Caspiano
Ivanaaaa
sample.pdf
Title A systematic review of cohort studies on the association ofsmoking with all-cause and lung cancer mortality in China
Author(s) Kong, Linyan; 孔林燕
Citation
Issued Date 2014
URL http://hdl.handle.net/10722/206969
Rights The author retains all proprietary rights, (such as patent rights)and the right to use in future works.
1
Abstract of project entitled
A systematic review of cohort studies on the association of
smoking with all-cause and lung cancer mortality in China
Submitted by
Kong, Lin-Yan
for the Degree of Master of Public Health
at The University of Hong Kong
in August 2014
Background
Smoking is a well-established causal risk factor of premature death. The
prevalence of smoking has been estimated to be more than 50% in Chinese men.
However, previous reviews of the association between smoking and mortality
from all-causes and lung cancer were mainly relied on developed countries. The
current systematic review of cohort studies aims at summarizing the existing
studies on the association of smoking with all-cause and lung cancer mortality in
China.
Methods
Articles published from 1980 to 2014 were searched systematically in databases
including PubMed, EMBASE, China National Knowledge Infrastructure (CNKI)
and Google scholar. Main results of all studies were extracted and summarized.
Results
A total of 14 cohort studies examining the association of smoking with all-cause
and lung cancer mortality in Chinese populations were identified. Compared with
never smoking, current smoking was associated with higher risks of all-cause and
lung cancer mortality in all studies. The relative risks (RR) for current smokers
2
were from 1.20 to 2.29 for all-cause mortality and from 2.44 to 9.40 for lung
cancer mortality. Former smokers also showed higher RRs for all-cause mortality
(RR=1.20-1.48) and for lung cancer mortality (RR=2.06-6.50) compared with
never smokers. Furthermore, dose-response associations of increasing smoking
categories with all-cause mortality and lung cancer mortality were observed in
most of the studies.
Conclusions
The risk estimates for all-cause and lung cancer mortality from smoking in China
were lower than those from the western countries suggesting that the tobacco
epidemic is at an early stage in China. Further large cohort studies giving updated
risk estimates are warranted for advocating stringent tobacco control policies in
China.
3
A systematic review of cohort studies on the association of smoking
with all-cause and lung cancer mortality in China
by
Kong, Lin-Yan
A project submitted in partial fulfillment of the requirements for
the Degree of Master of Public Health
at The University of Hong Kong
August 2014
i
Declaration
I declare that the project and the research work thereof represents my own work,
except where due acknowledgment is made, and that it has not been previously
included in a thesis, dissertation or report submitted to this University or to any
other institution for a degree, diploma or other qualifications.
Signed
---------------------------------------------------------------------
Kong Lin-Yan
ii
Acknowledgments
I would like to express my gratitude to Dr. XU, Lin for her supervision of this
project. Without her patience, guidance and valuable comments, I could not have
completed this project.
I would also like to thank my family for their support during my postgraduate
studies in public health.
iii
Contents
Declaration i
Acknowledgments ii
Contents iii
List of Figures and Tables iv
Chapter 1 1
Introduction 1
1.1 The tobacco epidemic in China 1
1.2 Smoking and health 1
Chapter 2 3
Methods 3
2.1 Search strategy 3
2.2 Inclusion criteria 3
2.3 Exclusion criteria 3
2.4 Definitions of smoking status in the included studies 4
2.5 Definitions of outcomes 4
2.6 Quality Assessment 4
Chapter 3 6
Results 6
3.1 Overview 6
3.2 Identification and Selection of Articles 6
3.3 Quality of reviews 7
3.4 Subject Characteristics 8
3.5 Mortality outcomes 13
Chapter 4 16
Discussion 16
4.1 Overview 16
4.2 Main findings 16
4.3 Limitations and strengths 18
4.4 Comparison with the western studies 19
4.5 Policy implications and recommendations for further studies 20
Chapter 5 21
Conclusion 21
References 22
iv
List of Figures and Tables
Figures
Figure1.Summary of article selection process 9
Tables
Table 1.Quality Items Used 5
Table2. Quality Assessment of Studies Included 10
Table3. Characteristics of the subjects at baseline 11
Table4. Baseline characteristics of the study participants by smoking status in four
high quality cohort studies 12
Table5. Relative risk (RR)† of all-cause or lung cancer of death by smoking status
at baseline 14
1
Chapter 1
Introduction
1.1 The tobacco epidemic in China
Over the past 20 years, smoking prevalence in men is relatively high. According
to the 1984, 1996, 2002 and 2010 survey in China, the prevalence of cigarette
smoking in Chinese men was 61%, 63%, 57% and 53% respectively. However,
the prevalence of smoking in Chinese women was only 3.1% according to the
survey of 2010. Considering the pattern of tobacco epidemic in world, China is at
the early stage of the tobacco epidemic.
The national cigarette consumption in China increased 40% from 2000 to 2010.
The average number of cigarettes consumed per day per person was one in 1950,
two in 1970, four point three in 1980 and five in 1990. It then reached to more
than five point five in 2007. Based on experience from the developed countries, it
would estimate that the mortality attributable to smoking in China will increase in
the near future if current smoking patterns persist.
1.2 Smoking and health
In 1964, the U.S. Surgeon General Report about smoking and health has identified
smoking as a causal risk factor for lung cancer, chronic bronchitis, emphysema
and coronary heart disease.(1) In addition, the World Health Organization (WHO)
report clearly states that, cigarette smoking cause most cases of lung cancer and is
a major risk factor for all-cause mortality, particularly for men.(2) Since 1964, the
US government took a series of public health interventions to inform the
2
Americans the harmfulness of smoking, help smokers quitting smoking, and
prevent people who do not smoke from beginning to smoke. The other western
developed countries also showed similar situations over the past 25 years, when
the consumption of tobacco products was slowly decreasing. Furthermore, there is
a large body of evidence showing the benefits of quitting smoking in the West and
in developed countries in Asia, such as Singapore and Japan. However, quitting
smoking is uncommon in China, which the tobacco epidemic is at an early stage.
With a population of 1.3 million, China has been the largest producer and
consumer of tobacco in the world.(3, 4) Currently, there are about one million
deaths every year in china because of cigarette smoking.(5) If the current pattern
of smoking persists, the number of deaths attributable to smoking will increase to
3 million every year in China by 2050.(6, 7)
A four stage model of the tobacco epidemic was originally proposed by Lopez et
al. in 1994 to illustrate the connection between the prevalence of cigarette
smoking and its effects on population mortality, as had been experienced in
developed countries where smoking was prevalent about 10 to 20 years earlier in
men than in women.(8) Countries such as the US or UK where the tobacco
epidemic has reached the most advanced stage can provide more reliable evidence
of tobacco harm. But results from China, where is still at an early stage of tobacco
epidemic, are certainly under-estimated.(9, 10) Understanding the association
between smoking and the risk of lung cancer as well as all-cause mortality in the
Chinese population will help the government make effective smoking cessation
strategies aimed at reducing deaths attributable to smoking in China.
3
Chapter 2
Methods
2.1 Search strategy
Literature references were searched systematically, which were published from
1980 to 2014in PubMed, EMBASE, China National Knowledge Infrastructure
(CNKI) and Google scholar. Keywords of “(smok* OR tobacco OR cigarette)
AND cohort AND (mortality OR death) AND China” were used for the literature
search. The search was limited to studies published in English or Chinese
language and those with full-text. I also used Google scholar for further search.
Lastly, I also searched the reference lists of selected papers for additional article.
Information of each study, including the number of subjects, duration on
follow-up, exposure, outcome and covariate assessment, results and conclusion,
was extracted for further evaluation.
2.2 Inclusion criteria
The following are the inclusion criteria for studies included in the final review: (1)
the included studies were conducted in China and all participants were Chinese
people; (2) the smoking status can be categorized as current smokers and ever
smokers including former smokers and never smokers; (3) Outcome
measurements should include lung cancer and/or all-cause mortality; (4) The
eligible studies should be prospective cohort studies.
2.3 Exclusion criteria
4
Articles not being original research were excluded. Studies which were published
in language not in English or Chinese, or on the association of second-hand
smoking but not active smoking with mortality were also excluded.
2.4 Definitions of smoking status in the included studies
Never smokers were defined as persons who had never smoked a cigarette or who
smoked less than 100 cigarettes during their lives. Former smokers were those
who had smoked at least 100 cigarettes in their lives had quitted smoking for more
than one year at the time of assessment. Current smokers were defined as those
who had smoked at least more than 100 cigarettes during their lifetimes or at the
time of the interview. Current smoking was further classified into daily current
smokers (smoked every day) or non-daily current smokers in some studies.(11, 12)
In the current review, ever-smokers included former smokers and current smokers.
2.5 Definitions of outcomes
In order to assess the risk of smoking cigarettes, we summarized all-cause and
lung cancer mortality in all studies. All-cause mortality was defined as the total
number of deaths from any causes and the average number of the population over
the same period. Lung cancer mortality was defined by the International
Classification of Disease, Ninth Revision (ICD 9) codes 162 and Tenth Revision
(ICD 10) codes C33-C34 based on the number of death from lung cancer per
100,000 person-years.
2.6 Quality Assessment
Quality assessment was done for all selected articles. The quality item, which
5
Ievaluated for all cohort studies in this systematic review, was from the
Newcastle-Ottawa Scale.(13) The scale is divided into three categories in terms of
selection, comparability and outcome, including a total of 8 items, which are
illustrated in Table 1 below. After careful examination, all the selected articles
were graded based on the score of each quality indicator. A total score for each
article was calculated by adding scores of each indicator, with the highest score
being nine and the lowest being zero. Results of the quality assessment of the
selected studies were shown in the results section of the current systematic review.
Table 1.Quality Items Used
A. Selection: Each item is one point
1. Representativeness of the exposed cohort
2. Selection of the no exposed cohort
3. Ascertainment of exposure
4. Demonstration that outcome of interest was not present at start of study
B. Comparability: Two points
5. Comparability of cohorts on the basis of the design or analysis
C. Outcome: Each item is one point
6. Assessment of outcome
7. Was follow-up long enough for outcomes to occur
8. Adequacy of follow up cohorts
6
Chapter 3
Results
3.1 Overview
I performed extensive literature search and carefully screened the studies for their
eligibility to be included in the systematic review. A total of 14 cohort studies
were identified for further evaluation of the association of smoking with all-cause
and lung cancer mortality in China. Information including the subject
characteristics including the year of baseline survey, geographic and age at study
entry, duration of follow-up and sample size considered was extracted from each
study. Moreover, the quality of the included studies was examined in detail using
the Newcastle-Ottawa Scale
3.2 Identification and Selection of Articles
An overview of the search process was shown in a flow diagram (Figure 1).
Using the keywords of “(smok* OR tobacco OR cigarette) AND cohort AND
(mortality OR death) AND China”, a total of 254 articles were identified from
PubMed, 29 articles from EMBASE databases and 23 from CNKI. These articles
were then screened based on titles and abstracts according to the inclusion and
exclusion criteria. A total of 48articles from PubMed without full-text and 4
articles not written in English or Chinese were excluded. In addition, after
reviewing the titles and abstracts, 166 articles from PubMed, 14 articles from
CNKI and 23 articles from EMBASE databases were determined to be irrelevant
and were excluded. Hence, there are 51 papers (36 articles from PubMed, 6
articles from EMBASE databases and 9 articles from CNKI) included for further
7
full-text scrutiny for relevance. Of these 51 articles, 33 studies were excluded as
duplicates and 5 studies were excluded because the exposure was second-hand
smoking rather than active smoking. An in-depth scrutiny of the remaining studies
identified a total of 13 cohort studies to be included in the systematic review. In
order not to avoid missing any relevant articles, a manual search from references
in each selected study was performed and one additional article was identified
from the bibliographic search. Finally, a total of 14 cohort studies were included
in the systematic review.
3.3 Quality of reviews
A summary of the quality assessment of the eligible studies was shown in Table 2.
The fourteen eligible papers included in the systematic review were evaluated for
their quality using the criteria defined in the methods section. For the
representativeness of the cohort, three studies including studies by Zhang et
al.(14), Zhang et al.(15) and Wang et al.(16) were given 0 point because these
studies were not conducted in community-based populations which may to some
extent limit the generalizability. Those studies in which smoking was treated as a
categorical variable (Never smoker/Ever-smokers; Never smoker/Former
smoker/Current smoker) were given 1 point while three studies conducted by Gu
et al.(10), He et al. (17) and Lam et al.(18) , were given 2 points because these 3
studies also provided information for the dose-response associations of increasing
pack-years with all all-cause mortality or lung cancer mortality. For the outcome
measurement, studies by Wang et al.(16) and Yuan et al.(19) had zero point
because they did not provide information on the lost to follow up. The overall
quality of the fourteen studies was rated as moderate to good.
8
3.4 Subject Characteristics
An overview of the subject characteristics of the studies included in the systematic
review is shown in Table 3 and Table 4.The present review contains 14 cohort
studies in China and consists of a total number of 693,995 Chinese people
(including 542,230 men and 151,765 women) with the duration of follow up
ranged from 5.4 years to 17 years. The baseline surveys of these cohort studies
were conducted from 1972 to 2000. The age of the participants in these studies
was at least 30 years.
Participants in two of the studies (10, 20) were throughout the country. Four
studies were from the Xi’an,(17, 18, 21, 22) three studies were from Shanghai,(19,
23) three studies were from Guangzhou,(14, 15, 24) and the remaining two studies
were from Shantou(16) and Beijing.(25) Although there were five studies included
both men and women, there were more men than women included in the current
review. The average age of participants in most studies ranged from 40.8 years to
62.9 years. In addition, I selected four studies, (10, 17, 18, 26) which were ranked
as high quality, to further examine the characteristics of never, former and current
smokers at baseline. Generally, former smokers tended to be older than never
smokers and current smokers, while current smokers were more likely to consume
alcohol and less likely to be obese than never smokers. As a whole, before
adjustment for potential confounder current smokers were less likely to have
hypertension than never smokers, probably because current smokers were younger
and had lower body mass index.
9
Figure1.Summary of article selection process
306 articles identified from PubMed (254),
EMBASE (29) and CNKI (23)
254 papers screened
52 records excluded using PubMed
filter: full text or not written in
English or Chinese
203 papers were excluded based on
titles of abstracts
38 articles excluded
5: subjects: second-hand smoking
8: analysis of the same sample
3: the literature in English and
Chinese is the same
22: Duplicates from PubMed and
EMBASE
13 articles included in the systematic review
14 articles included in the final systematic
review
1 article included from bibliographic
search
51 papers assessed for eligibility
10
Table2. Quality Assessment of Studies Included
Studies Quality Items
1 2 3 4 5 6 7 8 Overall Scores
Gu D (10) B B B B A B B B 9
He Y (17) B B B B A B B B 9
Chen Z (23) B B B B B B B B 8
Lam T (18) B B B B A B B B 9
Lam T (24) B B B C B B B B 7
Niu S (20) B B B C B B B B 7
Yuan J (19) B B B C B B B C 6
Zhang Z (27) B B C C B B B B 5
Zhang W (15) C B B B B B B B 7
Zhang W (14) C B B B B B B B 7
Wang J (22) B B C B B B B B 7
Wang W (16) C B C B B B B C 5
Chen J (25) B B C C B B B B 6
He Y (21) B B B B B B B B 8
A=2(Best); B=1(Good); C=0(Bad)
11
Table3. Characteristics of the subjects at baseline
First author Year Geographic location No. of participants
(M/F)
Lost to follow –up, % Follow-up duration, years Age, year
Gu D(10) 1991 nationally 83,533/86,388 7.10 8 40 years or order
He Y(17) 1994 Xi’an 961/533 - 17 -
Chen Z(23) 1997-1998 Shanghai 6,494/2,857 4.00 16 35-64 years
Lam T(18) 1987 Xi’an 1268/0 2.00 12 62.9 (mean)
Lam T(24) 1992 Guangzhou 81,344/0 - 7 40.8(mean)
Niu S(20) 1987 nationally 224,500/0 2.00 9 40 or over
Yuan J(19) 1986 Shanghai 18,244/0 - 5 45-64 years
Zhang Z(27) 1972 Shanghai 957/667 5.00 12 -
Zhang W(15) 1989-1992 Guangzhou 46,987/31,302 4.50 8 43.5 (mean)
Zhang W(14) 1989-1992 Guangzhou 36,730/24,918 0.05 7 30 or above
Wang J(22) 1990 Xi'an 13,000/0 4.00 10 -
Wang W(16) 1991 Shantou 2,133/0 - 9 15 or above
Chen J(25) 1979-1980 Beijing 5,137/0 2.94 11 18 or above
He Y(26) 1987 xi'an 1,268/0 2.00 11 62.9 (mean)
12
Table4. Baseline characteristics of the study participants by smoking status in
four high quality cohort studies
Author,
Year
Smoking
status
Age
mean(SD),
years
Alcohol
consumption
(%)
Hypertension
(%)
BMI
mean(SD),
kg/m 2
Lam
T(18)
Never
smokers
62.5 (5.2) 33.2 22.4 24.3 (3.0)
Former
smokers
64.0 (5.3) 61.8 26.5 24.6 (3.1)
Current
smokers
62.2 (4.8) 56.3 20.8 24.0 (3.1)
He
Y(17)
Never
smokers
60.2 (5.0) 8.9 30.4 23.5 (2.7)
Former
smokers
60.8 (5.2) 18.6 19.5 23.0 (2.7)
Current
smokers
61.2 (5.4) 30.4 31.3 22.6 (3.1)
He
Y(21)
Never
smokers
62.5 (5.2) 33.2 22.4 24.3 (3.0)
Former
smokers
64.0 (5.3) 61.8 26.5 24.6 (3.1)
Current
smokers
62.2 (4.8) 56.3 20.8 34.0 (3.1)
Gu
D(10)
Men
Never
smokers
57.3(11.1) 19.1 32.2 23.1(3.5)
Former
smokers
60.9(10.3) 36.3 23.9 22.6(3.7)
Current
smokers
54.2(9.9) 49.0 20.6 22.0(3.2)
Women
Never
smokers
57.7(11.0) 1.8 27.9 22.8(3.9)
Former
smokers
64.2(9.9) 7.6 28.0 22.9(4.7)
Current
smokers
59.3(9.5) 9.2 24.4 22.4(4.2)
13
3.5 Mortality outcomes
Data on all-cause death and lung cancer of death were available in 10 and 11
studies respectively. Table 5 shows that current smokers had significantly higher
risk of all-cause and lung cancer mortality than never smokers. Compared with
never smokers, current smokers had an increased risk of deaths from all-cause
(RR=1.20 to 2.29) and lung cancer (RR= 2.44-9.40) while the risks were lower for
all-cause (RR=1.20 to 1.48) and lung cancer mortality (RR= 1.06-6.50) for former
smokers. In addition, three studies (10, 14, 28) summarized the relative risk on
men and women for all-cause and/or lung cancer mortality separately. In male
smokers, the RRs ranged from 1.19 to 1.54 for all-cause mortality and ranged
from 2.52 to 7.4 for lung cancer mortality, while in female smokers, the RRs
ranged from 1.18 to 1.33 for all-cause mortality.
14
Table5. Relative risk (RR)† of all-cause or lung cancer of death by smoking status at baseline
Author, Year Smoking status All-cause mortality Lung cancer mortality
No. of deaths RR 95% CI P No. of deaths RR 95% CI P
Gu D (10) Non-smokers 3841 (M) 1.0 - - - - - -
6195 (F) 1.0 - - - - - -
All smokers 6190 (M) 1.10-1.26 - <0.001 - - - -
1519 (F) 1.22-1.38 - <0.001 - - - -
He Y (17) Never smoker 238 1.00 - - 15 1.0 - -
Former smoker 84 1.32-1.48 - <0.001 8 1.06-2.96 - <0.001
Current smoker 166 1.49-2.29 - <0.001 22 2.44-5.34 - <0.001
Chen Z (23) Nonsmokers - 1.0 - - 1.0 - -
Ever-smokers - 1.4 - <0.001 - 3.8 - <0.001
Lam TH (18) Never smoker 75 1.0 - 6 - - -
Former smoker 119 1.22 0.90-1.65 0.20 19 2.26 0.87-5.83 0.09
Current smoker 105 1.50 1.11-2.04 0.01 15 2.36 0.90-6.21 0.08
Lam TH (24) Ever-smokers 858 1.24 1.07-1.44 <0.01 90 1.41 0.90-2.22 >0.05
Niu S (20) Non-smokers 2236 1.0 - - - - - -
Ever smoker 6697 1.19 1.13-1.25 <0.001 - - - -
Yuan J(19) Never smoker - 1.0 - - - 1.0 - -
Former smoker - 1.4 - <0.01 - 6.5 - <0.01
Current smoker - 1.2-1.6 - <0.01 - 3.6-9.4 - <0.05
Zhang Z(27) Never smoker - - - - 3 - - -
Ever smokers (men) - - - - 15 8.5 1.6-45.2 <0.05
Ever smokers (women) - - - - 2 11.4 2.4-53.5 <0.001
Zhang W(15) Never smoker 451 1.0 - - - 1.0 - -
15
Ever smokers 722 1.22 1.07-1.40 <0.001 219 3.32 2.16-5.09 <0.001
Zhang W(14) Never smoker - 1.0 - - - - - -
Ever smokers(men) 955 1.22 <0.001 - - - -
Ever smokers(women) 13 1.18 - - - -
Wang J(22) Never smoker - - - - 6 1.0 -
Ever smokers - - - - 42 3.03 - <0.001
Wang W(16) Never smoker 52 1.0 1 1.0 - -
Ever smokers 181 1.91 1.40-2.62 <0.05 14 5.54 0.73-42.15 <0.05
Chen J(25) Never smoker - - - - 18 1.0 - -
Ever smokers - - - - 80 3.45 1.58-7.73 <0.05
He Y(26) Never smoker 75 1.0 - - 6 1.0 - -
Former smoker 119 1.22 0.90-1.65 0.20 19 2.26 0.87-5.83 0.09
Current smoker 105 1.50 1.11-2.04 0.01 15 2.36 0.90-6.21 0.08
†: Most of the RR above adjusted for age, body mass index, regular alcohol consumption and hypertension at baseline.
16
Chapter 4
Discussion
4.1 Overview
To my knowledge, this is the first systematic review on the association of smoking
on all-cause and lung cancer mortality in prospective cohort studies in China.
Given the growing prevalence of smoking in China, the growing mortality
attributable to smoking remains a major public health problem in recent decades.
In 2010, Li et al estimated 52.9% of men and 2.4% of women were current
smokers.(29) Moreover, it was different from the smokers in the west that 83% of
Chinese smokers did not want to quit smoking. Among all smokers, less than 10%
of them had intention to quit and only 4% had quitted successfully as defined by
quitting for more than two years at the time of assessment.(30) In addition, about
half of them quitted smoking because of illness (sick-quitter effect).(17) Smoking
cessation in older people, professionals, highly educated population was more
common.(3) This systematic review critically evaluated the association of
smoking with all-cause mortality and lung cancer mortality in China. A summary
of the results, strengths and limitations, comparison between selected studies,
implications for further studies and recommendations for health policymakers was
provided.
4.2 Main findings
Fourteen cohort studies examining the association of smoking with mortality in
Chinese populations were identified and included in the current review. Regarding
17
to the representativeness of the study sample, there were two studies (10, 20)
which based on nationally representative populations while other studies were
based on samples selected from one city of China. About the results, there were
consistent trend that both current and former smoking increased the risk of
all-cause mortality and lung cancer mortality in China, although the relative risk
was not statistically significant in two studies (RR=2.36 for lung cancer mortality
from current smoking, and RR=2.26 for lung cancer and 1.22 for all-cause
mortality from former smoking, P values for the above >0.05).(18, 26) In current
smokers, compared to never smokers, the RRs ranged from 1.20 to 2.29 for
all-cause mortality and ranged from 2.44 to 9.40 for lung cancer mortality, while
in former smokers, the RRs ranged from 1.32 to 1.48 for all-cause mortality and
ranged from 1.06 to 6.50 for lung cancer mortality. The RR for all-cause mortality
associated with smoking in our review was lower than the recently published large
cohort studies in the US and UK (31-33). Compared with never smokers, the RR
for all-cause mortality for current smokers was about 3.0 in US and UK and 1.5 in
China and for former smokers was 1.7 in US and UK, including British Doctors
study and million women study, and 1.3 in China.(10, 31-35) This results may be
due to a shorter duration of smoking exposure and lower numbers of cigarettes
smoke exposure (pack-years) in China.(36) Gu D et al. (10) and He Y et al. (17)
reported significantly dose-response associations of increasing smoking categories
with all-cause mortality and lung cancer mortality (P for trend from 0.010 to
<0.001). The study by He Y et al.(17) showed that the RRs for all-cause and lung
cancer mortality was 1.48 and 1.06 for former smokers who smoked less than 20
pack-years, 1.32 and 2.96 for former smokers who smoked 20 or more pack-years,
1.49 and 2.44 for current smokers who smoked less than 35 pack-years and 2.99
18
and 5.34 for current smokers who smoked 35 or more pack-years, respectively.
4.3 Limitations and strengths
This current systematic review has several limitations. First, the age ranges of
most of the 14 studies were majorly focused on middle age adults. Whether the
smokers started smoking at young age was not clear. The estimated risks would be
substantially underestimated for those starting smoking at young age. Second,
although the systematic searchwas performed in 3 databases including PubMed,
EMBASE and CNKI, followed by a complete bibliographic search for references
from each study, I still cannot guarantee that all relevant studies were found. In
particular, I did not seek additional unpublished reports from experts. Third, the
14 cohort studies included in the current review had varying follow-up periods
and age ranges, which may lead to a greater heterogeneity. Fourth, each study was
adjusted for a different set of covariates, which might have also lead to
heterogeneity and make the results difficult to interpret. In addition, disparities in
smoked products (e.g. cigars, pipes, cigarillos) would affect risk estimates.
However, as most of the smokers smoked cigarettes in China and only less than
5% smoked cigars or pipes,(37) this could not fully explain the lower risk
estimates for smoking in China compared with the Western populations. Another
limitation is the lack of information on duration of smoking or age at smoking
initiation in most of the included studies. Given China has a relatively shorter
period of development, Chinese adults are inevitably have a shorter duration of
exposure than individuals from the developed countries, which may mainly
contribute to the lower risk estimate.
19
There are several strengths in this systematic review. One is that including studies
in the review are prospective population-based cohort studies with relatively large
sample size. Therefore, such systematic review would be useful in China, where it
is essential to accelerate the pace of smoking cessation to avoid the increasing
burden of smoking-attributable disease.
4.4 Comparison with the western studies
In our review, the RR of lung cancer among male smokers was 2.26-9.4, which is
much lower than that reported in Western populations (typically a 20-fold RR of
male lung cancer) (6). Moreover, the average lung cancer mortality rate in
middle-aged men among lifelong nonsmokers in the United States is 8/100,000
(38), which is about two times higher that observed in these studies in China (10).
Because Chinese people had a shorter exposure to tobacco smoke than western
populations given the short period of development in China. In addition, due to
the effects of more prolonged smoking, the increasing RR for lung cancer
mortality attributable to smoking in China may exceed those now seen in the US
in the future. Although the beneficial effect of smoking cessation for the risk of
all-cause mortality and lung cancer mortality has been well-established, the
reasons for quitting smoking varied between developed and developing countries.
In developed countries, smokers may quit smoking so as to reduce disease risks
because of the awareness and knowledge of tobacco harm. Whereas in developing
countries where people are less health conscious and lack of awareness of the
hazardness of tobacco, quitting smoking is usually due to the presence of
illness.(39, 40) Therefore, the health benefits of quitting smoking in China are
often inevitably underestimated because of this “sick-quitter” effect.
20
4.5 Policy implications and recommendations for further studies
The risk estimates of lung cancer or all-cause mortality from smoking were much
lower than those from western populations, suggesting that China may be still at
an early stage of tobacco epidemic. In the developed countries, people have
realized the hazardness of the tobacco and the benefits of the quitting smoking,
which will reduce disease risks. In China, people are lack of the knowledge of the
hazardness of smoking. The findings of my review show that cigarette smoking is
a major preventable cause of death in China. On the basis of observed risks, we
estimate that smoking increased the risk of all-cause mortality and lung cancer
mortality in China. Evidence from the local populations will have very important
implication on the estimation of current and future disease burden from tobacco in
China and countries at early stages. Smoking cessation should be common in
China. Given China is at an early stage of tobacco epidemic and the full effect of
smoking on mortality is not yet evident, further high quality large-scale
prospective cohort studies are needed to update the risk estimates.
21
Chapter 5
Conclusion
In conclusion, this systematic review showed that smoking increase the risk of
all-cause and lung cancer mortality. Compared to never smokers, the RRs for
all-cause and lung cancer mortality for current smokers ranged from 1.2 to 2.29
and 2.44 to 9.40, respectively, which were all lower than those shown in the
Western countries including the British Doctors study (35) or the million women
study (32) in the UK or the National Health Interview Survey in the US (33).
Compared with the developed countries, the tobacco epidemic is at an early stage
in China and other middle- and low-income countries where the prevalence of
smoking has been increasing in the past three decades. As there are several
decades of delay between the peak of smoking-attributable deaths and the peak of
smoking prevalence, without active smoking cessation strategies, there would be a
sharp increase in the burden of disease in the next decades. Given the
well-documented tobacco harm as well as the benefits of quitting, stringent
tobacco control policies should be urgently introduced in China.
22
References
1. Holford TR, Meza R, Warner KE, Meernik C, Jeon J, Moolgavkar SH, et al.
Tobacco control and the reduction in smoking-related premature deaths in the
United States, 1964-2012. JAMA : the journal of the American Medical
Association. 2014;311(2):164-71.
2. Organization WH. WHO framework convention on tobacco control. Geneva:
WHO, 2003. 2007.
3. Yang G, Fan L, Tan J, Qi G, Zhang Y, Samet JM, et al. Smoking in China:
findings of the 1996 National Prevalence Survey. JAMA : the journal of the
American Medical Association. 1999;282(13):1247-53.
4. Gu D, Wu X, Reynolds K, Duan X, Xin X, Reynolds RF, et al. Cigarette
smoking and exposure to environmental tobacco smoke in China: the international
collaborative study of cardiovascular disease in Asia. American journal of public
health. 2004;94(11):1972-6.
5. Peto R, Chen Z-M, Boreham J. Tobacco: the growing epidemic in China.
CVD Prevention and Control. 2009;4(1):61-70.
6. Peto R, Lopez A, Boreham J, Heath C, Thun M. Mortality from tobacco in
developed countries, 1950–2000. Indirect estimates from natio. 1994.
7. Peto R, Lopez AD, Boreham J, Thun M, Heath C, Jr., Doll R. Mortality from
smoking worldwide. British medical bulletin. 1996;52(1):12-21.
8. Lopez AD, Collishaw NE, Piha T. A descriptive model of the cigarette
epidemic in developed countries. Tobacco control. 1994;3(3):242.
9. Lam TH, He Y, Li LS, Li LS, He SF, Liang BQ. Mortality attributable to
cigarette smoking in China. JAMA. 1997;278(18):1505-8.
10. Gu D, Kelly TN, Wu X, Chen J, Samet JM, Huang JF, et al. Mortality
attributable to smoking in China. The New England journal of medicine.
2009;360(2):150-9.
11. Schoenborn CA, Adams PE. Health behaviors of adults: United States,
2005-2007. Vital and health statistics Series 10, Data from the National Health
Survey. 2010(245):1-132.
12. Schane RE, Ling PM, Glantz SA. Health effects of light and intermittent
smoking: a review. Circulation. 2010;121(13):1518-22.
13. Wells G, Shea B, O’connell D, Peterson J, Welch V, Losos M, et al. The
Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised
studies in meta-analyses. 2000.
14. Zhang W, Jiang C, Hing L, Yin H, Chen Q, Liu W, et al. [A prospective
cohort study on the comparison of risk of occupational dust exposure and smoking
to death]. Zhonghua liu xing bing xue za zhi= Zhonghua liuxingbingxue zazhi.
2004;25(9):748-52.
15. Zhang W, Jiang C, Lam T, Chen Q, Ho S, Liu W, et al. A prospective cohort
study on the effect of occupational dust exposure and smoking on mortality.
Zhonghua lao dong wei sheng zhi ye bing za zhi= Zhonghua laodong weisheng
zhiyebing zazhi= Chinese journal of industrial hygiene and occupational diseases.
2002;20(5):365-8.
16. Wang W, Wen YB, Chen KC, She HZ, Zhao HZ, et al. A prospective cohort
study on the association of smoking with chronic disease in Male of Shantou.
Guangdong Medical Journal. 2004;25.
23
17. He Y, Jiang B, Li LS, Li LS, Sun DL, Wu L, et al. Changes in smoking
behavior and subsequent mortality risk during a 35-year follow-up of a cohort in
Xi'an, China. American journal of epidemiology. 2014;179(9):1060-70.
18. Lam TH, He Y, Shi QL, Huang JY, Zhang F, Wan ZH, et al. Smoking,
quitting, and mortality in a Chinese cohort of retired men. Annals of epidemiology.
2002;12(5):316-20.
19. Yuan JM, Ross RK, Wang XL, Gao YT, Henderson BE, Yu MC. Morbidity
and mortality in relation to cigarette smoking in Shanghai, China. A prospective
male cohort study. JAMA. 1996;275(21):1646-50.
20. Niu SR, Yang GH, Chen ZM, Wang JL, Wang GH, He XZ, et al. Emerging
tobacco hazards in China: 2. Early mortality results from a prospective study.
BMJ (Clinical research ed). 1998;317(7170):1423-4.
21. He Y, Taihing L, Shi Q, Huang J, Zhang F, Wan Z, et al. [A prospective
study on smoking, quitting and mortality in a cohort of elderly in Xi'an, China].
Zhonghua liu xing bing xue za zhi = Zhonghua liuxingbingxue zazhi.
2002;23(3):186-9.
22. Wang JJ GH, He YY, Ma JG, Guo XR, Liu CZ, Zhang TT et al. A
Prospective Study of Relationship Between Smoking and the List of Death
Reasons. China Public Health. 2000;16.
23. Chen ZM, Xu Z, Collins R, Li WX, Peto R. Early health effects of the
emerging tobacco epidemic in China. A 16-year prospective study. JAMA : the
journal of the American Medical Association. 1997;278(18):1500-4.
24. Lam TH, Jiang CQ, Ho SY, Zhang WS, Liu WW, He JM. Smoking and
mortality in 81,344 drivers in Guangzhou, China. Occupational and environmental
medicine. 2002;59(2):135-8.
25. Chen J-c, Wu X-g, Duan X-f, Yu H, Zheng Y. A Prospective Study of the
Effect of Smoking and Body Mass Index on the Risk of Lung Cancer in Male
Workers of Beijing Steel Industry. CHINESE JOURNAL OF PREVENTION
AND CONTROL OF CHRONIC NON COMMUNICABLE DISEASES.
2006;14(5):311.
26. He Y, Taihing L, Shi Q, Huang J, Zhang F, Wan Z, et al. [A prospective
study on smoking, quitting and mortality in a cohort of elderly in Xi'an, China].
Zhonghua liu xing bing xue za zhi= Zhonghua liuxingbingxue zazhi.
2002;23(3):186-9.
27. Zhang ZF, Yu SZ, Li WX, Choi BC. Smoking, occupational exposure to
rubber, and lung cancer. British journal of industrial medicine. 1989;46(1):12-5.
28. Zhang WS, Jiang CQ, Hing L, Yin HS, Liu WW, He JM, et al. [A prospective
cohort study on comparison of risk of death of respiratory system diseases
between occupational dust exposure and smoking]. Zhonghua lao dong wei sheng
zhi ye bing za zhi = Zhonghua laodong weisheng zhiyebing zazhi = Chinese
journal of industrial hygiene and occupational diseases. 2006;24(6):364-7.
29. Li Q, Hsia J, Yang G. Prevalence of smoking in China in 2010. The New
England journal of medicine. 2011;364(25):2469-70.
30. Yang G. Report on the 1996 nationwide survey of smoking prevalence. China
Science and Technology Press, Beijing. 1997.
31. Thun MJ, Carter BD, Feskanich D, Freedman ND, Prentice R, Lopez AD, et
al. 50-Year Trends in Smoking-Related Mortality in the United States. New
England Journal of Medicine. 2013;368(4):351-64.
32. Pirie K, Peto R, Reeves GK, Green J, Beral V, Million Women Study C. The
21st century hazards of smoking and benefits of stopping: a prospective study of
24
one million women in the UK. Lancet. 2013;381(9861):133-41.
33. Jha P, Ramasundarahettige C, Landsman V, Rostron B, Thun M, Anderson
RN, et al. 21st-Century Hazards of Smoking and Benefits of Cessation in the
United States. New England Journal of Medicine. 2013;368(4):341-50.
34. Foody JM, Cole CR, Blackstone EH, Lauer MS. A propensity analysis of
cigarette smoking and mortality with consideration of the effects of alcohol. The
American journal of cardiology. 2001;87(6):706-11.
35. Doll R, Peto R, Boreham J, Sutherland I. Mortality in relation to smoking: 50
years' observations on male British doctors. BMJ (Clinical research ed).
2004;328(7455):1519.
36. Hu T-w. Tobacco control policy analysis in China: economics and health:
World Scientific; 2008.
37. Yu JJ, Mattson ME, Boyd GM, Mueller MD, Shopland DR, Pechacek TF, et
al. A comparison of smoking patterns in the People's Republic of China with the
United States. An impending health catastrophe in the middle kingdom. JAMA :
the journal of the American Medical Association. 1990;264(12):1575-9.
38. Shopland DR, Eyre HJ, Peachacek TF. Smoking-attributable cancer mortality
in 1991: is lung cancer now the leading cause of death among smokers in the
United States? Journal of the National Cancer Institute. 1991;83(16):1142-8.
39. Nakamura K, Huxley R, Ansary-Moghaddam A, Woodward M. The hazards
and benefits associated with smoking and smoking cessation in Asia: a
meta-analysis of prospective studies. Tobacco control. 2009;18(5):345-53.
40. Yang G, Ma J, Chen A, Zhang Y, Samet JM, Taylor CE, et al. Smoking
cessation in China: findings from the 1996 national prevalence survey. Tobacco
control. 2001;10(2):170-4.
