University of Southern California Dissertations and Theses

Familiality and environmental risk factors of peptic ulcer: A twin study

(USC Thesis Other)

Familiality and environmental risk factors of peptic ulcer: A twin study

PDF

Download

Open document

Flip pages

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 of computer primer.
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. Bach
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 Zecb 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.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FAMILIALITY AND ENVIRONMENTAL RISK FACTORS OF PEPTIC ULCER
-A TWIN STUDY
BY
HAITAO WANG
A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(Applied Biometry and Epidemiology)
May 1998
Copyright 1998 HaitaoWang
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 1391105
UMI Microform 1391105
Copyright 1998, 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.
U N IV ERSITY O P S O U T H E R N C A L IF O R N IA
T H E G RADUATE S C H O O L .
U N IV ER SITY R A R K
•- OS A N G ELE S. C A L IF O R N IA S O O D 7
This thesis, written by
H a i ta o Wang
under the direction of £_£!* Thesis C om m ittee,
and approved by all its m em bers, has been pre
sented to and accepted by the D ean o f The
Graduate School, in partial fu lfillm en t of ike
requirements for the degree of
Master of Science
T)ni, ^ 7, 1998
THES lOMMITTEE
Ckcir,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
CONTENTS PAGE
ACKNOWLEDGMENTS...................................................................................iii
LIST OF TABLES................................................................................................v
LIST OF FIGURES.............................................................................................vi
INTRODUCTION............................................................................................. 1
CHAPTER 1. EPIDEMIOLOGY OF PEPTIC ULCER: GENETIC AND
ENVIRONMENTAL RISK FACTORS......................................3
CHAPTER 2. RESEARCH QUESTIONS.........................................................17
CHAPTER 3. DATA STUDY VARIABLES, AND METHODS................. 19
CHAPTER 4 RESULTS ...................................................................................22
CHAPTER 5. DISCUSSION............................................................................. 32
BIBLIOGRAPHY................................................................................................35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Acknowledgments
Many people helped me in the development o f this thesis. Dr. Thomas Mack,
Chair of my thesis committee, gave me invaluable guidance throughout this study. Ann
Hamilton and Wendy Cozen provided precious help in data management, literature
research and use of methodology. Stanley Azen, Director of the Biometry Division,
helped me arrange my course work semester by semester in this degree program. David
M. Heer, Director of the USC Population Research Laboratory, provided me scholarships
throughout this degree program. All of these are indispensable for the fulfillment of this
thesis. I am also grateful to Myles Cockbum, who provided detailed comments on the
earlier draft of this thesis.
Special thanks go to my wife — Jean. In more ways than I could ever express, she
made it possible for me to simultaneously work on three degree programs, and finished
them earlier than I had expected.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
TABLES PAGE
TABLE 1. DEMOGRAPHIC CHARACTERISTICS AND RISK OF PEPTIC
ULCER..........................................................................................................22
TABLE 2. COTWIN’S DISEASE STATUS, ZYGOSITY AND RISK OF PEPTIC
ULCER..........................................................................................................23
TABLE 3. SMOKING, ALCOHOL, ASPIRIN USE, EXERCISE, DEPRESSION
AND RISK OF PEPTIC ULCER.................................................................24
TABLE 4. DIETARY FACTORS AND RISK OF PEPTIC ULCER........................... 25
TABLE 5. MULTIPLE LOGISTIC REGRESSION PREDICTING RISK OF
PEPTIC ULCER...........................................................................................27
TABLE 6. MULTIPLE LOGISTIC REGRESSION PREDICTING RISK OF
PEPTIC ULCER: WITH FOCUS ON INTERACTION EFFECTS
BETWEEN CIGARETTE SMOKING AND COTWIN’S DISEASE
STATUS....................................................................................................... 29
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
FIGURES PAGE
FIGURE 1. RELATIVE RISKS OF PEPTIC ULCER BY ZYGOSITY
AND COTWIN’S DISEASE STATUS................................................ 30
FIGURE 2. RELATIVE RISKS OF PEPTIC ULCER BY SMOKING
STATUS AND COTWIN’S DISEASE STATUS................................30
FIGURE 3.1 RELATIVE RISKS OF PEPTIC ULCER BY SMOKING
STATUS AND COTWIN’S DISEASE STATUS,
FRATERNAL TWINS............................................................................31
FIGURE 3.2 RELATIVE RISKS OF PEPTIC ULCER BY SMOKING
STATUS AND COTWIN’S DISEASE STATUS,
IDENTICAL TWINS.............................................................................. 31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
INTRODUCTION
Peptic ulcer is a circumscribed loss of tissue in those parts of the gastrointestinal
tract exposed to acid and pepsin. The majority of peptic ulcers occur either in the stomach
(called gastric ulcers) or in the first portion of the small intestine (called duodenal ulcers).
The classic symptom of a peptic ulcer is epigastric abdominal pain which is worse at night
and is relieved by meals. Some ulcer cases present with less classic abdominal pain or with
complications such as bleeding in the gastrointestinal tract, perforation, or obstruction of
gastric emptying (Rotter et al., 1992).
Peptic ulcer was a rare condition prior to 1900 in the Western world. However,
peptic ulcer mortality increased consistently in the first half of this century in the United
States, reaching its highest level in the 1950s (about 5 per 100,000). Since then peptic
ulcer mortality has been declining constantly, with the age adjusted mortality rate reaching
approximately 2 per 100,000 in the 1980s (Kurata et al., 1983). The overall estimates of
ulcer prevalence, however, have not decreased. Data from the 1990 National Health
Interview Survey indicated that the ulcer prevalence had decreased for males but increased
for females, with their combined annual rates remaining steady at approximately 2%
during the past three decades in the United States (National Center for Health Statistics,
1991). It is estimated that peptic ulcer disease now affects as many as 10 percent o f people
in the United States at some time in their lives (National Institutes of Health, 1994).
Both gastric and duodenal ulcer rates increase rapidly with age. Ulcer prevalence
reaches its highest level among people aged over 65 years. The gender difference in ulcer
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
rate has decreased during the past 30 years. During this period male to female ratio has
decreased from approximately 2:1 to 1:1 for ulcer prevalence, hospitalization and
mortality (Rotter et al., 1992).
Both genetic and environmental factors may contribute to the risk of peptic ulcer.
It has been suggested that genetic factors like blood group O, hyper-PG, rapid gastric
emptying tend to predispose an individual to peptic ulcer disease (Rotter et al., 1992).
Major environmental factors of peptic ulcer include cigarette smoking, aspirin,
nonsteroidal anti-inflammatory drugs and possibly psychological stress, salt, low-fiber diet
and alcohol (Garland et al., 1992). The etiology of these risk factors, however, is still
unclear. Most importantly, genetic and environmental risk factors were rarely studied
simultaneously, and little effort has been made to examine the possible gene/environmental
interactions in peptic ulcer.
In this study, I will analyze the risk factors of peptic ulcer and possible
gene/environmental interactions using data from the California Twin Program. I will first
give a comprehensive review of previous epidemiologic studies of peptic ulcer, then I
conduct an analysis of the effects of suspected risk factors on the disease, and explore
possible gene/environmental interactions on peptic ulcer.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 1. EPIDEMIOLOGY OF PEPTIC ULCER: ENVIRONMENTAL AND
GENETIC DETERMINANTS
Genetic factors
It has been suggested that a simple mendelian inheritance patter does not exist for
all ulcer diseases. Polygenic inheritance has been proposed, but the hypothesis of genetic
heterogeneity seems to better fit the observed inheritance patterns (Rotter et al., 1992).
The genetic heterogeneity hypothesis implies that peptic ulcer results from a group of
distinct diseases with different etiologies and pathological mechanisms, having both
genetic and environmental components (Lam et al., 1980). This concept does not preclude
polygenic inheritance as one of a number of inheritance patterns (Rotter et al., 1981).
Given that some genetic factors are suspected to predispose an individual to peptic ulcer,
such as blood group and secretory status, one interest in genetic epidemiology is the
estimation of heritability— the proportion of total phenotypic variation attributable to all
genotypic differences among individuals. Studies of heritability in peptic ulcer can be
summarized into two categories based on the methodologies they used. They are family
studies and twin studies.
Family studies. The initial evidence of genetic factors in peptic ulcer was the
observation of an increased familial aggregation, i.e., a higher frequency of the disease in
relatives of patients than in the general population. Most reports indicated a positive
family history in 20%-50% of individuals with peptic ulcer compared with the 5%-15% in
controls (Rotter, 1980; McConnell, 1980). Ostensen and colleagues (1985) showed that
the risks of developing ulcers in patients with positive family histories were higher than in
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
those without family occurrence of ulcers. The relative risks were 1.5 for males (for
duodenal and gastric ulcer), 1.5 for female for duodenal ulcer and 2.0 for female for
gastric ulcer. These findings, however, did not constitute a clear demonstration that peptic
ulcer is determined by genetic factors, because family members share not only genes, but
also household milieu and many other environmental conditions. This problem can
partially be solved in studies comparing the resemblance in pairs of spouses, parent-child,
and brothers and sisters (sibs). Twin studies represent a powerful approach in resolving
the question of the relative influence o f genetics and environments.
Twin Studies. The classical twin design was introduced more than a century ago by
the pioneering research of Francis Galton (1869). It still provides one of the most efficient
methods of investigating the contribution of shared genes and shared environments to
familial traits. The basic principle underlying the twin study is straightforward. Identical
(monozygotic [MZ]) twins share all o f their genes and part of their rearing environment,
while fraternal (dizygotic [DZ]) twins, on average, share only one half of their genes and
part of their rearing environment. Consequently, the difference in similarity between the
members of MZ pairs versus those of DZ pairs reflects the extent to which a trait is
genetically determined. If there is no difference in similarity between MZ and DZ twin
pairs, it suggests that the trait arises solely from environmental effects. On the other hand,
any MZ/DZ difference suggests that the trait may be heritable within the population under
study.
Methods for assessing twin similarity and comparing the similarity across zygosity
groups vary in complexity and informativeness. One of the simplest methods is to compare
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the mean intrapair difference in MZ twins with that in DZ twins through t-tests when the
measurement of the trait is continuous, or to compare concordance rates for MZ twins and
DZ twins through chi-square tests when the measurement is dichotomous. This simple
method determines only whether MZ twin similarity exceeds DZ twin similarity, without
the quantification of the heritability of the trait or of the proportion of the variance which
is due to genetic differences between individuals. Several methods have been used to
estimate the heritability of traits. The “classical” estimate of heritability, calculated as
twice of the difference between the MZ and DZ intraclass correlations, has been popular
for many years and employed extensively in studies of different diseases (for derivation of
this formula, see Falconer, 1981). Although the computational simplicity of the classical
heritability estimate has great appeal, there are limitations to this estimate. This estimate is
best suited for the comparison of like-sex MZ and DZ twins. When extended to the
comparison of opposite-sex pairs it could result in an underestimate of heritability,
particularly if different genes have different effects across gender.
A major methodological improvement over the classical heritability estimate has
been the adoption of model-fitting approaches for twin data (Neale and Cardon, 1992). In
these approaches, estimates of genetic and environmental effects are derived by “fitting”
expectations in terms of these parameters to observed summary statistics (such as twin
variance and intrapair covariance). Model fitting is typically carried out by the method of
weighted least squares or maximum likelihood and results in chi-square statistics that can
be used to assess the goodness of fit of the models. To select the best model for any given
data set, alternative genetic and environmental models are fitted and the chi-square
5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
statistics of each are compared. Model fitting methods can be extended to test for
differences across gender in the genetic and environmental control of a trait and to explain
change and continuity in longitudinal data sets. This method, however, is also subject to
some limitations. Model fittings are usually based on the multivariate normality
assumption, which usually requires a large sample size. When the disease trait is
dichotomous, an underlying normal scale for the trait is required to validate the model
fitting. This condition is very difficult to satisfy, especially when the trait under study is a
rare condition.
For peptic ulcer, the most frequently used method is comparing concordance rates
between MZ twins and DZ twins. It has been shown consistently that the concordance for
ulcer among monozygotic twin pairs is greater than for dizygotic twin pairs (Rotter et al.,
1992; Almy et al., 1975, Eserhard, 1968, Pollinet et al., 1969). A detailed study even
revealed that the ulcer site was usually concordant when both members of a twin pair were
affected (Gotlieb-Jensen, 1972). One can speculate from the twin studies that a large part
of the increased familial aggregation in peptic disease is due to genetic factors.
The classical twin study has been criticized for its limitations in dealing with the
circumstance that MZ twins are more likely to share environmental exposures than DZ
twins. This is due to the fact that MZ twins are treated more similarly than DZ twins, and
they tend to have a closer life style than DZ twins. Data from twins reared apart can avoid
this problem to some extent. A study of adult twins using the data of the Swedish
Adoption/Twin Study of Aging found that the intraclass correlation for peptic ulcer
disease for MZ (apart), MZ (together), DZ (apart), and DZ (together) were 0.35, 0.34,
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0.07, and 0.12 respectively (Malaty et al., 1997). This pattern of correlations indicates a
role of genetic effects on liability to peptic ulcer. The correlation coefficient for MZ
(apart) twins (0.3 5) provides the best single estimate of the relative importance of genetic
effects (heritability).
Environmental factors
Accepted environmental risk factors for peptic ulcer include cigarette smoking,
aspirin use and H. Pylori (Kurata, 1993; Kurata and Nogawa, 1997). Less conclusive
associations have been reported for alcohol and coffee consumption, other dietary factors
and occupational and emotional stress (Garland et al., 1992).
Cigarette Smoking. Studies have consistently shown that smoking has been linked
to higher ulcer mortality, incidence, complications, and recurrence. The mortality rate for
peptic ulcer is approximately twice as great among smokers as among nonsmokers,
despite the presence of competing risks due to other health effects of smoking (McCarthy,
1984; Piper, 1980; Stemmermann et al., 1989). Several hospital- and population-based
studies indicated that ulcers were also about twice as prevalent among smokers as among
nonsmokers (Isenberg, 1984; Elashoff, 1980; Kurata et al., 1992; Anda et al., 1990). A
dose-response relationship between peptic ulcers and cigarette exposure, as measured by
number of cigarettes smoked, duration of smoking, and inhalation habits, was established
(Friedman et al., 1974). Nicotine and tar levels seemed less important predictors of peptic
ulcer than the number of cigarettes smoked (Petitti et al., 1982). Some recent studies
reported results for duodenal (DU) and gastric ulcers (GU) separately. A 20-year
7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
prospective study o f Japanese men enrolled in the Honolulu Heart Program showed that
cigarette smoking was associated with increased risks for both GU and DU
(Stemmermann et al., 1989). Smoking was more strongly related to GU than DU, and the
dose response effect in pack-years was observed only for GU. In a study of patients
observed in an endoscopy clinic in Japan, the risk factor most strongly associated with
ulcer disease was cigarette smoking (Watanabe et al., 1992). The risk ratio was 3.32 for
GU (95% Cl, 2.34 tO 4.71) and 2.99 for DU (95% Cl, 2.02 to 4.42). The higher risk
ratios for GU in these studies are different from other previous studies which reported
significant findings more often for DU then GU. Perhaps these contradictory findings
reflect differences in the background prevalence rates of GU and DU in the various
populations that have been studied (Kurata, 1993). GU is more common than DU in
Japanese population, while the prevalence of DU is higher than GU in Western
populations.
In addition, smoking also impairs ulcer healing and predisposes to ulcer relapse
(Korman et al., 1983; Korman et al., 1980; Massarat et al., 1981; Sonnerberg et al., 1981).
Smoking also interferes with medical treatment for both gastric and duodenal ulcers by
diminishing the effectiveness of histamine H2-receptor antagonists (Elashoff and
Grossman, 1980; Kikendall et al.,1984). The mechanism is not completely understood,
however. Duodenal ulcer recurrence is more frequent in smokers than in nonsmokers
(Sontag et al., 1984). As to ulcer complications, smoking has been reported to be
associated with perforated duodenal ulcer (Smedley et al., 1988). Furthermore, time
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
trends for cigarette smoking are consistent with the variation in ulcer mortality, especially
for DU (Kurata, 1986; Centers for Disease Control, 1988)
These epidemiological results are supported by the biological evidence. Biological
evidence suggests several possible mechanisms whereby smoking might enhance ulcer
pathogenesis. These include increased basal and maximal acid secretion, increased reflux
of duodenal contents back into stomach possibly damaging the gastric mucosal barrier,
and decreased pancreatic bicarbonate secretion impairing neutralization of gastric acid in
the duodenum (Whitfield and Hobsley, 1987; Kaufmann et al., 1990, Corinaldesi et al.,
1989; Haruma et al., 1992; Ainsworth et al., 1992). Smoking has also been reported to
significantly reduce gastric and duodenal prostaglandin production (Cryer, 1992; Quimby
et al., 1986; McCready et al., 1985). Since prostaglandin can decrease gastric acid
secretion and increase duodenal bicarbonate secretion, reduced prostaglandin
concentrations in smokers might explain the observed increase in gastric acid secretion and
reduction in bicarbonate secretion (Kurata, 1993).
Nonsteroidal anti-inflammatory Drugs (NSAIDs). The association between
chronic aspirin intake and peptic ulcer has been repeatedly observed. Levy (1974) found a
3-fold increase in gastric ulcers in patients who were taking aspirin regularly compared
with those who did not take aspirin. An early case-control study of patients who recovered
from myocardial infarction showed that the incidence of peptic ulcer was 6 times higher in
patients taking aspirin compared with those taking placebo (Aspirin Myocardial Infarction
Study Group, 1980). These early evidence was supported by some later experimental
studies. A 5-year, double-blind, placebo-controlled trial showed an ulcer risk ratio of 1.2-
9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.8 for subjects who took aspirin 325 mg every other day (Steering Committee of the
Physicians’ Health Study Research Group, 1989). A 3-year trial using a higher dose (1 g
daily) found a 6-11 fold increased risk in aspirin users (Kurata and Abbey, 1990). Other
case-control studies or cohort studies showed that aspirin users were 2 to 7 times more
likely to develop peptic ulcer than nonusers (Isenberg et al., 1991; Kurata et al., 1992;
Holvoet et al., 1991).
As for other nonsteroidal anti-inflammatory drugs (NSAIDs), several endoscopic
studies have shown significant degree of gastric mucosal hemorrhage and ulceration in
normal volunteers taking NSAIDs even for brief period of time (Lanza et al., 1981; Loebl
and Craig, 1977). Evidence indicated that NSAIDs might cause duodenitis, cause ulcer,
exacerbate peptic ulcer disease, or induce ulcer complications (Collier and Pain, 198S;
Walt et al., 1986; Somerville et al., 1986; Thompson et al., 1980). Some cohort and case-
control studies indicated that, for nonaspirin NSAIDs, odds ratios for all cases combined
varied from 4.1 to 8.9. When grouped according to type of NSAID, odds ratios for
nonaspirin NSAIDs had a larger range of 2.3 to 20.6. Odds ratios for nonaspirin NSAID
use were 4.3 to 6.5 for DU and 5.5 to 9.0 for GU (Griffin et al., 1991; Laporte et al.,
1991; Holvoet et al., 1991; Bloom, 1989; Fries, 1991). Recent epidemiological studies
showed a pattern of raised risk for both DU and GS associated with NSAID use. A meta
analysis suggested that 24% of the peptic ulcer risk in the general population could be
attributable to NSAIDs. The summary NSAID risk ratio for the 12 case-control and
cohort studies published from 1990 to 1997 was 4.0 (C.I., 3.8-4.3) (Kurata and Nogawa,
1997).
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Experimental laboratory studies suggest that there are at least two ways in which
NSAIDs might cause injury to the gastric mucosa. One way is that NSAIDs inhibit the
synthesis of endogenous prostaglandins (Richardson, 1989; Soil, 1991, Kurata and Haile,
1984). Various studies in animals and humans have shown that prostaglandins have a role
in mucosal protection, especially against GU. A second mechanism involves disruption of
the mucosal barrier. Changes in ionization caused by interaction of NSAIDs with the
various pH environments of the gastric mucosa may damage the mucosa when there is
excessive concentration of NSAIDs in the cells (Kurata, 1993).
It was noted that the patients most likely to develop ulcers when treated with
NSAIDs are those with blood group O (Semble et al., 1987). If this finding is true, it
would be an typical demonstration of a genetically influenced susceptibility o f the
deleterious effect of an environmental agent.
Helicobacter pylori. In the early 1980s, Marshall and Warren isolated a spiral
urease-producing organism nestled in the narrow interface between the gastric epithelial
cell surface and the overlying mucus gel. In their early studies, the presence of this
organism was shown to be highly correlated with antral gastritis as well as with gastric
and duodenal ulcers, and eradication of this organism effectively eliminated ulcer
recurrences (Mashall and Warren, 1984). The organism was named Campylobacter pylori
and later renamed Helicobacter pylori (H. pylori). Such studies have given rise to the
hypothesis that H. pylori is a major etiologic factor in peptic ulcer disease and that
diagnosis and eradication of the organism are necessary for optimal therapy of the disorder
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(NIH, 1994; Podolski, 1996).
H. Pylori infection is common in the United States: about 20 percent of people
under 40 and half o f people over 60 are infected with it (Podolski, 1996). H. pylori and
the presence of H. pylori-associated gastritis are strongly predictive of peptic ulcer
disease. Gastric H. Pylori is found in >90% of duodenal ulcer patients and 70-90% of
gastric ulcer patients (Blaser 1987; Graham 1989; Rauws et al., 1988). Clinical drug trials
where H. pylori is eradicated reported reduced ulcer recurrence (gastric and duodenal
ulcer), more rapid healing (duodenal ulcer) , and a higher rate of healing (duodenal ulcer)
(Mashall et al., 1988; Coghlan et al., 1987; Rauws and Tytgat, 1990; Graham et al.,
1991). The actual pathogenic mechanisms, however, have not been clearly demonstrated.
Furthermore, there has not been an explanation as to why, while H. pylori infection is
very common (about one third of asymptomatic study volunteers are infected), most
people who have H. pylori never develop an ulcer. H. pylori appears to be an important
factor contributing to the development of an ulcer (except for Zollinger-Ellison and
NSAID-associated cases), but not the only factor. It seems that H pylori is a necessary but
not sufficient condition for an ulcer to develop. Perhaps it is a cofactor that works
synergistically with other risk factors (Kurata, 1993).
Coffee. Physiological studies showed that both regular and decaffeinated coffee
stimulated excessive gastric acid secretion, increased esophageal reflux, and exacerbated
gastrointestinal symptoms (i.e., dyspepsia) (Cohen and Booth, 1975). But no conclusive
epidemiological evidence was found to support the role of coffee in the etiology of peptic
ulcer. A follow-up of Harvard graduates found that habitual coffee consumption in college
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
was the best predictor of peptic ulcer in later life (Paffenbarger et al., 1974). Like many
early epidemiological studies on the relationship between coffee consumption and peptic
ulcer, this study failed to control for smoking status, which is strongly associated with
coffee consumption (Kurata, 1982). A study of a health maintenance organization found
no relationship between coffee use and history of peptic ulcer, after controlling for
cigarette smoking (Friedman et al., 1974). Most other studies stratified on smoking failed
to an association between coffee drinking and ulcer disease ( Jorgensen and Gyntelberg,
1976).
Alcohol. Physiological studies indicated that concentrated ethanol lowered
mucosal resistance to gastric acid, but moderate alcohol intake suppressed gastric acid
production (Sleisenger and Fordtran, 1983). Moderate alcohol consumption does not
appear to increase peptic ulcer incidence, healing time, or recurrence (Isenberg 1986;
Friedman et al., 1974). Patients with alcohol cirrhosis of the liver, however, were found
to have an excess risk of peptic ulcer (Boyd et al., 1983). The association between alcohol
abuse and gastritis also illustrates the detrimental effect of chronic alcohol exposure on the
gastric mucosa (Garland et al., 1992).
Diet and eating habits. Most studies failed to implicate type of food as a cause of
ulcer disease. However, amelioration of symptoms in ulcer patients on a bland diet has
been widely recognized (Spiro, 1981; Welsh, 1977). It is believed that small, bland meals
might reduce the secretion of acid and pepsin, buffer the acid sereted into the stomach,
reduce the gastric motor activity, and maintain the resistance of the gastric mucosa. The
clinical value of the bland, low-fiber diets, however, is poorly documented. A cohort study
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of Harvard alumni showed that milk consumption in college was inversely related to
development of peptic ulcer in later life (Paffenbarger et al., 1974). Current milk-drinking
habits were not examined in this study, however. Surveys in India, Bangladesh, and South
Africa described elevated incidence of duodenal ulcer in areas with a high starch diet of
rice or yams (Tovey 1979; Tovey and Tunstall, 1975), while regions o f low incidence
corresponded to areas producing wheat or com. An ecological study in Israel reported a
significant inverse association between dietary vitamin A and mortality from peptic ulcer
(Palgi, 1981). A controlled trial in Norway demonstrated that duodenal ulcer patients on a
high-fiber diet experienced fewer ulcer recurrences than did those on a low-fiber diet and
suggested that a diet rich in fiber may protect against the development o f peptic ulceration
(Ryding and Berstad, 1984).
Stress. The role of stress in peptic ulcer disease has been controversial. It was
reported that emotional change could result in changes in gastric mucosa (Peters and
Richardson, 1983), but these changes did not necessarily lead to peptic ulceration. Most
case-control studies examining the effects of stressful life events on peptic ulcer disease
found no differences in ulcer frequencies between cases and controls (Piper et al., 1978;
Piper et al., 1978; Thomas et al., 1980). Feldman and associates (1986) reported that
peptic ulcer was not associated with the number of life events occurring in the previous
year, but they found that ulcer patients perceive their events more negatively, and that
ulcer patients also had significantly more personality disturbance and tended to be
hypochondriac, complainers, pessimistic, and excessively dependent. Combining
psychological and physiological approaches, it was reported that there was a significant
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
relationship between serum PG I and PGII and impaired coping ability and negative
personality traits such as hostility or resentment (Walker et al., 1988). These observations
suggest that, in certain individuals, hyperpepsinogenemia may be the mediating link
between psychological stress and peptic ulcer disease. It is possible that chronic emotional
stress acts via the vagus nerve to cause parietal cell and chief cell hypoplasia with resultant
increased maximal acid output and hyperpepsinogenemia (Rotter et al., 1992).
Despite some observed associations between ulcer and stress, an underlying causal
order is very difficult to establish. Problems with the study of psychological factors as the
cause of ulcer include definition of the theoretical model, definition and measurement of
psychological stress, differences in responses of individuals to stressful situations, and
problems of blinding in experimental design (Sturdevant, 1976). One approach to studying
the role of stress in the development of ulcer has been to study occupations considered to
be stressful. Air traffic controllers have been shown to have moderately elevated rates of
peptic ulcer (Cobb and Rose, 1973). Foremen and executives have a higher frequency of
ulcer than expected, possibly related to the role conflicts of middle management (Dunn
and Cobb, 1962). Sonnenberg and Haas (1986) reported that manual labor as compared
with sedentary labor carried a 1.6 times greater risk for gastric ulcer and a 2 times greater
risk for duodenal ulcer in a survey of more than 70,000 German employees receiving
occupational health-related check-ups. Shift workers in a large study in Japan were shown
to be at increased risk for both gastric and duodenal ulcer (Segawa et al., 1987). The
authors postulated that sleep disturbances in shift workers increased ulcer risk.
Unfortunately, most of these studies were not controlled for cigarette smoking, which is
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
significantly associated with self-reported subjective occupational stress indicators. There
are marked individual differences in the tendency to increase or decrease cigarette
smoking in response to varying levels of stress (Convay et al., 1981; Cummings, 1983).
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 2. RESEARCH QUESTIONS
The literature indicates that peptic ulcer is not one disease but a heterogeneous
group of disease that share a common end point. There are clearly genetic factors that
predispose an individual to the disease, as shown by the results of twin studies and family
studies, as well as the biological evidence of the roles played by some genetic traits such as
blood group O, hyper-PG I, and rapid gastric emptying in the development of peptic
ulcer. On the other hand, there are definitely environmental factors that can cause peptic
ulcer, mainly smoking, NSAIDs and H. Pylori. Studies of genetic factors and
environmental risk factors, however, were usually conducted separately, because data
used in genetic studies are usually short of detailed environmental exposure information,
and data used in studying environmental factors often lack detailed family history
information. Consequently, little effort has been made to investigate the potential
gene/environmental interactions. The few studies that did examine the interactions
between certain genetic traits and environmental risk factors support the hypothesis that
the pathogenesis and the natural history of ulcer are influenced by genetic and
environmental interactions. Semble and colleagues (1987) demonstrated that among
rheumatoid arthritis patients on NSAIDs, those with blood group O were more likely to
develop an ulcer than were patients with other blood groups. Walker and coworkers
(1988) found a significant relationship between serum pepsinogen and two psychological
mediating factors, negative personality traits and copying ability, suggesting that these
chronic emotional stress can interact to cause peptic ulcer. Obviously, future studies need
to be conducted to examine environmental factors together with genetic effects, so as to
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
delineate gene/environment interactions.
In the following analysis, I will simultaneously examine the genetic and
environmental determinants of peptic ulcer using twin data, and explore possible
gene/environmental interactions. I hypothesize that genetic factors confer differing
susceptibility among individuals to environmental factors.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 3. DATA, VARIABLES AND METHODS
The design of this study is based on data from the California Twin Program
available in the USC Norris Cancer Center. The California Twin Program was based on
the 1908-1982 California multiple birth file including 265,000 individuals. This file was
linked to the California Department of Motor Vehicles file o f active drivers’ licenses, and
nearly 102,000 native twin residents with current addresses were identified. In 1992, 16-
page questionnaires were sent to those 38,000 individuals on the roster bom before 1955.
There are nearly 20,000 completed responses representing over 12,000 pairs of twins.
The respondents represent the various parts of California in numbers roughly equal to the
native resident population. By ethnicity, whites, Asians, Amerindians are appropriately
represented, but Latinos and blacks are somewhat underrepresented. Whites in high social
class tracts are slightly over-represented, and older women are under-represented. The
questionnaire covered basic social and demographic variables, zygosity, family and
medical history, disease status, smoking and alcohol use history, lifestyle and dietary
preferences. Respondents were asked relevant information about their co-twins in many
questionnaire items (Mack, 1996).
This analysis will make use of a subsample which includes complete twin pairs (i.e.
both of the twins participated in the survey). After unlike-sex pairs and twins with
unknown zygosity are excluded, the resultant sample size is 4046, including 2021 fraternal
pairs and 2025 identical pairs. One twin is randomly selected from each twin pair as the
study subject, thus the total number of subjects used in this analysis is 4046.
Genetic implications of peptic ulcer are evaluated through the comparison of the
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
effects of cotwin’s disease status on the subject’s disease status across zygosity. If the
disease is genetically related, we would expect a higher familial effect in identical twins
than in fraternal twins. The effect of cigarette smoking on peptic ulcer is examined in
terms of smoking status (never, ever and <1 pk/day, and ever and >1 pk/day), age started
smoking (<=19 and >19), and kinds of cigarette smoked (non-filter versus filter, low-tar
versus regular, short versus long). Individuals who have not smoked more than 100
cigarettes during his/her entire life are defined as non-smokers. Alcohol drinkers are
defined as those who have drunk more than 10 drinks in his/her entire life. Drinkers are
further separated into two categories according to the amount they drink (<3 drks/day,
and > 3 drks/day). Both the amount of cigarette smoking and alcohol consumption refer to
the period when the maximum amount was used. Aspirin use is defined based on the
respondent’s reported length of aspirin use (never/only a short period, for over 1 yr, and
for over 5 yrs). Respondents were asked about their food preference and whether they eat
large portions of certain foods. Foods suspected to be related to peptic ulcer are
examined. They are: coffee, milk, potato/rice/pasta (low-fiber), bacon/hot dog/sausage
(high salt), beefrhamburger/lamb (high fat), cheese (diary product). Behavioral factors
examined include whether the subject reported having depressed for more than 3 weeks at
a time and whether he/she participates in regular physical exercise.
The effect of each independent variable is evaluated based on the odds ratios at
different exposure levels. These odds ratios and 95% confidence intervals are obtained
through maximum-likelihood point estimates with logistic regression analysis. Study
variables are examined individually and then simultaneously to avoid confounding effect.
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Cotwin’s disease status is used as a independent variable then its interaction effects with
other risk factors are examined to explore potential gene/environmental interactions.
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 4. RESULTS
The sample used in this analysis includes 980 male fraternal twins, 1041 female
fraternal twins, 9S2 male identical twins, and 1073 female identical twins. The
percentages of subjects who were aged 37-44, 45-54, and 55-86 were 46%, 38%, and
16% respectively. The mean age of the subjects was 47. Over 88% o f the subjects were
whites.
Table 1 shows the relationship between some demographic characteristics and risk
of peptic ulcer. The data show that the risk of peptic ulcer increases with age. People 45
years or older are 40% more likely to have peptic ulcer than those aged 37 to 44
(OR=l .43, 95% Cl, 1.09-1.87). Risk of peptic ulcer tends to decrease with the increase
of education. This may reflect a socioeconomic gradient of peptic ulcer. Zygosity, gender
and race (non-white versus white) are not associated with the risk of peptic ulcer.
Table 1. Demographic Characteristics and Risk of Peptic Ulcer
N w/o ulcer________N w ulcer___________ OR__________ 95% C.L
Age
<45 2016 101 1.00
>=45 1800 129 1.43 (1.09, 1.87)
Gender
Male 1818 114 1.00
Female 1998 116 0.93 (0.71, 1.21)
Zygosity
Fraternal 1899 122 1.00
Identical 1917 108 0.88 (0.67, 1.15)
Race
Non-White 431 24 1.00
White 3385 206 1.09 (0.71, 1.69)
Yrs of Schooling
12 or less 969 72 1.00
13-16 1853 111 0.81 (0.59, 1.09)
17 or more 994 47 0.63 (0.44, 0.93)
Bold value indicates that the OR is significantly different from 1.0 (p<0.05).
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2 shows the effects of cotwin’s disease status by zygosity. For the total
sample, a person is 2.2 time more likely to have ulcer if his/her cotwin has ulcer. This
could be caused either by shared genes or shared environment. However, the odds ratio
for identical twins (OR=2.72, 95% Cl, 1.47-5.03) is higher than that for fraternal twins
(OR=1.77, 95% Cl, 0.99-3.50), suggesting a genetic relevance o f peptic ulcer disease,
although the possible effect of more similar MZ twin environment can not be ignored.
N w/o ulcer N w ulcer OR 95% C. I.
Total
Twin w/o ulcer 3633 207 1.0
Twin w ulcer 183 23 2.21 (1.40, 3.48)
Fraternal
Twin w/o ulcer 1808 112 1.0
Twin w ulcer 91 10 1.77 (0.99, 3.50)
Identical
Twin w/o ulcer 1825 95 1.0
Twin w ulcer 92 13 2.72 (1.47, 5.03)
Bold value indicates that the OR is significantly different from 1.0 (p<0.05).
Table 3 presents the effects of smoking, alcohol consumption, aspirin use, physical
exercise, and depression on the risk of peptic ulcer. A dose-response effect of smoking is
observed. Heavy smokers (> 1 pk/day) show a 2-fold increased risk of peptic ulcer than
non-smokers (95% Cl, 1.53-2.80). Those who started smoking earlier (<=19 years old)
tend to have higher disease risk compared to those who started later. As for kinds of
cigarettes, it seems that non-filter cigarette smokers are the most vulnerable group, as
shown by their 2.3 fold increased risk relative to non-smokers and 1.5 increased risk
compared to filter smokers. Length of cigarettes and length of tars are not found to make
a difference in disease risk. For alcohol consumption, although heavy drinkers (> 3
drks/day) are 35% more likely to have peptic ulcer than non-drinkers, this effect is not
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
significant (95% C L, 0.84-2.18). Moreover, drinking in moderate amount does not
increase the risk of peptic ulcer at all (OR=0.99). Use of Aspirin tends to increase the risk
of peptic ulcer. Using aspirin for over 5 years increases ulcer risk by 45% (OR=l .45, 95%
Cl, 1.03-2.05). People engaged in regular exercise are less likely to have peptic ulcer,
although the effect is not strong (OR=0.73,95% Cl, 0.56-0.95). Severe depressors
(depressed for more than three weeks) are twice as likely to have peptic ulcers than others
(OR=2.08, 95% Cl, 1.44-3.00).
Taritle^^iniokinggAlcojiojj^AsgirinJUsegJIxercisegJ)^^
N w/o ulcer N w ulcer OR 95% C.L
Cigarette Smoking
Smoking Status
Never smoked 2046 95 1.0
Ever smoked, < 1 pk/day 854 47 1.19 (0.83, 1.70)
Ever smoked, > 1 pk/day 916 88 2.07 (1.53, 2.80)
Age started smoking
<=19 1072 91 1.83 (1.36, 2.46)
>19 657 44 1.44 (0.99, 2.08)
Cigarette filter
Non-filter 159 17 2.30 (1.34, 3.95)
Filter 1580 115 1.57 (1.18, 2.07)
Cigarette tar
Low-tar 790 57 1.68 (1.20, 2.36)
Regular 958 76 1.71 (1.25. 2.33)
Cigarette length
Short 908 68 1.61 (1.17, 2.22)
Long 763 63 1.78 (1.28, 2.47)
Alcohol Consumption
Non-drinker 411 22 1.0
Drinker, < 3 drks/day 2011 107 0.99 (0.62, 1.59)
Drinker, > 3drks/day 1394 101 1.35 (0.84, 2.18)
Aspirin Use
Never /only for short time 3126 173 1.0
For over 1 year 155 14 1.63 (0.93, 2.88)
For over 5 years 535 43 1.45 (1.03, 2.05)
Regular physical exercise?
No 1681 120 1.0
Yes 2095 109 0.73 (0.56, 0.95)
Depressed for > 3 weeks?
No 3414 188 1.0
Yes 332 38 2.08 (1.44, 3.00)
Bold value indicates that the OR is significantly different from 1.0 (p<0.05).
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4 shows the associations between dietary factors and peptic ulcer risk.
Eating large portions of bacon/hot dog/sausage or cottage cheese significantly increase
risk of peptic ulcer by 55-65%, while coffee, milk, potato/rice/pasta, and
beeffhamburger/lamb, are not found to be significantly related to ulcer risk.
N w/o ulcer N w ulcer OR
Most likely to choose coffee
as hot beverage?
No
Yes
Eat large portions of the
following food?
Milk
No
Yes
Potatoes/Rice/Pasta
No
Yes
Bacon/Hot dog/Sausage
No
Yes
Beef/Hamburger/Lamb
No
Yes
Cottage Cheese
No
Yes
1123
2693
3375
441
2840
976
3515
301
3092
724
3608
208
63
167
206
24
171
59
203
27
180
50
210
20
1.0
1.11
1.0
0.89
1.0
1.01
1.0
1.55
1.0
1.19
1.0
1.65
95% C. L
(0.82, 1.49)
(0.58, 1.38)
(0.74, 1.36)
(1.02, 2.36)
(0.86, 1.64)
(1.02, 2.67)
Bold value indicates that the OR is significantly different from 1.0 (p<0.05).
To reduce potential confounding effects among different risk factors, a multiple
logistic regression model is specified to include cotwin’s disease status, cigarette smoking,
alcohol, aspirin use and dietary factors found to be significantly related to the risk of
peptic ulcer in the univariate analysis (i.e. eating large portions of bacon/hot dog/sausage,
and eating large portions of cottage cheese). Age, gender are included as control variables.
Years of schooling is also controlled because education could be a confounder in many
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ways. For example, people with lower education tend to smoke more, and more likely to
have peptic ulcer (See Table 1). Failure to control for the effect of education would
overestimate the effect of smoking on the disease. Results from the multiple regression
model are generally consistent with those found in the univariate analysis, except that the
effect of heavy drinking (>3 drks/day), which was found to be related to a 35% increased
risk of peptic ulcer, virtually disappears in the multiple regression model. The bivariate
association between alcohol consumption and peptic ulcer may be caused by the
association between smoking and alcohol use. Moreover, effects of aspirin use, bacon/hot
dog/sausage, and cottage cheese lose their statistical significance. This may be due to the
reduced statistical power in the multivariate analysis compared with the univariate
analyses. In this multiple regression model, the only statistically significant factors are
cigarette smoking and cotwin’s disease status. Heavy smokers (>1 pk/day) are associated
with a 1.8 fold increased risk of peptic ulcer (95% Cl, 1.28-2.49), and one is twice more
likely to have peptic ulcer if his/her cotwin had the disease (95% Cl, 1.26-3 .17), after
controlling for other risk factors.
The effect of cotwin’s disease status in the multiple regression model provides a
clue for genetic relevance of this disease. Since fraternal twins share one-half of their
genes and identical twins have the same genotype, it is expected that the effect of cotwin’s
disease status is greater in identical twins than that in fraternal twins, if the disease is
genetically related. To test this speculation, Model 2 adds zygosity and an interaction term
between zygosity and cotwin’s disease status. Although the odds ratio for the interaction
term (identical cotwin with ulcer) was not statistically significant (p>0.05), the additional
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
45% increase in ulcer risk for identical twins when their cotwins have the disease suggests
that identical twins receive a higher influence from cotwin’s disease status, after
controlling for other environmental and behavioral risk factors. Figure 1 shows the
relationship between cotwin’s status and risk of peptic ulcer by zygosity, which is
calculated from Model 2. It is shown that identical twins have a 2.4 fold (2.08/0.87)
increased risk if their cotwins have the disease, compared with 1.65 in fraternal twins. This
effect is comparable to the effect shown in Table 2. It seems that the familiality of peptic
ulcer is independent of other risk factors.
Table S. Logistic Regression Predicting Risk of Peptic Ulcer*
OR
MODEL 1
95% C.L OR
MODEL 2
95% C.L
Cigarette smoking
Never smoked 1.00 1.00
Ever smoked, < lpk /day 1.08 (0.75, 1.57) 1.08 (0.75, 1.57)
Ever smoked, > lpk /day 1.78 (1.28, 2.49) 1.76 (1.26. 2.46)
Alcohol consumption
Non-drinker 1.00 1.00
Drinker, < 3 drks /day 0.92 (0.57, 1.48) 0.91 (0.57, 1.47)
Drinker, > 3 drks /day 1.05 (0.64, 1.73) 1.04 (0.64, 1.72)
Aspirin use
Never /only for short time 1.00 1.00
For over 1 year 1.41 (0.79, 2.51) 1.42 (0.79, 2.52)
For over 5 years 1.29 (0.91, 1.84) 1.30 (0.91, 1.85)
Eat large portions of
Bacon/Hot dog/Sausage
No 1.00 1.00
Yes 1.37 (0.87, 2.16) 1.37 (0.87, 2.16)
Eat large portions of
Cottage cheese
No 1.00 1.00
Yes 1.37 (0.82, 2.89) 1.37 (0.82. 2.29)
Co-twin with peptic ulcer
No 1.00 1.00
Yes 1.99 (1.26, 3.17) 1.65 (0.83, 3.28)
Identical twins -
No 1.00
Yes 0.87 (0.66, 1.16)
Identical + cotwin w ulcer
-
No 1.00
Yes 1.45 (0.57. 3.68)
* Other control variables include age, gender and education.
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
To explore potential gene/smoking interaction in peptic ulcer, Model 3 adds an
interaction term between smoking and cotwin’s disease status. Since light smokers are not
found to be significantly different from nonsmokers in ulcer risk, smoking status is
recoded as heavy smokers (> lpk / day) versus light or nonsmokers. An insignificant odds
ratio of 1.23 is found for the interaction term, indicating some additional increased risks
for those who are heavy smokers and meanwhile, having a family history of peptic ulcer.
Figure 2 shows calculated odds ratios by smoking status and cotwin’s disease status,
which indicates that family history exerts a somewhat higher effect on peptic ulcer among
smokers compared with that among nonsmokers or light smokers.
Further, if there exists a gene/smoking interaction, we would expect that the
interaction effect between smoking and cotwin’s disease status is higher in identical twins
than in fraternal twins. Therefore in Model 3, separate analyses are conducted for subjects
with different zygosity. The odds ratios for the interaction term is higher in identical twins
than in fraternal twins. Figure 3 shows the calculated effects of cotwin’s disease status, by
smoking status and zygosity. It is shown that identical twins have a 6-fold elevated risk of
peptic ulcer if they are heavy smokers and their cotwins have the disease, while the
relevant relative risk in fraternal twins is only 1.8. This suggests the existence of
gene/smoking interaction for peptic ulcer.
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 6. Logistic Regression Predicting Risk of peptic ulcer: With Focus on the Interaction Effect
B e t w e e i ^ C j g a r e t t e S n i o k j n j a n ^ ^ i
MODEL3
Total Fraternal twins Identical twins
OR 95% C.L OR 95 C.L OR 95% C.L
Heavy smoker (>1 pk/day)
No 1.00 1.00 1.00
Yes 1.69 (1.23,2.31) 1.68 (1.12,2.54) 1.72 (1.06,2.80)
Alcohol consumption
Non-drinker 1.00 1.00 1.00
Drinker, < 3 drks /day 0.92 (0.57,1.48) 1.02 (0.49,2.12) 0.87 (0.46,1.63)
Drinker, >3 drks /day 1.05 (0.64,1.73) 1.44 (0.68,3.06) 0.78 (0.39,1.54)
Aspirin use
Never/only for short time 1.00 1.00 1.00
For over 1 year 1.41 (0.79,2.51) 1.03 (0.40,2.67) 1.83 (0.87,3.85)
For over 5 years 1.29 (0.91,1.84) 1.50 (0.94,2.40) 1.11 (0.65,1.89)
Eat large portions of
Bacon/Hot dog/Sausage
No 1.00 1.00 1.00
Yes 1.37 (0.87,2.16) 1.67 (0.94,2.40) 0.99 (0.45,2.15)
Eat large portions of
Cottage cheese
No 1.00 1.00 1.00
Yes 1.37 (0.82,2.28) 1.52 (0.94,2.98) 1.07 (0.44,2.59)
Co-twin with peptic ulcer
No 1.00 1.00 1.00
Yes 1.81 (0.97,3.38) 1.92 (0.84,4.40) 1.73 (0.67,4.48)
Heavy smoker+cotwin w
ulcer
No 1.00 1.00 1.00
Yes 1.25 (0.49,3.19) 0.56 (0.12,2.54) 2.02 (0.56,7.36)
* Other control variables include age, gender and education.
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 1. Relative Risks of Peptic Ulcer by Zygosity and
Cotwin's Disease Status8
Z5
2.08
1.65
1.5
0.8!
0.5
Fraternal
Zygosity
□ Cotwin w/o ulcer
□ Cotwin w ulcer
aAdjusted for age, gender, education, smoking, alcohol, aspirin use,and dietary factors.
Figure 2. Relative Risks of Peptic Ulcer by Smoking Status and
Cotwin's Disease Status8
Nonsmokers/ light smokers Heavy smokers
□Cotwin w/o ulcer
□ Cotwin w ulcer
8 Adjusted for age, gender, education, alcohol, aspirin use,and dietary factors.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 3.1 Relative Risks of Peptic Ulcer by Smoking Status
and Cotwin's Disease Status, Fraternal Twins"
6
5
4
3
2
1
Q J
1.92
1 RR
1 .8 1
Nonsmokers/ light smokers Heavy smokers
□ Cotwin w/o ulcer
□ Cotwin w ulcer
Figure 3.2 Relative Risks of Peptic Ulcer by Smoking Status
and Cotwin's Disease Status, Identical Twins*
Nonsmokers/ light smokers
6.01
Heavy smokers
□ Cotwin w/o ulcer
□Cotwin w ulcer
a Adjusted for age, gender, education, alcohol, aspirin use and dietary factors.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 5. DISCUSSION
Twin studies provided a useful methodology to explore gene/environment
interactions without getting into specific genetic traits This population-based twin study
presented a comprehensive analysis of environmental risk factors of peptic ulcer together
with the familiality of this disease. The results can be summarized as the following:
Among the variables studied, cigarette smoking and family history seemed to be
two major factors associated with peptic ulcer disease. The effect of alcohol consumption,
although found to be significant in the univariate analysis, disappeared after controlling for
smoking status and other risk factors. Long-term aspirin use tended to increase the risk of
peptic ulcer, but the effect found was much lower than that suggested in the literature. For
dietary factors, salty food (bacon, hot dog, sausage) and cheese tended to somewhat
increase ulcer risk. Coffee, low-fiber food, or fat were not found to be related to peptic
ulcer. Physical exercise was negatively related with peptic ulcer, while depression tended
to be positively associated with the disease.
If these results largely replicated previous findings in the literature, the
gene/smoking interaction in peptic ulcer found in this analysis added some new
understanding of this disease. The higher level of smoking/cotwin’s disease status
interaction in identical twins than that in fraternal twins provided a strong signal of
gene/environmental interaction in the development of peptic ulcer. This effect, however,
did not achieve a substantial statistical significance in this analysis (p>0.0S). Therefore,
further studies need to be done to substantiate this finding. These studies may include
more detailed genetic epidemiological analysis and biological research exploring potential
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
interactions between smoking and those genetic traits known to be related to peptic ulcer
disease.
Although this study provided a comprehensive understanding of the risk factors of
peptic ulcer disease, the underlying causality of many associations between the disease and
risk factors, can not be established based on the cross-sectional data used in this analysis.
For example, if peptic ulcer interferes with a person’s physical activity and psychological
well-being, then the causal order for the associations found between physical exercise as
well as depression and peptic ulcer would be very unclear. Moreover, if ulcer patients tend
to restrict their risk behaviors due to cognitive effects, the effects of smoking, alcohol,
drugs and dietary factors may have been underestimated in this analysis.
An important factor related to peptic ulcer disease, Helicobacter pylori, was
unfortunately missing in this analysis, because this information was not available in the
survey. A recent meta-analysis of peptic ulcer showed that nearly 50% of the risk of peptic
ulcer can be attributable to H. pylori in the general population (Kurata, 1997), and the
eradication of H. Pylori has become one of the standard treatments of peptic ulcer disease
(NIH, 1994). Medical examination is necessary to find whether one is inflected with H.
pylori, so surveys based on self-administered questionnaires can not obtain this
information. Nonsteroidal anti-inflammatory drugs use has been regarded as the second
risk factor of peptic ulcer. However, only the effect of aspirin use was examined in this
study. With these important information missing, this study is far from complete.
Another limitation of this analysis is that duodenal ulcer and gastric ulcer were not
analyzed separately, because information on ulcer location was not available in the survey.
33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The literature showed that some ulcer risk factors tended to act differently for DU and GU
(Kurata, 1993), and these two ulcers tended to have distinct etiology. So the combined
analysis of DU and GU in this study can only provide a rough epidemiological description
of peptic ulcer.
This study can be significantly improved if detailed medical history information,
such as time of the disease onset, location of ulcer (DU or GU), NS AID use, blood type,
become available in the future. These information can possibly be obtained through
subjects’ medical record or through a medical examination. The latter can also provide
data on H. pylori as well as on some genetic traits suspected to be related to peptic ulcer.
With these additional information, a more detailed analysis of risk factors of peptic ulcer
can be conducted, and more importantly, the causality problem existing in the current
analysis could be solved.
34
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
BIBLIOGRAPHY
Adams, P.F., Bensen, V. (1991). Current estimates from the National health Interview
Survey, 1989-1990. National Center for Health Statistics Vital and Health
Statistics, series 10. DHHS publication nos (PHS)90-1504, (PHS)92-1509,
Hyattsville, MD.
Ainsworth, M.A., Hogan, D.L., & Koss, M.A. (1992). Cigarette smoking inhibits acid-
stimulated human proximal duodenal mucosal bicarbonate secretion.
Gastroenterology, 102, A32 (suppl; abstr).
Almy T.P., Mendeloff, A. I., & Rice, D. (1975). Prevalence and significance of digestive
diseases. Gastroenterology, 68, 1351-1371.
Anda, R.F., Williamson, D.F., Escobedo, L.G., et al. (1990). Smoking and the risk of
peptic ulcer among women in the United States. Archives o f Internal Medicine,
150, 1437-1441.
Aspirin Myocardial Infarction Study Group.(1980).A randomized, controlled trial of
aspirin in persons recovered from myocardial infarction. Journal o f American
Medical Association, 243: 661-669.
Blaster, M. (1987). Gastric compylobacter-like organisms, gastritis, and peptic ulcer
disease. Gastroenterology, 93, 371-383.
Bloom, B.S. (1989). Risk and cost of gastrointestinal side effects associated with
nonsteroidal anti-inflammatory drugs. Archives o f Internal M edicine , 149, 1019-
1022.
Boyd, E.J.S., Wilson, J.A., & Wormsley, K.G. (1983). Smoking impairs therapeutic
gastric inhibition. Lancet, 1, 95-97.
Center for Disease Control. (1991). Cigarette smoking among adults— United States,
1988. MWR40, 757-759, 765.
Collier, D.S.J, & Pain, J.A. (1985). Nonsteroidal anti-inflammatory drugs and peptic ulcer
perforation. Gut, 26, 359-363.
Conn, H.O., & Blitzer, B.I. (1976). Medical progress: non-association of
adrenocorticosteroid therapy and peptic ulcer. New England Journal o f Medicine,
294, 473-479.
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Cobb, S. & Rose R. (1973). Hypertension, peptic ulcer and diabetes in air traffic
controllers, Journal of American Medical Association 224, 489-492.
Coghlan, J.G., Gilligan, D., Humphries, H. et al. (1987) Campylobacter pylori and
recurrence of duodenal ulcers — A 12 month follow-up study. Lancet 2, 1109-
1111.
Convay, T.L., Ward, H.W., Vickers, R.R., & Rahe, R.H. (1981). Occupational stress and
variation in cigarette, coffee, and alcohol consumption. Journal o f Health and
Social Behavior 22, 155-165.
Cryer, B., Lee, E., & Feldman M.(1992). Factors influencing gastroduodenal mucosal
prostaglandin concentrations: Roles of smoking and aging. Annals o f Internal
Medicine, 116, 636-640.
Cummings, S. (1983). Stress, cigarettes, and ulcers. Gastroenterology, 85, 1232.
Dunn, J.P. & Cobb, S. (1962). Frequency of peptic ulcer among executives, craftsmen,
and Formen. Journal o f Occupational Medicine, 4, 343-348.
Eberhard, G. (1968). Peptic Ulcer in Twins— a Study in personality, heredity, and
environment. Acta Psychiatric Scandinavia [Suppl], 205, 7-118.
Elashoff, J.D., & Grossman, M.I. (1980). Smoking and duodenal ulcer. Gastroenterology,
79, 81.
Elashoff, J.D., & Grossman, M.I. (1980). Trends in hospital admissions and death rates
for peptic ulcer in the United States from 1970-78. Gastroenterology, 78, 280-
285.
Falconer, D.S. (1981). Introduction to Quantitative Genetics. (2nd ed.). New York:
Ronald Press.
Feldman, M., Walker, P., Green, J., & Weigarden, K. (1986). Life events stress and
psychological factors in men with peptic ulcer disease. Gastroentrology, 91,1370-
1379.
Friedman, G.S., Siegelaub, A.B., & Seltzer, C.C. (1974). Cigarettes, alcohol, coffee, and
peptic ulcer. New England Journal o f Medicine, 290, 469.
Fries, J.F., Williams, C.A., Bloch, D.A., etal. (1991). Nonsteroidal anti-inflammatory
drug- associated gastropathy: Incidence and risk factor models. American
Journal o f Medicine, 91, 213-222.
36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Galton, F. (1869). Heredity genius: A n inquiry into its laws and consequences,
Macmillan, London.
Garland, F.C., Garland, C.F., & Gorham, E.D. (1992). Gastrointestinal tract disorders. In
J.M. Last & R.B. Wallace (Eds), Public health and preventive medicine (13th ed.
pp. 897-912). Norwalk: Appleton & Lange.
Gotlieb-Jensen, K. (1972). Peptic ulcer: genetic and epidemiologic aspects based on twin
studies. Copenhagen, Munksgaard..
Graham, D.Y. (1989). Campylobacter pylori and peptic ulcer disease. Gastroenterology
96, 615-625.
Graham, D.Y., Lew, G.M., Evans D.G., et al. (1991). Effect of triple therapy (antibiotics
plus bimuth) on duodenal ulcer healing. Annals o f Internal Medicine, 115, 266-
269.
Griffin, M.R., Piper, J.M., Daugherty. J.R., et al. (1991). Nonsteroidal anti-inflammatory
drug use and increased risk for peptic ulcer disease in elderly persons. Annals o f
Internal Medicine, 114, 257-263.
Haruma, K., Fujimura, J., Hata, J., et al. (1992). Effect of Cigarette smoking on
antroduodenal mortality. Gastroenterology, 102, A457 (suppl; abstract).
Holvoet, J., Terriere, L., Van Hee, W. et al. (1991). Relation of upper gastrointestinal
bleeding to non-steroidal anti-inflammatory drugs and aspirin: A case-control
study. Gut, 32, 730-734.
Isenberg, J.I. (1984). Peptic ulcer: Epidemiology, nutritional aspects, drugs, smoking,
alcohol and diet. Journal o f Family Practice, 18,141-151.
Kaufmann, D., Wilder-Smith, C.H., Kempf, M., et al. (1990). Cigarette smoking,
gastric acidity and peptic ulceration— What are the relationships? Digestive
Disease Science 35, 1482-1487.
Kikendall, J.W., Eval, J., Johnson, L.F. (1984). Effect of cigarette smoking on
gastrointestinal physiology and non-neoplastic digestive disease. Journal o f
Clinical Gastroenterology, 6, 65-78.
Korman, M.G., Hansky, J., Evan, E.R., & Schmidt G.T. (1983). Influences of smoking
on healing and relapse of duodenal disease. Gastroenterology, 85, 871-874.
37
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Korman, M.G., Shaw, R.G., Hansky, J., Schmidt, G.T., & Stem, A.I. (1981). Influence of
cigarette smoking on healing and rate of duodenal ulcer in response to cimetidine
or high dose antacid. Gastroenterology, 80,1451-1453.
Kurata, J.H., Elashoff, J.D., Haile, B.M., et al. (1983). A reappraisal of twin trends in
ulcer disease: Factors related to changes in ulcer hospitalization and mortality
rates. American Journal o f Public Health, 73, 1066-72.
Kurata, J.H., Elashoff, J.D., Nogawa, A.N., et al.(1986) Sex and smoking differences in
duodenal ulcer mortality. American Journal o f Public Health, 76, 700-702.
Kurata, J.H., Nogawa, AN., Abbey, D.E., et al. (1992). A prospective study of risk for
peptic ulcer disease in Seventh-Day Adventists. Gastroenterology, 102, 902-909.
Kurata, J.H., & Haile, B.M. (1984). Epidemiology of peptic ulcer disease. Clinical
Gastroenterology, 13, 289-307.
Kurata, J.H., Abbey, D.E. (1990). The effect of chronic aspirin use on duodenal and
gastric ulcer hospitalizations. Journal o f Clinical Gastroenterology, 12, 260-266.
Kurata, J.H. & Nogawa A.N. (1997). Meta analysis of risk factors for peptic ulcer.
Journal o f Clinical Gastroenterology, 24, 2-17.
Kurata, J.H. (1993). Epidemiology: Peptic ulcer risk factors. Seminars in Gastrointestinal
Disease, 4, 2-12.
Lam, S.K., Isenberg, J.I., Grossman, M.I., et al. (1980). Gastein secretion is abnormally
sensitive to exogenous gastrin released after peptone test meals in duodenal ulcer
patients. Journal o f Clinical Investigation, 65, 555.
Lanza, F.I., Royal, G.L., Nelson, R.S., Chen, T.T., Seckman, C.E., Rack, M.E. (1981). A
comparative endoscopic evaluation of the damaging effects of non-steroidal anti
inflammatory agents on gastric and duodenal damage. American Journal o f
Gastroenterology, 75, 17-20.
Loebl, D.H., Craig, R., & Culie, D.D. (1977). Gastrointestinal blood loss: effect of aspirin,
fenoprofen, and acetaminophen in rheumatoid arthritis as determined by sequential
gastroscopy and redioactive fecal markers. Journal o f American M edical
Association, 237, 976-981.
Laporte, J.R., Came, X., Vidal, X., et al. (1991). Upper gastrointestinal bleeding in
relation to previous use of analgesics and non-steroidal anti-inflammatory drugs.
Lancet 337, 85-89.
38
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Levy, M. (1974). Aspirin use in patients with major upper gastrointestinal bleeding and
peptic ulcer disease. New England Journal o f Medicine, 290, 1158-1162.
Mack, T.M. (1996). Unpublished document. Norris Cancer Comprehensive Center,
University of Southern California.
Malaty, H.M., Isaksson, I., Engtrand, L., Graham, D.Y., & Pedersen, N.L. (1997). Are t
here genes for peptic ulcer disease in common with genes for Helicobactor pylori
infection: A twin study. American Journal o f Epidemiology, 146, S42 (SER
Abstracts).
Massarat, S., & Eisenmann, A. (1981). Factors affecting the healing rate of duodenal and
pyloric ulcers with low-dose antacid treatment. Gut, 22: 97-102.
McCarthy, D.M. (1984). Smoking and ulcer— time to quit New England Journal o f
Medicine, 311, 726-727.
McConnell, R.B. (1980). Peptic ulcer, early genetic evidence— families, twins, and
markers. In: J.L. Rotter, I.M. Samloff, & D.L. Rimoin (Eds). The Genetic and
Heterogeneity o f Common Gastrointestinal Disorders (pp 1-41). New York:
Academic.
Mashall, B.J., & Warren, J.R. (1984). Unidentified curved bacilli in the stomach of
patients with gastritis and peptic ulceration. Lancet, 1, 1311-1314.
Marshall, B.J., Goodwin, C.S., Warren, J.R., et al. (1988). Prospective double blind trail
of duodenal ulcer relapse after eradication of Campylobacter pylori. Lancet, 2,
1437-1442.
McConness, R.B. (1966). Gastric and duodenal ulcer. In The Genetics o f
Gastrointestinal Disorders. New York: Oxford University Press.
McCready, D.R., Clar, L., Cohen, M.M. (1985). Cigarette smoking reduces human gastric
luminal prostaglandin E2. Gut, 26, 1192-1196.
National Institutes of Health, Office of the Director^1994). NIH Consensus Statement:
Helicobactor pylori in peptic ulcer disease, Vol. 12, No. 1.
Neale, M.C., & Cardon, L.R. (1992). Methodology fo r genetic studies o f twins and
fam ilies. Boston: Kluwer Academic Press.
Ostensen, H., Gudmundsen, T.E., Ostensen, M., Burhol, P.G., Bonnevie, O. (1985).
Smoking, alcohol, coffee, and familial factors: any associations with peptic ulcer
disease? Scandinavia Journal o f Gastroenterology, 20, 1227-1235.
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Palgi, A. (1981). Association between dietary changes and mortality: Israel 1949 to 1977:
A trend-free regression model. American Journal o f Clinical Nutrition, 34, 1569-
1583.
Peters, M.N. (1983). Richardson CT: Stressful life events, acid hypersecretion, and ulcer
disease. Gastroenterology, 84, 114-119.
Petitti, D.B., Friedman, G.D., & Kahn, W. (1982). Peptic ulcer disease and the tar and
nicotine yield of currently smoked cigarettes. Journal o f Chronicle Diseases, 35,
503-507.
Piper, D.W. (1980). The treatment of chronic peptic ulcer. Front Gastrointestinal
Research, 6, 109.
Piper, D.W., Greig, M., Shinners, J., Thomas, J., Crawford, J. (1978). Chronic gastric
ulcer and stress: a comparison of an ulcer population with control population
regarding stressful life events over a lifetime. Digestion, 18, 303-309.
Piper, D.W., McIntosh, J.H., Ariotti, D.E., Calogiuri, J.V., Brown, R.W., & Shy, C.M.
(1981). Life events and chronic duodenal ulcer: a case control study. Gut,
22:1011-1017.
Podolski, J.L. (1996). Recent advances in peptic ulcer disease: H pylori infection and its
treatment. Gastroenterology Nursing, 19(4), 128-136.
Pollin, W., Allen, M.G., Hoffer, A., Stabenau, J.R., & HrubecZ.(1969). Psychopathology
in 15,909 pairs of veteran twins: Evidence for a genetic factor in the pathogenesis
of schizophrenia and its relative absence in psychoneurosis. American Journal o f
Psychiatry, 126, 597-610.
Quimby, G.F., Bonnice, C.A., Burstein, S.H., et al. (1986). Active smoking depresses
prostaglandin synthesis in human gastric mucosa. Annals o f Internal Medicine,
104: 616-619.
Rauws, E.A., Tytgat, G.N. (1990). Cure of duodenal ulcer associated with eradication of
Helicobacter pylori. Lancet, 335, 1233-1235.
Rauws, E.A., Langenberg, W., HouthofF, H., et al.(1988). Campylobacter pyloridis-a
ssociated chronic active antral gastritis: A prospective study of its prevalence and
the effects of antibacterial and antiulcer treatment. Gastroenterology, 94, 33-40.
40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Richardson, C.T. (1989). Gastric ulcer. In M.H. Sleisenger, & J.S. Fordtran (Eds):
Gastrointestinal Disease: Pathophysiology, Diagnosis, Management (4th Ed., pp
879-909). Philadephia: Saunders.
Rotter, J.I., Shohnt, T., and Petersen, G.M. (1992). Peptic ulcer disease. In R.A. King,
J.I. Rotter, A.R. Motulsky (Eds.), The Genetic Basis o f Common Diseases. New
York: Oxford University Press.
Rotter, J.I. (1981). Genetic Aspects of Ulcer Disease. Comparative Therapy, 7, 16-25.
Rotter, J.I. (1980). Peptic ulcer disease— more than one gene, more than one disease.
Progress in Medical Genetics, 4, 1-58.
Rotter, J.I., Samloff, I.M., & Romoin, D.L. (Edsj. The Genetics o f Heterogeneity o f
Common Gastrointestinal Disorders, pp31-41. New York: Academic.
Segawa, K., Nakazawa, S., Tsukamoto, Y., et al. (1987). Peptic ulcer is prevalent among
shift workers. Digestive Disease Science, 32, 449-453.
Semble, E.L., Turner, R.A., Wu, W.C. (1987). Clinical and genetic characteristics of
upper gastrointestinal disease in rheumatoid arthritis. Journal o f Rheumatology,
14, 692-699.
Smedley, F., Hichish, T., Taube, M., Yale, C., leach, R., & Wastell, C. (1988). Perforated
duodenal ulcer and cigarette smoking. Journal o f Social Medicine, 81, 92-94.
Sleisenger, M.D., & Fordtran, J.S. (1983J. Gastrointestinal disease, pathophysiology,
diagnosis, and treatment (pp655). Philadephia: W.B. Saunders.
Sonnenberg, A., Muller-Lissner, S.A, Vogel, E., Schmid, P., Gonvers, J., Peter, P.,
Strohmeyer, G., Blum, A.L. (1981) Predictors of duodenal ulcer healing and
relapse. Gastroenterology, 81, 1061-1067.
Sontage, S., Graham, D.Y., Belsito, A., Weiss, J., Farley, A., & Grunt, R. (1984).
Cimetidne, Cigarette smoking, and recurrence of duodenal ulcer. New England
Journal o f Medicine, 311, 689-693.
Soil, A. (1991). Nonsteroidal anti-inflammatory drugs and peptic ulceration. Annals o f
Internal Medicine, 114,307-319.
Somerville, K., Faulkner, G., Langman, M. (1986). Non-steroidal anti-inflammatory drugs
and bleeding peptic ulcer. Lancet, 1, 462-464.
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Sonnenberg, A., & Haas, J. (1986). Joint effects of occupation and nationality on the
prevalence of peptic ulcer in German Workers. British Journal o f Medicine, 43,
490-493.
Spiro, H.M. (1981). Is milk all that bad for the ulcer patient? Journal o f Clinical
Gastroenterology, 3, 219.
Stemmermann, G.N., Marcus, E.B., & Buist, A.S. (1992). Relative impact of smoking and
reduced pulmonary function on peptic ulcer risk: A prospective study of Japanese
and Koreans in Japan. Journal o f Clinical Gastroenterology, 15, 68-74.
Stemmermann, G.N., Marcus, E.B., Buist, A.S., & MacLean, C.J. (1989). Relative impact
of smoking and reduced pulmaonary function on peptic ulcer risk.
Gastroenterology, 96, 1419-1424.
Steering Committee of the Physicians’ Health Study Research Group. (1989). Final report
on the aspirin component of the ongoing Physicians’ Health Study. New England
Journal o f Medicine, 321, 129-135.
Sturdevant, R.A.L. (1976). Epidemiology of peptic ulcer: Report of a conference.
American Journal o f Epidemiology, 104, 9-14.
Thompson, M.R., Morris, A.F., Hughes, A.D. (1980). Indomethacin and perforated
duodenal ulcer. British M edical Journal, 280, 1618.
Thomas, J., Greig, M., Piper, D.W. (1980). Chronic gastric ulcer and life events.
Gastroenterology, 78, 905-911.
Tovey, F.I. (1979). Peptic ulcer in India and Bangladesh. Gut, 20, 329-347.
Tovey, F.I., Tunstall, M. (1975). Progress report: Duodenal ulcer in black populations in
Africa south of the Sahara, Gut, 16, 564-576.
Walker, P., Luther, J., Samloff, I.M., Feldman, M.(1988). Life events, stress and
psychological factors in men with peptic ulcer disease. Gastroenterology, 94, 323.
Walt, R., Logan, R., Katschinski, B., Ashley, J., Langman, M. (1986). Rising frequency of
ulcer perforation in elderly people in the United Kingdom. Lancet, 1, 489-492.
Welsh, J. (1977). Diet therapy and peptic ulcer disease. Gastroenterology, 72, 740.
Whitfield, P.F., Hobsley, M. (1987). Comparisons of maximal gastric secretion in smokers
and nonsmokers with and without duodenal ulcer, Gut, 28, 557-560.
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IMAGE EVALUATION
TEST TARGET (Q A -3 )
r
< 9
s *
150mm
IM rfG E . Inc
1653 East Main Street
Rochester. NY 14609 USA
Phone: 716/482-0300
Fax: 716/288-5989
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Descriptive epidemiology of thyroid cancer in Los Angeles County, 1972-1995

PDF

A comparative study of environmental factors associated with multiple sclerosis in disease-discordant twin pairs

PDF

Evaluation of the accuracy and reliability of self-reported breast, cervical, and ovarian cancer incidence in a large population-based cohort of native California twins

PDF

An exploration of nonresponse with multiple imputation in the Television, School, and Family Project

PDF

Immune recovery vitritis in AIDS: Incidence, clinical predictors, sequellae, and treatment outcomes

PDF

Determinants of body mass in healthy adult identical twins

PDF

Associations of physical fitness and other risk factors with cardiovascular and non-violent mortality

PDF

Rates of cognitive decline using logitudinal neuropsychological measures in Alzheimer's disease

PDF

The effect of family structure on risk of familial and sporadic self-reported asthma in a cohort of California twin pairs

PDF

Imputation methods for missing data in growth curve models

PDF

Associations of weight, weight change and body mass with breast cancer risk in Hispanic and non-Hispanic white women

PDF

The Effect Of Non-Sampling Measurement Errors On Relative Risk Estimates Of Coronary Heart Disease In Framingham Women

PDF

Do reproductive factors have the same effect on breast cancer risk in African-American and White women?

PDF

Evaluation of the efficacy of two lipid-lowering treatments using serial quantitative coronary angiography: Two- and four-year treatment results

PDF

A joint model for Poisson and normal data for analyzing tumor response in cancer studies

PDF

Analysis of gene-environment interaction in lung cancer

PDF

A study of pediatric oncology nurses' attitudes to and knowledge of genetic testing

PDF

Influence of body weight and weight change on carotid IMT and IMT progression: A pooled analysis using subject-level data from four clinical trials

PDF

Developing a case report form for assessing the risk factors for cervicitis among women in rural El Salvador

PDF

Construction and validation of multivariable population pharmacokinetic models: Utility of allometric forms to predict the pharmacokinetic disposition of gentamicin in pediatric patients with app...

Asset Metadata

Creator Wang, Haitao (author)

Core Title Familiality and environmental risk factors of peptic ulcer: A twin study

School Graduate School

Degree Master of Science

Degree Program Applied Biometry

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

Tag biology, biostatistics,biology, genetics,health sciences, public health,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Mack, Thomas (committee chair), Cozen, Wendy (committee member), Hamilton, Ann S. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-23376

Unique identifier UC11337888

Identifier 1391105.pdf (filename),usctheses-c16-23376 (legacy record id)

Legacy Identifier 1391105.pdf

Dmrecord 23376

Document Type Thesis

Rights Wang, Haitao

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