes were randomly selected.
In summary, at the third-year of the CSCS survey, we obtained the 9th
graders data from both urban and rural areas. The 9th graders in China are senior
students in middle school who are just about to transit into high school. Ninth grade
may be a key period for the severer smoking behavior that will happen in high school.
Therefore, we decided to take advantage of a study of the smoking behavior among
both urban and rural senior middle school students to determine whether the students
who were at same grade, but lived at the areas with different social economic status
(SES) – urban (higher SES) and rural (lower SES), had differences in smoking
behavior.

Data Collection
We obtained written parental consent from each participating student prior to
data collection. Trained data collectors explained the study to the students in their
classes, and students were asked to provide verbal informed consent prior to
completing the survey. All consent procedures and survey instruments were
approved by the Institutional Review Boards of the University of Southern California
and, each of the seven participating Chinese cities.

7

Student participants completed a self-administered structured questionnaire,
which took approximately 45 minutes to complete. Questionnaire items included,
demographics (e.g. gender, age), information about health education received, and
risk behaviors, including tobacco and alcohol use, sexual activity, and mental health.
Middle and high school students completed the structured questionnaire
during class. Classroom teachers were not present during the survey period. Prior to
the survey for the students, students took home consent forms and questionnaires for
their parents to complete, and then returned them to their classroom teachers in
sealed envelopes. All data collection methods and procedures were exactly the same
in both urban and rural areas.

Measures
The smoking behaviors of interest in this analysis were lifetime smoking,
past-month smoking, a whole cigarette use, and the initiation age for smoking a
whole cigarette. Lifetime smoking was assessed with the question, “Have you ever
tried smoking, even a few puffs” (response options were “yes” and “no”). Past-
month smoking was assessed with the question, “During the past 30 days, on how
many days did you smoke cigarettes?”. Because the responses to this item deviated
significantly from a normal distribution, they were recoded to 0=”0 days” and 1=1 or
more days. A whole cigarette use (“yes” and “no”) and its initiation age were
assessed with the question, “Have you ever tried cigarette smoking, even a few

8

puffs?”. The responses were between the ages of 8-17. Ages less than 8 were
counted as 8 and ages elder than 17 were counted as 17.

Data Analysis
The demographic characteristics of the rural and urban samples were
compared across the four regions using chi-square tests (for categorical variables)
and Analysis of Variance (for continuous variables). Chi-square analyses were also
applied to examine variation in tobacco use across the four regions.
Because the original study used a group-randomized sampling strategy,
multilevel logistic regression models were adopted to control for the interdependence
among observations due to school affiliation. We fitted two types of multilevel
models – Model A and Model B. In model A, we used random-intercept model
without considering any potential confounders. In model B, we included parental
smoking and peer smoking as covariates. Parental smoking and peer smoking had
been found to be important risk factors for adolescent smoking (Grenard et al., 2006;
Hesketh et al., 2001) The purpose of this analytical strategy was to examine whether
the difference in smoking behavior between the rural and urban students was due to
the confounding of these two individual level covariates.. The details for these two
models are at below.
The Level-1 model for students, or with-in-school, model:
Model A: Y
ij
= β
0 j
+ e
ij
,

9

Model B: Y
ij
= β
0 j
+ β
1j
X
ij
+ e
ij
,
Where in this study, Y
ij
stands for the smoking behavior such as life-time
smoking, past month smoking, or whole cigarette use of student i in school j. In
model B, we additionally include X
ij
that represents the potential confounders that
consist of adolescents’ age (continuous variable), (1=yes, 0=no), and parental
smoking (1=yes, 0=no). Parents’ tobacco smoking was determined by either the
father or mother who had ever smoked 100 cigarettes. We estimated only study
design effect adjusted, and both study design effect and multivariable adjusted OR
and 95% CI of smoking behaviors for rural versus urban area.

Model A and B used the same Level-2 model, or between-school model, in
which intercept ( β
0j
) is outcome:
β
0j
= γ
00
+ γ
01
Z
j
+ u
0j
,

where
β
0j
is parameters specified in Level-1 model;
Z
j
is a dummy coded variable that indicates the adolescent in rural or urban
area.
γ
00
is the average intercept ( β
0j
) across the Level-2 units (schools);
u
0j
is the residual of intercept ( β
0j
)
;
u
0j
~ N(0
,
τ
00
2
);

The Level-1 and Level-2 equations can be combined to yield a reduced form:

10

Model A: Y
ij
= γ
00
+ γ
01
Z
j
+ u
0j
+ e
ij
,

Model B:

Y
ij
= γ
00
+ γ
01
Z
j
+ β
1j
X
ij
+ u
0j
+ e
ij
,

Significant test of each parameter was based on the criterion two sided Ps<0.05
or the 95% CIs excluded 1.0 to reject null hypotheses. All analyses were performed
with the SAS statistical package (Version 9.13, SAS Institute, Cary, NC, USA).

11

RESULTS

Participants
A total of 3849 senior middle school students completed the survey (62 urban
middle schools and 14 rural middle schools). For 2631 urban senior middle school
students, their initial participant rate at CSCS baseline survey was more than 94%.
Their two-year follow-up rate was 87%. For 1158 rural middle school students (9
th

graders), their participant rate was approximately 99%.
However, we were unable to include 31 students in the analyses because of
missing information on gender (23 Urban and 8 rural students). Consequently the
analyses included in this study only consisted of 3818 students (urban:2660, rural:
1158).

Demographic Characteristics
The demographic characteristics of the 3818 middle school students were
listed by urban and rural area in Table 1. Overall, the percentage of male students
was 49.92% and 51.38% among urban and rural sample, respectively, which did not
statistically significantly vary across the four regions in either urban or rural area.
The urban students’ mean age was 14.73 (standard deviation [SD] = 0.74) while rural
students’ mean age was 15.04 (SD = 0.82). The mean ages varied across the
different regions in both urban and rural area (all Ps < 0.005). Overall peer smoking
rate across regions was 28.76% in urban area and 37.64% in rural area. Peer smoking

12

rate was statistically significantly varied across the four regions (all Ps < 0.0001) in
either urban or rural areas. In addition, parental smoking rate was on average
63.03% in urban area and 63.17% in rural areas. Parental smoking rate varied
significantly across the four regions in urban area but not in rural area (P = 0.052).
Table 1. Demographic Characteristics of Urban & Rural 9
th
Graders by Region
Regions
Total
(N=3849)
Northeast
(N=996)
Central
(N=586)
Southwest
(N=1122)
Coastal
(N=1145)
Difference
across
region
Urban students
N (including
subjects with
missing data)
2683 722 432 766 763

Gender
Male
Female
1328 (49.92)
1332 (50.08)
352 (49.09)
365 (50.91)
211 (48.84)
221 (51.16)
380 (50.13)
378 (49.87)
389 (51.66)
364 (48.34)
χ
2
=1.30

Mean age ( SD),
yrs
14.73 (0.74) 14.86
(0.67)
14.27
(0.69)
14.53
(0.65)
15.06 (0.72) F=154.61
**

Peer smoking 28.76 25.31% 20.88% 40.76% 24.40% χ
2
= 77.80
**

Parental smoking 63.03% 59.42% 62.40% 66.53% 63.30% χ
2
= 7.69

Rural students
N (including
subjects with
missing data)
1166 274 154 356 382

Gender
Male
Female
595 (51.38)
563 (48.62)
133 (49.08)
138 (50.92)
75 (49.02)
78 (50.98)
179 (50.28)
177
(49.72)
205 (54.23)
173 (45.77)
χ
2
=1.99

Mean age ( SD),
yrs
15.04 (0.82) 14.95
(1.02)
14.95
(0.68)
15.03
(0.75)
15.16 (0.74) F=4.33
**

Peer smoking 37.64% 21.98% 37.01% 47.03% 40.42% χ
2
= 43.06
**

Parental smoking 63.17% 55.92% 62.31% 68.06% 64.60% χ
2
= 8.83
*

*: p<0.05
**: p<0.01

13

Lifetime Smoking
Overall, lifetime smoking prevalence was 23.20%, 16.22%, 32.98%, 51.19%
for urban girls, rural girls, urban boys, and rural boys, respectively (Table 2).
Differences on lifetime smoking prevalence were detected across four regions . After
adjusting for age, peer smoking, and parental smoking, rural boys were 1.89 (95% CI
= 1.36-2.62) times more likely to ever smoke a cigarette than urban boys while rural
girls were 0.64 (95% CI = 0.42-0.97) times less likely to ever smoke a cigarette than
urban girls. The same pattern of the difference between urban and rural was
observed in all the regions except for girls in central area (OR = 1.48, 95% CI =
0.72-3.04). Both boys and girls in urban southwest region and boys in rural
southwest region had a higher lifetime smoking prevalence than those in the other
three regions (all P < 0.001).
Table 2. Lifetime Smoking among Urban & Rural 9
th
Graders by Gender and
Region
Regions

Total
Northeast Central Southwest Coastal
Difference
across
region

(N=735) (N=157) (N=98) (N=299) (N=201)

Male

Urban
434
(32.98%)
109
(31.05%)
54
(26.06%)
166
(44.03%)
105
(27.46%)
X
2
=30.94
**

Rural
301
(51.19%)
48
(36.36%)
44
(57.14%)
133
(63.84%)
96
(47.52%)
X
2
=25.13
**

Comparison
(rural vs.
urban)

Adjusted
1
OR
(95%CI)
2.14
(1.47-
3.12)
1.34
(0.75- 2.36)
3.80
(2.06-7.03)
2.29
(1.22-4.30)

2.31
(1.10-4.88)

Adjusted
2
OR
(95%CI)
1.89
(1.36- 2.62)
1.34
(0.78- 2.31)
3.91
(1.84- 8.31)
1.50
(0.74-3.04)
1.96
(1.19-3.23)

14

Table 2, Continued
Total Region
Northeast Central Southwest Coastal

Difference
across
region
(N=396) (N=99) (N=62) (N=162) (N=73)
Female
Urban
306
(23.20%)
78
(21.37%)
41
(18.64%)
125
(33.33%)
62
(17.24%)
X
2
=31.95
**

Rural
90
(16.22%)
21
(15.22%)
21
(28.77%)
37
(21.39%)
11
(6.43%)
X
2
=24.02
**

Comparison
(rural vs.
urban)

Adjusted
1
OR
(95%CI)
0.63
(0.42-0.96)
0.66
(0.31-
1.43)
1.76
(0.96-
3.25)
0.56
( 0.27-1.16)
0.56
(0.27- 1.16)

Adjusted
2
OR
(95%CI)
0.64
(0.42-0.97)
0.75
(0.36- 1.57)
1.48
(0.72-3.04)
0.61
(0.29-1.26)
0.31
(0.15-0.63)

1
adjust for design effect by using multilevel random intercept model
2
additionally adjusted for age, peer smoking, and parental smoking
*: p<0.05
**: p<0.01

Past Month Smoking
Overall, past month smoking prevalence was 13.46%, 17.80%, 7.02%, 4.58%
for urban girls, rural girls, urban boys, and rural boys, respectively (Table 3). After
adjusted for age, peer smoking, and parental smoking, rural boys were 1.28 (95% CI
= 0.93-1.77) times more likely to smoke than urban boys while rural girls were 0.69
(95% CI = 0.39-1.17) times less likely to smoke than urban girls. The same pattern
in the difference between urban and rural was observed in all the regions.

15

Table 3. Past Month Smoking among Urban & Rural 9
th
Graders by Gender and
Region
Regions

Total
Northeast Central Southwest Coastal
Difference
across Region

(N=284) (N=65) (N=35) (N=120) (N=64)

Male

Urban
179
(13.46%)
48
(13.68%)
22
(10.43%)
78
(20.58%)
31
(7.97%)
X
2
=28.25
**

Rural
105
(17.80%)
17
(12.78%)
13
(16.88%)
42
(23.73%)
33
(16.26%)
X
2
=6.92

Comparison
(rural vs.
urban)

Adjusted
1
OR
(95%CI)
1.38
(0.97-1.95)
0.95
(0.47-1.93)
1.75
(0.83-3.68)
1.20
(0.78-1.84)
2.13
(1.06-4.27)

Adjusted
2
OR
(95%CI)
1.28
(0.93-1.77)
1.08
(0.55-2.11)
2.12
(0.80-5.60)
0.86
(0.49-1.52)
2.16
(1.18-3.94)

(N=120) (N=39) (N=11) (N=44) (N=26)

Female

Urban
93
(7.02%)
30
(8.24%)
9
(4.09%)
34
(9.02%)
20
(5.49%)
X
2
=7.33

Rural
27
(4.58%)
9
(6.52%)
2
(2.67%)
10
(5.81%)
6
(3.49%)
X
2
=2.65

Comparison
(rural vs.
urban)

Adjusted
1
OR
(95%CI)
0.68
(0.41-1.13)
0.79
(0.34-1.84)
0.64
(0.13-3.06)
0.65
(0.20-1.62)
0.62
(0.24-1.58)

Adjusted
2
OR
(95%CI)
0.69
(0.39-1.17)
0.85
(0.35-2.05)
0.58
(0.09-3.67)
0.62
(0.24-1.63)
0.68
(0.26-1.76)

1
adjust for design effect by using multilevel random intercept model
2
additionally adjusted for age, peer smoking, and parental smoking
*: p<0.05
**: p<0.01

Ever Smoked a Whole Cigarette
Overall, the percentage of the students who had reported having ever smoked
a whole cigarette was 18.09%, 35.81%, 9.10%, 5.37% for urban girls, rural girls,
urban boys, and rural boys, respectively (Table 4). After adjusted for age, peer

16

smoking, and parental smoking, rural boys were 2.04 (95% CI = 1.27- 3.26) times
more likely to smoke than urban boys while rural girls were 0.56 (95% CI = 0.30-
1.04) times less likely to smoke than urban girls. The same pattern in the difference
between urban and rural was observed in all the regions. The students in southwest
region had a higher smoking prevalence than those in the other three regions (all P <
0.05).
Table 4. Ever Smoked a Whole Cigarette among Urban & Rural 9
th
Graders by
Gender and Region
Total Regions Difference
across region
Northeast Central Southwest Coastal
Male

Urban
240
(18.09%)
60
(17.05%)
24
(11.37%)
112
(29.63%)
44
(11.40%)
X
2
=52.32
**

Rural
212
(35.81%)
24
(18.05%)
31
(39.74%)
91
(51.41%)
66
(32.35%)
X
2
=38.59
**

Comparison (rural vs. urban)
Adjusted
1
OR
(95%CI)
2.51
(1.51-4.16)
1.06
(0.40-2.82)
5.14 (2.75-
9.59)
2.65
(1.18-5.96)
3.70
(1.64-8.38)

Adjusted
2
OR
(95%CI)
2.04
(1.27- 3.26)
0.93
(0.35-2.44)
7.12
(2.69-18.87)
1.52
(0.66-3.49)
2.95
(1.63-5.33)

Female

Urban
120
(9.1%)
28
(7.7%)
10
(4.5%)
62
(16.4%)
20
(5.5%)
X
2
= 36.63
**

Rural
30
(5.37%)
6
(4.35%)
4
(5.33%)
17
(9.71%)
3
(1.75%)
X
2
=11.19
*
Comparison (rural vs. urban)
Adjusted
1
OR
(95%CI)
0.58
( 0.32- 1.07)
0.55
(0.19-1.55)
1.19
(0.36-3.93)
0.60
(0.25-1.46)
0.31
(0.08-1.20)

Adjusted
2
OR
(95%CI)
0.56
(0.30-1.04)
0.66
(0.26-1.72)
0.75
(0.17-3.31)
0.57
( 0.22-1.49)
0.27
(0.08-0.99)

1
adjust for design effect by using multilevel random intercept model
2
additionally adjusted for age, peer smoking, and parental smoking
*: p<0.05
**: p<0.01

17

Onset Age for Smoking a Whole Cigarette
The average age of initial smoking among urban boys was 12.15 and 12.37
among rural boys instead. The average age of initial smoking among urban girls was
11.99 and 13.13 among rural girls. The average age of initial smoking seemed to
be older in costal region than the other regions.
Table 5. Smoking Onset Age of Urban & Rural 9
th
Graders by Gender and Region
Regions

Total
Northeast Central Southwest Coastal
Difference
across
region
Male

Urban
11.98 (2.46) 11.95 (2.32) 11.26 (2.12) 11.91 (2.43) 12.61 (2.84) F=1.60

Rural
12.42 (2.65) 13.08 (2.65) 11.36 (2.61) 12.06 (2.75) 13.19 (2.30) F=4.77
**

Difference
between
urban and
rural
t = -1.82
P = 0.07
t = -1.94
P = 0.06
t = -0.14
P = 0.89
t = -0.40
P = 0.69
t = -1.14
P = 0.26

Female
Mean (SD) Mean (SD) Mean (SD) Mean (SD)

Mean (SD)

Urban
11.91 (2.52) 11.50 (2.99) 12.50 (2.07) 11.57 (2.19) 13.25 (2.67) F=2.82
*

Rural
12.33 (2.59) 13.67 (2.34) 11.50 (3.11) 11.65 (2.45) 14.67 (1.53) F=2.09

Difference
between
urban and
rural
t = -0.82
P = 0.41
t =-1.66
P = 0.11
t = 0.71
P = 0.49
t =- 0.13
P = 0.89
t = -0.88
P = 0.39

*: p<0.05
**: p<0.01

18

DISCUSSION
This study found that rural boys were on average more likely to smoke
compared to urban boys while rural girls were less likely to smoke than urban girls.
After adjusting for age, peer and parental smoking, same patterns have consistently
observed. Rural boys’ smoking initiation age was not statistically significantly
different from urban boys’, but it appeared to be older than urban boys that of.
Across the four regions, both boys and girls in southwest region had higher smoking
prevalence rates than those in the other three regions in both urban and rural areas.
We found that rural boys were on average more likely to smoke compared to
the urban boys while rural girls were less likely to smoke than urban girls. These
findngs are consistent with a previous study conducted between December 1998 and
January 1999 in Wuhan (Unger et al., 2001), which represents central region in the
current study. The finding is also consistent with another study that conducted in
Zhejiang, China (Hesketh et al., 2001). The consistent pattern of difference in
adolescents’ smoking behavior between urban and rural area may provide important
guideline for prevention program development in China.
In our data, rural boys touched cigarettes later than urban boys, but they were
more likely to have ever smoked in their lives by the time they were in the 9
th
grade.
These findings are consistent with the previous studies conducted in China(Zhou et
al., 2006). This is also consistent with the current social environment in China. With

19

China's reform and modernization, more Western culture and lifestyles are being
introduced to urban area of China earlier than to rural area. Many Chinese
adolescents in urban area blindly worship Western lifestyles and imitate things they
see in movies, magazines, and the internet. They treat smoking as a cool and
fashionable habit. However, China's rural areas are relatively poor. Therefore, the
adolescents in rural area are less vulnerable to the impact of Western culture,
parental and peer smoking behavior could be the main access for them to learn about
the world. The previous studies(Grenard et al., 2006; Hesketh et al., 2001; Lai et al.,
2004) found that parent smoking and peer smoking are main risk factors for
adolescents’ smoking. On the other hand, the adolescents in rural area accept the
education that cigarettes are harmful relatively less than those living in cities(Sun
and Ling, 1997). That could be the reason that more adolescents in rural area have
smoked in their lifetimes.
Across the four regions of China covered by our survey, boys and girls in
southwest region had a higher smoking prevalence than those in the other three
regions. The severe smoke behavior was also observed among high school students,
college students, and adults in the southwest area (Anderson Johnson et al.,
2006). The severe smoking behavior in the southwest area could be due to the strong
tobacco production in this region. Yunnan Tobacco is a major cash crop and is also
important export commodities (Eng, 1999). Yunnan Yuxi is a well-known village of

20

cigarettes (Wang, 1992). This may provide a very convenient access to cigarette
smoking for people living in Southwest region.
Several limitations of this study must be considered. The data used for this
analysis are self-reported. This was the first time for rural students to participate in
the survey, but third time for the urban students because they were involved in our
longitudinal study. We do not know whether this could results in any bias and do not
have validation on data obtained from rural youth. However, the opposite direction
in the differences of smoking behavior between urban and rural area for boys and
girls prompted us to believe that the difference in smoking behaviour we observed
was not biased. Another limitation is that our sample size is relatively small for
studying girls’ smoking behavior. Also, the geography areas this study covered did
not included Northwest and Southern China. These could impact the generalization
of our findings.
Even with these potential limitations, the current analysis provides
informative data for developing and delivering effective smoking intervention
programs for the adolescents in China. The smoking intervention should target on
both urban and rural boys, but more efforts should be on rural boys, especially those
who live in southeast region of China.

21

BIBLIOGRAPHY
Anderson Johnson, C., Palmer, P. H., Chou, C.-P., Pang, Z., Zhou, D., Dong, L.,
Xiang, H., Yang, P., Xu, H., Wang, J., et al. (2006). Tobacco use among youth and
adults in Mainland China: the China Seven Cities Study. Public Health 120, 1156-
1169.

Eng, I. (1999). Agglomeration and the Local State: The Tobacco Economy of
Yunnan, China. Transactions of the Institute of British Geographers 24, 315-329.

Gao, Y. T., McLaughlin, J. K., Blot, W. J., Ji, B. T., Benichou, J., Dai, Q., and
Fraumeni, J. F., Jr. (1994). Risk factors for esophageal cancer in Shanghai, China. I.
Role of cigarette smoking and alcohol drinking. International Journal of Cancer 58,
192-196.

Grenard, J. L., Guo, Q., Jasuja, G. K., Unger, J. B., Chou, C.-P., Gallaher, P. E., Sun,
P., Palmer, P., and Anderson Johnson, C. (2006). Influences affecting adolescent
smoking behavior in China. Nicotine & Tobacco Research 8, 245-255.

Gu, D., Wu, X., Reynolds, K., Duan, X., Xin, X., Reynolds, R. F., Whelton, P. K.,
He, J., and for the Inter, A. C. G. (2004). Cigarette Smoking and Exposure to
Environmental Tobacco Smoke in China: The International Collaborative Study of
Cardiovascular Disease in Asia. Am J Public Health 94, 1972-1976.

Hesketh, T., Ding, Q. J., and Tomkins, A. (2001). Smoking among youths in China.
American Journal of Public Health 91, 1653-1655.

Hu, T. W., and Tsai, Y. W. (2000). Cigarette consumption in rural China: survey
results from 3 provinces. Am J Public Health 90, 1785-1787.

Lai, M. K., Ho, S. Y., and Lam, T. H. (2004). Perceived peer smoking prevalence
and its association with smoking behaviours and intentions in Hong Kong Chinese
adolescents. Addiction 99, 1195-1205.

Li, H. Z., Fish, D., and Zhou, X. (1999). Increase in cigarette smoking and decline of
anti-smoking counselling among Chinese physicians: 1987-1996. Health Promot Int
14, 123-131.

Sun, W. Y., and Ling, T. (1997). Smoking behavior among adolescents in the city,
suburbs, and rural areas of Shanghai. American Journal of Health Promotion 11,
331-336.

22

Sung, H. Y., Wang, L., Jin, S., Hu, T. W., and Jiang, Y. (2006). Economic burden of
smoking in China, 2000. Tob Control 15, i5-11.

Unger, J. B., Yan, L., Chen, X., Jiang, X., Azen, S., Qian, G., Tan, S., Jie, G., Sun, P.,
ChunHong, L., et al. (2001). Adolescent smoking in Wuhan, China: baseline data
from the Wuhan Smoking Prevention Trial. American Journal of Preventive
Medicine 21, 162-169.

Wang, Y. (1992). Yunnan yuxi juanyan chang fazhan shi (Light from Yuxi: a history
of Yunnan's Yuxi cigarette factory) Yunnan renmin chuban she, Kunming (Kuming:
Yuxi Zhi Guang).

Weng, X. Z., Hong, Z. G., and Chen, D. Y. (1987). Smoking prevalence in Chinese
aged 15 and above. Chinese Medical Journal 100, 886-892.

Woodward, M., Lam, T. H., Barzi, F., Patel, A., Gu, D., Rodgers, A., Suh, I., and
Asia Pacific Cohort Studies, C. (2005). Smoking, quitting, and the risk of
cardiovascular disease among women and men in the Asia-Pacific region.
International Journal of Epidemiology 34, 1036-1045.

Yang, G., Fan, L., Tan, J., Qi, G., Zhang, Y., Samet, J. M., Taylor, C. E., Becker, K.,
and Xu, J. (1999). Smoking in China: Findings of the 1996 National Prevalence
Survey. JAMA 282, 1247-1253.

Yang, Y., Jiang, Y., Yang, X., Deng, Y., Li, J., Guo, Z., Lin, D., Wang, J., Chen, Q.,
and Yu, F. (2004). [Analysis on main factors for successful quitting--study on the
one-year follow-up for Chinese 'Quit and Win' in 2002]. Wei Sheng Yen
Chiu/Journal of Hygiene Research 33, 478-480.

Yuan, J. M., Ross, R. K., Wang, X. L., Gao, Y. T., Henderson, B. E., and Yu, M. C.
(1996). Morbidity and mortality in relation to cigarette smoking in Shanghai, China.
A prospective male cohort study. JAMA 275, 1646-1650.

Zhang, H., and Cai, B. (2003). The impact of tobacco on lung health in China.
Respirology 8, 17-21.

Zhou, X., Su, Z., Deng, H., Xiang, X., Chen, H., and Hao, W. (2006). A comparative
survey on alcohol and tobacco use in urban and rural populations in the Huaihua
District of Hunan Province, China. Alcohol 39, 87-96.

Abstract (if available)

Abstract This study examined the smoking behavior among 3818 middle school students (9th graders') in either urban or rural area across four geographic regions in China. After adjusting for age, peer and parental smoking, rural boys demonstrate more severe smoking behaviors compared to urban boys (odds ratio [OR] =1.89, 95% confidence interval [CI] =1.36-2.62 for lifetime smoking

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Risk factors associated with smoking initiation among Chinese adolescents: a matched case-control study

PDF

Exploring the association of number of cigarettes smoked and confidence to quit smoking in Korean American emerging adults: a multilevel modeling approach

PDF

Measurement invariance across cultures: a comparison between Chinese adolescents in China and in U.S.

PDF

Association between latchkey status and smoking behavior in middle school children

PDF

Personal hair dye use and risk of B-cell non-Hodgkin’s lymphomas among adult women in Los Angeles County

PDF

The vicious cycle of inactivity, obesity, and metabolic health consequences in at-risk pediatric populations

PDF

Examination of sun awareness behavior, knowledge, and attitudes before and after a sun awareness program in 4th to 8th grade students

PDF

Assessment of the mortality burden associated with ambient air pollution in rural and urban areas of India

Asset Metadata

Creator Zhou, Yan (author)

Core Title Comparison of senior middle school students' smoking behavior between urban and rural area in China

School Keck School of Medicine

Degree Master of Science

Degree Program Biostatistics

Publication Date 07/25/2007

Defense Date 07/02/2007

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag middle school student,OAI-PMH Harvest,smoking behavior

Place Name China (countries)

Language English

Advisor Azen, Stanley Paul (committee chair), Chou, Chih-Ping (committee chair), Unger, Jennifer B. (committee member)

Creator Email zhouyan@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m663

Unique identifier UC1502735

Identifier etd-Zhou-20070725 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-526997 (legacy record id),usctheses-m663 (legacy record id)

Legacy Identifier etd-Zhou-20070725.pdf

Dmrecord 526997

Document Type Thesis

Rights Zhou, Yan

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu