Epidemiology Report: Analysis of GER and Esophageal Adenocarcinoma

This report provides a detailed analysis of an epidemiological study investigating the relationship between gastroesophageal reflux (GER) and esophageal adenocarcinoma. The study, conducted by Terry et al. (2000), examines the association between dietary factors and the risk of gastric cardia and adenocarcinoma. The report outlines the study design, population, and methodology, including the use of a nation-based population study in Sweden. It evaluates the internal and external validity of the study, discussing potential confounding factors, chance variations, and biases, such as recall bias. The report also assesses the causal association between exposure and outcome using the Bradford-Hill criteria, concluding that a causal relationship is not plausible. Furthermore, the report discusses the generalizability of the study's findings to wider populations and considers the limitations of the study, such as the lack of consistency with previous research and potential for misclassification. The report highlights the need for further research to clarify the complex relationship between dietary factors, GER, and esophageal cancer.

Running head: EPIDEMIOLOGY
Epidemiology
Name of the Student
Name of the University
Author note

1EPIDEMIOLOGY
1) Overview of the selected paper
Gastroesophaeal reflux (GER) is a digestive disease that affects the lower esophageal
sphincter (LES), which is a ring of muscles between the stomach and the esophagus (Cook et al.,
2014). GER has been found to a one of the primary cause of the esophageal adenocarcinoma.
There are several dietary food products like chocolate, mint products, dietary fatty foods, coffee,
onions, tomatoes and citrus fruits that are found to be partially responsible for the GER, which is
associated with the relaxation of the lower sphincter muscles (Gibson & Shepherd, 2012).
Aim of the study
This study conducted by Terry et al., (2000), aims to prove whether the dietary foods
items are associated with the risk of gastric cardia and adenocarcinoma. Thus, there was a clear
statement of the aims of research.
Study design and study population
In order to prove the fact that dietary factor is not associated with the development of
Adenocarcinoma, the Author Terry et al., (2000), had conducted a nation based population study
in Sweden, where 185 and 258 cases of esophageal Adenocarcinoma and gastric cardia have
been taken. 815 were taken as control. The population study contained Swedish population
below the age of 80 years in between 1 December 1994 and 31 December 1997. The patients
who had been newly identified with the case of Adenocarcinoma were also considered. All the
patients with squamous cell carcinoma have been avoided, as the risk of squamous cell
carcinoma is not related to GER. All the cases have been identified anatomically and histological
by a study pathologist. Therefore, all the cases have been classified based on the site of the tumor
and the histological type. Non-participation in the cases is mainly due to poor clinical conditions
or immediate death after the diagnosis. There was also nonparticipation among the controls due

2EPIDEMIOLOGY
to their unwillingness to participate. Seven subjects were removed from the study due to their
errors in the dietary information and finally the study was conducted with 185 and 258 cases of
esophageal adenocarcinoma and cardia adenocarcinoma and 815 controls. Thus, this qualitative
method is the right methodology for addressing this research goal.
Pattern of the study
The questionnaire for the interview contained questions related to the dietary habits. 63
food items and beverages have been considered for the study and the participants were enquired
about the food habits 20 years back to avoid any political risk factors. Questions regarding the
meal timings and the average size of the meal were asked by considering the photographs of the
standard Swedish 7 common dishes. The reflux symptoms have been classified as hear burn at
least once per week. The duration, intensity and the frequency of the reflux are also considered
(Gibson & Shepherd, 2012). Thus, it can be seen that all the parameters for the study had been
properly considered in this paper. The correlation between the reflux symptoms and the foods
items have been studied only in the control subjects. The data collection method was sufficient
although the author might have taken up a large population to avoid the chance variation.
Findings
From the population study, it was found that the GER symptoms were not related to the
dietary factors. The dietary factors were previously associated with reduction in the mean basal
pressure of the LES, but it is not related to the risks of adenocarcinoma. As per the population
study, the dietary factors associated with LES relaxation are not related to esophageal
malignancy in any way. As per the statistical data although, transient LES is found to be closely
linked to the disease, the dietary factors have no association with the development of cancer.

3EPIDEMIOLOGY
2) Appraisal of the internal validity
The exposure of this study is the dietary factors that are being predicted to have a relation
with adenocarcinoma that is the outcome of this study.
The confounding factor in this study is the dietary supplements that in real did not have
any relation with the development of malignancy. A confounder is normally a variable that
actually affects the different variables in the study that affects the actual relationship. There are
several ways to control the different confounding factors like Restriction, matching and
randomization. In this study, the author has chosen the randomization method to eliminate the
confounding factors. It can be found from the study that the adjustments due to the confounding
factors did not affect the age-adjusted data.
Presence of chance variation in the study may lead to erroneous results. It is evident from
the study that the response rate of this study is quite low and the sample population is also large.
Chance variation in a randomized control study can be managed by the increasing the population
of the participants.
The participants were subjected interviews taken by the professional interviewers from
Statistics Sweden. It should be noted that 10 dietary factors were thought to be associated with
LES relaxation in the laboratory settings. An increase in prevalence of gastric cardia was
observed due to chocolate intake (Zhang, 2013). No trends of Adenocarcinoma were observed
associated to the dietary foods (Gibson & Shepherd, 2012). The meal timing and the portion size
of the meals were also found to be unrelated to Adenocarcinoma (Zhang, 2013). It should be
noted that the individuals having a tendency of GER, normally consume less amount of those
food that causes the symptoms, which actually weakens the association of the reflux symptoms

4EPIDEMIOLOGY
with the food causing LES relaxation (Ganz et al., 2014). Thus the presence or absence of GER
does not validate the point that food causing LES relaxation would increase the risk of Adeno
carcinoma, although avoidance of food is a very narrow range of exposure in the participants
(Jarosz & Taraszewska, 2014).
The non-association of cancer with food can be due to certain factors. The food that is
responsible for temporary reflux or relaxed LES in the laboratory settings might not be sufficient
to give rise to significant chronic reflux (Napier, Scheerer & Misra, 2014).
Another factor that might have caused bias in the study is that, in most of the laboratory
study the portion size of the food was moderate in size that might not be successful for producing
an acid reflux in the laboratory. Another possible explanation is that, the severity of the
temporary refluxes can be different from the actual GER taking place in a body (Napier,
Scheerer & Misra, 2014). Although the relation of dietary food and GER remains elusive,
transient LES was found to be more closely associated to the growth of a disease. It was found
that the food-induced regurgitation has a low harm rate than the LES relaxations (Wheeler &
Reed, 2012). It should be noted that the data used in the study was limited and non-differential
measurement error of exposure probably would have weakened the findings of the study. A high
degree of differential participation with reference to exposure might have caused theoretical bias
and have given results towards or far from unity.
3) Causal association between exposure and outcome
The aim of epidemiological studies is to assess the cause of any disease and as most of
the studies are observational instead of experimental, there are many possibilities for observed
association that are taken into consideration (Gage, Munafò & Davey Smith, 2016). The

5EPIDEMIOLOGY
inference of cause-effect relationship exists in case-control studies and bias, confounding affects
the results. The association between risk factor and exposure to disease do not infer causal
relationship (Vandenbroucke & Pearce, 2015). In the given article, causal association can be
studied using Bradford-Hill criteria that provide a framework for the assessment of an observed
association to be causal. In case-control studies, the chance of showing causal relationship is
less, as it is majorly prone to bias. As discussed above in internal validity, there is bias in
selection, incident cases, sources of cases and control selection. The recall bias is the main
reason for less causal relationships studied in case control studies where cases showed certain
exposures than controls due to the fact about what may have caused the disease (Vandenbroucke,
Broadbent & Pearce, 2016). Although, the response rates were high in the study as compared to
previous studies of dietary factors in oesophageal adenocarcinoma, there was high degree of
differential misclassification (recall bias) as the response rates in cases and controls were related
to dietary factors that relaxes LES. This can be the main reason for the negative findings
obtained in the study. The hypothesis regarding the categorization of tertiles, foods containing
coffee, chocolate, and mint, public does not know portion size and time of last daily meal and
this may have elicited differential recall between case and control participants in the study.
Taking into consideration, the criterion used for finding causal relationship suggests that
there was no association between dietary factors that causes LES relaxation and risk for gastric
cardia and oesophagus adenocarcinoma. The strength of association between the risk factor and
outcome is weak, as findings showed no relationship between dietary factors that causes LES
relaxation and risk for gatric cardia or adenocarcinoma of oesophagus. There is no consistency of
findings as the study was conducted among Swedish population and so, this finding cannot be
observed in other populations. Another criterion for the study of causal relationship is specificity

6EPIDEMIOLOGY
and in the given article, the study lacked one to one relationship between cause and outcome
(Richmond et al., 2014). Coherence is also not well defined in this study, as the current
knowledge of biology of disease does not agree with relationship (Oakes, 2017). Previous
studies on dietary factors showed strong relationship between foods that relaxes LES and risk for
oesophageal adenocarcinoma and gastric cardia, however, this study showed no such
associations between LES relaxing foods and oesophageal malignancy. There is no temporal
sequence of association, as exposure did not precede outcome. In the given article, the recruited
sample was already having the disease (oesophageal cancer and gastric cardia) before the study
for outcomes. The biological plausibility criterion is also not fulfilled in this study, as biological
mechanism was not explained in the study for the increased risk for oesophageal
adenocarcinoma and gastric cardia. Therefore, from the above discussion, it can be inferred that
causal relationship between exposure and outcome is not plausible and no association between
LES relaxing foods and risk for oesophageal malignancy.
4) External validity
Observations in epidemiological studies can be evaluated in terms of external validity in
case-control studies. It is defined as the ability of study to generalize result findings in the wider
population. The cause-effect relationship in a case-control study is evaluated externally as it can
be generalized to other population settings (Vandenbroucke et al., 2014. In the given article, the
findings do not align with previous studies and does not hold for other conditions or populations.
Although, the samples are large, the participants are recruited from Sweden population register
that is a single geographic location. According to Zhang, (2013) the incidence of gastric cardia
and oesophageal adenocarcinoma is highly prevalent in Sweden due to gastro-oesophageal
reflux.

7EPIDEMIOLOGY
The avoidance of foods that causes LES relaxation among the sufferers can be a reason
for narrowing of exposure range. The previous research examined dietary factors and risk for
malignancy showing strong relationship, but the study by Terry et al., (2000) showed no
association between these factors. Food items like citrus foods, garlic and onions promote reflux,
however, might have anti-carcinogenic properties negating reflux harmful effects. The results of
the study do not contradict previous findings that showed strong relationship between reflux
symptoms and adenocarcinoma malignancy. In the general population, people are not aware of
consumption of reflux-promoting foods and this suggests that it is not a health issue in universal
population (Boeckxstaens et al., 2013). The condition of reflux symptoms is perhaps a
consequence or explanations for the above disease condition. The results obtained from the study
showed no positive association between LES relaxing foods and risk for malignancy. There are
implications for future studies and then, it would be possible to apply the findings to source
population from which the study is derived.
The discussion is explained in an explicit manner demonstrating the findings of the study
conducted in Sweden. The random population in Sweden comprising of adults with self-reported
recurrent GER symptoms two decades before interview were no associated with lower basal
pressure or dyspepsia of LES. The results showed no consistency that dietary factors are not
associated with adenocarcinoma risk of oesophagus or gastric cardia. The strengths and
limitations of the study are also explained in the discussion. The strength is population-based
design with large sample size that allowed large comparative study. Moreover, controls were
recruited from random sample and cases were newly diagnosed oesophageal cancer cases that
were classified rigorously based on site of tumour and histological type.

8EPIDEMIOLOGY
The discussion also explained the possible reasons for no association between dietary
factors and risk for malignancy. The food consumption rendering temporary GER or LES
relaxation in lab experiments may not be enough in frequency or quantity to cause reflux disease.
In past studies, food items as onions, chocolates, fat or orange juice that induced reflux
symptoms were moderate and considered only one portion of the study. This illustrates that
temporary reflux caused by these food items may be different from inflammation characteristic
GER (Banovcin et al., 2016). The weak association between food items causing LES relaxation
may be due to less consumption of these foods among individuals predisposing to reflux
symptoms (Cook et al., 2014). The authors explained the avoidance of these food items might be
a reason for narrowing of range of exposure. Lastly, from the study findings they concluded that
dietary factors associated with transient GER and LES relaxation is not associated with any kind
of risk for gastric cardia or oesophageal malignancy.

9EPIDEMIOLOGY
References
Banovcin Jr, P., Halicka, J., Halickova, M., Duricek, M., Hyrdel, R., Tatar, M., & Kollarik, M.
(2016). Studies on the regulation of transient lower esophageal sphincter relaxations
(TLESRs) by acid in the esophagus and stomach. Diseases of the Esophagus, 29(5), 484-
489.
Boeckxstaens, G., El-Serag, H. B., Smout, A. J., & Kahrilas, P. J. (2014). Symptomatic reflux
disease: the present, the past and the future. Gut, 63(7), 1185-1193.
Cook, M. B., Corley, D. A., Murray, L. J., Liao, L. M., Kamangar, F., Ye, W., ... & Chow, W. H.
(2014). Gastroesophageal reflux in relation to adenocarcinomas of the esophagus: a
pooled analysis from the Barrett’s and Esophageal Adenocarcinoma Consortium
(BEACON). PLoS One, 9(7), e103508.
Gage, S. H., Munafò, M. R., & Davey Smith, G. (2016). Causal inference in developmental
origins of health and disease (DOHaD) research. Annual review of psychology, 67, 567-
585.
Ganz, R. A., Peters, J. H., Horgan, S., Bemelman, W. A., Dunst, C. M., Edmundowicz, S. A., ...
& Schlack-Haerer, S. C. (2013). Esophageal sphincter device for gastroesophageal reflux
disease. New England Journal of Medicine, 368(8), 719-727.
Gibson, P. R., & Shepherd, S. J. (2012). Food choice as a key management strategy for
functional gastrointestinal symptoms. The American journal of gastroenterology, 107(5),
657-666.

10EPIDEMIOLOGY
Jarosz, M., & Taraszewska, A. (2014). Risk factors for gastroesophageal reflux disease: the role
of diet. Przeglad gastroenterologiczny, 9(5), 297.
Napier, K. J., Scheerer, M., & Misra, S. (2014). Esophageal cancer: A Review of epidemiology,
pathogenesis, staging workup and treatment modalities. World journal of gastrointestinal
oncology, 6(5), 112.
Oakes, J.M., 2017. Methods in social epidemiology, John Wiley & Sons, (Vol. 17), 150-158.
Richmond, R. C., Al-Amin, A., Smith, G. D., & Relton, C. L. (2014). Approaches for drawing
causal inferences from epidemiological birth cohorts: a review. Early human
development, 90(11), 769-780.
Terry, P., Lagergren, J., Wolk, A., & Nyrén, O. (2000). Reflux-Inducing Dietary Factors and
Risk ofAdenocarcinoma of the Esophagus and Gastric Cardia. Nutrition and
cancer, 38(2), 186-191.
Vandenbroucke, J. P., Broadbent, A., & Pearce, N. (2016). Causality and causal inference in
epidemiology: the need for a pluralistic approach. International journal of
epidemiology, 45(6), 1776-1786.
Vandenbroucke, J. P., von Elm, E., Altman, D. G., Gøtzsche, P. C., Mulrow, C. D., Pocock, S. J.,
... & STROBE Initiative. (2014). Strengthening the Reporting of Observational Studies in
Epidemiology (STROBE): explanation and elaboration. International journal of
surgery, 12(12), 1500-1524.

11EPIDEMIOLOGY
Vandenbroucke, J., & Pearce, N. (2015). Point: incident exposures, prevalent exposures, and
causal inference: does limiting studies to persons who are followed from first exposure
onward damage epidemiology?. American journal of epidemiology, 182(10), 826-833.
Wheeler, J. B., & Reed, C. E. (2012). Epidemiology of esophageal cancer. Surgical Clinics of
North America, 92(5), 1077-1087.
Zhang, Y. (2013). Epidemiology of esophageal cancer. World journal of gastroenterology:
WJG, 19(34), 5598.