| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Department of Epidemiology (R.D.G.), Mailman School of Public Health, Columbia University, New York, New York; and the Department of Psychiatry (M.B.S.), University of California–San Diego, La Jolla, California.
Address reprint requests to: Renee D. Goodwin, PhD, 1051 Riverside Dr., Unit 43, New York, NY 10032. Email: rdg66{at}columbia.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: Data were drawn from the National Comorbidity Survey, a representative household survey of the adult population of the United States (N = 8098). Multivariate logistic regression analyses were used to determine the relationship between GAD and self-reported ulcer, controlling for differences in sociodemographic characteristics and psychiatric and medical comorbidity.
RESULTS: GAD was associated with a significantly increased risk of self-reported PUD (odds ratio = 2.8, 95% confidence interval = 1.4–5.7; p = .0002) after adjusting for differences in sociodemographic characteristics, comorbid mental disorders, and physical morbidity. Further analyses revealed a dose-response relationship between number of GAD symptoms (odds ratio = 1.2, 95% confidence interval = 1.1–1.4; p = .001) and increased risk of self-reported PUD.
CONCLUSIONS: These findings are consistent with and extend previous clinical and epidemiologic data, providing evidence of a dose-response relationship between GAD and self-reported PUD among adults in the general population. The mechanism of this association remains unknown. Future work investigating the relationship between onset of GAD and development of PUD in prospective, longitudinal, epidemiologic data with objective measures of physical health status and mental health may be useful in improving our understanding of this link.
Key Words: generalized anxiety disorder, • anxiety, • ulcer, • Helicobacter pylori, • epidemiology, • comorbidity.
Abbreviations: GAD = generalized anxiety disorder;; PUD = peptic ulcer disease.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Preclinical studies have also suggested an association between chronic stress and anxiety-like states and increased risk of gastric ulcers (7, 8). Research using animal models shows that chronic stress and anxiety are associated with increased frequency of gastric ulcerations in mice (8, 9). Specifically, data from animal models suggest that stress-related gastric ulceration is associated with a dopaminergic-sensitivity factor in mice, which may have relevance to mental disorders in humans (9). These data may offer an alternative approach to conceptualizing the association between chronic anxiety and PUD, where the co-occurrence of the two reflects a shared underlying vulnerability (genetic or environmental) rather than a cause-and-effect relationship.
Over the last several years the focus of research on the causes and treatments for PUD has largely abandoned the psychosomatic/psychological arena and instead focused on immunological/inflammatory pathways and genetic approaches (10, 11). The recent identification of Helicobacter pylori(H. pylori) as an infectious cause of PUD has been considered by many to disprove the possibility that psychiatric disorder plays a role in the pathogenesis of PUD (12). There is strong clinical evidence documenting an association between H. pyloriand PUD: H. pyloriis present in the large majority of patients with PUD; eradication of H. pylori results in resolution of the gastritis; and injection of H. pylori into animals causes gastric ruptures (9). A causal relationship between H. pylori and PUD, however, is difficult to establish from the available data, in part because of the lack of an animal model. In addition, although nearly all PUD patients have H. pylori, only a small percentage of individuals with H. pylori develop ulceration. The available data on the relationship between H. pylori and PUD show that unidentified host characteristics, strain variability, and other factors clearly play a role in the pathogenesis of peptic ulcer disease as well (12). Moreover, recent laboratory-based studies suggest that another possible pathway for this association could be related to the hypothesis that stress, which is associated with generalized anxiety disorder (GAD) and PUD, is immunomodulatory and may cause a shift in inflammatory reactions to infections with organisms normally controlled by TH1 lymphocytes/cytokines, such as H. pylori (13, 14). To date the determinants of PUD and the possible role of anxiety or other mental disorders in the risk of PUD remain unknown.
Previous studies have consistently described a link between severe, persistent anxiety and increased risk of PUD in clinical populations. Two main methodological constraints of previous studies, however, have limited our understanding of how anxiety and PUD may be related. First, previous studies have not investigated the specificity of the relationship between anxiety and PUD. Therefore, it is not possible to determine whether ulcer is associated with any mental disorder or if the ulcer is more strongly related to particular chronic anxiety states as previous data suggest. Second, previous studies have not examined the nature of the relationship between PUD and anxiety disorders in a community-based sample, making it unclear whether clinical findings result from an exposure-disease relationship or from selection into treatment bias.
The objective of the study reported here was to determine the relationship between generalized anxiety disorder and odds of PUD by overcoming these previous methodological limitations. First, this study examined the relationship between generalized anxiety disorder, a chronic anxiety disorder with a high rate of somatic symptoms, and self-reported PUD among adults in the general population. Second, the study determined the specificity and strength of the relationship between GAD and self-reported PUD. Third, the study determined the dose-response relationship between GAD and self-reported PUD. Based on previous clinical findings (2–5), we hypothesized that GAD, as a representation of chronic anxiety, is associated with a significantly increased odds of self-reported PUD among adults in the community.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Diagnostic Assessment
Psychiatric diagnoses were generated from a modified version of the World Health Organization (WHO) Composite International Diagnostic Interview (17), a structured interview designed for use by trained interviewers who are not clinicians. WHO field trials (18) and National Comorbidity Survey clinical reappraisal studies (19–21) documented acceptable reliability and validity of all these diagnoses. Psychiatric disorders examined include current (12-month) major axis I DSM-III-R mental disorders (alcohol dependence, substance dependence, major depression, dysthymia, panic attack, panic disorder, GAD, agoraphobia, social phobia, specific phobia, bipolar disorder, posttraumatic stress disorder, and nonaffective psychosis). Information on physical illness was obtained through self-report with the question "Have you experienced any of these health problems in the past 12 months?" with "ulcer" included as 1 of the 15 possible disorders. Written informed consent was obtained from each participant after the survey had been fully explained.
Analytic Strategy
First, differences in sociodemographic characteristics were compared between individuals with and without self-reported PUD using F-based tests of independence. Second, the rates of each mental disorder (12-month) were compared between those with and without PUD using the same method. Third, separate multiple logistic regression analyses were used to determine the relationship between GAD and the odds of PUD, adjusting for differences in demographic characteristics and all other mental and physical comorbidity. Adjusted odds ratios (ORs) with 95% CIs were computed to describe the association between GAD and the odds of PUD. Fourth, multivariate logistic regression analyses were used to compute ORs (with 95% CIs) describing the relationship between number of GAD symptoms from the list of 23 possible somatic GAD symptoms (eg, restless; bothered by tense, sore, or aching muscles; keyed up or on edge; particularly irritable) and the likelihood of PUD.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
PUD and Mental Disorders
Each mental disorder, with the exception of dysthymia, was associated with a significantly increased odds of PUD after adjusting for differences in sociodemographic characteristics (Table 1). GAD was associated with an almost five-fold increase in likelihood of PUD compared with that among patients without GAD. Also, bipolar disorder was associated with an approximately eight-fold increase in odds, and those with agoraphobia reported rates more than four times higher than that among those without agoraphobia.
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The mechanism of association between GAD and PUD remains unknown. It may be that GAD leads to the development of peptic ulceration. This hypothesis is consistent with the "stress" model of gastrointestinal ulceration, using GAD as a marker for chronic stress (9). Moreover, there are some data to suggest that GAD has a relatively early onset (23), suggesting that GAD may be a long-standing stressor that precedes the onset of PUD, though we had no data on age of onset of PUD to either confirm or disprove this hypothesis in this investigation. It is also conceivable that having PUD leads to an anxiety disorder. Alternatively, it is possible that a third factor (either environmental or genetic) is associated with increased vulnerability to the co-occurrence of GAD and ulcer. Finally, the possibility that individuals with GAD simply overreport physical symptoms is another possible explanation, though it does not explain the specificity of these findings.
The specificity of the association between GAD and PUD, compared with hypertension and diabetes, is of interest especially with regard to the potential mechanism of this association. PUD is frequently associated with painful physical symptoms and emotional distress; therefore, it has been suggested that treatment focus on psychosocial factors and physical symptoms rather than on an ongoing search for structural disease (24) without regard for other factors. GAD is associated with increased reporting of physical symptoms and is frequently characterized by physiologic anxiety symptoms, overreporting of medical problems, and associated with psychosocial distress. It is conceivable that GAD be treated when it co-occurs with PUD as it may play a role in the etiology of PUD, as it is currently understood, and especially in the absence of structural evidence of disease. Future studies that examine whether and to what degree treatment of GAD influences the onset, persistence, and outcome associated with PUD may reveal important evidence about the mechanism of this observed link as well as about whether treatment of mental disorder can be effective in the prevention of physical illness.
There are several significant limitations of this study that should be considered carefully when interpreting these results. First, data on PUD and other medical conditions were obtained from self-reports. It is therefore possible that respondents with chronic anxiety were reporting their experience of gastrointestinal complaints rather than diagnosable PUD per se. Yet, self-report is the most commonly used method for obtaining data on physical health in epidemiologic research (25), and the prevalence found in this study is consistent with previous reports (22). Second, mental disorder diagnoses were obtained with a structured lay-administered interview rather than a clinician assessment. There is considerable evidence, however, documenting the validity and reliability of this instrument in the community (18–21). Third, these data are based on retrospective reports, which are subject to memory bias. Finally, there was no available information on the temporal relationship between GAD and PUD, which would be more informative in furthering current understanding of this link.
The identification of some degree of specificity in the relationship between mental disorders and physical illnesses at the population level, which is consistent with and reflects a good fit with animal models (9), can help to improve our understanding of the interrelationship between different physical and psychiatric systems. In light of recent data suggesting that PUD is caused by exposure to an infectious agent (H. pylori) (1, 11, 12), these data add an interesting perspective. If indeed H. pylori were the sole cause of PUD, it seems unlikely that this sort of association would emerge in the data. Considering that the consensus on the research to date is that the association between H. pylori and PUD is an association (1) and not causation and that there are other factors that contribute to the risk of PUD, these data may reveal an individual factor associated with increased susceptibility to PUD in the population. If these findings are replicated, it would also be important to consider the possible link between treatment of GAD and the response and outcomes of medical treatment for PUD as well as outcomes associated with other medical morbidity.
From a clinical perspective, increased awareness of the likelihood that patients seeking help for PUD may be at increased risk for concurrent GAD may improve rates of identification and treatment of this common yet frequently unrecognized anxiety disorder (26), especially in general medical settings, where it is most likely to be seen (23). This information is especially relevant in light of recent clinical data suggesting that antidepressant treatment is associated with improvement among patients with other gastrointestinal problems (eg, irritable bowel syndrome) (27) as well as previous clinical data showing that high levels of anxiety are associated with poorer gastric ulcer healing (28). Also, psychotropic medication with proven efficacy in treating GAD (eg, antidepressants) may have potential for effective use with patients with PUD, perhaps in combination with medications that eradicate H. pylori. Future studies that use prospective, longitudinal, epidemiologic data to determine the relationship between GAD and the risk of peptic ulcer disease, with data on objective measures of physical health and information on the ages of onset of both GAD and PUD, should enhance our understanding of the specific relationship revealed in the current study.
Received for publication September 18, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. D. Goodwin, K. M. Keyes, M. B. Stein, and N. J. Talley Peptic Ulcer and Mental Disorders Among Adults in the Community: The Role of Nicotine and Alcohol Use Disorders Psychosom Med, May 1, 2009; 71(4): 463 - 468. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Sareen, F. Jacobi, B. J. Cox, S.-L. Belik, I. Clara, and M. B. Stein Disability and poor quality of life associated with comorbid anxiety disorders and physical conditions. Arch Intern Med, October 23, 2006; 166(19): 2109 - 2116. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. Hoge, J. E. Oppenheimer, and N. M. Simon Generalized Anxiety Disorder Focus, July 1, 2004; 2(3): 346 - 359. [Full Text] [PDF]
|
|
|
|
 |
|
 J. Arehart-Treichel It's Back to the Future For Ulcers and Anxiety Psychiatr News, February 7, 2003; 38(3): 21 - 21. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
