University of Southern California Dissertations and Theses

A prospective cohort study of morbidity and mortality among middle-aged men in Shanghai, China

(USC Thesis Other)

A prospective cohort study of morbidity and mortality among middle-aged men in Shanghai, China

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may be
from any type o f computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in reduced
form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6” x 9” black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly to
order.
UMI
A Bell & Howell Information Company
300 North Zeeb Road, Ann Arbor MI 48106-1346 USA
313/761-4700 800/521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
A PROSPECTIVE COHORT STUDY OF MORBIDITY AND MORTALITY
AMONG MIDDLE-AGED MEN IN SHANGHAI, CHINA
by
Jian-Min Yuan
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Epidemiology)
May 1996
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 9636387
UMI Microform 9636387
Copyright 1996, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized
copying under Title 17, United States Code.
UMI
300 North Zeeb Road
Ann Arbor, MI 48103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UNIVERSITY OF SOUTHERN CALIFORNIA
THE GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES, CALIFORNIA 90007
This dissertation, written by
Jian-Min Yuan
under the direction of h.^.f........ Dissertation
Committee, and approved by all its members,
has been presented to and accepted by The
Graduate School in partial fulfillm ent of re
quirements for the degree of
DOCTOR OF PHILOSOPHY
Dean of Graduate Studies
D a te ....^ P T .h } ...? .^ .!...} .^ .? .9 .........
DISSERTATION COMMITTEE
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 1
To my parents, my wife, Jin-Hong,
and my daughter, Jacqueline,
for their support and understanding
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
in
ACKNOWLEDGMENTS
I would like to thank members of my Doctoral Committee for their advice and
suggestions throughout the course of my research and in the preparation of this
manuscript. Thanks go to Brian Henderson, Peter Jones, John Peters, Ronald Ross, and
Mimi Yu for their helpful comments and expertise in their respective fields. Special
thanks and gratitude are extended to Ronald Ross and Mimi Yu for their invaluable
suggestions and unending support provided throughout my doctoral study.
I thank the staff at the Department of Epidemiology and the Shanghai Cancer
Registry for their assistance in data collection and management and in verifying cancer
diagnosis. Special appreciation goes to Yu-Tang Gao and Ji-Tao Tu for their support and
encouragement.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IV
TABLE OF CONTENTS
LIST OF TABLES............................................................................................................................... vi
VITA..................................................................................................................................................... viii
ABSTRACT OF DISSERTATION.................................................................................................. x
PREFACE............................................................................................................................................. xii
CHAPTER 1 INTRODUCTION...................................................................................................... 1
CHAPTER 2 MATERIAL AND METHODS...............................................................................7
I. Study D e sig n.................................................................................................................................... 7
II. Study Population..........................................................................................................................7
III. D ata Collection..........................................................................................................................8
rv. D a ta b a se E stab lish m en t and M an agem en t.................................................................12
V. Follow-up.......................................................................................................................................13
VI. Statistical A nalysis................................................................................................................14
CHAPTERS CHARACTERISTICS OF THE STUDY POPULATION............................. 16
I. Demographic Param eters........................................................................................................ 16
II. Tobacco U s e ................................................................................................................................ 21
III. Alcohol U se............................................................................................................................... 21
IV. Medical Histo ry......................................................................................................................25
CHAPTER 4 MORTALITY AND MORBIDITY IN RELATION TO CIGARETTE
SMOKING............................................................................................................................................ 27
I. Introduction..................................................................................................................................28
II. Subjects and Methods..............................................................................................................29
III. Resu lts..........................................................................................................................................31
IV. D iscussion....................................................................................................................................41
CHAPTER 5 MORTALITY IN RELATION TO ALCOHOL DRINKING......................47
I. Introduction..................................................................................................................................48
II. Subjects and Methods..............................................................................................................49
III. Re su l t s..........................................................................................................................................51
IV. D iscussion....................................................................................................................................65
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 6 SERUM TESTOSTERONE AND HEPATOCELLULAR
CARCINOMA..........................................................................................................................70
I. In t r o d u c t i o n ...................................................................................................................................................... 71
n. M a t e r i a l a n d M e t h o d s .............................................................................................................................73
III. Re su l t s........................................................................................................................................75
rv. D i s c u s s i o n .........................................................................................................................................................80
REFERENCE............................................................................................................................82
APPENDIXES..........................................................................................................................92
A p p e n d ix A . Q u e s t io n n a ir e .......................................................................................................................... 9 2
A p p e n d ix B . In f o r m e d C o n s e n t F o r m ................................................................................................. 121
A p p e n d ix C . F o l l o w - u p Q u e s t i o n n a i r e .............................................................................................122
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
V I
LIST OF TABLES
T a b l e 1 - 1 A v e r a g e a n n u a l a g e - s t a n d a r d iz e d in c id e n c e r a t e s o f t h e e ig h t
COMMONEST CANCERS IN CHINESE MEN OF SHANGHAI AND
n o n -H is p a n ic -s u r n a m e d w h it e m e n o f L o s A n g e l e s C o u n t y , 1 9 8 3 -1 9 8 7 .............. 2
T a b l e 3-1 S t u d y p o p u l a t io n b y r e s id e n t ia l a r e a s a t r e c r u it m e n t , S h a n g h a i
C o h o r t S t u d y 1 9 8 6 -1 9 8 9 ...................................................................................................................... 17
T a b l e 3 -2 D e m o g r a p h ic c h a r a c t e r is t ic s a t r e c r u it m e n t , S h a n g h a i C o h o r t
S t u d y 1 9 8 6 -1 9 8 9 ......................................................................................................................................... 18
T a b l e 3 -3 T o b a c c o u s e a t r e c r u it m e n t , S h a n g h a i C o h o r t S t u d y 1 9 8 6 -1 9 8 9 .........2 2
T a b l e 3 -4 A l c o h o l u s e a t r e c r u it m e n t , S h a n g h a i C o h o r t S t u d y 1 9 8 6 -1 9 8 9 .........23
T a b l e 3 -5 M e d ic a l h is t o r y a t r e c r u it m e n t , S h a n g h a i C o h o r t S t u d y
1 9 8 6 - 1 9 8 9 ....................................................................................................................................................... 26
T a b l e 4-1 N u m b e r o f s u b je c t s b y a g e a n d c ig a r e t t e s m o k in g s t a t u s a t
r e c r u i t m e n t ..................................................................................................................................................33
T a b l e 4 -2 A g e - a d ju s t e d c a n c e r in c id e n c e r a t e s a n d r e l a t iv e r is k s b y
c ig a r e t t e s m o k in g s t a t u s a t r e c r u it m e n t ............................................................................. 34
T a b l e 4 -3 A g e - a d ju s t e d m o r t a l it y r a t e s a n d r e l a t iv e r is k s b y c ig a r e t t e
s m o k in g s t a t u s a t r e c r u it m e n t ......................................................................................................38
T a b l e 5-1 V a r io u s a t t r ib u t e s o f a l c o h o l c o n s u m p t io n a t r e c r u i t m e n t ................. 52
T a b l e 5 -2 S e l f- r e p o r t e d c h a r a c t e r is t ic s b y a l c o h o l c o n s u m p t io n s t a t u s
AT RECRUITMENT........................................................................................................................................... 54
T a b l e 5-3 T o t a l m o r t a l it y b y a l c o h o l c o n s u m p t io n s t a t u s a t r e c r u i t m e n t ......56
T a b l e 5 -4 M o r t a l it y f r o m c a n c e r b y a l c o h o l c o n s u m p t io n s t a t u s a t
RECRUITMENT..................................................................................................................................................59
T a b l e 5 -5 M o r t a l it y f r o m n o n - c a n c e r - r e l a t e d c a u s e s b y a l c o h o l
CONSUMPTION STATUS AT RECRUITMENT...........................................................................................61
T a b l e 5 -6 T o t a l m o r t a l it y b y c o n s u m p t io n s t a t u s o n s p e c if ic t y p e s o f
ALCOHOLIC BEVERAGES AT RECRUITMENT........................................................................................ 6 4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
vil
T a b l e 6-1 A g e - a d ju s t e d m e a n l e v e l o f s e r u m t e s t o s t e r o n e a c c o r d in g t o
HBV SERUM MARKERS IN 382 CONTROL SUBJECTS, A NESTED CASE-CONTROL
STUDY OF HCC.................................................................................................................77
T a b l e 6 -2 A g e - a d ju s t e d g e o m e t r ic m e a n l e v e l o f s e r u m t e s t o s t e r o n e in
CASES AND CONTROLS, A NESTED CASE-CONTROL STUDY OF HCC..............................78
T a b l e 6-3 R i s k o f HCC a c c o r d in g t o l e v e l o f s e r u m t e s t o s t e r o n e a n d
SERUM HBsAG s t a t u s , A NESTED CASE-CONTROL STUDY OF H CC............................79
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
V lll
VITA
June 28, 1958 Bom in Shanghai, China
1978-1983 Bachelor of Medicine, Shanghai Medical University, Shanghai,
China
1983-1986 Master of Public Health, Shanghai Cancer Institute/Shanghai
Medical University, Shanghai, China
1986-1988 Research Fellow, Department of Epidemiology, Shanghai Cancer
Institute, Shanghai, China
1988-1992 Assistant Professor, Department of Epidemiology, Shanghai
Cancer Institute, Shanghai, China
1986, 1989 Visiting Scholar, Department of Preventive Medicine, University
of Southern California, Los Angeles, California, USA
1992-1996 Research Fellow and graduate student in Ph.D. program in
Epidemiology of the Department of Preventive Medicine,
University of Southern California, Los Angeles, California, USA
Publications
Yuan J-M, Yu MC, Ross RK, Gao Y-T, Henderson BE. Risk factors for breast cancer in
Chinese women in Shanghai. Cancer Res 48: 1949-1453,1988.
Yuan J-M, Gao Y-T. The progress on research for the etiology of breast cancer (review).
Shanghai Tumor 6:316-318,1988. (in Chinese.)
Shu X-0, Brinton LA, Gao Y-T, Yuan J-M. Population-based case-control study of ovarian
cancer in Shanghai. Cancer Res 49: 3670-3674,1989.
Shu X-0, Gao Y-T, Yuan J-M, Zeigler RG, Brinton LA. Dietary factors and epithelial
ovarian cancer. B rJ Cancer 59: 97-101,1989.
Bernstein L, Yuan J-M, Ross RK, Pike MC, Hanisch R, Lobo R, Stanczyk F, Gao Y-T,
Henderson BE. Serum hormone levels in premenopausal Chinese women in Shanghai and
white women in Los Angeles: results from two breast cancer case-control studies. Cancer
Causes Control 1: 51-58,1990.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IX
Howe GR, Hirohata T, Hislop TG, Iscovich JM, Yuan J-M, Katsouyanni K, Lubin F,
Marubini E, Medan B, Rohan T, Toniolo P, Yu S. Dietary factors and risk of breast cancer:
combined analysis of twelve case-control studies. JN atl Cancer Inst 83:561-569,1990.
Yuan J-M, Gao Y-T. Epidemiology of breast cancer. In: Shen Z-Z (ed). Breast Cancer, pp.
21-57. Shanghai Scientific and Technical Press, Shanghai, 1990. (in Chinese.)
Yuan J-M, Gao Y-T. Prevention of breast cancer. In: Shen Z-Z (ed). Breast Cancer, pp.
480-487. Shanghai Scientific and Technical Press, Shanghai, 1990. (in Chinese.)
Yuan J-M, Gao Y-T. Risk factors for breast cancer related to the menopausal status at
diagnosis. Shanghai Tumor 11: 199-203,1991. (in Chinese.)
Ross RK, Yuan J-M, Yu MC, Wogan GN, Qian G-S, Tu J-T, Groopman JD, Gao Y-T,
Henderson BE. Urinary aflatoxin biomarkers and risk of hepatocellular carcinoma. Lancet
339: 943-946,1992.
Ross RK, Yu MC, Henderson BE, Yuan J-M, Qian GS, Tu J-T, Gao Y-T, Wogan GN,
Groopman JD. Aflatoxin biomarkers (Letter). Lancet 340: 119,1992.
Qian G-S, Ross RK, Yu MC, Yuan J-M, Gao Y-T, Henderson BE, Wogan GN, Groopman
JD. A follow-up study of urinary markers of aflatoxin exposure and liver cancer risk in
Shanghai, People's Republic of China. Cancer Epidemiol Biomarkers Prev 3:3-10,1994.
Yuan J-M, Wang Q-S, Ross RK, Henderson BE, Yu MC. Diet and breast cancer in
Shanghai and Tianjin, China. B rJ Cancer 71:1353-1358,1995.
Yuan J-M, Ross RK, Stanczyk FZ, Govindarajan S, Gao Y-T, Henderson BE, Yu MC. A
cohort study of serum testosterone and hepatocellular carcinoma in Shanghai, China. Int J
Cancer 63:491 -493, 1995.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT OF DISSERTATION
Professor Mimi C. Yu, Chairperson
In the mid-1980s, a prospective cohort study of 18,244 middle-aged men in
Shanghai, China, was initiated to investigate the etiology of cancer. Exposure
information and blood and urine samples were collected from each study participant. The
cohort was followed via annual visits. By February 1995,694 incident cancer cases and
1,157 deaths had been identified during 122,300 cumulative person-years of follow-up.
Smoking was a risk factor for both morbidity and mortality in middle-aged men in
Shanghai. Compared with lifelong nonsmokers, the relative risks in heavy smokers ( ^ 0
cigarettes/day) were 2.2 for any incident cancer, 9.4 for lung cancer, and 6.7 for head and
neck cancer. Heavy smokers also exhibited an 1.6-fold excess risk for mortality overall,
and a 2- to 3-fold excess risk for fatal heart disease. Among the study subjects, 36% of
all cancer cases and 21% of all deaths could be attributed to cigarette smoking.
Light-to-moderate drinking was related to a reduced risk of overall mortality and
mortality due to ischemic heart disease. Compared with lifelong nondrinkers, a
statistically significant 19% reduction in overall mortality and significant 36% reduction
in ischemic heart disease mortality was observed in light-to-moderate (1-28 drinks/week)
drinkers. This protective effect was not restricted to any specific types of alcoholic
beverages.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
XI
A nested case-control study within the cohort was conducted to examine the
association between serum testosterone and hepatocellular carcinoma. Seventy-six
incident cases of hepatocellular carcinoma and 410 matched control subjects were
assessed for serum testosterone and serologic markers of hepatitis B and C infections.
The mean level of testosterone in hepatitis B surface antigen (HBsAg)-positive controls
was 21% higher relative to HBsAg-negative controls and the difference was statistically
significant (two-sided p value [p]=0.0006). Relative to controls, hepatocellular
carcinoma cases had a significantly higher mean level of testosterone (570 vs. 485 ng/dl,
p=0.0005), but the difference was explicable on the basis of a higher proportion of
HBsAg-positive individuals among cases than controls (p=0.42 after adjustment for
HBsAg status).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Xll
PREFACE
I began my collaboration in cancer research with Drs. Brian E. Henderson, Ronald
K. Ross, and Mimi C. Yu of the Norris Comprehensive Cancer Center, University of
Southern California in 1986. As a graduate student in the Masters program in Cancer
Epidemiology at the Shanghai Cancer Institute during 1984-1986,1 designed, initiated
and subsequently completed a population-based case-control study of breast cancer
among Chinese women in Shanghai. I then spent 8 months here at the Norris
Comprehensive Cancer Center to analyze the study data under the supervision of Dr.
Mimi C. Yu in 1986. Meanwhile, Drs. Ross, Henderson and Yu were working with
investigators at the Shanghai Cancer Institute to start a prospective, epidemiologic study
of diet and cancer in middle-aged men in Shanghai.
Upon my return to Shanghai in the late 1986,1 began to implement the cohort
study in the city of Shanghai. As the project manager, I was in charge of training the
interviewers, setting up coding and processing rules for the study questionnaire, setting
up laboratory procedures to process the serum and urine samples collected from study
subjects, working with local government officials in cohort recruitment, maintaining the
database and serum/urine bank in Shanghai, and setting up the follow-up procedures.
Between 1986 and 1989, 18,244 middle-aged men were recruited into the cohort.
In addition to an in-person interview, we collected blood and urine samples from each
participant. Several hypotheses relating to the etiology of cancer have been tested.
Among them, results from this prospective study have provided the most direct evidence
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Xlll
for an etiological role of aflatoxin in hepatocellular carcinogenesis in human and
demonstrated a strong modifying effect of hepatitis B infection on the association
between dietary aflatoxin exposure and risk of hepatocellular carcinoma.
For my dissertation research, I have used data from this cohort study to examine
the following three issues: (1) morbidity and mortality in relation to cigarette smoking;
(2) mortality in relation to alcohol drinking; and (3) Serum testosterone level and
hepatocellular carcinoma risk in men.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 1 INTRODUCTION
The People’s Republic of China offers special opportunities for conducting
prospective cohort study aimed at understanding causes of cancer and other diseases.
Among the advantages are high participation rates, relatively low costs, ease to follow up
due to relative immobility of the population, and the opportunity to investigate the
etiology of cancers which are uncommon in the United States and other Westem
populations. Table 1-1 lists the age-standardized (world population) rates of eight
leading cancer sites in Chinese men of Shanghai alongside the rates in non-Hispanic-
sumamed white (white) men of Los Angeles during 1983-1987 (Parkin et al. 1992).
Cancers of the stomach, liver, and esophagus are more common in Chinese men than in
white men. The initial objectives of the cohort study were (1) to determine the
independent and interactive roles of hepatitis B virus (HBV), hepatitis C virus (HCV),
and aflatoxin exposure in the etiology of hepatocellular carcinoma (HCC); (2) to
determine the independent and interactive roles of salt intake, vitamin C deficiency,
intake of N-nitroso compounds and infection with the bacterium Helicobacter pylori (H.
pylori) in the etiology of stomach cancer; and (3) to determine the role of retinol, beta-
carotene, and other carotenoid deficiencies in the etiology of cancers of the lung and
stomach and to determine any interaction between smoking and deficiency of these
micronutrients in the etiology of these two cancers.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 1-1 Average annual age-standardized incidence rates of the eight commonest
cancers in Chinese men of Shanghai and non-Hispanic-sumamed white men of Los
Angeles County, 1983-198?'
Cancer site Chinese men
of Shanghai
(per 100,000)
Non-Hispanic-
sumamed white men
of Los Angeles
(per 100,000)
Lung 53 56
Stomach 52 7
Liver 31 2
Esophagus 15 4
Colon 9 27
Rectum 9 13
Bladder 7 18
Pancreas 6 7
' Source: Parkin et al., 1992.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3
Using the data of this cohort study, I have addressed the following three specific
issues:
(1) To prospectiveiy evaluate the effect of smoking on morbidity and mortality in
middle-aged Chinese men, in whom the smoking prevalence in the early 1980s is
similar to the peak rate of smoking in U.S. men during the 1950s;
(2) To prospectiveiy evaluate the relationship between alcohol intake and overall
mortality as well as cause-specific mortality in middle-aged Chinese men; and
(3) To determine the role of testosterone in the etiology of HCC in middle-aged
Chinese men, a high-risk population for this malignancy.
Background and Significance
1. Smoking and risk of cancer and death
Cigarette smoking is an important cause of morbidity and mortality in Westem
populations. This knowledge has evolved firom numerous epidemiological investigations,
which in general show great consistency in results (Centers for Disease Control, 1989).
These studies have established tobacco smoking as an important cause o f cancer,
cardiovascular disease, and chronic obstructive pulmonary disease in Westem
populations (Doll and Peto, 1976; Rogot and Murray, 1980; Hammond and Seidmen
1980; Carstensen et al., 1987; Baron et al., 1993; Freund et al., 1993; Paganini-Hill and
Hsu, 1994).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
Cigarette smoking is becoming increasingly common in China, especially among
men. In 1980,69% of Chinese men aged 20 years or above were current smokers
(Collaborative Group for Investigating Prevalence of Smoking, 1987), a rate that is
similar to the peak rate of smoking in U.S. men during the 1950s (Hammond and
Garfinkel, 1961; Centers for Disease Control, 1989). However, there has been scant
information on smoking habit and subsequent health risk in Chinese men. It is
particularly important to evaluate the cigarette smoking and cancer incidence and
mortality relationship in China, given the large population, the high prevalence of
smoking, and the potential for an expanded market for tobacco manufacturers in that
country.
2. Alcohol intake and mortalitv
It has long been recognized that alcohol can cause or is related to sudden death,
(ie., poisoning, accidents, or violence), and that long-term use o f alcohol can increase the
incidence of hepatic cirrhosis and of certain types of cancer. In recent years, however,
evidence has emerged that the regular consumption of small to moderate amounts of
alcohol may reduce ischemic heart disease (IHD) risk and overall mortality. In
prospective studies conducted in Westem populations, light-to-moderate drinkers of
alcoholic beverages of either sex have been found to have lower overall mortality rate
than either non-drinkers or heavy drinkers (Marmot et al., 1981; Kozarevic et al., 1982;
Colditz et al., 1985; Boffetta and Garfinkel, 1990; De Labry et al., 1992; Farchi et al.,
1992; Klatsky et al., 1992; Cullen et al., 1993; Doll et al., 1994; Gronbaek et al., 1994;
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
Fuchs et al., 1995; Serdula et al., 1995). The observed protective effect of drinking on
overall mortality is due largely to a reduced risk of fatal IHD which accounts for roughly
one-third of total deaths in such Westem populations. High consumption of wine has
been suggested to be responsible for the low risk for IHD in France (Renaud and De
Lorgeril, 1992) and recent findings from a prospective study in Copenhagen support this
hypothesis; wine intake but not intake of beer or hard liquor was found to be protective
against overall mortality in Danish people (Gronbaek et al., 1995).
In this cohort, IHD accounts for only 9% of all deaths (Yuan et al., 1996). Also,
grape wine is a minor source of ethanol among study subjects who drink regularly.
Therefore, our study population is a valuable one to further evaluate the relationship
between moderate alcohol intake and total mortality as well as IHD death.
3. Semm testosterone level and HCC
In all populations with reliable cancer statistics, males are at least 2 to 3 times
more likely to develop HCC than females (Parkin et ai., 1992). Using data from a cohort
study in Taiwanese men, Yu and Chen (1993) demonstrated a significant excess of HCC
cases with high serum testosterone at recruitment relative to control subjects. The
experimental data have shown that various species of female rodent were less susceptible
to the carcinogenic effect of aflatoxins relative to their male counterparts (Wogan, 1992).
Liver microsomes from several strains of male rat have been observed to form more
aflatoxin metabolites (including DNA adducts) than their female counterparts.
Furthermore, aflatoxin B, metabolism was reduced when microsomes from castrated
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
male rats were used, while testosterone treatment of castrated immature male and
immature female rats led to enhanced aflatoxin Bi metabolism by their microsomes
(Gurtoo and Motycka, 1976).
Data from this cohort of middle-aged men in Shanghai, a high-risk population for
HCC, enabled us to further evaluate the relationship between serum testosterone level and
subsequent risk of HCC after adjustment for HBV infection.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 2 MATERIAL AND METHODS
I. Study Design
This is a prospective cohort study in which all exposure information was collected
from study subjects before they developed clinical cancer or died. The study subjects
were actively followed on an annual basis to identify cases of cancer and death.
II. Study Population
Shanghai is a metropolitan city located on the central-eastern coast of China.
There are about seven million permanent residents in the city of Shanghai, which is
divided into 12 districts. Each of these districts contains a number of “street
committees.” Each street committee is consisted of a number of “neighborhood
committees.”
The Shanghai Cancer Registry, a population-based registry covering the seven
million residents of Shanghai city and administratively a part of the Department of
Epidemiology, Shanghai Cancer Institute, was established in 1963. The registry was
started as a result of regulation on cancer notification issued by the Shanghai Municipal
Bureau of Public Health. According to this regulation, each medical facility in Shanghai
is required to report all new cases of cancer. Notification forms completed by physicians
or medical clerks are routinely forwarded to the cancer registry. In addition, each district
of the city has its own Vital Statistics Unit. The cancer registry together with district vital
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8
statistics units are sources which we have used to identify incident cases of cancer and
death occurring among cohort members.
1. Identification of target population
We selected four small, geographically defined communities which scatter over a
wide area of the city of Shanghai (northeast, northwest, central, and south-central areas)
as our target population. We relied on a Street Committee-based network to establish the
cohort. The “street” hospitals provided us with temporary offices to enroll study subjects
and the “street” police stations provided us with complete rosters of all residents living
within their boundaries. A list of name and address of all men aged 45-64 years who
were residents of the four selected communities were abstracted manually from these
rosters.
2. Selection criteria of studv subjects
Men aged between 45 and 64 years during 1986-1989, who registered as
permanent residents in the selected communities, were eligible subjects for our cohort
study. Based on the seleted street hospitals, we invited all eligible men to participate in
the study. All men with a history of cancer were excluded from this cohort.
III. Data Collection
L-Qpestionnairs
A structured questionnaire (Appendix A) was used to acquire and record interview
information from participants. The questionnaire included the following sections.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
(1) Background; Name, gender, present address, home telephone number,
workplace, work telephone number, date of birth, name of interviewer, and date and
duration (in minutes) of interview.
(2) General Information: Birthplace, age began living in Shanghai, marital status,
level of education, size of household, monthly income of household, height, usual adult
weight, and current and usual occupations.
(3) Use of Tobacco: A participant was asked whether he had ever smoked at least
one cigarette a day for six months or longer. If the answer was yes, the respondent was
asked to provide the age at which he started to smoke regularly, the number of years he
had smoked, and the number of cigarettes he usually smoked in a day. If he had stopped
smoking, age at which he quit the habit was recorded. Similar questions were asked
about tobacco smoking by a water or dry pipe.
(4) Use of Alcohol: A participant was asked whether he had ever drunk alcoholic
beverages at least once a week for six months or longer. If the answer was yes, the
subject was asked to provide the age at which he started to drink regularly, the number of
years he had been drinking regularly, and the usual amounts of beer, wine, and hard
liquor he consumed in a week. If he had stopped drinking, the age at quitting was
recorded.
(5) Diet: Frequencies of intake of 45 common food items, which included com,
peanuts, rice, various sea foods, meats, eggs, dark green leafy and other vegetables, fresh
fruits, and different preserved foods. For each food item, the usual number of times per
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
day, week, month or year the subject consumed this food during adulthood was asked.
For a seasonal food item, the intake frequency when the food was in season and the
number of months the food was in season were recorded.
(6) Medical History: A participant was asked whether he was ever told by a
physician to have had any one of 20 illnesses, including liver disease, diabetes, gastric
and duodenal ulcer, gallbladder disease, tuberculosis, high blood pressure, asthma,
emphysema, hemorrhoid, diverticulosis, polyposis coli, ulcerative colitis, adenomatous
polyps, schistosomiasis and other parasitic diseases, and cancer. For each of these listed
illnesses, age when the illness was first diagnosed and methods used to treat the illness
were recorded.
2. Interview
Retired nurses were hired by the Shanghai Cancer Institute to conduct in-person
interviews and to collect blood and urine samples. Each nurse had to pass the 2-week
training course before she started field work.
Before the interview began, the participant was asked to sign an informed consent
form (Appendix B). The interview then was administered by the nurse using a structured
questionnaire (Appendix A).
Because most o f the participants in this cohort study were employed during the
day, the interviews were mainly conducted between 4:30 p.m. and 9:30 p.m. in
participants’ homes or our local offices.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11
3. Collection of blood and urine samples
At the completion of the interview, the nurse collected a 10 ml blood sample
using a vacutainer and auto-separator blood tube. The tube was labeled with a
preassigned and preprinted study number and placed on an ice pack in a styrofoam
container. In addition, the interviewee was asked to provide a urine specimen in a 500 ml
prelabeled container.
4. Processing ofblood and urine samples
Each evening at the completion of her daily workload (about 3-4 interviews), the
nurse had to send all blood and urine specimens directly to the Institute or to our local
office to be transferred to the Institute at about 10 p.m. Laboratory technicians
centrifuged and stored 1 x 2 ml serum sample at -70 °C and 2 x 1.5 ml serum samples at
-20 °C and refrigerated the urine specimens for processing in the next morning.
Urine samples were aliquoted into 2 vials with 8 ml untreated urine for each vial
and 1 vial with 25 ml alkalinized urine and stored at -20 °C. Twenty-five ml of the
remaining urine sample was centrifuged for 10 minutes and the supernatant transferred to
a 50 ml beaker. Two hundred ul of a tritiated chromatographic standard in 50% ethanol
and 1.25 ml of methanol was added to the urine. The urine sample was then loaded into a
syringe and pushed through the C,g Sep-Pak cartridge at a flow rate of 2-3 ml per minute.
Afterwards, a 10 ml solution of 5% methanol/95% water was loaded into the syringe and
pushed through the Sep-Pak at a comparable flow rate. The Sep-Pak was placed back
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
into its foil packet, sealed with tape to prevent drying, and stored at 4 °C. All remaining
urine was discarded.
IV. Database Establishment and Management
Questionnaires were checked for completeness and coded by the interviewers.
The non-precoded information (such as birthplace, occupation, etc.) was coded and any
special situations regarding coding was written in a code book by a coder. The
questionnaires then were rechecked manually by a data manager for internal consistency.
If necessary, respondents were recontacted by the interviewer to clarify responses and/or
resolve inconsistencies. Information on a questionnaire was then entered into personal
computers by two different data clerks. The two computer files were compared by
computer programs to correct keypunching errors. After the errors were corrected, the
internal consistency of the questionnaire data was checked again by a computer program.
Any errors found would be resolved and then corrected.
All serum and urine storage tubes and Sep-Paks were prelabeled with each study
subject’s identification number. The identification numbers were entered in the master
log book with locations of the stored serum and urine samples.
The freezers storing serum and urine samples were checked twice a day by a
laboratory technician and each time, the temperature of the freezer was recorded. An
alarm was installed to monitor the electricity supply and freezer function. Whenever
power supply from the city was cut off, a backup generator provided electrical supply to
the freezers.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
V. Follow-up
The cohort was followed for occurrence of cancer and death in a variety of ways:
Cancer notifications were obtained from the population-based cancer registry.
Cancer notifications contain name and address, date of birth, cancer site, subsite (if any),
histological type, date and basis of diagnosis, name of attending physician and the name
of medical facility to which the physician affiliates. Linkage between the cohort files and
cancer registry records was conducted every three months by computer program as well
as manually.
Copies o f death certificates on men who were residents of the communities
targeted for our cohort were routinely ascertained from the Vital Statistics units of the
districts. Death certificates contain name and address, date of birth, date and cause of
death, and if death occurred in a hospital, the name of the hospital. The death certificates
were matched against the master file of the cohort by computer linkage and manually.
As a final means of maximizing identification of new cancers and deaths among
cohort members and of minimizing losses to follow-up, we attempted to recontact each
surviving cohort member on an annual basis. This was accomplished by retired nurses,
who visited the last known address of each living cohort member and recorded details of
the interim heath history of the cohort member (Appendix C). For cohort members who
had moved, information was sought on new address from relatives or neighbors; for
individuals who had returned to their respective places of origin outside Shanghai after
retirement, information on health status was sought by mail, from relatives, or from
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
employers who were paying retirement pensions. For persons who failed to be located by
these means, we attempted to obtain residence and health-related information from the
local police department.
We also attempted to review the medical records o f ^1 cancer patients and deaths
occurring among cohort members from various clinics and hospitals to verify the
diagnosis of cancer and death.
VI. Statistical Analysis
1. Cohort analyses
Classical methods for cohort analysis were used. Pot oach individual, man-years
of follow-up were counted from the date of recruitment to tb® date of cancer diagnosis,
death or loss to follow-up, whichever occurred first. Cancer incidence or mortality rates
were calculated as the number of cases divided by cumulative man-years. At a given
exposure level, the age-adjusted rate was calculated as sum of the product of the age-
specific rate and the proportion of cumulative man-years w ith^ that age subgroup
relative to the entire cohort which served as the internal standard.
Poisson regression method was used to examine the relationship between
exposure status at recruitment and subsequent risk of cancer/death among cohort
members (Breslow and Day, 1987). Relative risks and their corresponding 95%
confidence limits were computed for various levels of exposure in reference to the non
exposure group. To control for potential confounding and/of modifying effect, covariates
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15
representing various strata of the confounders were added to the Poisson regression
model.
2. Case-Control Analysis
A nested case-control study design was used to examine the relationship between
serum testosterone level and risk of hepatocellular carcinoma (HCC). Controls were
matched to each case of HCC by age (within one year), time ofblood sample collection
(within one month) and neighborhood of residence. The conditional logistic regression
method was used to examine the relationship between HCC and markers of HBV/HCV
infection and serum testosterone level (Breslow and Day, 1980). Odds ratio for matched
sets was computed to estimate the relative risk for HCC for subjects at various exposure
categories in reference to the non-exposure group. The analysis of covariance method
was used to compare testosterone levels across various categories of subjects as defined
by their hepatitis serum marker and case/control status while controlling for age a:
recruitment (Winer, 1971).
In both cohort and case-control analyses, 2-sided tests were used to test the
hypotheses of interest. For a given test, two-sided p values under 0.05 were considered to
reject the null hypothesis.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
CHAPTER 3 CHARACTERISTICS OF THE STUDY POPULATION
Between January 1986 and September 1989, we completed interviews on 18,244
middle-aged male residents of four small communities in the city of Shanghai. We also
collected blood and urine samples from each study participant. Half of the study subjects
were from the central area of the city (Jungian District). The remaining subjects were
from the northeast (Hongkou District, 22%), the northwest (Putuo District, 17%), and the
south-central areas (Luwan District, 11%) of Shanghai (table 3-1). About 80% of eligible
subjects enrolled in the study.
I. Demographic Parameters
The cohort members are characterized according to demographic parameters
(table 3-2). The age distribution at study entry was shifted slightly toward older ages;
57% of these men were age 55 or older and only 43% under age 55 at recruitment. Forty-
one % of the cohort was bom in the metropolitan area of Shanghai. As expected, the
cohort was relatively uneducated by Western standards, with nearly 30% of cohort
members having either no formal education or at primary school level. Nineteen % of the
cohort had a college education or above (table 3-2).
Twenty-seven % of the cohort men reported that their usual occupation was
production and construction work, and 21% reported professional or technical jobs as
their usual occupation. As expected, Chinese men (mean height of 169 cm and mean
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17
Table 3-1 Study population by residential areas at recruitment, Shanghai Cohort Study
1986-1989
Number %
Total number of subjects 18,244 100.0
Residential areas in Shanghai
Central area (Jingan District) 9,220 50.5
Northeast area (Hongkou District) 3,945 21.6
Northwest area (Putuo District) 3,009 16.5
South-central area (Luwan District) 2,070 11.4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18
Table 3-2 Demographic characteristics at recruitment among middle-aged men, Shanghai
Cohort Study 1986-1989
Characteristics Number %
Age (year)
<44 420 2.3
45-49 3,062 16.8
50-54 4,390 24.1
55-59 5,323 29.2
60-64 4,718 25.9
>65 331 1.7
Mean 55.3
Birthplace
Shanghai 7,403 40.6
Jiangsu Province 5,688 31.2
Zhejiang Province 3,549 19.5
Other places 1,604 8.7
Level of education
No formal schooling 880 4.8
Primary school 4,320 23.7
Junior middle school 5,330 29.2
Senoir middle school 3,205 17.6
Technical institute 1,077 5.9
College or above 3,432 18.8
Marital status
Married 17,240 94.5
Widowed 538 2.9
Divorced 240 1.3
Single 226 1.3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 3-2 (continued)
19
Characteristics Number %
Usual occupation
Production/construction worker 4,876 26.7
Professional/technical worker 3,846 21.1
Sales/service worker 2,684 14.7
Clerical worker 2,456 13.5
Government administrator 1,889 10.4
Stationary/transportation engine operator 1,000 5.5
All others 1,493 8.1
L Tsual adult weight (kg)
^ 9 714 3.9
50-54 2,275 12.5
55-59 3,223 17.7
60-64 4,128 22.6
65-69 3,212 17.6
70-74 2,455 13.5
75-79 1,243 6.8
>80 994 5.4
Mean 63.1
Adult height (cm)
<159 729 4.0
160-164 2,802 15.4
165-169 6,111 33.5
170-174 5,984 32.8
175-179 2,139 11.7
>180 479 2.6
Mean 168.7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
Table 3-2 (continued)
Characteristics Number %
Body mass index (kg/m^)
<20 4,560 25.0
20-<22 4,643 25.4
22-<24 4,298 23.6
24-<26 2,890 15.8
26-<28 1,245 6.8
^ 8 608 3.4
Mean 22.2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
body mass index of 22 kg/m^) were both shorter and leaner than U.S. men.
II. Tobacco Use
Men who had smoked at least one cigarette a day for six months or longer were
defined as cigarette smokers. Among 18,244 men, 50% were current smokers and 7%
were ex-smokers. Among ever smokers, 2% started smoking under age 15 years whereas
9% began to smoke regularly at £^e 40 or older (mean age = 25). The average daily
“dose” of cigarettes was 16 cigarettes a day. Most smokers had been smoking for many
years (table 3-3). Only 2% of study subjects ever smoked a pipe, and all pipe smokers
also were cigarette smokers.
III. Alcohol Use
A man who had ever drunk alcoholic beverages at least once a week for six
months or longer was defined as a regular drinker of alcohol. Among cohort men, 43%
were regular drinkers; almost all of them were current drinkers at recruitment (table 3-4).
Among ever drinkers, the mean age at starting to drink was 30 years; 17% started their
drinking habit before age 20 and 25% began to drink weekly at age 40 or older. Fifteen
% of drinkers had drunk for less than 10 years and 17% for 40 or more years, with a mean
of 26 years. On average, regular drinkers consumed 17 drinks per week (one drink was
defined as 360 g of beer, 103 g of wine, or 30 g of hard liquor) (Adames, 1975); one-third
of them drank 7 or less drinks per week and 17% consumed 28 or more drinks per week.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
Table 3-3 Tobacco use at recruitment, Shanghai Cohort Study 1986-1989
Number %
Cigarette smoking
Never 7,787 42.7
Ever 10,457 57.3
Ex-smoker 1,256 6.9
Current smoker 9,201 50.4
Ever smoker 10,457 100
Age started to smoke regularly (year)
<14 216 2.0
15-19 2,209 21.1
20-24 3,399 32.5
25-29 1,883 18.0
30-34 1,266 12.1
34-39 579 5.5
>40 905 8.8
Mean 25.2
Number of years of smoking cigarette
<9 562 5.4
10-19 1,114 10.7
20-29 3,117 29.8
30-39 3,567 34.1
>40 2,097 20.0
Mean 29.6
Number of cigarettes/day
<9 1,665 15.9
10-19 3,559 34.0
20-29 4,323 41.3
^ 0 910 8.8
Mean 16.2
Ever smoked a pipe 321 1.8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 3-4 Alcohol use at recruitment, Shanghai Cohort Study 1986-1989
23
Number %
Alcohol drinking
Never 10,471 57.4
Ever 7,773 42.6
Ex-drinker 383 2.1
Current drinker 7,390 40.5
Ever drinker 7,773 100
Age started to drink regularly (year)
<19 1,318 17.0
20-29 2,940 37.8
30-39 1,549 19.9
40-49 1,160 14.9
>50
Mean
806
29.7
10.4
Number of years of drinking
<9 1,157 14.9
10-19 1,216 15.6
20-29 1,923 24.7
30-39 2,127 27.4
>40
Mean
1,350
26.0
17.4
Number of drinks/week
1-7 2,473 31.8
8-14 1,892 24.3
15-21 1,128 14.5
22-28 991 12.7
29-42 752 9.7
>43
Mean
537
16.9
7.0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
Table 3-4 (continued)
Number %
Daily ethanol intake (g)
<10 2,000 25.7
10- <20 1,635 21.0
20- <30 1,024 13.2
30- <40 925 11.9
40- <50 890 11.4
^ 0 1,299 16.8
Mean 30.7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
In terms of daily ethanol intake, one-fourth of regular drinkers reported consuming less
than 10 g of ethanol per day while 17% of drinkers consumed 50 g or more of ethanol per
day. The mean daily intake of ethanol for ever drinkers was 31 g (table 3-4).
rv. Medical History
A positive medical history was defined as having the illness diagnosed by a
medical doctor. The proportions of cohort men who reported a positive history of 20
selected medical conditions are presented in table 3-5. Among cohort members, 25%
reported a history of hypertension, 17% of tuberculosis, and 12% of hepatitis.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
26
Table 3-5 Medical history at recruitment, Shanghai Cohort Study 1986-1989
Number %
Ever diagnosed by a medical doctor of:
High blood pressure 4,492 24.6
Tuberculosis 3,132 17.2
Emphysema 769 4.2
Asthma 708 3.9
Hepatitis 2,236 12.3
Hepatic cirrhosis 206 1.1
Other liver disease 493 2.7
Gallstone 658 3.6
Other gallbladder disease 477 2.6
Gastric ulcer 1,738 9.5
Duodenal ulcer 2,097 11.5
Diverticulosis 58 0.3
Polyposis 42 0.2
Ulcerative colitis 136 0.7
Adenomatous polyps 31 0.2
Diabetes 235 1.3
Schistosomiasis 584 3.2
Other parasitic disease 511 2.8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
CHAPTER 4 MORTALITY AND MORBIDITY IN RELATION TO CIGARETTE
SMOKING*
Summary
To evaluate prospectively the health risk of cigarette smoking in middle-aged men
in Shanghai, China, data from a prospective cohort study involving 18,288 male residents
in the city o f Shanghai were analyzed. The follow-up of this cohort was accomplished
through routine review of the population-based Shanghai Cancer Registry, routine review
of death certificates, and annual recontact of all surviving cohort members. By
September 30, 1993, 852 deaths and 554 incident cancer cases were identified during the
follow-up period which averaged about 5.4 years per subject. The overall incidence rate
for cancer was 568 per 100,000 man-years with the three leading sites being lung
(146/100,000), stomach (116/100,000), and liver (81/100,000). Forty-one percent of all
deaths were due to cancer. Stroke was the most frequent cause of death unrelated to
cancer, with an age-adjusted rate that was 4.2 times higher than that of U.S. white men
(201 versus 48 per 100,000), followed by ischemic heart disease, with an age-adjusted
rate one-fifrh that of U.S. white men (69 versus 366 per 100,000). Compared with
lifelong non-smokers, the relative risks in heavy smokers (20 or more cigarettes per day)
after adjustment for alcohol consumption were 2.2 for any incident cancer, 9.4 for
incident lung cancer, 6.7 for head/neck cancer, and 1.8 for liver cancer. In terms of
*This paper is in press in the Journal of American Medical Association.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
mortality, heavy smokers were at a 60% greater risk of death relative to lifelong non-
smokers; there was a 2.3-fold excess risk of death from cancer and 2 to 3 fold excess risk
of death from heart disease. In conclusion, cigarette smoking is an important predictor of
risk of cancer and mortality in Shanghainese men. Among the study subjects, 36% of all
cancer cases and 21% of all deaths could be attributed to cigarette smoking.
1 . Introduction
Cigarette smoking is an important cause of morbidity and mortality in Western
populations. This knowledge has evolved from numerous epidemiological investigations,
which in general show great consistency in results (Centers for Disease Control, 1989).
These studies have established tobacco smoking as an important cause of cancer,
cardiovascular disease, and chronic obstructive pulmonary disease (COPD) in Western
populations (Hammond, 1966; Kahn, 1966; Doll and Peto, 1976; Hammond and Seidmen
1980; Rogot and Murray, 1980; Garfinkel, 1985; Carstensen et al., 1987; Centers for
Disease Control, 1989; Baron et al., 1993; Freund et al., 1993; Pagnini-Hill and Hsu,
1994). In the United States in 1990, approximately 20% o f all deaths (7% from cancer,
8% from cardiovascular disease, and 4% from COPD) could be attributed to cigarette
smoking (Centers for Disease Control, 1989).
Cigarette smoking is becoming increasingly common in China, especially among
men. In 1984, 69% of Chinese men aged 20 years or above were current smokers
(Collaborative Group for Investigating Prevalence of Smoking, 1987), a rate that is
similar to the peak rate of smoking in U.S. men during the 1950s (Hammond and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
Garfinkel, 1961; Centers for Disease Control, 1989). Peto and his colleague has
estimated that about 2 million tobacco-related deaths (including 900,000 from lung
cancer) will occur in China in the year 2025 (Peto, 1986; Novomy and Peto, 1988).
There has been scant information on smoking habits and subsequent health risk in
Chinese men. In this report, we present rates of cancer incidence and cause-specific
mortality according to smoking status of a cohort of 18,244 middle-aged men in
Shanghai, China.
II. Subjects and Methods
Between January 1986 and September 1989, we invited all eligible male residents
of four small, geographically defined communities scattered over a wide area of the city
of Shanghai to take part in a prospective, epidemiological study of diet and cancer. The
eligibility criteria were ages 45-64 years and no history of cancer. In addition to
collecting a 10 ml blood sample and a single void urine sample, we interviewed each
subject in person using a structured questionnaire which included level of education,
usual occupation, current height and weight, history of tobacco and alcohol use, current
diet (intake frequencies of 45 food items), and medical history. During the 3-year
recruitment period, 18,244 men (about 80% of eligible subjects) took part in the study.
The prevalence of cigarette smoking among cohort members was virtually identical to
that reported from a city-wide survey conducted during 1982 (Deng and Gao, 1985), thus
providing evidence that our cohort is representative of the general population of Shanghai
men in terms of cigarette smoking habits.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
Details of the follow-up procedures for this cohort have been described previously
(Ross et al., 1992). We have conducted routine review of death certificates and cancer
reports from the population-based Shanghai Cancer Registry, and all surviving members
of the cohort have been contacted on an aimual basis. By September 30, 1993 (the cut-off
time point for this analysis), only 50 subjects have been lost to follow-up.
In addition to overall cancer incidence rates, age-adjusted incidence rates for
cancer sites with more than 15 cases were calculated for all subjects as well as for the
following cigarette smoking categories: lifelong non-smokers, ever smokers, ever
smokers of less than 20 cigarettes per day, and ever smokers of 20+ cigarettes per day
(due to the extremely small number of ex-smokers in the study population, ex- and
current smokers could not be analyzed separately). The same algorithm was applied to
the calculation of mortality rates. All rates were adjusted for age using the man-year
distribution of the entire cohort as the internal standard (the proportions of total man-
years in the age groups <55, 55-59, 60-64, and 65+ years were 0.28,0.28, 0.28, and 0.16,
respectively). As a comparison, cancer incidence rates in non-Hispanic white males of
Los Angeles and mortality rates in U.S. white males (National Center for Health
Statistics, 1993) are presented in this paper alongside the Shanghai rates.
We used the Poisson regression method (Breslow and Day, 1987) to examine the
relationship between cigarette smoking status at recruitment and subsequent risk of
cancer/death among cohort members. Relative risks (RRs) and their corresponding 95%
confidence limits (CLs) for cancer/death were computed for various categories of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
smokers in reference to lifelong non-smokers with and without adjustment for alcohol
consumption (among cohort subjects, cigarette smoking was positively correlated with
alcohol consumption. Fifty-five percent of non-drinkers were lifelong non-smokers while
only 15% of heavy drinkers—40+ g per day of ethanol intake—had never smoked
cigarettes regularly). There was no substantive change between unadjusted and adjusted
RRs. Only the adjusted RRs are presented in this report. The computation of population
attributable risks and 95% CLs used the methods as described in Leung and Kupper
(1981). All quoted p values are 2-sided. RRs with 2-sided p values under 0.05 were
considered to be significantly different fi'om 1.0.
III. Results
At recruitment, 50% of study subjects were current smokers, half of whom
smoked 20 or more cigarettes per day. Few subjects were ex-smokers (7%). The
remaining 43% of cohort members had never smoked cigarettes regularly. Among
current smokers, 11% started smoking under age 18 years, and 46% began the habit at
age 25 years or older. On average, current smokers had smoked for 29 years and smoked
about 16 cigarettes per day. Smokers who started smoking at an early age tended to be
heavier smokers. Among current smokers who began smoking under age 20 years, 65%
smoked 20 or more cigarettes each day. In contrast, only 35% of current smokers who
began smoking after age 29 years smoked 20 or more cigarettes per day. Ex-smokers
were comparable to current smokers in terms o f age starting to smoke regularly, number
of cigarettes smoked per day and duration of smoking. On average, ex-smokers gave up
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
32
smoking 8 years prior to study enrollment. Frequency distributions o f current smokers
were similar across the different age groups (table 4-1).
As of September 30, 1993, cohort members had contributed 98,267 man-years of
observation (an average of 5.4 years of follow-up per subject). Five hundred fifty-four
incident cancer cases and 852 deaths had occurred among study subjects. We attempted
to review the medical records of all cancer cases and deaths from various clinics and
hospitals to verify the diagnosis of cancer and death. O f the 554 incident cancer cases,
67% were confirmed histopathologically; only 1% of the diagnoses were based solely on
clinical observations. Of the 197 stroke deaths, 82% had been hospitalized prior to death,
while no evidence of medical care could be established for 9% of these patients. The
comparable figures for ischemic heart disease were 79% and 14%, respectively.
Table 4-2 presents the age-adjusted incidence rates for various cancers among
cohort members along with the comparable rates in Los Angeles non-Hispanic white
(white) men. The overall rate of cancer in the cohort (568; 95% CL = 520, 615 per
100,000) was about half the corresponding rate in Los Angeles whites (1,094/100,000).
Lung (146; 95% CL = 122,169 per 100,000) was the most common cancer site among
study subjects, followed by stomach (116; 95% CL = 94, 137 per 100,000) and primary
liver (81; 95% CL = 63, 99 per 100,000) whereas in Los Angeles whites, prostate
(279/100,000), lung (213/100,000), and colon (85/100,000) were the three most common
cancers, respectively. Rates in Shanghai Chinese were 2 to 8 fold higher than in Los
Angeles whites for cancers of the esophagus, stomach and liver. In contrast, the prostate
cancer incidence rate in Shanghai Chinese (7/100,000) was only 2.5% that in Los
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
C D
■ D
O
Q .
C
g
Q .
■ D
CD
Table 4-1 Number of subjects (%) by age and cigarette smoking status at recruitment
o'
3
O
8
( O '
3 "
i
3
CD
3.
3 "
CD
CD
" O
O
Q .
O
3
■ D
O
Age (years) Non-smokers Ex-smokers
Current smokers,
<20 cigarettes
per day
Current smokers,
20+ cigarettes
per day
All
subjects
<50 1,604(46) 118(3) 970(28) 790 (23) 3,482 (100)
50-54 1,883 (43) 212(5) 1,140 (26) 1,155 (26) 4,390(100)
55-59 2,260 (42) 379 (7) 1,314(25) 1,370 (26) 5,323 (100)
60+ 2,040 (40) 547(11) 1,206(24) 1,256 (25) 5,049(100)
Total 7,787 (43) 1,256 (7) 4,630 (25) 4,571 (25) 18,244(100)
CD
Q .
■ D
CD
C /)
C /)
W
w
■o
1
1
1
■ g
3
C/)
C/)
o'
3
2 ,
CO
8
Table 4-2 Age-adjusted^ cancer incidence rates and relative risks (RRs) by cigarette smoking status at recruitment
Cancer sites (ICD-9)
No. of
cancer
cases
Non-
smokers
Incidence rates (per 100.0001
Smokers, Smokers,
Ever- <20 cigarettes 20+ cigarettes
smokers per day per day
All
subjects
Incidence rates
of non-Hispanic
white males of
Los Angeles'*
C O
s
1 All sites (140-208) 554 369 709 588 825 568 1094
CD
Adjusted R R ® 1.0 1 9 ' 1.6* 2.2*
n
c
p.
I T
Head/neck (140-149,161) 28 10 43 32 53 29 72
CD
O
Adjusted R R * ^ 1.0 5.2* 3.8* 6.7*
■o
1 Esophagus(150) 24 17 30 21 38 25 13
a
O
Adjusted R R * ^ 1.0 1.4 1.0 1.7
■o
o
Stomach (151) 113 97 130 133 128 116 23
I T
CT
O
Adjusted R R ® 1.0 1.4 1.4 1.3
Û .
Colon (153) 26 25 28 40 17 27 85
?
o
5 -
Adjusted R R * ^ 1.0 1.1 1.5 0.7
i
Rectum (154) 31 41 25 22 27 32 41
J h '
( / )
o'
Adjusted RR*' 1.0 0.6 0.5 0.7
3
Liver(155) 79 59 96 97 95 81 10
Adjusted R R * ^ 1.0 1.8* 1.8* 1.8*
C D
■ D
O
Q .
C
g
Q .
■a
CD
C/)
w
o'
3
0
3
CD
8
( O '
3 "
1
3
CD
" n
c
3.
3 "
CD
CD
■ D
O
Q .
C
a
o
3
■ D
O
CD
Q .
■ D
CD
C /)
C /)
Table 4-2 (continued)
Cancer sites (ICD-9)
No. of
cancer
cases
Incidence rates (ner 100.000) Incidence rates
of non-Hispanic
white males of
Los Angeles'*
Non-
smokers
Ever-
smokers
Smokers,
<20 cigarettes
per day
Smokers,
20+ cigarettes
per day
All
subjects
Pancreas (157) 21 15 26 21 31 22 25
Adjusted R R * ^ 1.0 1.8 1.5 2.1
Lung (162) 142 32 226 119 331 146 213
Adjusted R R * ^ 1.0 6.5* 3.6* 9.4*
Bladder(188) 16 12 19 22 17 16 53
Adjusted R R ® 1.0 1.9 2.1 1.7
All others 74 62 86 82 89 76 550
Adjusted R R ® 1.0 1.4 1.4 1.4
“ To the man-year distribution of the entire cohort (<50, 55-59, 60-65, 65+ years).
^ Age-adjusted rates per 100,000 in non-Hispanic white males (aged 45-69 years) of Los Angeles County, California, 1988-
1992. Age-adjustment was according to the man-year distribution of the Shanghai cohort.
RRs were adjusted for age and alcohol consumption (non-drinker, <20 g and 20+ per day of ethanol intake).
* 2-sided p values less than 0.05, test for RR = 1.0.
L / \
36
Angeles whites. Cancers of the head/neck (including mouth, nose, pharynx and larynx),
colon, and bladder were relatively uncommon in Shanghai (one-third to one-half the
corresponding rates in Los Angeles whites). Rates between Shanghai Chinese and Los
Angeles white men were comparable for cancers of the rectum and pancreas. Lung
cancer incidence in Shanghai Chinese was 69% that in Los Angeles white men. We
examined cancer incidence rates among study subjects in relation to their duration of
follow-up, and noted no difference in incidence rates between groups with varying years
of follow-up.
Cigarette smoking was significantly associated with an increase in risk for cancer
overall (RR = 1.9; 95% CL = 1.6,2.3), and for cancers of the lung (RR = 6.5; 95% CL =
3.6, 11.6), head/neck (RR = 5.2; 95% CL = 1.8,15.1), and liver (RR = 1.8; 95% CL =
1.1,3.0). Non-significant increases in risks were noted for cancers of the bladder (RR =
1.9; 95% CL = 0.8,2.3), pancreas (RR = 1.8; 95% CL = 0.7,4.8), esophagus (RR = 1.4;
95% CL = 0.6, 3.6), and stomach (RR =1.4; 95% CL = 0.9, 2.1). A significant dose-
response relationship (ie., increasing risk with increasing amount smoked) was observed
for cancer overall, and for cancers of the lung and head/neck; a non-significant dose-
response relationship was noted for cancers of the esophagus and pancreas (table 4-2).
We also examined cancer mortality rates in relation to cigarette smoking status at
recruitment. The results were comparable to those based on cancer incidence (data not
shown).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37
For cancers of the lung, head/neck and primary liver, we further examined the
possible relationship between risk and age at starting to smoke after controlling for
amount smoked per day and amount of alcohol consumed per day. We noted a
significant increase in risk of lung cancer in those who began smoking at younger age.
Relative to smokers who began the habit at age 25 years or older, those who smoked
comparable amounts but began smoking at an earlier age experienced a 2.4-fold increase
in lung cancer risk (p = 0.0001). For cancers of the head/neck and primary liver, the
comparable risk ratios were 1.3 (p = 0.51) and 1.5 (p = 0.14), respectively.
Table 4-3 shows the age-adjusted mortality rates for the cohort along with the
comparable rates in U.S. white men. The overall mortality rate in the cohort (867; 95%
CL = 809,925 per 100,000) was 61% that in U.S. whites. Cancer accounted for 41% of
all deaths. Stroke (n = 197) was the most frequent cause of death unrelated to cancer,
accounting for 23% of deaths. The age-adjusted stroke mortality rate was 4.2 times
higher than that in U.S. white men (201 versus 48 per 100,000). The second most
frequent cause of non-cancer deaths in the cohort was ischemic heart disease (n = 68);
however, the age-adjusted mortality rate was only one-fifth that in U.S. white men (69
versus 366 per 100,000). Mortality related to other forms of heart disease also was
uncommon compared to the U.S. white experience (35 versus 120 per 100,000). Mortality
from injuries or accidents, another important category of deaths, also was less common in
Shanghai Chinese than U.S. whites (27 versus 81 per 100,000).
We examined mortality rates among cohort members in relation to their duration
of follow-up, and noted a deficit in the number of total deaths during the first year of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
C D
■ D
O
Q.
C
8
Q.
■ D
CD
Table 4-3 Age-adjusted“ mortality rates and relative risks (RRs) by cigarette smoking status at recruitment
C/)
C/)
8
ci'
Cause of death (ICD-9)
All causes
Adjusted RR*'
No. of
deaths
___________Mortality rates (per 100.000)__________________ Mortality
Smokers, Smokers, rates of
Non- Ever- <20 cigarettes 20+ cigarettes All U.S. white
smokers smokers per day per day
852 707
1.0
981
1.4*
831
1.2 *
1127
1.6*
subjects males
867 1410
3
3 "
CD
Cancer (140-208)
Adjusted RR*'
346 230
1.0
438
1.9*
338
1.5*
533
2.3*
352 464
CD
■ D
O
Q .
C
a
o
3
" O
o
Stroke (430-438) 197 181 215
Adjusted RR*' 1.0 1.1
Ischemic heart disease (410-414) 68 49 82
Adjusted RR*' 1.0 2.0*
203
1.1
78
1.8
227
1.2
89
2 . 1*
201
69
48
366
CD
Q .
■ D
CD
C /)
C /)
Other heart diseases (391-398,
401,402,404,415-417,420-429)
Adjusted RR*'
COPD** (490-496)
Adjusted RR*'
34
53
22
1.0
45
1.0
44
2.4*
60
1.4
32
1.8
36
0.8
55
3.0*
82
2.0*
35
54
120
81
Liver cirrhosis (571)
Adjusted RR*'
32 46
1.0
23
0.5
32
0.7
13'
0.3
33 39
U )
00
CD
■ D
O
Q.
C
g
Q.
■ D
CD
C/)
W
o"
3
O
3
CD
8
ci'
O
3.
3 "
CD
CD
■ D
O
Q.
C
a
O
3
■ D
O
CD
Q.
■ D
CD
C/)
( / )
Table 4-3 (continued)
Cause of death (ICD-9)
No. of
deaths
Mortalitv rates foer 100.000') Mortality
rates of
U.S. white
m ales**
Non-
smokers
Ever-
smokers
Smokers,
<20 cigarettes
per day
Smokers,
20+ cigarettes
per day
All
subjects
Diabetes (250) 17 20 16 21 11' 17 28
Adjusted R R * ^ 1.0 0.9 1.2 0.6
Injuries/accidents (E800-E999) 26 20 32 28 36 27 81
Adjusted RR“ 1.0 1.4 1.4 1.5
Other causes 79 93 72 64 80 80 183
Adjusted R R ® 1.0 0.8 0.7 0.8
“ To the man-year distribution of the entire cohort (<55, 55-59, 60-64, 65+ years).
* * Age-adjusted rates per 100,000 in U.S. white males (aged 45-69 years) in 1989 (National Center for Health Statistics, 1993).
Age-adjustment was according to the man-year distribution of the Shanghai cohort.
RRs were adjusted for age and alcohol consumption (non-drinker, <20 g and 20+ g per day of ethanol intake).
* * COPD: chronic obstructive pulmonary disease, ie., chronic bronchitis and emphysema.
‘ Rate was based on fewer than 5 deaths.
* 2-sided p values less than 0.05, test for RR = 1.0.
W
V O
40
follow-up relative to subsequent years (ie., terminally ill patients were less likely to enroll
in the study compared to other residents of Shanghai). Total mortality rate in the first
year of follow-up was 526/100,000 relative to 916/100,000 for all subsequent years.
However, there was no difference in mortality rates between two, three or longer years of
follow-up. Given the differential rate of mortality between the first and later years of
follow-up, we repeated all analyses described in the previous paragraph with the
exclusion of all deaths and man-year contribution from the cohort during the 12 months
post enrollment. Mortality rates derived from the latter analysis were only about 5-6%
higher than the corresponding rates based on all deaths and total follow-up experience.
For example, overall mortality rate from the second analysis was 916/100,000 vs.
867/100,000 from the original analysis. Comparable mortality rates for stroke and
ischemic heart disease were 212 vs. 201/100,000 and 73 vs. 69/100,000, respectively.
Overall mortality was 40% higher in smokers compared to lifelong non-smokers
(table 4-3). Smokers were 90% more likely to die from cancer than never smokers.
Significant increases in RRs of 2.0 (95% CL = 1.1, 3.8) for COPD (ie., chrortic bronchitis
and emphysema), 2.1 (95% CL =1.2,3.9) for ischemic heart disease, and 3.0 (95% CL =
1.3, 7.0) for other forms of heart disease were noted in one pack or more per day smokers
versus never smokers. Smokers also exhibited a non-significant increase in risk of death
from injuries or accidents (RR = 1.4) relative to never smokers. There was a modest,
non-significant increase in stroke deaths among one pack or more per day smokers versus
never smokers (RR = 1.2). A significantly reduced risk of death from liver cirrhosis was
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
observed among heavy smokers, but the number of deaths in this category was very small
(n = 4). Increased risks of overall mortality, and mortality from cancer, heart disease, and
COPD were positively and significantly related to increasing number of cigarettes
smoked per day. Similar results were obtained when we restricted the analysis to deaths
and man-years of follow-up which occurred at least 12 months post enrollment.
For causes of mortality exhibiting a significant association with cigarette
smoking, we further examined the possible relationship between risk and age at starting
to smoke after controlling for amount smoked per day and amount of alcohol consumed
per day. We noted a significant increase in risk of COPD in those who began smoking at
younger age. Relative to smokers who began the habit at age 25 or older, those who
smoked comparable amounts but began smoking at an earlier age experienced a 3.7-fold
increase in COPD risk (p = 0.008). For all cause mortality, cancer deaths and heart
disease deaths, the comparable risk ratios were 1.4 (p = 0.0002), 1.5 (p = 0.005) and 1.4
(p = 0.16), respectively.
IV. Discussion
This study provides one of the first systematic evaluations of cigarette smoking,
cancer incidence, and cause-specific mortality in China. In fact, our study is one of only
a handful of such investigations ever conducted in a developing country. Such evaluation
is particularly timely given the large population in China, the high prevalence of smoking
among men, and the potential for an expanded market for multinational tobacco
manufacturers. In China, the per capita consumption of cigarettes has increased threefold
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
between 1950’s and 1987, from about 500 to 1,748 (Yu and Mattson et al., 1990). A
national survey conducted in the mid-1980’s found that 69% of Chinese men aged 20
years or older were current smokers (Collaborative Group for Investigating Prevalence of
Smoking, 1987). The tar yield and nicotine contents of cigarettes produced by tobacco
manufacturers in Shanghai remained high in the 1980s. Based on the measurements
made in 1983 for some common brands of cigarettes, the tar yield ranges from 19 to 32
mg per cigarette, and the nicotine contents ranges from 0.8 to 1.6 mg per cigarette. The
tobacco manufactures in Shanghai began producing filter cigarettes in 1959, but the
proportion was relatively low, accounting for 0.2% of total production of the
manufacturers in Shanghai in 1967,2.0% in 1975, and 25.8% in 1980 (Gao, 1986). The
present study has a number of strengths for evaluating smoking/disease relationships,
including (1) the prospective study design which minimizes exposure (ie., smoking and
drinking status) misclassification due to recall bias, (2) a negligible loss to follow-up
(only 50 out of the 18,244 subjects could not be accounted for), and (3) the relatively
large numbers of cases identified to date (852 deaths and 554 incident cancers,
respectively).
Consistent with published data from the population-based cancer registry covering
the city of Shanghai (Parkin et al., 1992), the overall cancer incidence rate (568 per
100,000) in our cohort is only half the rate of comparable non-Hispanic white males in
Los Angeles. As expected, cancers of the liver, stomach, and esophagus are more
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
common in Shanghai while cancers of the prostate, bladder, colon, head/neck, and lung
are more common in Los Angeles whites.
In our cohort, 36% (95% CL = 28%, 45%) of all incident cancers could be
attributed to cigarette smoking. O f all malignancies, limg cancer shows the greatest risk
among smokers. Seventy-eight percent (95% CL = 65%, 88%) of lung cancer in middle-
aged men in Shanghai could be prevented if smoking was eliminated from this
population. In comparison, 45% of all cancer deaths and 90% of lung cancer deaths in
U.S. men are estimated to be related to cigarette smoking (Shopland et al., 1991).
Cigarette smoking is also an important predictor of overall mortality in middle-aged
Shanghainese men. Twenty-one percent (95% CL = 14%, 29%) of total mortality could
be attributed to smoking whereas the comparable figure of smoking-attributable mortality
in the U.S. is about 20% (Centers for Disease Control, 1993).
In the present study, cigarette smoking was foimd to be positively related to liver
cancer risk, independent of alcohol consumption. This finding is consistent with results
of case-control studies of hepatocellular carcinoma that we have conducted among
Chinese in Hong Kong (Lam et al., 1982) and among black and white residents of Los
Angeles who are at low-risk for the disease (Yu et al., 1991). Cigarette smoking was a
risk factor for hepatocellular carcinoma in these populations, independent of hepatitis B
and/or C infection and heavy alcohol drinking.
Our results are consistent with previous reports that stroke is considerably more
common in China, Japan, and other Asian countries than in North America (Chen et al..
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
1992; Kodama, 1993; Thom, 1993). Among study subjects, one o f every four deaths was
due to stroke. It is unlikely that inter-country variation in accuracy of diagnosis can
account for a rate differential of the magnitude we observed between Shanghai Chinese
and U.S. whites.
Although smoking has been found to be a risk factor for stroke in various Western
populations (Abbott et al., 1986; Coditz et al., 1988; Wolf et al., 1988), the present study
suggests that it may not be a major predictor of risk for fatal stroke in Shanghai. We
recently demonstrated that hypertension is the major risk factor associated with stroke
death in this very high-risk population (Ross et al., 1996). However, cigarette smoking,
as in Western populations (Doll and Peto, 1976; Rogot and Murray, 1980; Carstensen et
al., 1987; Willett et al., 1987; Ragland and Brand, 1988; Dagenais and Robitaille et al.,
1990; Dagenais and Ahmed et al., 1990; Menotti et al., 1990; Freund et al., 1993), is an
important risk factor for ischemic heart disease in Shanghai. Among study subjects, 30%
(95% CL = 12%, 60%) of ischemic heart disease deaths could be attributable to cigarette
smoking.
Cigarette smoking exerts adverse effects on the structure and function of the
bronchial airways and can lead to COPD. Several large cohort studies conducted in
Western populations have consistently shown smoking to be a strong risk factor for death
from COPD, with a relative risk of about 10 in smokers versus non-smokers (Doll and
Peto, 1976; Rogot and Murray, 1980; Carstensen et al., 1987). This contrasts with a 40%
increase in COPD death among ever smokers and a 2-fold increased risk among heavy
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
45
smokers (20+ cigarettes per day) relative to never smokers in our cohort. A likely
explanation for this apparent discrepancy is the considerably later age at starting to smoke
among Chinese men relative to their Western counterparts. Close to half of the smokers
in our cohort began smoking after age 25 whereas virtually all smokers in Occidental
populations begin to smoke before age 20 (Centers for Disease Control, 1989). Our data
show a very strong effect of age at starting to smoke on the smoking/COPD association.
There was a close to 4-fold difference in risk of COPD death between smokers who
began the habit before age 25 years relative to those who smoked comparable amounts
but began the habit at age 25 years or older.
The burden of tobacco-related disease/mortality to be experienced by today’s
young Chinese men will surpass estimates derived firom the present study. About half of
our cohort of middle-aged men are smokers, who, on average, started smoking at age 25
years and smoked 16 cigarettes per day. In contrast, among the adult males (aged 20+
years) in China today, 69% are smokers with three-fourths of smokers starting to smoke
under age 25, although they smoked a similar number of cigarettes daily as that in our
cohort (Collaborative Group for Investigating Prevalence of Smoking, 1987). Our data
have shown strong and independent effects of age at starting to smoke and amount
smoked per day on disease risk in Chinese men. Comparisons of relative risks of
smoking and disease morbidity/mortality in U.S. men (with virtually all smokers
beginning to smoke before age 20) versus Shanghai Chinese men (with half of the
smokers starting after age 25 years) consistently show higher risk levels in the former
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
46
population (Centers for Disease Control, 1989). For example, the relative risk of death
from all causes in U.S. smokers vs. never smokers is 2.3 compared to 1.4 in Shanghai
Chinese. For risk of COPD death which has an especially strong relationship with age
starting to smoke, the relative risk in U.S. smokers vs. never smokers is 9.7 compared to
1.4 in Shanghai Chinese. Thus, there is an urgent need for anti-smoking campaigns and
related activities in China.
In summary, the results of the present study provide a systematic evaluation of
smoking and subsequent health risk in China. Smoking was shown to be an important
predictor of risk of cancer and mortality in middle-aged men in Shanghai. Among the
study subjects, 36% of cancer morbidity and 21% o f overall mortality were attributable to
cigarette smoking.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
CHAPTER 5 MORTALITY IN RELATION TO ALCOHOL DRINKING^
Summary
Previous studies in Western populations have noted that light-to-moderate alcohol
drinkers have a lower risk of mortality than abstainers or heavy drinkers. The observed
protective effect on total mortality is due largely to a reduced risk of fatal ischemic heart
disease (IHD) which accounts for roughly one-third of all deaths in such populations.
The present study examined the relationship between alcohol intake and mortality in a
cohort of middle-aged Chinese men in whom IHD accounts for only 9% of all deaths.
Between 1986 and 1989,18,244 middle-aged male residents of Shanghai enrolled in a
prospective study of diet and cancer. Information on alcohol use was collected from each
participant through an in-person interview. The cohort members have been actively
followed via annual visits. By February 28,1995, a total of 1,157 deaths (including 482
from cancer, 259 from stroke, and 101 from IHD) had been identified during the follow-
up period (an average of 6.7 years per subject). Compared with lifelong non-drinkers, a
statistically significant 19% reduction in overall mortality risk was observed in light-to-
moderate drinkers (1-28 drinks per week) after adjustment for age, level of education and
cigarette smoking. This protective effect was not restricted to any specific type of
alcoholic beverages. Although light-to-moderate drinking was significantly associated
with a 36% reduction in IHD death, it had no effect on stroke death which is the leading
' This paper is in review.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48
cause of death in this study population, accounting for 22% of total mortality. As
expected, heavy drinking (29 or more drinks per week) was significantly associated with
increased risks o f death fi-om cancer of the upper aerodigestive tract, stroke, and hepatic
cirrhosis. In conclusion, Light-to-moderate drinking is associated with a reduced risk of
IHD death as well as overall mortality in middle-aged Chinese men. This survival benefit
is not restricted to any specific type of alcoholic beverages.
I. Introduction
In prospective cohort studies conducted in Western populations, light-to-moderate
drinkers of alcoholic beverages of either sex have been found to have lower overall death
rates than either non-drinkers or heavy drinkers (Marmot et al., 1981; Kozarevic et al.,
1982; Colditz et al., 1985; Gordan and Doyle, 1987; Boffetta et al., 1990; De Labry et al.,
1992; Farchi et al., 1992; Klatsky et al., 1992; Cullen et al., 1993; Doll et al., 1994;
Gronbaek et al., 1994; Fuchs et al., 1995; Rhem et al., 1995; Serdula et al., 1995). The
observed protective effect of drinking on overall mortality is due largely to a reduced risk
of fatal ischemic heart disease (IHD) which accounts for roughly one-third of all deaths in
such populations. High consumption of wine has been suggested to be responsible for the
low risk for IHD in France (Renaud and De Lorgeril, 1992) and recent findings from a
prospective study in Copenhagen support this hypothesis; wine intake but not intake of
beer or hard liquor was found to be protective against overall mortality in Danish people
(Gronbaek et al., 1995). On the other hand, several studies reported no differential effects
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
of wine versus other types of alcoholic beverages on risk of death (Yano et al., 1977;
Friedman and Kimball, 1986; Stampfer et al., 1988; Rimm et al., 1991).
The present study examined prospectively the relationship between alcohol
consumption and mortality in a large cohort of middle-aged Chinese men in Shanghai,
China. In this cohort, IHD accounts for only about 9% of all deaths (Yuan et al., 1996).
Also grape wine is a minor source of ethanol among study subjects who drink regularly.
Therefore, our study population is a valuable one to further evaluate the relationship
between moderate alcohol intake and overall as well as cause-specific mortality.
n. Subjects and Methods
Between January 1986 and September 1989, we invited all eligible male residents
of four small, geographically defined communities scattered over a wide area of the city
of Shanghai to participate in a prospective, epidemiological study of diet and cancer. The
eligibility criteria were ages 45-64 years and no history of cancer. In addition to
collecting a 10 ml blood sample and a single void urine sample, we interviewed each
subject in person using a structured questionnaire which included level of education,
usual occupation, current height and weight, history of tobacco and alcohol use, current
diet (intake frequencies of 45 food items), and medical history. During the 3-year
recruitment period, 18,244 men (about 80% of eligible subjects) enrolled in the study.
In terms of alcohol consumption, we asked each participant whether he had ever
drunk alcoholic beverages at least once a week for six months or more. If the answer was
yes, he was asked to provide the age at which he started to drink regularly and the usual
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
amount of consumption of beer, wine, and hard liquor separately. If the subject was an
ex-drinker, the age at which he stopped drinking was recorded. One drink was defined as
360 g o f beer (12.6 g of ethanol), 103 g of wine (12.3 g o f ethanol), or 30 g o f hard liquor
(12.9 g of ethanol) (Adames, 1975).
Details of the follow-up procedures for this cohort have been described previously
(Ross et al. 1992). We conducted routine review of death certificates and cancer reports
from the population-based Shanghai Cancer Registry, and all surviving members of the
cohort were contacted on an annual basis. To date, only 102 subjects have been lost to
follow-up.
For each man, man-years of follow-up were counted from the date of recruitment
to February 28, 1995, or the date of death or loss to follow-up, whichever occurred first.
Overall mortality rates were calculated according to the following seven categories of
alcohol consumption; lifelong non-drinkers, drinkers of 8-14, 15-21,22-28,29-42,
>43 drinks per week (due to the small number of ex-drinkers in the study population, ex-
and current drinkers could not be analyzed separately). In cause-specific analyses,
regular drinkers were grouped into two categories (1-28, and ^ 9 drinks per week). All
rates were adjusted for age using the man-year distribution of the entire cohort as the
internal standard (the proportion of total man-years in the age groups <54, 55-59,60-64,
and >65 years were 0.26,0.27,0.28, and 0.19, respectively).
We used the Poisson regression method to examine the relationship between
alcohol consumption at recruitment and subsequent risk o f death among cohort members
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
(Breslow and Day, 1987). Relative risks (RRs) and their corresponding 95% confidence
limits (CLs) were computed for various categories of drinkers in reference to lifelong
non-drinkers. To examine the relationship between a specific type of alcoholic beverages
and overall mortality while controlling for the consumption of the other two types, a
covariate representing the total ethanol intake fi'om the latter was added to the list of
confounding variables, which, in combination, defined the various strata of the regression
model. The other confounders adjusted for in the regression model were age (<54, 55-59,
60-64, and >65 years), cigarette smoking (never-smoker, ever-smoker of <19 cigarettes
per day, and ever-smoker of ^ 0 cigarettes per day), and level of education (primary
school or below, secondary school, and college or higher). All quoted p values are 2-
sided. RRs with 2-sided p values under 0.05 were considered to be significantly different
from 1.0.
III. Results
At recruitment, 57% (n = 10,471) of cohort members had never drunk alcoholic
beverages regularly, 41% (n = 7,390) were current drinkers, and the remaining 2% (n =
383) were ex-drinkers. Among current drinkers, about one-third drank 7 or fewer drinks
per week and 15% consumed 29 or more drinks per week. Level of drinking was
negatively associated with age at starting. Drinkers of 7 or less drinks per week, on
average, began drinking around age 33 years while those who consumed 43 or more
drinks per week had a mean age at starting of 23 years. Heavy drinkers preferred hard
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
■o
I
I
C/)
o'
3
CD
8
ë'
O
3
CD
C
p.
3 "
CD
0
■o
1
c
a
o
3
■o
o
&
o
c
■o
CD
( / )
W
o'
3
Table 5-1 Various attributes of alcohol consumption at recruitment
Ex Current drinkers (no. of drinks oer week 1 All
drinkers <7 8-14 15-21 22-28 29-42 >43 drinkers
No. of subjects (%) 383 (5) 2.354 (30) 1,809(23) 1,077(14) 949 (12) 712(9) 489 (6) 7,773 (100)
Mean no. of drinks per week 21.4 3.5 10.4 17.3 23.9 34.5 61.7 16.9
Mean age at starting(year) 25.8 32.9 31.0 28.3 28.2 26.0 23.4 29.7
Mean duration (year) 25.6 22.5 24.8 28.1 28.1 30.0 32.8 26.0
Daily ethanol intake (g) 39.1 6.3 18.9 31.0 43.7 62.7 112.6 30.7
Amount from beer 2.7 2.1 4.9 3.6 3.9 4.6 7.9 3.8
Amount from wine 8.2 2.9 7.1 19.7 8.6 19.7 26.1 10.2
Amount from hard liquor 28.3 1.3 6.9 7.8 31.1 38.4 78.6 16.7
L A
to
53
liquor to beer or wine. On avenge, 67% of ethanol consumed by heavy drinkers ( ^ 9
drinks per week) was derived from hard liquor, while the comparable percentage was
21 among light drinkers drinks per week). Ex-drinkers were comparable to current
drinkers in terms of quantity consumed, age at starting and duration of drinking (table 5-
1). On average, ex-drinkers gave up the habit 5.9 years prior to study enrollment.
Table 5-2 shows selected baseline characteristics. There was a weak, positive
relationship between age and alcohol consumption. Twenty-seven % of non-drinkers and
32% of heavy drinkers ( ^ 9 drinks per week) were aged 60 or older at recruitment. Mean
values of body mass indices were similar across the different categories o f alcohol
consumption. There was an inverse relationship between level of education and alcohol
consumption. Thirty-one % of non-drinkers attended college while only 7% of heavy
drinkers attained such level of education. Drinking and smoking habits were closely
related among study subjects. Fifty-five % o f non-drinkers were lifelong non-smokers
while only 12% of heavy drinkers had never smoked cigarette regularly. The proportions
of individuals who had a history of diabetes (1%), emphysema (4%), hepatic cirrhosis
(1%), or high blood pressure (25%) at recruitment were similar across different categories
of alcohol consumption (table 5-2).
As of February 28,1995, cohort members had contributed 122,300 man-years of
observation (an average of 6.7 years of follow-up per subject). During the follow-up,
1,157 of the men died (946 per 100,000 man-years). There were 482 deaths from cancer,
259 deaths from stroke, and 101 deaths from IHD. We attempted to review the medical
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
■ o
o
D.
C
g
Q.
■ O
CD
Table 5-2 Self-reported characteristics by alcohol consumption status at recruitment
C/)
C/)
CD
8
S'
p -
3 "
CD
CD
■ o
O
Q .
C
a
o
3
■ o
o
CD
Q .
■ o
CD
C / )
C / )
Non
drinkers
(n =
10,471)
<7
(n =
2,473)
8-14
(n =
1,892)
No. of drinks per week
15-21 22-28
(n = (n =
1,128) 991)
29-42
(n =
752)
>42
(n =
537)
Total
subjects
(n =
18,244)
Age 60 years or older (%) 27 26 29 30 31 31 34 28
Body mass index (mean, kg/m^)“ 22.1 22.1 22.2 22.4 22.2 22.4 22.2 22.2
Level of education (%)
Primary school or below 23 28 31 40 44 44 54 29
Secondary school 46 48 91 46 45 46 40 47
College or higher 31 24 18 14 11 9 6 25
Cigarette smoking (%)
Non-smoker 55 37 29 26 17 13 12 43
£19 cigarettes/day 24 39 39 34 34 28 23 29
^ 0 cigarettes/day 22 25 32 40 50 59 65 29
Prior medical condition (%)
Diabetes 1 1 1 1 1 1 1 1
Emphysema 4 3 5 4 4 3 4 4
Hepatic cirrhosis 1 1 1 1 1 2 1 1
High blood pressure 24 25 24 29 23 27 27 25
The weight in kilograms divided by the square of the height in meters
55
records of all cancer cases and deaths. Sixty % of deaths from all cancer sites were
confirmed histopathologically, including 93% of colon and rectum, 88% of stomach,
81%of upper aerodigestive tract, 67% of lung, and 13% of liver (97% of the remaining
cases exhibited elevated serum alpha-fetoprotein and/or a positive liver scan at
diagnosis). Of all stroke deaths, 85% had been hospitalized prior to death while no
evidence of medical care immediately prior to death was established for 9% of these
patients. The comparable figures, respectively, were 82% and 13% for IHD; 85 and 11%
for chronic bronchitis and emphysema; and 94% and 1% for hepatic cirrhosis.
Light-to-moderate drinkers were at a reduced risk of death relative to non
drinkers. After adjusting for age, education and cigarette smoking, RR of death in
drinkers consuming 28 or fewer drinks per week as compared with lifelong non-drinkers
was 0.81 (95% CL: 0.71,0.92). On the other hand, men who consumed 43 or more
drinks per week (ie., 6 or more drinks a day) experienced a borderline significant, 30%
excess risk (RR = 1.30; 95% CL: 1.00,1.68) relative to lifelong non-drinkers. Excluding
men who gave up drinking or who had potentially life-threatening illnesses (ie., diabetes,
emphysema, or hepatic cirrhosis) did not change the relation between alcohol
consumption and overall mortality (table 5-3).
Previously, we had noted a deficit in the number of deaths during the first year of
follow-up relative to subsequent years (Yuan et al., 1996). In other words, terminally ill
patients were less likely to participate in the cohort study relative to other eligible
subjects. We repeated the analyses with the exclusion of all deaths and man-year
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7 D
CD
■ D
O
Q.
C
3
Q.
■ D
CD
C/)
C/)
8
( § ■
3
3 "
CD
CD
" O
O
Q.
C
O
■ D
O
CD
Q.
■ D
CD
e n
e n
Table 5-3 Total mortality by alcohol consumption status at recruitment
Non- No. of drinks per_week______________________ Total
drinkers <7 8-14 15-21 22-28 29-42 >43 subjects
Entire cohort
No. of deaths 638 132 111 76 70 64 66 1,157
Mortality rate' 944 801 829 964 974 1,211 1,651 946
RR^ 1.00 0.80* 0.80* 0.84 0.82 1.00 1.30
Excluding ex-drinkers
No. of deaths 638 116 95 68 63 62 55 1,097
Mortality rate' 944 745 742 921 926 1,241 1,505 919
§ RR^ 1.00 0.75 0.71 0.80 0.78 1.02 1.17
Excluding those with serious illness^
No. of deaths 522 110 89 66 60 55 59 961
Mortality rate' 838 712 718 890 889 1,117 1,584 847
RR^ 1.00 0.80* 0.77* 0.88 0.83 1.02 1.38*
L A
O N
CD
■ D
O
Q .
C
g
Q .
■ D
CD
( / >
W
o'
3
3
CD
8
( O '
3 "
i
3
CD
" n
c
3.
3 "
CD
CD
" O
O
Q .
■ D
CD
( / )
( / )
Table 5-3 (continued)
Non No. of drinks ner week Total
drinkers <7 8-14 15-21 22-28 29-42 >43 subjects
Excluding the first vear of follow-up'*
No. of deaths 590 125
Mortality rate* 981 849
RR^ 1.00 0.82'
104
859
0 80'
70
983
0.84
67
1,054
0.85
59
1,254
0.99
61
1,726
1.29
1,076
985
' Rates per 100,000 man-years were adjusted for age according to the man-year distribution of the entire cohort (<34, 55-59,
60-64, and >65 years).
a ^ Relative risk, adjusted for age (<34, 55-59,60-64, and >65 years), level of education (primary school or below, secondary
= > school, and college or higher), and cigarette smoking (nonsmokers, smokers of <19 cigarettes per day, and smokers of >20
o cigarettes per day).
CD 3
Q . History of diabetes, emphysema, or hepatic cirrhosis.
All deaths and man-year contribution from the cohort during the 12 months post enrollment were excluded.
* 2-sided p values less than 0.05, and * * 2-sided p values less than 0.01, test for RR = 1.0. Table 4
L A
58
contribution from the cohort during the first 12 months of follow-up. The relationship
between alcohol consumption and risk of death remained unchanged (table 5-3). We also
compared the alcohol/risk association among study subjects by duration o f follow-up (<4
years versus longer duration of follow-up); no difference was detected (data not shown).
The relationship between alcohol consumption and mortality risks for cancer
overall, and for the five commonest cancer sites are presented in table 5-4. Cancer
accounted for 42% of all deaths. After adjustment for age, level of education and
cigarette smoking, light-to-moderate drinkers (1-28 drinks per week) exhibited a non
significant 15% reduction in risk of death from any cancer relative to lifelong non
drinkers (table 5-4). As expected, heavy drinkers ( ^ 9 drinks per week) exhibited a
significant, 3.7-fold increased risk for cancer of the upper aerodigestive tract, mainly
esophageal cancer (n = 22). A statistically non-significant 30-40% increase in risks of
death from cancers of the stomach, colon and rectum was observed in heavy drinkers.
Table 5-5 presents the association between alcohol consumption and mortality
from all non-cancer-related causes, and from the five commonest non-cancer-related
specific causes. Stroke was the leading cause of non-cancer death in the study
population, accounting for 22% o f total deaths, while IHD accounted for only 9% o f total
mortality. Compared with lifelong non-drinkers, light-to-moderate drinkers (1-28 drinks
per week) experienced a statistically significant 21% reduction in rate of mortality from
all non-cancer-related causes (RR = 0.79; 95% CL: 0.66, 0.93), and a significant 36%
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
59
Table 5-4 Mortality from cancer by alcohol consumption status at recruitment
Cancer site Non- 1-28 drinks ^ 9 drinks Total
(ICD-9) drinkers per week per week subjects
All sites
No. of deaths 254 172 56 482
Mortality rate' 376 379 599 394
RR- 1.00 0.85 1.08
Upper aerodigestive tract
(141,143-146,148-150, 161)
No. of deaths 10 12 9 31
Mortality rate' 15 27 94 25
RR^ 1.00 1.39 3.73“
Stomach (151)
No. of deaths 47 31 10 88
Mortality rate' 70 69 104 72
RR^ 1.00 0.94 1.38
Colon and rectum (153, 154)
No. of deaths 17 8 3 28
Mortality rate' 25 18 32 23
RR^ 1.00 0.72 1.33
Liver(155)
No. of deaths 58 30 9 97
Mortality rate' 86 65 104 79
RR^ 1.00 0.67 0.91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
Table 5-4 (continued)
Cancer site
(ICD-9)
Non
drinkers
1-28 drinks
per week
^ 9 drinks
per week
Total
subjects
Lung (162)
No. of deaths 70 54 20 144
Mortality rate’ 104 119 212 118
RR’ ’ 1.00 0.79 0.93
All other sites
No. of deaths 52 37 5 94
Mortality rate’ 77 82 53 77
RR^ 1.00 0.98 0.56
* Rates per 100,000 man-years were adjusted for age according to the man-year
distribution of the entire cohort (<54, 55-59, 60-64, and >65 years).
^ Relative risk, adjusted for age (<54, 55-59, 60-64, and >65 years), level of education
(primary school or below, secondary school, and college or higher) and cigarette
smoking (nonsmokers, smokers o f <19 cigarettes per day, and smokers of ^ 0
cigarettes per day).
2-sided p values less than 0.01, test for RR = 1.0.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
61
Table 5-5 Mortality from non-cancer-related causes by alcohol consumption status at
recruitment
Non-cancer related
cause (ICD-9)
Non
drinkers
1-28 drinks
per week
^ 9 drinks
per week
Total
subjects
All non-cancer-related causes
No. of deaths 384 217
Mortality rate’ 568 482
RR^ 1.00 0.79*
74
805
1.17
675
552
Ischemic heart disease
(410-414)
No. of deaths
Mortality rate’
RR^
62
92
1.00
30
66
0.64*
9
96
0.82
101
83
Stroke (430-438)
No. of deaths
Mortality rate
RR^
130
193
1.00
94
207
0.99
35
378
1.62*
259
212
Chronic bronchitis and
emphysema (490-496)
No. of deaths
Mortality rate’
RR^
35
52
1.00
22
48
0.82
5
51
0.72
62
51
Hepatic cirrhosis (571)
No. of deaths
Mortality rate’
R R ^
23
34
1.00
6
14
0.46
6
65
2.99*
35
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
62
Table 5-5 (continued)
Non-cancer related
cause (ICD-9)
Non
drinkers
1-28 drinks
per week
^ 9 drinks
per week
Total
subjects
Injury/accident (800-999)
No. of deaths 23 11 7 41
Mortality rate* 33 25 81 34
R R * * 1.00 0.69 1.96
All other non-cancer-related
causes
No. of deaths 111 54 12 177
Mortality rate* 164 121 135 145
RR^ 1.00 0.69* 0.66
* Rates per 100,000 man-years were adjusted for age according to the man-year
distribution of the entire cohort (<54, 55-59, 60-64, and >65 years).
^ Relative risk, adjusted for age (<54, 55-59,60-64, and >65 years), level of education
(primary school or below, secondary school, and college or higher) and cigarette
smoking (nonsmokers, smokers of <19 cigarettes per day, and smokers of >20
cigarettes per day).
* 2-sided p values less than 0.05, and 2-sided p values less than 0.01, test for RR = 1.0.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
63
reduction in risk for fatal IHD (RR =0.64; 95% CL; 0.41,0.99). Light-to-moderate
drinking also was related to lower mortality from non-cancer-related causes excluding
EHD (RR = 0.82; 95% CL: 0.68, 0.98) or excluding all specific causes listed in the table
(mainly other forms of heart disease and diabetes) in the study population (table 5-5). On
the other hand, light-to-moderate drinking offered no protection from risk of stroke death.
In fact, heavy drinking ( ^ 9 drinks per week) was associated with a significant 1.6-fold
excess in risk of stroke death (95% CL: 1.09,2.40). Adjustment for history of
hypertension slightly diminished the effect of heavy drinking on risk of fatal stroke (RR =
1.50; 95% CL: 1.01,2.22). Heavy drinking also was positively associated with death due
to hepatic cirrhosis (RR = 2.99; 95% CL: 1.12, 7.94) and injury/accident (RR = 1.96;
95% CL: 0.79,4.85).
We repeated all cause-specific analyses with the exclusion of ex-drinkers, men
with pre-existing serious illnesses, or deaths and man-year contribution from the cohort
during the first 12 months post enrollment. The exclusions did not alter any of the
relationships described above. We also repeated all cause-specific analyses for the five
commonest cancer sites after exclusion of cancer deaths without histopathologic
confirmation and for the five commonest non-cancer-related causes after exclusion of
deaths lacking evidence of medical care immediately prior to death. No material changes
in results were observed.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
64
Table 5-6 Total mortality by consumption status on specific types of alcoholic beverages
at recruitment
Type of alcoholic
beverages
Non
drinkers
No. of drinks ner week
8-14 15-28 ^ 9
Beer
No. of deaths 638 143 40 7
1
Mortality rate^ 944 856 808 1,141 —
RR^ 1.00 0.87 0.77 0.84 -----
Wine
No. of deaths 638 114 73 71 36
Mortality rate^ 944 757 1,031 1,053 1,586
RR^ 1.00 0.69” 0.89 0.86 1.20
Hard liqwar
No. of deaths 638 59 65 72 78
Mortality rate^ 944 761 1,028 963 1,541
RR] 1.00 0.72* 0.93 0.79 1.15
* No deaths in this category.
^ Rates per 100,000 man-years were adjusted for age according to the man-year
distribution of the entire cohort (<54,55-59,60-64, and >65 years).
^ Relative risk, adjusted for age (<54, 55-59,60-64, and >65 years), education (primary
school or below, secondary school, and college or higher), cigarette smoking
(nonsmokers, smokers of <19 cigarettes per day, and smokers of >20 cigarettes per
day), and consumption level of the other two types of alcoholic beverages.
* 2-sided p value less than 0.05, and * * 2-sided p value less than 0.01, test for RR = 1.0.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
65
We also examined the relationship between specific types o f alcoholic beverages
and overall mortality. Among regular drinkers, 3,500 drank beer, 4,341 drank wine, and
3,723 drank hard liquor. Irrespective of the type of alcoholic beverages consumed,
moderate drinkers exhibited a reduced risk of death relative to non-drinkers. The
reduction in risk was statistically significant for drinkers of wine or hard liquor
consuming no more than 7 drinks per week compared to lifelong non-drinkers (table 5-6).
On the other hand, there was a non-significant, 15-20% increase in mortality risk among
men who drank either wine or hard liquor excessively (29 drinks or more per week).
Very few study subjects (n = 18) reported consuming 29 or more drinks o f beer per week.
IV. Discussion
In the present study, we observed that Chinese men who consumed no more than
28 drinks per week experienced a statistically significant, 19% reduction in mortality
relative to lifelong non-drinkers. The observed association was not explicable by
confounding factors such as level of education, cigarette smoking, or age. Consistent
with findings in Western populations (Yano et al., 1977; Kittner et al., 1983; Colditz et
al., 1985; Friedman and Kimball, 1986; Boffetta et al., 1990; Rimm et al., 1991; Klatsky
et al., 1992; Suh et al., 1992; Cullen et al., 1993; Doll et al., 1994; Serdula et al., 1995),
we observed a statistically significant, 36% reduction in IHD mortality among light-to-
moderate (1-28 drinks per week) drinkers relative to lifelong non-drinkers. However, the
protective effect of moderate drinking on total mortality in Shanghai men was not
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 6
confined to IHD which accounts for only 9% o f total deaths. We also observed a
significant, negative association between moderate drinking and mortality other than IHD
(RR = 0.83; 95% CL: 0.72,0.95).
Bias is unlikely to explain the observed associations. The prospective study
design precluded recall bias. A structured questionnaire was used to administer the in-
person interviews to ensure that all subjects were asked identical questions. There is
evidence that the collected information is reasonably reliable; diseases known to be
positively associated with heavy alcohol intake in Western populations (ie., upper
aerodigestive cancer, liver cirrhosis, and injury/accident) were found to exhibit results as
predicted in this study population. It is also important to note that loss to follow-up in the
present study has been negligible (only 102 out o f 18,244 subjects have been lost during
the study period). Finally, it is unlikely that the observed association between alcohol
intake and cause-specific mortality were due to misclassification of cause of death. A
high proportion (82%) of cause of death among all cases was medically confirmed
immediately prior to death, and exclusion of those without such confirmation had no
effect on the observed association between alcohol intake and cause-specific mortality.
Shaper and his colleagues (1988) pointed out that non-drinkers might include ex
drinkers who had quit drinking due to ill health. Such individuals would be expected to
have an increased risk for mortality. This was apparently not the case in the present
study. We were able to separate ex-drinkers firom lifelong non-drinkers, and with or
without ex-drinkers in the analysis, the results were essentially unchanged. Another
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
67
proposed argument to explain the relatively high mortality rate in abstainers is that such
individuals carry a greater “burden of ill health” due to lifelong health problems than
light-to-moderate drinkers, regardless of their previous drinking status (Shaper, 1990).
We excluded all subjects with a history of potentially life-threatening illnesses from the
analysis and no change in the results was observed. Our results also were unaffected by
duration of follow-up, supporting the notion that other as yet unidentified bias is unlikely
to be responsible for the observed association.
A number of hypotheses have been proposed to explain the protective effect of
moderate drinking on IHD risk. Several investigators have found that moderate alcohol
consumption is associated with increased concentration o f high-density lipoprotein
cholesterol (Gaziano et al., 1993; Linn et al., 1993; Lavy et al., 1994; Marques-Vidal et
al., 1995) which in turn has been shown to be related to a reduced risk of IHD. As much
as one-half of the protective effect of alcohol on IHD may be accounted for by the
elevated concentrations of high-density lipoprotein cholesterol (Criqui et al., 1987; Suh et
al., 1992; Paunio et al., 1994). In addition, small amounts of ethanol have been shown to
increase the concentration of plasminogen activator inhibitor, to decrease the
concentration of fibrinogen, and to reduce the tendency o f platelets to aggregate (Pikaar
et al., 1987; Hendriks et al., 1994; Ridker et al., 1994). These hematological effects of
alcohol are believed to reduce the possibility of clot formation and thus reduce the risk of
IHD.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 8
An ecological study suggested that high consumption o f wine might be
responsible for the lower IHD risk in France (Renaud and De Lorgeril, 1992). Recent
findings from a prospective study in Copenhagen have shown that wine drinkers had a
50% reduction in mortality relative to nondrinkers (Gronbaek et al., 1995). In the same
study, beer drinking was not related to mortality risk and spirit consumption increased the
risk (Gronbaek et al., 1995). However, several studies have reported no differential
effects of wine relative to other types of alcoholic beverages on mortality risk (Yano et
al., 1977; Friedman and Kimball, 1986; Stampfer et al., 1988; Rimm et al., 1991). In our
study, we did not separate the grape wine from other wines drunk by Chinese. But, data
from a case-control study of nasopharyngeal carcinoma conducted in the same population
in the late 1980s indicate that few Chinese men drank grape wine regularly. Among male
controls in that study, 6% who drank any wine regularly reported drinking grape wine;
the rest of them drank only rice wine. Our data did not show that intake of wine was
more beneficial in reducing risk of death than consumption of beer or hard liquor.
A higher risk for fatal stroke in heavy drinkers than abstainers has been observed
by others as well (Kono et al., 1986; Boffetta et al., 1990; Ridgers et al., 1993; Doll et al.,
1994; Goldberg et a;., 1994; Iso et al., 1995). Alcohol has been found to raise blood
pressure (Maheswaran et al., 1991; Marmot et al., 1994) and hypertension is a strong
predictor of fatal stroke (Ross et al., 1996). However, the increased risk for fatal stroke in
heavy drinkers was not explicable by self-reported hypertension in the present study. The
anti-clotting effects of alcohol (possibly benefit to IHD) could also increase the risk of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
69
hemorrhagic stroke (Stmapfer et al., 1988; Klatsky et al., 1990; Giroud et al., 1995; Iso et
al., 1995), which is the predominant form among Chinese (Chen et al., 1992).
Alcohol intake is a strong risk factor for cancer of the upper aerodigestive tract in
this population, as in others (Boffetta et al., 1990; Doll et al., 1994). The observed 30-
40% elevation in mortality risk for cancers of the stomach, colon and rectum in heavy
drinkers of Shanghai men is comparable to findings firom other populations (Longnecker
et al., 1990; Wu-Williams et al., 1990; Jedrychowski et al., 1993; Meyer and White,
1993; Goldbohm et al., 1994; Vaughan et al., 1995). Unlike results in Western males (Yu
et al., 1991), heavy drinking was unrelated to risk of liver cancer in this study population.
Previously, we have shown that hepatitis B virus infection and dietary aflatoxin exposure
are the two major risk factors for this malignancy in the study population and most cases
can be attributed to these two factors alone or in combination (Ross et al., 1992; Qian et
al., 1994).
In summary, light-to-moderate drinking (1-28 drinks per week) was associated
with a statistically significant 19% reduction in total mortality and a significant 36%
reduction in IHD death in middle-aged men in Shanghai. This protective effect on
overall mortality risk was not restricted to any specific type of alcoholic beverages.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
70
CHAPTER 6 SERUM TESTOSTERONE AND HEPATOCELLULAR CARCINOMA^
Summary
A nested case-control study of hepatocellular carcinoma (HCC) based on a cohort
of 18,244 middle-aged men in Shanghai, China, who had been followed for an average of
5.3 years, was conducted. Our hypothesis dealt with the possible role o f testosterone in
the etiology of HCC, which shows a minimum of a 2- to 3-fold male excess in all
populations world-wide. Seventy-six incident cases of HCC and 410 control subjects
drawn from the cohort and individually matched to the cases by age (within 1 year), time
of blood sample collection (within 1 month) and neighborhood of residence were
assessed for serum hepatitis B surface antigen (HBsAg), antibodies to the hepatitis B core
antigen (anti-HBc), antibodies to the hepatitis B surface antigen (anti-HBs), antibodies to
the hepatitis C virus (anti-HCV), and testosterone. Among controls, serum testosterone
levels were similar between those who had no markers of hepatitis B virus (HBV)
infection, those who were positive for anti-HBs only and those who were positive for
anti-HBc but negative for HBsAg. However, the geometric mean level o f testosterone in
HBsAg-positive controls was 21% higher relative to HBsAg-negative controls and the
difference was statistically significant (2-sided p =0.0006). Relative to controls, HCC
cases had a significantly higher mean level of testosterone at the time o f recruitment (570
' This paper was published in the International Journal of Cancer: 63, 491-493, 1995.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
vj. 485 ng/dl, 2-sided p =0.0005), but the difference was explicable on the basis of a
higher proportion o f HBsAg-positive individuals among cases than controls ip = 0.42
after adjustment for HBsAg status).
I. Introduction
Hepatocellular carcinoma (HCC) exhibits wide variation in incidence world-wide.
While it is an uncommon cancer in the United States and most of Western Europe, it
represents the 3rd and 4th commonest cause of cancer death in Chinese men and women,
respectively (Li et al., 1979; Parkin et al., 1992). There is considerable geographical
variation in HCC incidence within China; the high-risk regions cluster mostly along the
coastline bordering the East and South China Seas, which include the metropolitan area
of Shanghai (Li et al., 1979).
In all populations with reliable cancer statistics, males are at least 2 to 3 times
more likely to develop HCC than females. In Shanghai, the male:female ratio in age-
adjusted incidence is 2.9 (30.6/100,000 in men vj. 10.7/100,000 in women). Yu and
Chen (1993) hypothesized that testosterone plays a role in human HCC and is at least
partly responsible for the observed male excess. Using data from a cohort study of
Taiwanese men (another Chinese population at high risk for HCC), these 2 investigators
demonstrated a significant excess of HCC cases with high serum testosterone at
recruitment relative to controls, who were randomly selected from HCC-free cohort
subjects and matched to cases by age, date of blood collection and township of residence.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
72
Relative to individuals with serum testosterone in the lower 2 tertiles, those with values
in the top tertile showed a 4.1-fold increased risk for HCC after adjustment for various
confounders, which included serum hepatitis B surface antigen (HBsAg) positivity,
serum antibodies to the hepatitis C virus (anti-HCV) positivity, vegetable intake and a
history of liver disease (Yu and Chen, 1993). There is also some support for this
hormonal hypothesis from experimental data. Various species of female rodent have
been shown to be less susceptible to the carcinogenic effect of aflatoxins relative to their
male counterparts (Wogan, 1992). Liver microsomes from several strains of male rat
have been observed to form more aflatoxin metabolites (including DNA adducts) than
their female counterparts. Furthermore, aflatoxin B, (AFBi) metabolism was reduced
when microsomes from castrated male rats were used, while testosterone treatment of
castrated immature male and immature female rats led to enhanced AFBi metabolism by
their microsomes (Gurtoo and Motycka, 1976).
During 1986-1989, we assembled a cohort of 18,244 middle-aged (45-64 years)
male residents of Shanghai to investigate the environmental etiology of several cancers
prevalent in the local population. Results so far have indicated that the hepatitis B carrier
state (reflected by the presence of HBsAg in serum) and exposure to dietary aflatoxin
(reflected by the presence of AFBi metabolites in urine) are the 2 major risk factors for
HCC in Shanghai men, accounting for at least three-fourths of cases (Ross et al., 1992;
Qian et al., 1994). In this report, we present the findings of our investigation on the
possible association between high serum testosterone and HCC development.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
73
II. Material and Methods
The design of the cohort study has been described previously (Ross et al., 1992;
Qian et al., 1994). Briefly, 18,244 mostly middle-aged (45-64 years) men residing in 4
geographically defined small areas of Shanghai were recruited between January 1986 and
September 1989. A 10 ml blood sample and a single void urine sample were collected at
recruitment, following an extensive in-person interview which emphasized dietary
exposures. Follow-up of this cohort has been accomplished through routine review of the
population-based Shanghai Cancer Registry records, routine review of death certificates
and annual recontact o f all surviving cohort members.
By September 30, 1993, the study had accumulated 97,592 man-years of
observation. Seventy-six incident cases of HCC had been identified, and the basis for
their diagnoses was as follows; histopathologic confirmation (n = 9), elevated serum
alpha-fetoprotein with consistent clinical and radiologic history (n = 21), positive
computerized axial tomography scan and/or ultrasonography with consistent clinical
history (n = 44) and death certificate only (n = 2).
Four hundred and ten controls were selected from the cohort to form the
comparison group. Multiple controls were matched to each case by age (within 1 year),
time of sample collection (within 1 month) and neighborhood of residence. Controls
were randomly selected from among all cohort members who were enrolled in the study
and had no history of HCC on the date of cancer diagnosis of the index case. Ten
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
74
controls were matched to each of the first 6 identified cases of HCC, while 5 controls
were matched to each of the remaining 70 cases.
Coded serum samples 6om HCC cases and controls were first tested for HBsAg
(AUSRIA, Abbott, North Chicago, IL), and the negative samples were further tested for
antibodies to the hepatitis B core antigen (anti-HBc) and antibodies to the hepatitis B
surface antigen (anti-HBs) (Corah, AUSab, Abbott). Coded serum samples fi^om all
subjects also were tested for anti-HCV using ELISA version 2.0 (Ortho, Raritan, NJ).
Testosterone was measured using a modified version of the procedure of
Goebelsmann et al. (1979). In the modified procedure, serum samples were extracted
with hexane:ethylacetate (1:1) prior to radioimmunoassay. Authentic testosterone was
purchased fi'om Steraloids (Wilton, NH) and 1,2-^H-testosterone was purchased firom
Dupont NEN (Boston, MA). Anti-testosterone-3-carboxymethyloxime-bovine serum
albumin serum was prepared in our laboratory.
Due to inadequate quantities of serum, values for one or more hepatitis B virus
(HB V)/hepatitis C virus (HCV) markers and/or testosterone were missing in a number of
subjects, as described below: HBsAg (1 control), Anti-HBc and/or anti-HBs (3 cases, 27
controls), anti-HCV (1 control) and testosterone (7 controls).
Testosterone levels among study subjects displayed a log normal distribution;
thus, formal statistical testing was performed on logarithmically transformed values of
testosterone, and geometric (as opposed to arithmetic) mean values are presented. The
analysis of covariance method was used (i) to compare testosterone levels across various
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
75
categories of control subjects as defined by their HBV serum marker status while
controlling for age at sample collection and (//) to compare testosterone levels between
HCC cases and controls while controlling for HBsAg status and age at sample collection
(Winer, 1971). The conditional logistic regression method was used to examine the
relationship between HCC risk and markers of HBV/HCV infections and serum
testosterone level (Breslow and Day, 1980). All p values quoted are 2-sided.
III. Results
HCV seropositivity was rare among cohort members; only 1 case and 1 control
tested positive for anti-HCV, yielding a RR of 5.0 (95% CL = 0.3, 79.9). Presence of
anti-HBs and anti-HBc was not associated with HCC risk. Relative to absence of all
HBV markers (HBsAg, anti-HBc or anti-HBs), the RRs for anti-HBs and anti-HBc
positivity were 1.0 (95% CL = 0.3,3.4) and 1.1 (95% CL = 0.3, 3.9), respectively. In
contrast, 50 HCC cases and 49 controls were positive for HBsAg, resulting in a 15.0-fold
increase in HCC risk (95% CL = 4.4, 51.6) relative to those negative for all HBV
markers.
We examined the relationship of serum testosterone by age at sample collection
among control subjects, and no association was observed {p = 0.55). However, to
facilitate comparison of results between our study and that of Yu and Chen (1993), age-
adjusted testosterone levels are presented below. We also examined, among controls.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
76
serum testosterone level by time of year the sample was collected. No seasonal
fluctuation in testosterone level was observed {p = 0.53/
Table 6-1 presents the geometric mean levels of serum testosterone in 4 mutually
exclusive groups of control subjects as defined by their HBV status:(/) subjects who
tested negative on all 3 markers, (//) subjects who were positive for anti-HBs but negative
for HBsAg and anti-HBc, (///) subjects who were negative for HBsAg but positive for
anti-HBc with or without Anti-HBs and (/v) subjects who were HBsAg-positive. There
was no difference in age-adjusted mean testosterone levels among the three subgroups of
subjects who were negative for HBsAg (/.e., groups 1-3; p = 0.65). In contrast, control
subjects positive for HBsAg exhibited a significantly higher mean testosterone level
compared to HBsAg-negative controls (574 vs. 475 ng/dl, p = 0.0006).
Relative to controls, HCC cases had a significantly higher age-adjusted mean
level of serum testosterone at the time of recruitment (570 vs. 485 ng/dl, p = 0.0005).
However, this difference was explicable on the basis of a higher proportion of HBsAg-
positive individuals among cases than controls. The 2-way analysis of covariance test
indicated a significant difference in testosterone levels between HBsAg-positive vs.-
negative individuals after controlling for case/control status (p = 0.0001) but no
difference in testosterone levels between cases and controls after controlling for HBsAg
status ip = 0.42; Table 6-2).
Table 6-3 presents the risk of HCC according to level of serum testosterone (in
tertiles) and serum HBsAg status. Cigarette smoking, a risk factor for HCC in Shanghai
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
77
Table 6-1 Age-adjusted mean level of serum testosterone according to HBV serum
markers in 382 control subjects, a nested case-control study of HCC*
HBV marker status Number o f controls Age-adjusted mean
testosterone (ng/dl)
None^ 44 479
Anti-HBs-positive only 51 456
Anti-HBc-positive but HBsAg-negative 239 479
HBsAg-positive 48 574
* Twenty-seven controls had missing values on I or more of the 3 HBV markers, and I
additional control had unknown serum testosterone. One-way analysis of covariance
with age at sample collection as the covariate: p (groups 1-4) = 0.006, p (groups 1-3) =
0.65 and p (HBsAg-positive vj. -negative) = 0.0006.
^ Absence of HBsAg, anti-HBc and anti-HBs.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
78
Table 6-2 Age-adjusted geometric mean level of serum testosterone in cases and
controls, a nested case-control study of HCC*
Cases Controls
Number
Age-adjusted
mean testosterone
(ng/dl)
Number
Age-adjusted
mean testosterone
(ng/dl)
All subjects 76 570 402 485
HBsAg-positive 50 633 48 574
HBsAg-negative 26 467 354 473
' Seven controls had unknown serum testosterone levels, and 1 additional control had
unknown HBsAg status. Two-way analysis of covariance with age at sample collection
as the covariate: p (HBsAg-positive vs. -negative) = 0.0001,/? (cases vs. controls) =
0.42.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
79
Table 6-3 Risk of HCC according to level of serum testosterone and serum HBsAg
status, a nested case-control study of HCC '
HBsAe-neeative HBsAg-positive
Ca/Co RR (95% CL) Ca/Co RR (95% CL)
1st tertile 10/124 1.0 8/9 8.5 (2.5, 28.9)
2nd tertile 7/122 0.8 (0.3,2.2) 11/13 9.4 (3.1,28.2)
3rd tertile 9/108 1.0 (0.4,2.8) 31/26 12.5 (5.1,31.0)
Eight controls with missing values (1 on HBsAg and 7 on serum testosterone) were
excluded from this analysis, which included duration o f cigarette smoking as a
covariate in the logistic regression model.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8 0
men, was adjusted for in the logistic regression analysis. There was no evidence o f an
association between serum testosterone and HCC risk among HBsAg-negative
individuals. Among those who were HBsAg-positive, there was a non-significant 50%
increase in risk between individuals in the highest vj. the lowest tertile.
rV. Discussion
Our results confirm the earlier observation of Yu and Chen (1993) of higher
serum testosterone levels in Chinese male HCC cases relative to controls. However,
unlike the Taiwan study, which demonstrated an increased testosterone level in cases
relative to controls irrespective of HBsAg status, the excess in cases in Shanghai with
high testosterone values was explicable on the basis of a higher proportion of HBsAg-
positive individuals in the case group. Since testosterone metabolism occurs primarily in
the liver, it would not be totally unexpected that chronic carriers of HBV would possess
higher and varying levels of circulating testosterone relative to unaffected individuals.
Hepatic enzymes, specifically steroid oxidoreductases, transform testosterone to different
reduced metabolites, which are then conjugated and excreted in urine (Briggs and
Brotherton, 1970). Alteration of these enzymes in liver disease may lead to lower
metabolism of testosterone and consequently higher circulating levels of this steroid.
There is evidence to suggest that HB-infected males are more likely than HB-
infected females to progress to the chronic carrier state. Lingao et al. (1986) reported that
among rural Filipinos with prior exposure to HBV (i.e., positive for any HBV markers).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
81
the age-adjusted HBsAg positivity rate in men (26%) was significantly higher than that in
women (18%). In Taiwan, although sero-prevalence of HB infection was identical
between two sexes (both at 50%), age-adjusted prevalence o f HBsAg was significantly
higher in men than women (12% vs. 9%) (Sung, 1981). Therefore, it is possible that
testosterone promotes progression to chronicity following primary HB infection,
therefore indirectly contributing to the development of HCC and the observed excess of
this disease worldwide.
The present data show that hepatitis C infection plays a negligible role in HCC
development in Shanghai, China; only 1 of 76 HCC cases and 1 of 409 control subjects
tested positive for anti-HCV. The serum samples from this study were submitted blindly
to the test laboratory and intermingled with serum samples from black and white patients
with HCC in Los Angeles. Prevalence of anti-HCV positivity among these recently
diagnosed HCC cases in Los Angeles was similar to that of cases diagnosed earlier from
the same case-control study (12/31 vs. 12/35 in males, and 8/27 V 5 . 3/16 in females; Yu
and Tong et al., 1990). Thus, there is no evidence that laboratory irregularities have
contributed to an under-estimation of the frequency of anti-HCV positivity in the study
population. Other studies conducted in high-risk but rural populations in close proximity
to Shanghai (Qidong and Haimen counties, which are about 60 km from Shanghai) have
reported anti-HCV positivity rates of 16-17% in HCC cases and 3-4% in control subjects
(Chen et al., 1994; Ye et al., 1994).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
82
REFERENCE
ABBOTT, R.D., YIN, Y., REED, D.M. and YANO, K., Risk of stroke in male cigarette
smokers. New Engl. J. Med., 315, 717-720 (1986).
ADAMES, C.F., Nutritive value of American foods in common unit. Agriculture
Handbook No. 456, p. 31, United States Department of Agriculture, Washington, DC:
United States Government Printing Office (1975).
BARON, A.E., FRANCESCm, S., BARRA, S., TALAMINI, R. and VECCHIA, C.L., A
comparison of the joint effects of alcohol and smoking on the risk of cancer across sites
in the upper aerodigestive tract. Cancer Epidemiol. Biomarkers Prev., 2,519-523 (1993).
BOFFETTA, P. and GARFINKEL, L., Alcohol drinking and mortality among men
enrolled in an American Cancer Society Prospective Study. Epidemiology, 1, 342-348
(1990).
BRESLOW, N.E. and DAY, N.E., Statistical methods in cancer research. Vol. 1, The
analysis o f case-control studies, lARC Scientific Publications 32, pp. 192-279, Lyon
(1980).
BRESLOW, N.E. and DAY, N.E., Statistical Methods in Cancer Research, Vol. 2, The
design and analysis o f cohort studies. lARC Scientific Publications 82, pp. 120-176,
Lyon (1987).
BRIGGS, M.H. and BROTHERTON, J. (eds.). Steroid biochemistry and pharmacology,
pp. 70-75, Academic Press, London (1970).
CARSTENSEN, J.M., PERSHAGEN, G. and EKLUND, G., Mortality in relation to
cigarette and pipe smoking: 16 years’ observation of 25000 Swedish men. J. Epidemiol.
Commun. Health, 41,166-172 (1987).
CENTERS FOR DISEASE CONTROL, Reducing the Health Consequences o f Smoking:
25 years ofProgress. A Report o f the Surgeon General 1989. DHHS Publications
(CDC)89-8411, pp. 37-102, Rockville, Maryland: US Department of Health and Human
Services (1989).
CENTERS FOR DISEASE CONTROL, Cigarette smoking-attributable mortality and
years of potential life lost - United States, 1990. Mortality Morbidity Weekly Report, 42,
645-649 (1993).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
83
CHEN, D., ROMAN, G.C., WU, G.-X., WU, Z.-S., YAO, C.-H., ZHANG, M. and
HIRSCH, R.P., Stroke in China (Sino-MONICA-Beijing Study) 1984-1986.
Neuroepidemiology, 11, 15-23(1992).
CHEN, G., ZI, X.-L., YU, S. Z., LU, J.L., CHEN, G.-C. and KANG, T.C., Case-control
study on the relation between hepatitis C and liver cancer in high-risk areas. Acta Oncol.
Sinica, 4, 30-32 (1994). (In Chinese with English abstract.)
COLDITZ, G.A., BRANCH, L.G., LIPNICK, R.J., WILLETT, W.C., ROSNER, B.,
POSNER, B. and HENNEKENS, C H., Moderate alcohol and decreased cardiovascular
mortality in an elderly cohort. Amer. Heart J., 109, 886-889 (1985).
COLDITZ, G.A., BONITA, R., STAMPFER, M.J., WILLETT, W.C., ROSNER, B.,
SPEIZER, F.E. and HENNEKENS, C.H., Cigarette smoking and risk of stroke in middle-
aged women. New Engl. J. Med., 318, 937-941 (1988).
COLLABORATIVE GROUP FOR INVESTIGATING PREVALENCE OF SMOKING,
Results of national sampling survey on prevalence of smoking. Chinese Med. J., 67, 229-
232 (1987). (in Chinese.)
CRIQUI, M.H., COWAN, L.D., TYROLER, H.A., BANGDIWALA, S., HEISS, G.,
WALLACE, R.B. and COHN, R., Lipoproteins as mediators for the effects of alcohol
consumption and cigarette smoking on cardiovascular mortality: results from the lipid
research clinics follow-up study. Amer. J. Epidemiol, 126, 629-637 (1987).
CULLEN, K.J., KNUIMAN, M.W. and WARD, N.J., Alcohol and mortality in
Busselton, Western Australia. Amer. J. Epidemiol, 137, 242-248 (1993).
DAGENAIS, G.R., ROBITAILLE, N.-M., LUPEIN, P.J., CHRISTEN, A., GINGRAS,
S., MOORJANI, S., MEYER, F. and ROCHON, J., First coronary heart disease event
rates in relation to major risk factors: Quebec cardiovascular study. Can. J. Cardiol, 6,
274-280 (1990).
DAGENAIS, G.R., AHMED, Z., ROBITAILLE, N.-M., GINGRAS, S., LUPEIN, P.J.,
CHRISTEN, A., MEYER, F. and ROCHON, J., Total and coronary heart disease
mortality in relation to major risk factors: Quebec cardiovascular study. Can. J. Cardiol
6, 59-65 (1990).
DE LABRY, L.O., GLYNN, R.J., LEVENSON, M R., HERMOS, J.A., LOCASTRO,
J.S. and VOKONAS, P.S., Alcohol consumption and mortality in an American male
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
84
population: recovering the U-shaped curve—findings fi-om the Normative Aging Study.
J. Stud. Alcohol, 53, 25-32 (1992).
DENG, J. and GAO, Y.T., The prevalence of smoking habit among 110,000 adult
residents in Shanghai urban area. Chinese J. Prev. Med., 5, 271-274 (1985). (in Chinese.)
DOLL, R. and PETO, R., Mortality in relation to smoking: 20 years’ observations on
male British doctors. Br. M ed J., 2,1525-1536 (1976).
DOLL, R., PETO, R., HALL, E., WHEATLEY, K. and GRAY, R., Mortality in relation
to consumption of alcohol: 13 years’ observations on male British doctors. Br. Med. J.,
309,911-918(1994).
FARCHI, G., FIDANZA, F., MARIOTTI, S. and MENOTTI, A., Alcohol and mortality
in the Italian rural cohorts of the seven countries study. Int. J. Epidemiol, 21, 74-81
(1992).
FREUND, K.M., BELANGER, A.J., D AGOSTINO, R.B. and KANNEL, W.B., The
health risks of smoking. The Framingham Study: 34 years of follow-up. Ann. Epidemiol,
3,417-424(1993).
FRIEDMAN, L A. and KIMBALL, A.W., Coronary heart disease mortality and alcohol
consumption in Framingham. Amer. J. Epidemiol, 1241,481-489 (1986).
FUCHS, C.S., STAMPFER, M.J., COLDITZ, G.A., GIOVANNUCCI, E.L., MANSON,
I.E., KAWACHI, L, HUNTER, D.J., HANKINSON, S.E., HENNEKENS, C.H.,
ROSNER, B., SPEIZER, F.E. and WILLETT, W.C., Alcohol consumption and mortality
among women. New Engl J. Med., 332, 1245-1250 (1995).
GAO, Y.-T. Smoking and lung cancer in Shanghai. In: Tobacco: a major international
health hazard (ZARIDZE, D.G. and PETO, R., eds). lARC Scientific Publications 74,
pp. 115-121, Lyon (1986).
GARFINKEL, L., Selection, follow-up, and analysis in the American Cancer Society
Prospective Studies. Nat. Cancer Inst. Monogr., 67,49-52 (1985).
GAZIANO, J.M., BURING, I.E., BRESLOW, J.L., GOLDHABER, S.Z., ROSNER, B.,
VANDENBURGH, M., WILLETT, W. and HENNEKENS, C.H., Moderate alcohol
intake, increased levels of high-density lipoprotein and its subfi-actions, and decreased
risk of myocardial infarction. New Engl J. M ed, 329,1829-1834 (1993).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
85
GIROUD, M., CREISSON, E., FAYOLLE, H., ANDRE, N., BECKER, P., MARTIN, D.
and DUMAS, R., Risk factors for primary cerebral hemorrhage: a population-based study
— the stroke registry of Dijon. Neuroepidemiology, 14,20-26 (1995).
GOEBELSMANN, U., BERNSTEIN, G.S., GALE, J.A., KLETZKY, O.A.,
NAKAMURAR, R.M., COULSON, A.H. and KORELITZ, J.J., Serum gonadotropin,
testosterone, estradiol and estrone levels prior to and following bilateral vasectomy. In:
I.H. Lepow and R. Crozier (eds.). Vasectomy: immunologic and pathophysiologic effects
in animals and man, pp. 165-175, Academic Press, New York (1979).
GOLDBERG, R.J., BURCHFIEL, C.M., REED, D M., WERGOWSKE, G. and CHIU,
D., A prospective study of the health effects of alcohol consumption in middle-aged and
elderly men: the Honolulu Heart Program. Circulation, 89, 651-659 (1994).
GOLDBOHM, R.A., VAN DEN BRANDT, P.A., VAN’T VEER, P., DORANT, E.,
STURMANS, F. and HERMUS, R.J.J., Prospective study on alcohol consumption and
the risk of cancer of the colon and rectum in the Netherlands. Cancer Causes Control, 5,
95-104(1994).
GORDON, T. and DOYLE, J.T., Drinking and mortality. Amer. J. Epidemiol., 125, 263-
270 (1987).
GRONBAEK, M., DEIS, A., SORENSEN, T.I.A., BECKER, U., BORCH-JOHNSEN,
K., MULLER, C., SCHNOHR, P. and JESEN, G., Influence of sex, age, body mass
index, and smoking on alcohol intake and mortality. Br. Med. J., 308, 302-306 (1994).
GRONBAEK, M., DEIS, A., SORENSEN, T.I.A., BECKER, U., SCHNOHR, P. and
JESEN, G., Mortality associated with moderate intakes of wine, beer, and spirits. Br.
Med. J., 310,1165-1169 (1995).
GURTOO, H.L. and MOTYCKA, L., Effect of sex difference on the in vitro and in vivo
metabolism of aflatoxin Bj by the rat. Cancer Res., 36,4663-4671 (1976).
HAHEIM, L.L., HOLME, L, HJERMANN, I. and LEREN, P., Risk factors of stroke
incidence and mortality: a 12-year follow-up of the Oslo study. Stroke, 24,1484-1489
(1993).
KAHN, H.A., The Dorn study of smoking and mortality among US Veterans: Report on
8.5 years of observation. Nat. Cancer Inst. Monogr., 19,1-125 (1966).
HAMMOND, E.C.and GARFINKEL, L., Smoking habits of men and women. J. Nat.
Cancer Inst., 27,419-442 (1961).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
86
HAMMOND, E.C., Smoking in relation to the death rates of 1 million men and women.
Nat. Cancer Inst. Monogr., 19, 127-204(1966).
HAMMOND, B.C. and SEIDMEN, H., Smoking and cancer in the United States. Prev.
Med., 9,169-173 (1980).
HARMSEN, P., ROSENGREN, A , TSIPOGIANNI, A. and WILHELMSEN, L., Risk
factors for stroke in middle-aged men in Goteborg, Sweden. Stroke, 21,223-229 (1990).
HENDRIKS, H.F.J., VEENSTRA, J., VELTHUIS-TE WIERIK, E.J.M., SCHAAFSMA,
G. and KLUFT, C., Effect of moderate dose of alcohol with evening meal on fibrinolytic
factors. Br. Med. J., 308,1003-1006 (1994).
ISO, H., KITAMURA, A., SHIMAMOTO, T., SANKAI, T., NAITO, Y., SATO, S.,
KIYAMA, M., IIDA, M. and KOMACHI, Y., Alcohol intake and the risk of
cardiovascular disease in middle-aged Japanese men. Stroke, 26, 767-773 (1995).
JEDRYCHOWSKI, W., BOEING, H., WAHRENDORF, J., POPIELA, T., TOBIASZ-
ADAMCZYK, B. and KULIG, J., Vodka consumption, tobacco smoking and risk of
gastric cancer in Poland. Int. J. Epidemiol, 22, 606-613 (1993).
KLATSKY, A.L., ARMSTRONG, M.A. and FRIEDMAN, G.D., Risk of cardiovascular
mortality in alcohol drinkers, ex-drinkers and nondrinkers. Amer. J. Cardiol, 66,1237-
1242 (1990).
KLATSKY, A.L., ARMSTRONG, M.A. and FRIEDMAN, G.D., Alcohol and mortality.
Ann. Intern. Med., 117, 646-654 (1992).
KODAMA, K., Stroke trends in Japan. Ann. Epidemiol, 3, 524-528 (1993).
KOZAREVIC, D., DEMIROVIC, J., GORDON, T., KAELBER, C.T., MCGEE, D. and
ZUKEL, W.J., Drinking habits and coronary heart disease: the Yugoslavia cardiovascular
disease study. Amer. J. Epidemiol, 116, 748-758 (1982).
LAM, K.C., YU, M.C., LEUNG, J.W.C. and HENDERSON, B.E., Hepatitis B virus and
cigarette smoking: risk factors for hepatocellular carcinoma in Hong Kong. Cancer Res.,
42, 5246-5248 (1982).
LAVY, A., FUHRMAN, B., MARKEL, A., DANKNER, G., BEN-AMOTZ, A.,
PRESSER, D. and AVIRAM, M., Effect of dietary supplementation of red or white wine
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
87
on human blood chemistry, hematology and coagulation: favorable effect of red wine on
plasma high-density lipoprotein. Ann. Nutr. Metab., 38,287-294 (1994).
LEUNG, H.M. and KUPPER, L.L., Comparison of confidence intervals for attributable
risk. Biometrics, 37,293-302(1981).
LI, J. and 30 OTHERS (eds.). Atlas o f cancer mortality in the People's Republic o f
China, pp. vi, 39-46, China Map Press, Shanghai (1979).
LINGAO, A.L., DOMINGO, E.O., WEST, S., REYES, C M., GASMEN, S., VITERBO,
G., TTU, E. and LANSANG, M.A., Seroepidemiology of hepatitis B virus in the
Philippines. Amer. J Epidemiol, 123,473-480 (1986).
LINN, S., CAROLL, M., JOHNSON, C., FULWOOD, R., KALSBEEK, W. and
BRIEFEL, R., High-density lipoprotein cholesterol and alcohol consumption in US white
and black adults: data from NHANES II. Amer. J. Public Health, 83, 811-816 (1993).
LONGNECKER, M.P., ORZA, M.J., ADAMS, M.E., VIOQUE, J. and CHALMERS,
T.C., A meta-analysis of alcoholic beverage consumption in relation to risk of colorectal
cancer. Cancer Causes Control, 1, 59-68 (1990).
MAHESWARAN, R., GILL, J.S., DAVIS, P. and BEEVERS, D.G., High blood pressure
due to alcohol: a rapidly reversible effect. Hypertension, 17, 787-792 (1991).
MARMOT, M.G., ROSE, G., SHIPLEY, M.J. and THOMAS, B.J., Alcohol and
mortality: a U-shaped curve. Lancet, i, 580-583 (1981).
MARMOT, M.G., ELLIOTT, P., SHIPLEY, M.J., DYER, A.R., UESHIMA, H.,
BEEVERS, D.G., STAMLER, R., KESTELOOT, H., ROSE, G. and STAMLER, J.,
Alcohol and blood pressure: the INTERSALT study. Br. Med. J., 308, 1263-1267
(1994).
MARQUES-VIDAL, P., CAMBOU, J.P., NICAUD, V., LUC, G., EVANS, A.,
ARVEILER, D., BINGHAM, A. and CAMBIEN, F., Cardiovascular risk factors and
alcohol consumption in France and Northern Ireland. Atherosclerosis, 115,225-232
(1995).
MENOTTI, A., KEYS, A., KROMHOUT, D., NISSINEN, A., BLACKBURN, H.,
FIDANZA, F., GIAMPAOLI, S., KARVONEN, M.J., PEKKANEN, J., PUNSAR, S. and
SECCARECCIA, F., Twenty-five-year mortality from coronary heart disease and its
prediction in five cohorts of middle-aged men in Finland, The Netherlands, and Italy.
Prev. Med, 19, 270-278 (1990).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
88
MEYER, F. and WHITE, E., Alcohol and nutrients in relation to colon cancer in middle-
aged adults. Amer. J. Epidemiol., 138,225-236 (1993).
NATIONAL CENTER FOR HEALTH STATISTICS, Vital Statistics o f the United
States, 1989. Vol. 2, Mortality, Part A. DHHS Publications (PHS) 93-1101, Hyatts ville,
Maryland: US Department of Health Service (1993).
NOVOTNY, T.E. and PETO, R., Estimates of future adverse health effects of smoking in
China. Public Health Report, 103, 552-553 (1988).
PAGANINI-HILL, A. and HSU, G., Smoking and mortality among residents of a
California retirement community. Amer. J. Public Health, 84,992-995 (1994).
PARKIN, D.M., MUIR, C.S., WHELAN, S.L., GAO, Y.-T., FARLAY, J. and POWELL,
J. (eds). Cancer Incidence in Five Continents, Vol. 6. lARC Scientific Publications 120,
Lyon (1992).
PAUNIO, M., HEINONEN, O.P., VIRTAMO, J., KLAG, M.J., MANNINEN, V.,
ALBANES, D. and COMSTOCK, G.W., HDL cholesterol and mortality in Finnish men
with special reference to alcohol intake. Circulation, 90,2909-1918 (1994).
PETO, R., Tobacco: U.K. and China (letter). Lancet, 2, 1038 (1986).
PIKAAR, N.A., WEDEL, M., VAN DER BEEK, E.J., VAN DOKKUM, W., KEMPEN,
H.J.M., KLUFT, C., OCKHUIZEN, T. and HERMUS, R.J.J., Effects o f moderate alcohol
consumption on platelet aggregation, fibrinolysis, and blood lipids. Metabolism, 36, 538-
543 (1987).
QIAN, G. S., ROSS, R.K., YU, M.C., YUAN, J.-M., GAO, Y.-T., HENDERSON, B.E.,
WOGAN, G.N. and GROOPMAN, J.D., A follow-up study of urinary markers of
aflatoxin exposure and liver cancer risk in Shanghai, People’s Republic of China. Cancer
Epidemiol. Biomarkers Prev., 3,3-10 (1994).
RAGLAND, D R. and BRAND, R.J., Coronary heart disease mortality in the Western
Collaborative Group Study: Follow-up experience of 22 years. Amer. J. Epidemiol., 127,
462-475 (1988).
REHM, J. and SEMPOS, C.T., Alcohol consumption and all-cause mortality. Addiction,
90,471-480(1995).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
89
RENAUD, S. and DE LORGERIL, M., Wine, alcohol, platelets, and the French paradox
for coronary heart disease. Lancet, 339,1523-1526 (1992).
RIDKER, P.M., VAUGHAN, D.E., STAMPFER, M.J., GLYNN, R.J. and
HENNEKENS, C H., Association of moderate alcohol consumption and plasma
concentration of endogenous tissue-type plasminogen activator. J. Amer. Med. Assoc.,
272,929-933 (1994).
RIMM, E.B., GIOVANNUCCI, E.L., WILLETT, W.C., COLDITZ, G.A., ASCHERIO,
A., ROSNER, B. and STAMPFER, M.J., Prospective study of alcohol consumption and
risk of coronary disease in men. Lancet, 338,464-468 (1991).
RODGERS, H., AITKEN, P.D., FRENCH, J.M., CURLESS, R.H., BATES, D. and
JAMES, O.F.W., Alcohol and stroke: a case-control study of drinking habits past and
present. Stroke, 24, 1473-1477 (1993).
ROBBINS, A.S., MANSON, I.E., LEE, I.-M., SATTERFIELD, S. and HENNEKENS,
C H., Cigarette smoking and stroke in a cohort of U.S. male physicians. Ann. Intern.
Med., 120,458-462 (1994).
ROGOT, E. and MURRAY, J.L., Smoking and causes of death among U.S. Veterans: 16
years of observation. Public Health Report, 95,213-222 (1980).
ROSS, R.K., YUAN, J.-M., YU, M.C., WOGAN, G.N., QIAN, G.-S., TU, J.-T.,
GROOPMAN, J.D., GAO, Y.-T. and HENDERSON, B.E., Urinary aflatoxin biomarkers
and risk of hepatocellular carcinoma. Lancet, 339, 943-946 (1992).
ROSS, R.K., YUAN, J.-M., HENDERSON, B.E., PARK, J., GAO, Y.-T. and YU, M.C.,
A prospective evaluation of dietary and other predictors of fatal stroke in Shanghai,
China. Amer. J. Epidemiol, (submitted).
SEMENCIW, R.M., MORRISON, H.L, MAO, Y., JOHANSEN, H., DAVIES, J.W. and
WIGLE, D.T., Major risk factors for cardiovascular disease mortality in adults: results
from the nutrition Canada survey cohort. Int. J. Epidemiol., 17, 317-324 (1988).
SERDULA, M.K., KOONG, S.-L., WILLIAMSON, D.F., AND A, R.F., MAD ANS, J.H.,
KLEINMAN, J.C. and BYERS, T., Alcohol intake and subsequent mortality: Findings
from the NHANES I follow-up study. J. Stud. Alcohol, 56,233-239 (1995).
SHAPER, A.G., WANNAMETHEE, G. and WALKER, M., Alcohol and mortality in
British men: explaining the U-shaped curve. Lancet, ii, 1267-1273 (1988).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
90
SHAPER, A.G., Alcohol and mortality: a review of prospective studies. 3r. J. Addict.,
85, 837-847 (1990).
SHOPLAND, D.R., EYRE, H.J. and PECHACEK, T.F., Smoking-attributable cancer
mortality in 1991 : is lung cancer now the leading cause of death among smokers in the
United States? J. Nat. Cancer Inst., 83, 1142-1148 (1991).
STAMPFER, M.J., COLDITZ, G.A., WILLETT, W.C., SPEIZER, F.E. and
HENNEKENS, C.H., A prospective study of moderate alcohol consumption and the risk
of coronary disease and stroke in women. New Engl. J. Med., 319, 267-273 (1988).
SUH, I., SHATEN, J., CUTLER, J.A. and KULLER, L.H., Alcohol use and mortality
from coronary heart disease: the role of high-density lipoprotein cholesterol. Ann. Intern.
M ed, 116,881-887(1992).
SUNG, J.-L., Hepatitis B virus infection and its sequelae in Taiwan. Proc. nat. Sci.
Council, Part B: Basic science (Taiwan), 5,385-399 (1981).
THOM, T.J., Stroke mortality trends: an international perspective. Ann. Epidemiol., 3,
509-518(1993).
VAUGHAN, T.L., DAVIS, S., KRISTAL, A. and THOMAS, D.B., Obesity, alcohol, and
tobacco as risk factors for cancers of the esophagus and gastric cardia: adenocarcinoma
versus squamous cell carcinoma. Cancer Epidemiol. Biomarkers Prev., 4, 85-92 (1995).
WILLETT, W.C., GREEN, A., STAMPFER, M.J., SPEIZER, F.E., COLDITZ, G.A.,
ROSNER, B., MONSON, R., STASON, W. and HENNEKENS, C.H., Relative and
absolute excess risks of coronary heart disease among women who smoke cigarettes.
New Engl. J. M ed, 317, 1303-1309 (1987).
WINER, B.J., Statistical principles in experimental design (2nd ed.), pp. 149-260,
McGraw Hill, New York (1971).
WOGAN, G.N., Aflatoxins as risk factors for hepatocellular carcinoma in humans.
Cancer Res., 52, 2114S-2118S (1992).
WOLF, P.A., D AGOSTINO, R.B., KANNEL, W.B., BONITA, R. and BELANGER,
A.J., Cigarette smoking as a risk factor for stroke: The Framingham study. J. Amer. Med.
Assoc., 259, 1025-1029(1988).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
91
WU-WILLIAMS, A.H., YU, M.C. and MACK, TM ., Life-style, workplace, and stomach
cancer by subsite in young men of Los Angeles County. Cancer Res., 50,2569-2576
(1990).
YANO, K., RHOADS, G.G. and KAGAN, A., Coffee, alcohol and risk of coronary heart
disease among Japanese men living in Hawaii. New Engl. J. Med., 297,405-409 (1977).
YE, B., SHEN, J., XU, Y., NIU, J.-Y., CHEN, J.-G., ZHANG, B. C., LIU, B. and
JIANG, Y.-H., Etiologic study on the relationship between HBV, HCV and HCC.
Chinese J. Epidemiol., 15, 131-134 (1994). (In Chinese with English abstract.)
YU, J.J., MATTSON, M.E., BOYD, G.M., MUELLER, M.D., SHOPLAND, D.R.,
PECHACEK, T.F. and CULLEN, J.W., A comparison of smoking patterns in the
People’s Republic of China with the United States: An impending health catastrophe in
the middle Kingdom. J. Amer. Med. Assoc., 264, -.1575-1579 (1990).
YU, M.C., TONG, M.J., COURSAGET, P., ROSS, R.K., GOVINDARAJAN, S. and
HENDERSON, B E., Prevalence of hepatitis B and C \iral markers in black and white
patients with hepatocellular carcinoma in the United States. J. Nat. Cancer Inst., 82,
1038-1041 (1990).
YU, M.C., TONG, M.J., GOVINDARAJAN, S. and HENDERSON, B.E., Nonviral risk
factors for hepatocellular carcinoma in a low-risk population, the non-Asians of Los
Angeles County, California. J. Nat. Cancer Inst., 83,1820-1826 (1991).
YU, M.-W. and CHEN, C.-J., Elevated serum testosterone levels and risk of
hepatocellular carcinoma. Cancer Res., 52, 790-794 (1993).
YUAN, J.-M., ROSS, R.K., WANG, X.L., GAO, Y.T., HENDERSON, B.E. and YU,
M.C., Morbidity and mortality in relation to cigarette smoking: a prospective male cohort
study in Shanghai, China. J. Amer. Med. Assoc., (1996, in press).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
92
APPENDIXES
Appendix A.
Respondent Information Sheet
Shanghai Cohort Study, 198S
Rle = ABCDBF
Study Subject's Identification Number
(attach label here): ___________________________________ 1-5
Enter the following information prior to interviewing, and verify that
it is complete and accurate.
1. Name:
2. Sex: male 1 6
female ............. 2
3. Present address:
District: _________________________ 7-8
Street office:_______________________________________ 9-10
Neighborhood committee: ___________________________ 11-13
Household group:____________________________________
Street name and house number: ____________________ ___
Telephone num ber:__________________________________
4. Place of work: ___________________________________________
5. Work phone:_____________________________________________
6. Date of birth: 19 /_____ / 14-21
year month day
7. Age (in years):_________________________ 22-23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
93
Name of interviewer;
Interviewer identification num ber:______________________________ 24-26
Date of interview: 19 /_____ / 27-34
year month day
Time interview b eg an :__________________________________ am/pm
Time interview e n d e d :__________________________________ am/pm
Length of interview (in minutes) : ________________________________ 35-36
GENERAL INTRODUCTION:
Hello, my name is [YOUR NAME], May I please speak with [STUDY
SUBJECT'S NAME].
INTRODUCTION TO STUDY SUBJECT:
I am here on behalf of the First Shanghai Medical College. You have been
selected to help us in our health survey by answering some questions about
your health and habits, and by providing us with a blood and a urine specimen.
SECTION A: GENERAL INFORMATION
First, I would like to ask you some questions about your background.
A1. Where were you bom?
________________ I ____________ /__________________________ _37-38
County/City Province Country if not China
(if not Shanghai)
At what age did you move to Shanghai?
Age years 39-40
or 19 ”
A2. Are you currently married, divorced, widowed, or never married?
Married ......................... 1
Widowed ....................... 2
Divorced ....................... 3 _41
Never married .............. 4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
94
A3. What is the highest level of schooling you have completed?
No formal schooling .
Primary school .........
Junior middle school .
Senior middle school
Technical institute ...
University .................
0
1
2
3 42
4 —
5
A4. How many people are there in your household?
_________people 43-44
AS. What is the total monthly income of your household?
_________ yuan 45-48
A G . How tall are you?
_________ cm. 49-51
A7. What is your usual adult weight?
_________ jin( = 500g) 52-54
A3. What is your current occupation?
Is this your usual occupation?
55-57
Yes ................................ 1
No ................................. 2 _ 5 8
(If no)
What is your usual occupation?
59-61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
95
SECTIONS: U SE OF TOBACCO
Now I have some questions about your smoking habits.
81. Have you ever smoked at least one cigarette a day for six months
or longer?
Yes ................................ 1
No .................................. 2 (go to 88) _ 6 2
82. At what age did you start smoking cigarettes daily?
Age years 63-64
or 19 2ZZ “
83. Do you smoke cigarettes daily now?
Yes 1 (go to 85)
No .................................. 2 __65
84. At what age did you stop smoking cigarettes daily?
A ge____ years 66-67
or 19 ■"
85. Please tell me how many years you have smoked cigarettes, excluding
the years when you did not smoke.
_________ years__________________________________________________68-69
86. (Do/did) you usually roll your own cigarettes or buy them in a store?
Roll own ...................... 1
Store bought ............... 2 70
8oth .............................. 3 —
87. During most of your adult life when you smoked, how many store
bought cigarettes did you smoke on an average day?
_________ cigarettes __71-72
How many liang of cigarette tobacco did you use in an average
month?
_________ liang_________________________________________________73-74
88. Have you ever smoked any kind of pipe daily for six months or
longer?
Yes ................................ 1
No .................................. 2 (g o to 815) 75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
96
89. At what age did you start smoking a pipe daily?
Age____years 76-77
or 19 2 2 —
810. Do you smoke a pipe daily now?
Yes ............................... 1 (goto 812)
No ................................. 2 _ 7 8
811. At what age did you stop smoking a pipe daily?
Age years 79-80
or 19 2 Z ~
812. Please tell me how many years you have smoked a pipe, excluding
the years when you did not smoke.
_________years_________________________________________________81-82
813. (Do/did) you usually smoke a dry pipe or a water pipe?
Dry pipe ........................ 1
Water pipe ................... 2 83
8oth .............................. 3 —
814. On the average, during most of your adult life when you smoked a
pipe, about how many liang of tobacco did you smoke per month?
_________liang __84-85
815. Did you ever smoke anything else besides cigarettes or a pipe?
Yes ................................ 1
No ................................. 2 (go to Cl) 86
816. What did you smoke?
87
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
97
SECTION C: U SE OF ALCOHOL
Now, I have some questions about your drinking habits.
Cl. Have you ever drunk beer, wine, or spirits at least once a week
for 6 months or longer?
Yes ............................... 1
No .................................. 2 (go to 01} __88
C2. At what age did you start to drink at least once a week?
Age years 89-90
or 19 “
C3. Do you drink at least once a week now?
Yes ............................... 1 (go to C5)
No ................................. 2 _ 91
C4. At what age did you stop drinking at least once a week?
A ge____ years 92-93
or 19 ”
C5. Please tell me how many years you have drunk beer, wine or nease ten me now many years you nave drunic beer, wine
spirits at least once a week, excluding the years when you
did not drink.
_________ years __94-95
C6. During the years you were drinking at least once a week, how
many times per week did you drink beer on the average?
_________ times __96-97
And how many liang did you drink each time?
_________ liang __98-99
C7. On the average, how many times per week did you drink wine
such as rice wine and fruit wine?
_________ times 100-101
.4nd how many liang did you drink each time?
_________ liang 102-103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
98
C8. On the average, how many times per week did you drink spirits
such as white liquor?
_________ times 104-105
And how many liang did you drink each time?
liang 106-107
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
99
SECTION D: D IET
Rle = DIET.DBF
ID#; 1-5
Now, I would like to ask you some questions about your usual diet as
an adult.
D1. During most of your adult life, how many meals did you normally
eat on an average day?
number of meals
I will now read you a food group and then ask you how often you usually
ate these foods. Please tell me now often you ate these foods in number
of times per day, times per week, times per month or times per year. For
seasonal items, please also tell me how many months you ate the food
item during the year. There are no right or wrong answers. Please give
the most accurate answer you can.
D2. Corn
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ................ 3
Times per week ............ 4
Times per month .......... 5
Times per year .............. 6
times 8-9
D3. Corn oil
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ............... 3 10
Times per week ............ 4
Times per month .......... 5
Times per year .............. 6
times 11-12
On the average, how many liang of corn oil did your household
use each month?
_________ liang 13-14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
D4. Wheat and other grains including bread, steamed rolls, and noodles
Never eaten ................. 1
Don't know if eaten . . . 2
Times per day ............... 3__________________________________________IS
Times per week ............ 4
Times per month .......... 5
Times per year ............. 6
times 16-17
05. Dried peas or beans of any kind
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ............... 3 __18
Times per week ........... 4
Times per month .......... 5
Times per year ............. 6
times 19-21
D6. Peanuts
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ............... 3 __22
Times per week ........... 4
Times per month .......... 5
Times per year ............. 6
times 23-25
D7. Peanut oil
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ............... 3 __26
Times per week ........... 4
Times per month .......... 5
Times per year ............. 6
times 27-28
On the average, how many liang of peanut oil did your household
use each month?
_________ liang 29-30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
101
D8. P e a n u t b u tte r
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ............... 3 31
Times per week ............ 4 —
Times per month .......... 5
Times per year .............. 6
______________ times __32-33
D9. Rice
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ............... 3 3 4
Times per week ............ 4 —
Times per month .......... 5
Times per year ............. 6
______________ times 35-36
DIO. Fresh bean curd or bean curd skin
Never eaten ................. 1
Don't know if eaten . . . 2
Times per day ............... 3 3 7
Times per week ............ 4 —
Times per month .......... 5
Times per year .............. 6
______________ times 38-39
D ll. Fresh fish
Never eaten ................. 1
Don't know if eaten . . . 2
Times per day ............... 3 40
Times per week ............ 4 —
Times per month .......... 5
Times per year .............. 6
______________ times 41-42
D12. Fresh shrimp, clam, crab, or squid
Never eaten ................. 1
Don't know if eaten . . . 2
Times per day ............... 3 43
Times per week ............ 4 —
Times per month .......... 5
Times per year .............. 6
______________ times 44-45
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
D 13. Fresh fa tty p o rk
014. Fresh lean pork
015. Fresh chicken or duck
016. Cow or goat milk
102
Never eaten 1 . .
Oon't know if eaten . . . 2
Times per day ............... 3 46
Times per week ........... 4 —
Times per month .......... 5
Times per year ............. 6
______________ times 47-48
Never eaten ................. 1
Oon't know if eaten . . . 2
Times per day ............... 3 4 9
Times per week ........... 4 —
Times per month .......... 5
Times per year ............. 6
_______________times 50-51
Never eaten ................. 1
Oon't know if eaten . .. 2
Times per day ............... 3 52
Times per week ........... 4 —
Times per month .......... 5
Times per year ............. 6
times 53-54
Never eaten ................. 1
Oon't know if eaten . .. 2
Times per day ............... 3 55
Times per week ........... 4 —
Times per month .......... 5
Times per year ............. 6
times 56-57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
103
D17. Fresh e g g s o f a n y k in d
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ................ 3 58
Times per week ............ 4
Times per month .......... 5
Times per year .............. 6
times 59-60
D18. Liver of any kind
Never eaten ................. ^
Don't know if eaten . .. 2
Times per day ............... 3 61
Times per week ............ 4
Times per month .......... 5
Times per year .............. 6
________times 62-63
D19. Fresh vegetables of any kind
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ............... 3 64
Times per week ............ 4
Times per month .......... 5
Times per year .............. 6
times 65-66
D2G. Dark green leafy vegetables, such as spinach, winter cress,
or mustard greens
Never eaten .................... 1
Don't know if eaten . .. 2
Times per day .................. 3 __67
Times per week .............. 4
Times per month ............ 5
Times per year ................ 6
times 68-70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
104
021. Fresh green beans and peas
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day 3 71
Times per week ............ 4
Times per month .......... 5
Times per year ............. 6
times 72-73
022. Green squash and gourd such as 'koogua* and 'seequa'
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ............... 3 74
Times per week ............ 4
Times per month .......... 5
Times per year .............. 6
times 75-76
023. Light green and white vegetables such as cabbage,
cauliflower, and eggplant
Never eaten ................. 1
Don't know if eaten . . . 2
Times per day ............... 3 77
Times per week ............ 4
Times per month .......... 5
Times per year .............. 6
times 78-80
024. Carrots
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ............... 3 _81
Times per week ............ 4
Times per month .......... 5
Times per year ............. 6
times 82-83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
105
025. Tomatoes in season
Never eaten .................... 1
Don't know if eaten . . . 2
Times per day .................. 3 84
Times per week .............. 4
Times per month ............ 5
Times per year ................ 6
______________ times _85-86
______________ number of months of consumption 87-88
026. Pumpkin and other orange winter squash
Never eaten .................... 1
Oon't know if eaten . . . 2
Times per day .................. 3 89
Times per week .............. 4
Times per month ............ 5
Times per year ................ 6
______________ times __90-91
027. Red peppers (sweet or hot) in season
Never eaten ................... 1
Oon't know if eaten . . . 2
Times per day .................. 3 92
Times per week .............. 4 times per week ............ »
Times per month .......... 5
Times per year ............. 6
______________ times _93-94
______________ number of months of consumption _95-96
028. Green pepper (sweet or hot) in season
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ............... 3 _97
Times per week ............ 4
Times per month .......... 5
Times per year ............. 6
______________ times _98-99
______________ number of months of consumption 100-101
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
106
D 29. W h ite p o ta to
Never eaten ................. 1
Don't know if eaten ... 2
Times per day ............... 3 __102
Times per week ........... 4
Times per month ......... 5
Times per year ............. 6
times 103-105
D30. Pale sweet potato in season
Never eaten ................. 1
Don't know if eaten ... 2
Times per day ............... 3 106
Times per week ........... 4
Times per month ......... 5
Times per year ............. 6
______________ times 107-108
______________ number of months of consumption _109-110
D31. Dark sweet potato in season
Never eaten ................. 1
Don’ t know if eaten ... 2
Times per day ............... 3 111
Times per week ........... 4
Times per month ......... 5
Times per year ............. 6
______________ times 112-113
______________ number of months of consumption 114-115
D32. Fresh fruits of any kind
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ............... 3 116
Times per week ........... 4
Times per month .......... 5
Times per year ............. 6
times 117-118
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
107
D33. O ra n g e o r ta n g e r in e
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ............... 3 119
Times per week ........... 4
Times per month ......... 5
Times per year ............. 6
times 120-122
D34. Watermelon in season
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ............... 3 123
Times per week ........... 4
Times per month ......... 5
Times per year ............. 6
______________ times _ 1 24-125
______________ number of months of consumption __126-127
D35. Other melons such as honeydew in season
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ............... 3 128
Times per week ........... 4
Times per month ......... 5
Times per year ............. 6
______________ times _ 1 29-130
______________ number of months of consumption 131-132
D36. Loquat in season
Never eaten ................. 1
Don't know if eaten ... 2
Times per day ............... 3 __133
Times per week ........... 4
Times per month ......... 5
Times per year ............. 6
______________times 134-135
______________number of months of consumption 136-137
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
108
D 37. P ersim m on in se aso n
Never eaten ................. 1
Don't know if eaten . . . 2
Times per day ............... 3 138
Times per week ........... 4
Times per month .......... 5
Times per year ............. 6
______________ times _ 1 39-140
______________ number of months of consumption 141-142
038. Apricots in season
Never eaten ................. 1
Don't know if eaten . . . 2
Times per day ............... 3 143
Times per week ........... 4
Times per month ......... 5
Times per year ............. 6
______________ times 144-145
______________ number of months of consumption 146-147
D39. Other fruits such as apple, pear and banana
Never eaten ................. 1
Don't know if eaten . . . 2
Times per day ............... 3 148
Times per week ........... 4
Times per month .......... 5
Times per year ............. 6
___________ times 149-150
D40. Fermented bean curd (Fuyu)
Never eaten ................. 1
Don't know if eaten . . . 2
Times per day ............... 3 151
Times per week ........... 4
Times per month .......... 5
Times per year ............. 6
times 152-153
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
109
D41. F e rm e n te d b e a n p a s te s
Never eaten ................. 1
Don't know if eaten . . . 2
Times per day ............... 3_________________________________________154
Times per week ............ 4
Times per month .......... 5
Times per year ............. 6
times 155-156
D42. Salted vegetables or roots
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ............... 3 __157
Times per week ........... 4
Times per month .......... 5
Times per year ............. 6
times 158-159
D43. Salted fish
Never eaten ................. 1
Don't know if eaten . . . 2
Times per day ............... 3 __160
Times per week ............ 4
Times per month .......... 5
Times per year ............. 6
times 161-162
D44. Salted pork, shrimp or crab
Never eaten ................. 1
Don't know if eaten . .. 2
Times per day ............... 3 __163
Times per week ........... 4
Times per month .......... 5
Times per year ............. 6
times 164-166
D45. Sausage, ham or other cured meats
Never eaten ................... 1
Don't know if eaten
Times per day ........
Times per week
Times per month ..
Times per year ___
2
3 167
4 —
5
6
times 168-170
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
110
D46. Pickled vegetables
Never eaten ................. 1
171
CO LC» I • • • • • • • • • * I
Don't know if eaten . .. 2
Times per day ............... 3
Times per week ............ 4
Times per month .......... 5
Times per year ............. 6
times 172-173
D47. Salted eggs of any kind
Never eaten ................. 1
Don't know if eaten ... 2
Times per day ............... 3 __174
Times per week ............ 4
Times per month .......... 5
Times per year ............. 6
times 175-177
D48. Marinating sauce such as soy sauce and oyster sauce
Never eaten ................... 1
Don't know if eaten . .. 2
Times per day ................. 3 __178
Times per week .............. 4
Times per month ............ 5
Times per year ................ 6
times 179-180
D49. Sesame oil or rape seed oil
Never eaten ................. 1
Don't know if eaten ... 2
Times per day ............... 3 __181
Times per week ............ 4
Times per month ......... 5
Times per year ............. 6
______________ times __182-183
On the average, how many liang of these oils did your
household use each month?
______________ liangs _184-186
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I l l
D50. Soy bean oil
Never eaten ................. 1
187
Don't know if eaten ... 2
Times per day ............... 3
Times per week ........... 4
Times per month .......... 5
Times per year ............. 6
______________ times 189-190
On the average, how many liang of this oil did your
household use each month?
_______________liangs 191-193
051. Lard
Never eaten ................. 1
Don't know if eaten ... 2
Times per day ............... 3 194
Times per week ........... 4
Times per month ......... 5
Times per year ............. 6
_______________times 195-196
On the average, how many liang of these oils did your
household use each month?
_______________liangs 197-199
052. Did you take any pills, especially vitamin pills, at least once
a day for two weeks or longer during the past month?
Yes ................................ 1 200
No .................................. 2 —
If yes, pill name:
1. _________________________ _ 2 0 1-203
2. 204-206
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
112
S E C n C N E : MEDICAL HISTORY
FILEzDESE.DBF
ID # _ l-5
Now. I am going to read to you a list of health conditions. I would
like to know if you were ever told by a doctor that you had any one
of the following conditions.
El. Hepatitis
Yes ............................... 1
No ................................. 2 (go to E2) 6
Was it infectious hepatitis or was it serum hepatitis?
Infectious .................... 1
Serum ........................... 2 7
Don't k n o w ................... 9
At what age were you first diagnosed of hepatitis?
Age years 8-9
or 19 -----
Were you ever treated?
Yes ............................... 1
No ................................. 2 10
Please describe the treatments:
11-12
E2. Cirrhosis of the liver
Yes .................................. 1
No ................................... 2 (go to E3) __13
At what age were you first diagnosed of cirrhosis of the liver?
Age years 14-15
or 19 ___
Were you ever treated?
Yes .................................. 1
No .................................... 2 16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
113
Please describe the treatments:
________________________________________________________ _17-18
E3. Other liver diseases
Yes ................................ 1
No .................................. 2(gotoE4) __19
What was the name of the liver disease?
_ 20-21
At what age were you first diagnosed of this liver disease?
Age years 22-23
or 19 “
Were you ever treated?
Yes ................................ 1
No .................................. 2 __24
Please describe the treatments:
________________________________________________________ _25-26
E4. Diabetes
Yes ................................ 1
No .................................. 2(gotoE5) _27
At what age were you first diagnosed of diabetes?
Age years 28-29
or 19 — —
Were you ever treated?
Yes ................................ 1
No ................................. 2 _30
Please describe the treatments:
31-32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
114
ES. Gastric ulcer
Yes ................................ 1
No .................................. 2{gotoE6) _ 3 3
At what age were you first diagnosed of gastric ulcer?
A ge years 34-35
or 19 “
Were you ever treated?
Yes ................................ 1
No .................................. 2 36
Please describe the treatments;
Please describe the treatments:
37-38
E6. Duodenal ulcer
Yes ................................ 1
No ................................. 2(gotoE6) _ 3 9
At what age were you first diagnosed of duodenal ulcer?
Age years 40-41
or 19 ”
Were you ever treated?
Yes ................................ 1
No .................................. 2 42
43-44
E7. Gallstones
Yes ................................ 1
No .................................... 2(gotoE8) 45
At what age were you first diagnosed of having gallstones?
Age years 46-47
or 19 ----- “
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
115
Were you ever treated?
Yes ................................ 1
No .................................. 2 _ 4 8
Please describe the treatments:
49-50
E8. Other gallbladder conditions
Yes ................................ 1
No .................................. 2(gotoE9) __51
What was the name of the gallbladder disease?
____________________________________________________________________ _ 5 2
At what age were you first diagnosed ofthis gallbladder condition?
Age years 53-54
or 19 ___
Were you ever treated?
Yes ................................ 1
No .................................. 2 _55
Please describe the treatments:
56-57
E9. Tuberculosis
Yes ................................ 1
No .................................. 2(gotoE10) 58
At what age were you first diagnosed of tuberculosis?
Age years 59-60
or 19 ~ “
Were you ever treated?
Yes ................................ 1
No .................................. 2 61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
116
Please describe the treatments;
62-63
610. High blood pressure
Yes ................................ 1
No ................................ 2 (goto Ell) __64
At what age were you first diagnosed of high blood pressure?
Age years 65-66
or 19 2ZZ “
Were you ever treated?
Yes ................................ 1
No .................................. 2 67
Please describe the treatments:
68-69
E ll. Asthma
Yes .................................. 1
No .................................... 2 (go to 612) _ 7 0
At what age were you first diagnosed of asth*^^?
A ge years 71 •72
or 19 ----- ~
Were you ever treated?
Yes ................................... 1
No .................................... 2 73
Please describe the treatments:
E12. Emphysema
74-75
Yes ................................ 1
No .................................. 2{gotoE13) __76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
117
At what age were you first diagnosed of emphysema?
A ge years 77-78
or 19 2Z Z ~
Were you ever treated?
Yes ................................... 1
No .................................... 2 79
Please describe the treatments:
80-81
El 3. Hemorrhoid
Yes ................................... 1
No .................................... 2 (go to EU) _ 8 2
At what age were you first diagnosed of hemorrhoid?
Age years 83-84
or 19 —
Were you ever treated?
Yes ................................... 1
No .................................... 2 85
Please describe the treatments:
86-87
E14. Oiverticulosis
Yes ................................ 1
No .................................. 2 (gotoE l5) _ 8 8
At what age were you first diagnosed of diverticulosis?
Age years 89-90
or 19 2 % ~
Were you ever treated?
Yes ................................ 1
No .................................. 2 91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
118
Please describe the treatments:
92-93
E1S. Polyposis coli
Yes ................................ 1
No .................................. 2(gotoE16) 94
At what age were you first diagnosed of polyposis coli?
Age years 95-96
or 19 2 % ~
Were you ever treated?
Yes ................................ 1
No .................................. 2 97
Please describe the treatments:
98-99
E16. Ulcerative colitis
Yes ................................ 1
No .................................. 2(gotoE17) __100
At what age were you first diagnosed of ulcerative colitis?
Age years 101-102
or 19 ~
Were you ever treated?
Yes ................................ 1
No .................................. 2 _103
Please describe the treatments:
104-105
E17. Adenomatous polyps
Yes ................................ 1
No .................................. 2 (gotoE18) 106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Please describe the treatments;
Please describe the treatments:
E19. Other parasitic diseases
What was the name of this parasitic disease?
At what age were you first diagnosed of parasitic disease?
119
At what age were you first diagnosed of adenomatous polyps?
Age years 107-108
or 19 — “
Were you ever treated?
Yes .................................. 1
No .................................... 2 109
110-111
E18. Schistosomiasis
Yes .................................. 1
No .................................... 2 (go to E19) _112
At what age were you first diagnosed of schistosomiasis?
Age years 113-114
or 19 2 Z I
Were you ever treated?
Yes .................................. 1
No .................................... 2 115
116-117
Yes ................................ 1
No .................................. 2 (go to E20) 118
119-120
Age years 121-122
or 19 ----- —
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
120
Were you ever treated?
Yes ................................ 1
No .................................. 2 123
Please describe the treatments:
124-125
E20. Any kind of cancer
Yes ................................ 1
No .................................. 2 126
What kind of cancer did you have?
127-129
At what age were you first diagnosed ofthis cancer?
Age years 130-131
or 19 2 2 “
Were you ever treated?
Yes ................................ 1
No .................................. 2 132
Please describe the treatments:
133-134
E21. ABO blood types:
A ty p e ............................... 1 __135
B type ............................... 2
O ^ p e ............................. 3
AB type ........................... 4
E22. What was the time interval between your last meal and collection of
the urine specimen?
_______________ (min) _ 1 36-139
E23. What was the time interval between your last meal and collection of
the blood specimen?
(min) 139-141
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 2 1
Appendix B.
Ocpaitsicnt af Prcvtnhvc Median*
PaÀview Medical Building
(2U) 23V @500
A Cohocc Scud7 of Cancer In M iddled-aged Men l a Shanghai
Aonald X Ross, M.D. and Gao Yn-Tong, M.D., P rin c ip a l In v esclg aco rs
Shanghai Cancer In sclcu ce and CSC C ooprehenslve Cancer Cancer
IM PO SM EO C O N SEM T F O R M
Va a r e conducclng a ecudy In an e ffo rc eo le a m more about fa c to rs chat may.
a f f e c t th e h e a l t h o f p e rs o n s l i v i n g I n S h a n g h a i. Ve would l i k e your
p a r tic ip a tio n In th is stu d y . Your p a r tic ip a tio n In v o lv es w lthdraw lag one sm all
b lo o d sample from your arm by hypodermic needle and th e c o lle c tio n of a
u r in e sam ple. In a d d itio n you w ill be asked to com plete a s h o rt q u estio n n aire
w ith th e help of one o f our nurses on your h e a lth and d i e t . The ris k s and
d isc o m fo rts of th e blood c o lle c tio n a re m inim al. A sm all p a in fu l b ru ise may
o c c u r In ra re In sta n c e s from th e puncture and th e re I s a s lig h t psychological
hasard o f f a in tin g .
You w i l l be paid 13 yuan f o r your p a r tic ip a tio n In t h i s stu d y . Any questions
you may have w ill be answered as com pletely as p o s s ib le .
By s ig n in g b elo w , I c o n s e n t to p a r t i c i p a t e In th e s tu d y . I
u n d e rs ta n d t h a t any In f o r m a tio n o b ta in e d from me w i l l be k ep t
s t r i c t l y c o n f id e n tia l. I r e a liz e th a t I have th e r ig h t to deny o r
w ithdraw my consent to p a r tic ip a te a t any tim e.
S ig n atu re o f P a rtic ip a n t Dace
I c e r t i f y th a t I have review ed the c o n te n ts of t h i s form w ith the person
sig n in g above who. In my o p in io n , understood the e x p la n a tio n . I have explained
th e known s Id e - e f f e c ts and b e n e fits of th e stu d y . Any s ig n if ic a n t change In
th e n a tu r e of th e stu d y , from th a t d e sc rib e d above, w i l l be f u lly explained to
th e perso n sig n in g above.
S ig n atu re W itnessed Dace
Univeniiy of Southern CsUfomU School of Medicine Î02J Zonal Avenue Loe Angeles California *0033
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
122
Appendix C.
Follow-up Questionnaire for Shanghai Cohort Study
Street Name:______________________________________
Year: 19_________
_________ (1st, 2nd. 3rd, etc) follow-up survey
(This follow-up survey covers the period of the previous 12 months)
1. Subject's ID#: __________________________________ 1-5
Name:___________________
Last known address:__________________________________
Place of work:________________________________________
Office telephone # : __________________________________
2. Present address is
Same as (1) ....................................................... 0 6
Moved, still in Shanghai Urban A rea............... 1
Moved out of Shanghai Urban Area ............... 2
Changed, but only temporarily ...................... 3
New address is:__________________________________
3. Respondent is
Self .................................................................. 1 7
Other ................................................................ 2 “
Relationship to subject_____________________________ 8-9
4. Deleted.
5. Vital status of subject
Alive .................................................................. 1 10
Dead .................................................................. 2 * “
Unknown ......................................................... 3
a. Date of d e a th :_____ / /______ 11-18
month 3âÿ year
b. Cause of death:______________________ 19-21
c Name of hospital where subject died:_________________ 22-24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
123
2nd hospitalization:
D a te o f adm ission (yr, m o n t h ) : _________________ 57-60
Cause of admission:_______________________ 61-63
Name of hospital:_________________________
P a tie n t# :________________________________
3rd hospitalization:
Date of admission (yr, m onth):______________ 64-67
Cause of admission : _______________________ 68-70
Name of hospital:____________ ____________
P a tie n t# :________________________________
10. Was subject diagnosed with cancer during the past 12 months?
No .................................................................. O (gotoQ II) 71
Yes 1 —
a. Cancer site:____________ Subsite:_________ 72-76
b. Histology:_______________ 77-80
c Criteria of diagnosis (circle all that are applicable)
Pathology ............................................................ 1
Cytology ........................................................... 2
Surgery, but no pathology ............................. 3
Bone marrow smear ........................................ 4
Blood smear ..................................................... 5
Radioisotope scanning ..................................... 6 81-84
X -ray.................................................................. 7
CT .................................................................. 8
Ultrasonography .............................................. 9
Alpha fetoprotein ............................................... 10
Clinical examination only ................................. 11
d. Name of hospital:___________________________ 85-87
e. Patient#: _________________________________ 88-107
f. Date of diagnosis: / / 108-115
month 3ay year
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
124
d. Did su b je c t d ie o f ca n ce r?
No .................................................................. 0(gotoQ 6) 25
Yes .................................................................. ^ —
d1. Cancer site:____________ Subsite:_____________________ 26-30
d2. Histology:_________________________________ __31-34
d3. Criteria of diagnosis (circle all that are applicable)
Pathology ......................................................... 1
Cytology .......................................................... 2
Surgery, but no pathology ............................. 3
Bone marrow smear ........................................ 4
Blood smear ..................................................... 5
Radioisotope scanning .................................... 6 35-38
X-ray .................................................................. 7 “
CT .................................................................. 8
Ultrasonography ............................................. 9
Alpha fetoprotein ........................................... 10
Clinical examination only ............................... n
Postmortem inference .................................... 12
d4. Date of diagnosis: / /___________________ 39-46
month 3iÿ year
6. Did subject smoke regularly during the past 12 months?
No .................................................................. 0 47
Yes .................................................................. 1 —
7. Did subject drink regularly during the past 12 months?
No .................................................................. 0 48
Yes .................................................................. 1 —
8. Was subject ever hospitalized during the past 12 months?
No .................................................................. O(gotoQIO) 49
Yes .................................................................. 1 —
9. 1st hospitalization:
Date of admission (yr, m onth):______________ 50-53
Cause of admission : _______________________ __54-56
Name of hospital:_________________________
Patient#:
Reproduced with permission of the oopyright owner. Further reproduction prohibited without permission.
125
g. Treatment (circle all that are applicable)
Surgery only ..................................................... 1
Chemotherapy only ........................................ 2
Radiotherapy o n ly ............................................ 3
Chinese traditional medidne o n ly ................... 4 116
Surgery and chemotherapy.............................. 5 —
Surgery and radiotherapy ............................... 6
Surgery and Chinese traditional m edicine 7
Others (specify____________) ........................ 8
11. Did subject have any other serious illness?
No .................................................................. 0(gotoQ l2) 117
Yes .................................................................. 1 —
1 st serious illness:
Date of diagnosis (yr, m onth):______________ __1 18-121
Name of disease:____________________________________ 122-124
Name of hospital:_______________________
Patient#:
2nd serious illness:
Date of diagnosis (yr. m onth):______________ 125-128
Name of disease:_________________________ _129-131
Name of hospital:_______________________
P a tie n t# :______________________________
3rd serious illness:
Date of diagnosis (yr, m onth):______________ 132-135
Name of disease;_________________________ 136-138
Name of hospital:_______________________
P a tie n t# :______________________________
12. Did subject experience any tragedy during the past 12 months?
No .................................................................. 0(gotoQ 13) 139
Yes .................................................................. 1 —
Event(1): ______________________________ _140-142
Event (2 ):__________________ 143-145
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
126
13. Name of interviewer:_________________________ 146-148
14. Today's date: _____ / /______ 149-156
month 3iÿ year
(End of Interview)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Antioxidants and risk of myocardial infarction and cancer in a cohort of middle-aged finnish men: the kuopio ischemic heart disease risk factor study

PDF

Dietary fiber intake and atherosclerosis progression: The Los Angeles Atherosclerosis Study

PDF

Dietary fats, fat metabolizing genes, and the risk of breast cancer

PDF

Development and evaluation of standardized stroke outcome measures in a population of stroke patients in rural China

PDF

Androgen receptor gene and prostate -specific antigen gene in breast cancer

PDF

Determinants of the age at natural menopause: The multiethnic cohort

PDF

Family history, hormone replacement therapy and breast cancer risk on Hispanic and non-Hispanic women, The New Mexico Women's Health Study

PDF

An exploratory investigation of responses of young children to nonverbal auditory cues in messages

PDF

Assessment of fatigue as a late effect of therapy among survivors of childhood leukemia

PDF

Colorectal cancer risks in Singapore Chinese: Polymorphisms in the insulin-like growth factor-1 and the vitamin D receptor

PDF

A comparative study of the realm of meaning of four child prostitutes in Taiwan: A hermeneutic approach

PDF

Biopsychosocial factors in major depressive disorder

PDF

Construction validation of the Vocal Behavior Inventory

PDF

Genetic risk factors in breast cancer susceptibility: The multiethnic cohort

PDF

Child sexual abuse in a sample of male and female Hispanic and White nonclinical adolescents: Extending the reliability and validity of the Trauma Symptom Inventory (TSI)

PDF

Endometrial cancer in Asian migrants to the United States and their descendants

PDF

Association between body mass and benign prostatic hyperplasia in Hispanics: Role of steroid 5-alpha reductase type 2 (SRD5A2) gene

PDF

Risk factors for breast cancer according to estrogen and progesterone receptor status

PDF

beta3-adrenergic receptor gene Trp64Arg polymorphism and obesity-related characteristics among African American women with breast cancer: An analysis of USC HEAL Study

PDF

Attrition in a longitudinal drug use prevention study

Asset Metadata

Creator Yuan, Jian-Min (author)

Core Title A prospective cohort study of morbidity and mortality among middle-aged men in Shanghai, China

Degree Doctor of Philosophy

Degree Program Epidemiology

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

Tag health sciences, public health,OAI-PMH Harvest,sociology, ethnic and racial studies

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Yu, Mimi (committee chair), [illegible] (committee member), Jones, Peter (committee member), Peters, John (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c17-503621

Unique identifier UC11352104

Identifier 9636387.pdf (filename),usctheses-c17-503621 (legacy record id)

Legacy Identifier 9636387-0.pdf

Dmrecord 503621

Document Type Dissertation

Rights Yuan, Jian-Min

Type texts

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

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA