Psychosomatic Medicine Tips for Better Browsing
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Published online before print February 8, 2007, 10.1097/PSY.0b013e318031391d
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sayuk, G. S.
Right arrow Articles by Clouse, R. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sayuk, G. S.
Right arrow Articles by Clouse, R. E.
Psychosomatic Medicine 69:173-181 (2007)
© 2007 American Psychosomatic Society

ORIGINAL ARTICLES

Predictors of Premature Antidepressant Discontinuation in Functional Gastrointestinal Disorders

Gregory S. Sayuk, MD, Jill E. Elwing, MD, Patrick J. Lustman, PhD and Ray E. Clouse, MD

From the Division of Gastroenterology (G.S.S., J.E.E., R.E.C.) and the Department of Psychiatry (P.J.L., R.E.C.), Washington University School of Medicine, and the Department of Veterans Affairs Medical Center (P.J.L.), Saint Louis, Missouri.

Address correspondence and reprint requests to Ray E. Clouse, Division of Gastroenterology, Washington University School of Medicine, 660 South Euclid Avenue, Campus Box 8124, St. Louis, MO 63110. E-mail: rclouse{at}im.wustl.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
Objectives: To identify factors responsible for premature antidepressant discontinuation that would assist in designing management strategies for patients with functional gastrointestinal disorders. Antidepressants are being used increasingly to manage patients with functional gastrointestinal disorders; poor patient adherence to treatment regimens, particularly in the period shortly after antidepressant initiation, is common and interferes with success.

Methods: Clinical records were reviewed from 172 outpatients who attended a university-based practice and who had been prescribed antidepressants to manage their functional gastrointestinal symptoms. Survival analysis methods were used to determine independent predictors of premature antidepressant discontinuation (within 6 months of initiation). Logistic regression analyses were used to see if the same predictors were responsible for side effects or poor treatment response.

Results: Premature antidepressant discontinuation occurred in 41 (23.8%) subjects. Somatization features (state or trait) and history of depression or an anxiety disorder were the most significant predictors of premature discontinuation (p ≤ .01 for each). Advancing age and female sex also were independent predictors (p < .05 for each). Somatization features and psychiatric illness were each linked to poor treatment response, whereas somatization features most consistently were associated with antidepressant side effects.

Conclusions: Failure to maintain treatment occurs in nearly a quarter of outpatients given antidepressants for functional gastrointestinal disorders. Somatization features and history of depression or anxiety most significantly interfered with treatment by predicting side effects, poor treatment response, and premature antidepressant discontinuation. Management algorithms should include specific strategies targeted at patients with these risk factors for poor treatment adherence.

Key Words: functional gastrointestinal disorders • somatization • antidepressants • depression

Abbreviations: IBS = irritable bowel syndrome; HR = hazard ratio; OR = odds ratio; CI = confidence interval.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
Antidepressants remain the most commonly recommended central neuromodulators for functional gastrointestinal disorders, although their mechanism of action in these settings is poorly understood (1,2). At least one in eight patients with irritable bowel syndrome (IBS), the most prevalent of the functional gastrointestinal disorders, is prescribed an antidepressant at some point in clinical management (3,4). Use of antidepressants is problematic, however, as adherence to treatment recommendations is poor and long treatment courses typically are required. Side effects interfering with treatment occur in 24% to 63% of patients in clinical practice (3,5,6).

Patient attrition during the initiation phase of antidepressants is well recognized, and cautious medication introduction with slow dose incrementation and patient counseling has been recommended (7,8). These efforts may be better directed if risk factors for premature discontinuation (i.e., within the first 6 months of antidepressant introduction) were defined. It is possible that the presence of psychiatric illness, specifically depression, or prominent features of somatization are two such factors, as these features have been associated with poor treatment response and side effect reporting, respectively, in previous controlled antidepressant treatment trials (6,9,10). Other features affecting adherence during antidepressant initiation have not been determined nor have any factors been established in clinical settings wherein attempts are made to maintain treatment with dosage modifications and drug changes.

The purpose of this study was to examine predictors of premature antidepressant discontinuation in a university-based clinical outpatient practice using survival analysis methods. We were particularly interested in whether somatization features were predictors of antidepressant discontinuation, were independent of psychiatric illness (depression or anxiety disorders) in their observed effects on discontinuation, and were operational when measured in either state (i.e., current somatic symptoms) or trait fashion (i.e., history of functional diagnoses). The findings from this study could be used to identify independent risk factors for poor patient adherence and fashion optimal treatment strategies for patients being started on antidepressants for functional gastrointestinal disorders.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
Subjects
All patients seen in an outpatient university-based adult gastroenterology clinic during a two-year period from January 1, 2003, to January 1, 2005, provided the pool from which subjects were identified. Subjects were considered for inclusion in the study if they met the following criteria: a) a functional gastrointestinal disorder had been established as the primary diagnosis; b) at least 3 months of clinical follow-up were available; and c) an antidepressant had been initiated specifically to manage the functional symptoms. Functional gastrointestinal disorders were diagnosed using the Rome II criteria-based system (11). Review of clinical records for the purposes of this study was approved by the Washington University Human Studies Committee.

Record Review
Using an instrument prepared a priori for this study, data were extracted from the outpatient records by members of the investigative team who had not participated in the medical care of the patients. All clinical information, including records from office visits, notations of telephone contact, and written communications, were considered. Clinical features including demographic characteristics (age, sex, race), the primary functional gastrointestinal disorder, and presence of comorbid psychiatric illness (chart diagnoses) were recorded. For the purposes of this study, only major depression and anxiety disorders, the most common psychiatric diagnoses encountered in patients with functional gastrointestinal disorders, were included as comorbid psychiatric illness (12,13). Psychiatric diagnoses were not formally established using conventional criteria (14). Rather, diagnoses that had been assigned by the gastroenterologist or had been carried in the medical history by patient report or previous physician contact were accepted without consideration for the validity of the diagnosis. In this gastroenterology clinic, diagnoses typically are assigned using the criteria described in the Diagnostic and Statistical Manual of Mental Disorders, Version IV (DSM-IV), but a structured interview is not utilized.

Two measures of somatization were used. Each subject completed a self-reported review-of-systems checklist at the initial visit. This checklist establishes the presence of 60 individual symptoms across ten organ systems, and the subjects were instructed to endorse only those symptoms that were being "currently experienced." The sum of responses to a subset of 13 symptoms was recorded as the measure of somatization state, this subset representing 13 of the 15 symptoms included in a previously validated measure of somatic symptom severity (15). A second measure assessed somatization trait features as the sum of all documented patient-reported or physician-determined gastrointestinal and nongastrointestinal functional syndromes recorded in the medical history (e.g., fibromyalgia, chronic fatigue syndrome) other than the primary diagnosis. Because of the narrow distribution of responses on this measure (range 0–5) and the low median value (2), the measure was dichotomized so that the subjects with a higher degree of somatization trait had >2 additional functional diagnoses.

Antidepressant treatment history was documented, including final antidepressant type, dose, treatment duration, and number of sequential treatment trials. Antidepressant global treatment response was graded by the chart reviewer applying a previously reported Likert scale (0 = no significant improvement or worse; 1 = slight improvement, requiring treatment changes; 2 = moderate improvement, regimen stable but symptoms not completely resolved; 3 = clinical remission and complete patient satisfaction with therapy) to the final antidepressant trial (16,17). This method of grading treatment response from chart records has high interobserver agreement ({kappa} exceeding 0.8) (18). Scores <2 were regarded as an unsatisfactory global treatment response for the purposes of determining potential explanations for antidepressant discontinuation. Presence and severity of side effects from the prescribed antidepressants also were rated from the chart notations using a second Likert scale (0 = no reported side effect; 1 = mild side effects but able to continue medication; 2 = severe or intolerable side effects requiring medication discontinuation). A score of 2 was required to explain antidepressant discontinuation for side effects.

Dates of and principal reason for antidepressant discontinuation (as reported by the patient) were determined from the chart review. Discontinuation of antidepressants within 6 months of initiation for functional gastrointestinal symptoms was regarded as premature, unless the medication was stopped because it was regarded by the physician as no longer needed for symptom control. The threshold of 6 months was arbitrarily defined but derived from clinical practice. In contrast to the remission-induction and maintenance approach taken for depression management, the treatment of patients with functional gastrointestinal disorders typically is initiated at low antidepressant dosage and incremented slowly until a response is achieved and sustained (1,19). Consequently, treatment courses of 6 to 24 months typically are recommended (2,20).

Statistical Methods
Measures of central tendency are reported as mean ± standard error of the mean for continuous variables and median values with ranges for categorical data. Grouped data were compared using Student's t tests for continuous data and {chi}2 and Fisher's exact tests (for samples ≤5 subjects within a given cell) for binomial data. For nonparametric data, significance of between-group differences was assessed using Mann-Whitney U tests. Individual covariate correlations were determined by the calculation of Pearson coefficients. Kaplan-Meier plots were constructed to compare time to discontinuation of antidepressant therapy in relation to clinical variables. Cox proportional hazards models were used to determine independent predictors of premature treatment discontinuation within the 6-month period after antidepressant initiation. Both demographic and clinical variables were used as covariates in these models, and Wald statistics were used to determine statistical significance of the predictors. The same covariates were used in logistic regression analyses for predicting either unsatisfactory treatment response or side effects as the potential explanation for attrition. Model likelihood ratio {chi}2 tests were applied to determine overall model significance, Hosmer and Lemeshow tests were conducted to determine model goodness of fit, and calculation of Nagelkerke pseudo R2 model summary statistics was performed to assess the amount of variance explained by the model. All analyses were conducted using SPSS version 14.0 (SPSS Inc., Chicago, IL), with a p < .05 as a determinant of statistical significance in two-tailed testing.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 NOTES
 REFERENCES
 
Demographic and Clinical Characteristics of the Subjects
Antidepressant treatment had been offered to 172 patients who met the inclusion criteria. These subjects had been followed for a mean of 12.8 ± 1.4 months after antidepressant initiation. A summary of their baseline demographic and clinical characteristics is provided in Table 1. IBS was the most common functional gastrointestinal disorder, being present in more than a quarter of the subjects. A psychiatric diagnosis (depression or anxiety disorder) had been made in 61 (35.5%) of the subjects.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Clinical Characteristics of the Subjects

 

Subjects with a higher degree of somatization by the trait measure had been given more than three times as many functional diagnoses (other than their primary functional gastrointestinal disorder) as the remainder (3.1 ± 0.1 versus 0.9 ± 0.04, p = .001). The most common types of additional functional diagnoses were chronic headache (27.3%), other functional gastrointestinal disorders (21.5%), and chronic back pain unattributed to structural disease (12.2%). Scores on the somatization state and trait measures were modestly but significantly correlated (r = 0.45, p < .001). Although psychiatric comorbidity portended a higher scale score on the somatization state measure (5.6 ± 0.3 versus 3.9 ± 0.2, p < .001), it was not strongly associated with a higher number of functional diagnoses on the somatization trait measure (2.2 ± 0.2 versus 1.9 ± 0.1, p = .11).

Antidepressant Treatment and Outcome
Tricyclic antidepressants were the antidepressants most commonly prescribed for managing functional gastrointestinal symptoms in this subject group and were used in 84.3% of subjects (Table 2). Forty-one (23.8%) of the 172 subjects discontinued antidepressants within 6 months of initiation primarily because of either intolerable side effects or self-reported inefficacy. Fifteen of these subjects stopped therapy for side effects alone; nine subjects discontinued antidepressants because of inefficacy; and the remaining 17 subjects cited multiple reasons (including side effects and/or inefficacy) as the explanation for discontinuation.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Summary of Antidepressant Treatment

 

In univariate analyses, no significant age, gender, or race differences were observed in those subjects who discontinued antidepressants prematurely compared with those who continued therapy (Table 1). The distribution of functional gastrointestinal diagnoses likewise was similar between groups. Comorbid psychiatric diagnoses were more prevalent in those who discontinued antidepressants prematurely, the difference being attributed to higher rates of both depression and anxiety disorders in this group (Table 1). Similarly, somatization features by either measure were more pronounced in the subset with premature antidepressant discontinuation. Subjects who discontinued antidepressants prematurely underwent a greater number of antidepressant trials despite a markedly shorter duration of antidepressant therapy and, for the subset receiving tricyclic antidepressants, did not reach the same median daily dose as those who continued therapy (Table 2). Tricyclic antidepressants typically are tried initially in this outpatient practice; patients who tolerate these agents poorly are switched to other antidepressants in subsequent trials. Thus, the proportion concluding with a tricyclic antidepressant is lower in the subset that discontinued treatment prematurely compared with those who continued antidepressants (Table 2).

Premature Antidepressant Discontinuation by Survival Analysis
Kaplan-Meier plots demonstrating the time to discontinuation of antidepressants within the 6-month period after initiation are shown for the entire group and by sex in Figure 1. The difference between sexes did not reach statistical significance (female hazard ratio 1.88, 95% Confidence Interval 0.90–3.94, p = .09). By 6 months, 27.5% of female subjects had stopped the medications versus 16.1% of the male subjects.


Figure 19
View larger version (16K):
[in this window]
[in a new window]

  		Figure 1. Kaplan-Meier plots showing time to discontinuation of antidepressant therapy. Premature discontinuation (within the first 6 months after initiation) trended toward being more common in female subjects (p = .08 compared with males).

 

Cox proportional hazards models were constructed using the five clinical variables listed in Table 1 as potential predictors of premature antidepressant discontinuation (age, sex, functional gastrointestinal diagnosis, psychiatric comorbidity, and degree of somatization). Because of the correlation between somatization state and trait measures, two models were constructed, each only using one of these measures. Functional gastrointestinal disorder was dichotomized as IBS versus other diagnoses. In the model using the somatization state measure, advancing age, female sex, presence of psychiatric comorbidity, and greater number of somatization symptoms each independently predicted shorter time to discontinuation, higher degree of somatization being the most significant predictor (Table 3). When the somatization trait measure was used in the model, presence of psychiatric comorbidity and higher degree of somatization were the independent predictors, advancing age and female sex each demonstrating a trend toward significance (Table 3). Including type of antidepressant (tricyclic antidepressant versus others) in the models as an additional independent variable in a post hoc analysis did not add or subtract significant predictors from the original models, and the type of antidepressant was not predictive of premature discontinuation.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Predictors of Premature Antidepressant Discontinuation Using Cox Proportional Hazard Models

 

The overall effects of somatization on this outcome are shown by the Kaplan-Meier plots in Figure 2. For purposes of illustration and not statistical analysis, the somatization state measure was dichotomized using numbers of symptoms endorsed to a higher degree (≥6 of the 13 symptoms, n = 55) or lower degree of somatization (<6 symptoms, n = 108). (Data were missing for nine subjects on this measure.) Continuation of antidepressants past 6 months occurred in 84.3% to 87.2% of subjects with lower somatization scale scores, whereas only 58.2% to 65.1% of those with higher degrees of somatization by either measure continued the medications.


Figure 29
View larger version (20K):
[in this window]
[in a new window]

  		Figure 2. Kaplan-Meier plots showing time to discontinuation of antidepressant therapy in relation to degree of somatization on the state and trait measures. The somatization state measure was dichotomized into higher and lower degree of somatization for illustrative purposes here and in Figure 3. A higher degree of somatization by either measure predicted premature antidepressant discontinuation.

 


Figure 39
View larger version (20K):
[in this window]
[in a new window]

  		Figure 3. Kaplan-Meier plots showing the additive effect of a higher degree of somatization (state measure for these plots) and the presence of psychiatric comorbidity on premature antidepressant discontinuation. The effect of somatization on discontinuation is most pronounced in the first 2 months after antidepressant initiation.

 

Additive effects of somatization and psychiatric comorbidity also are apparent from the review of Kaplan-Meier plots (Figure 3). Subjects with both a higher degree of somatization (state measure shown in Figure 3 and psychiatric comorbidity were particularly vulnerable to early antidepressant discontinuation, especially in the initial 2 months of therapy. These subjects had only a 58.6% antidepressant continuation rate at 6 months compared with an 89.0% continuation rate by those with neither psychiatric illness nor a higher degree of somatization. The effect of female sex also was most evident in subjects with other risk factors for premature discontinuation; the Kaplan-Meier plots of female and male subjects with lower somatization scale scores and no psychiatric comorbidity were almost identical (not shown). The clinical record was further reviewed in the 17 subjects with a higher degree of somatization and psychiatric comorbidity who continued antidepressants to determine the explanation for slow attrition after the first 2 months after initiation. Eight (47.0%) of these patients tolerated antidepressants well because of a very slow titration to the final dosage. An additional four (23.5%) patients were successfully managed with nontricyclic antidepressant agents. Four (23.5%) benefited from referral to a mental health professional for adjunct nonpharmacological therapy. Almost uniformly, the subjects required frequent telephone interaction (in cases, weekly) and/or office visits during the early period of antidepressant initiation.

Given that the arbitrary threshold used for defining premature antidepressant discontinuation is longer than the conventional threshold for treatment failure in depression management, the analyses were repeated using a cutoff of 3 months for premature discontinuation. With this cutoff, 30 (17.4%) subjects discontinued therapy prematurely. The Cox regression analyses were repeated using the same variables (Table 3). In each analysis, the same independently significant predictors were found as seen in the 6-month cutoff used for premature discontinuation.

Potential Explanations for Attrition
Potential mechanisms for treatment failure were examined by evaluating the treatment response and side effects resulting from antidepressant therapy. Of 172 subjects, 72 (41.9%) were rated as having an unsatisfactory treatment response of their functional gastrointestinal symptoms to antidepressants. The patient subset with an unsatisfactory treatment response underwent a significantly greater number of sequential antidepressant trials (1.7 ± 0.1 versus 1.3 ± 0.1, p = .001) and reached a lower median daily dose of tricyclic antidepressant (in subjects receiving these agents) compared with their responsive counterparts (30 mg versus 50 mg, p = .03). Unsatisfactory treatment response was far more common in those subjects with premature antidepressant discontinuation than in those who maintained therapy (92.7% versus 26.0%, p < .001).

As for the Cox proportional hazards models, logistic regression analyses for predicting antidepressant inefficacy or intolerance were constructed using each somatization measure separately (Table 4). A higher degree of somatization by either measure was a strong predictor of unsatisfactory treatment response (odds ratio (OR) 1.31 for each additional state symptom; OR 4.78 for a higher degree of somatization trait). The presence of psychiatric comorbidity also independently predicted unsatisfactory treatment response in each model. The other measured clinical features, including age and sex, were not independent predictors of this potential explanation for antidepressant discontinuation.


View this table:
[in this window]
[in a new window]

  		TABLE 4. Clinical Characteristics Predicting Unsatisfactory Treatment Response or Side Effects from Antidepressant Therapy

 

By definition, intolerable side effects responsible for discontinuation were only observed in the group with premature discontinuation (78.0% of this group). Logistic regression analyses similar to those used to predict unsatisfactory treatment response showed again that a higher degree of somatization by either measure predicted intolerable side effects (Table 4). Female sex was a significant independent predictor when using the somatization state measure in the model and trended toward being a significant independent predictor with the somatization trait measure; increasing age also showed a trend toward significance in each model. Psychiatric comorbidity was not an independent predictor of side effects. Post hoc logistic regression analyses that included type of antidepressant (tricyclic antidepressant versus others) as an independent variable failed to demonstrate that this variable had predictive value toward unsatisfactory treatment response or intolerable side effects.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
NOTES
 REFERENCES
 
In this report, we found that premature antidepressant discontinuation occurred in 23.8% of a group of patients with functional gastrointestinal disorders treated with antidepressants. Of the clinical characteristics examined, somatization, psychiatric comorbidity, and, to a lesser degree, advancing age and female sex predicted this occurrence. The strongest predictors were higher degrees of somatization, as measured by either recent somatization symptoms (somatization state) or relevant historical diagnoses (somatization trait), and history of depression or an anxiety disorder. Both predicted subjective inefficacy, whereas somatization and female sex predicted intolerable side effects. Advancing age also trended toward independently predicting side effect reporting. A post hoc analysis confirmed clinical impressions that attrition by patients at greatest risk possibly can be attenuated through regular patient contact, careful dose adjustments, use of antidepressants with lower side effect profiles, and implementation of nonpharmacological therapies.

Evidence derived from meta analyses and more recent clinical trials supports the use of antidepressants in functional gastrointestinal disorders (19,21,22), but, as illustrated in a large, placebo-controlled trial of desipramine in women with functional bowel disorders, early attrition in the active treatment arm can prevent demonstration of a significant drug effect (6). Little is known of factors leading to antidepressant discontinuation in the functional gastrointestinal disorders. One conspicuous factor would be lack of adequate symptom response, and a quarter of IBS patients in the practice setting have cited inefficacy as the basis for medication discontinuation (5). Given the substantial burden of psychiatric comorbidity (1,23–25), one might hypothesize intuitively that a lack antidepressant response would be predicted by the absence of comorbid psychiatric illness. This is not, however, supported by the existing literature (1,21). To the contrary, the previously cited desipramine study noted that it was the subgroup of female IBS subjects with active depression symptoms that had no discernable improvement with active drug (6). These observations align themselves with the findings of this report, wherein comorbid psychiatric illness (depression or anxiety disorder) had a significant independent negative effect on both treatment response and continuation. Postulated mechanisms by which psychiatric illnesses interfere with medication responsiveness include the presence of comparatively greater degrees of central dysfunction in patients with comorbid disorders yielding less apparent antidepressant benefit, as well as the potential direct influences of untreated affective disorders on sensory hypervigilance and autonomic arousal (26).

One survey of IBS patients found that close to three fourths of respondents reported discontinuing their medications because of side effects (5). More than half of the dropouts in the active treatment arm of the multicenter desipramine study did so as a consequence of intolerable medication side effects (6). The same authors reported in preliminary form that the presence of somatization features predicted the experience of side effects at week 2 of the study (9). These preliminary data are corroborated by our findings that higher degrees of somatization predict intolerable side effects experiences and lead to early antidepressant discontinuation. To our knowledge, no other studies have reported specific patient characteristics that predict antidepressant side effects in patients with functional gastrointestinal disorders. Side effects are common, however, when antidepressants are used in psychiatric samples, the reported rates ranging from 25% to 60% (3,5,7). Side effects represent a major reason cited for antidepressant discontinuation in depressed patients, accounting for 25% to 50% of nonadherence in populations whose premature discontinuation rates average 40% (27–30). These rates may be underestimations; as many as two thirds of depressed patients stopped medications without informing their physicians in one study (31) and, in another, a quarter of the patients reported medication adherence to their physicians although no prescriptions were being filled (32). Similar to our observations in patients with functional gastrointestinal disorders, depressed patients with higher degrees of somatization are more resistant to treatment and experience greater numbers of side effects than their low somatization counterparts (33–35).

With the recognition that somatization disorder, a pronounced form of somatization, is present in more than a third of female IBS patients attending university-based clinics (36,37), somatization is gaining attention both as a risk for developing functional disorders and as an important factor influencing presentation and outcomes (12). Patient self-reporting of multiple gastrointestinal and nongastrointestinal symptoms has modest accuracy in predicting a functional over a structural diagnosis (38). Increased likelihood of postinfectious IBS and greater rectal sensitivity to balloon distention have been observed in subjects with high somatization scores, suggesting a mechanistic linkage of somatization to the pathophysiology behind functional gastrointestinal disorders (39,40). Additionally, a high degree of functional comorbidity is associated with reduced quality of life and greater healthcare seeking in IBS patients (41). In our study, somatization measures emerged as robust independent predictors of side effects, perceived medication inefficacy, and antidepressant discontinuation. Thus, somatization has relevance not only to the pathophysiology underlying functional gastrointestinal disorders but also to the treatment approach. Importance of somatization presentations that fall short of the criteria required for the DSM-IV somatoform disorders is being increasingly recognized (42).

Sex differences have been appreciated both clinically and physiologically in patients with functional gastrointestinal disorders. Greater degrees of depression, anxiety, and somatic complaints are detected in female IBS patients (43). Women with IBS also possess a different pattern of symptom reporting, having generally higher levels of gastrointestinal and nongastrointestinal symptoms compared with men with IBS (44). From a physiological perspective, female patients with IBS exhibit greater rectal hyperalgesia and demonstrate differences in autonomic tone and regional cerebral activation in response to visceral stimulation (45–48). Recent evidence has indicated sex differences in therapeutic responses, with female patients with IBS responding more favorably to alosetron than male subjects with IBS (49). In our study, the female sex was associated with a greater odds of antidepressant discontinuation and experiences of side effects—observations that were independent of psychiatric comorbidity and the degree of somatization, although the sex effect was seen predominantly in patients with a background of these other factors. Whether our findings relate to differences in symptom expression or whether they are a consequence of physiological characteristics unique to women warrants further exploration.

In summary, this study establishes that somatization, a history of depression or anxiety disorder, and, to a lesser degree, advancing age and female sex are relevant to early discontinuation of antidepressants in patients with functional gastrointestinal disorders. These factors affect the subjective experiences of side effects and perceived medication efficacy. An overlap of mood disorder with higher reporting of somatization symptoms was observed in our study population, consistent with prior medical literature (50,51). Although it is conceivable that either may be the primary underlying pathology (somatization producing certain symptomatic features on which mood disorder diagnoses were based or, alternatively, mood disorder leading to physical symptom reporting), the multivariate analyses supported the independent contribution of each to the outcomes. The relationship of somatization and mood disorders may be greater than we detected, as the method of diagnosing psychiatric comorbidity was suboptimal. In practice, functional somatic symptoms often obscure recognition of underlying depression because the clinician expects depressed patients to present with affective complaints (52–58). Limitations to our conclusions relate to the use of medical record review for the determination of psychiatric and functional diagnoses as well as the use of a subjective grading of treatment response. However, the rate of psychiatric illness was very similar to a 1-year rate established from the same practice using a structured diagnostic interview (36). The somatization trait measure correlated well with the findings from a self-report measure of somatization symptoms, and some data corroborate the reproducibility of the outcome rating method (59). Thus, these findings should form a sound basis for incorporating the discovered risk factors into management algorithms in clinical practice. They also should be considered in the design of clinical trials wherein early attrition is deleterious to understanding treatment effects.

In practice, particular attention to education, support, and dosage titration should be given to those individuals at high risk for attrition. An established body of literature suggests that patient education about potential antidepressant side-effect experiences (60–63) and clear communication regarding expected duration of therapy (64,65) may improve medication compliance when implemented in the treatment of mood disorders. Moreover, frequent healthcare provider visits or contacts (28,66,67) and adjunctive behavioral therapies (68) have been shown to be beneficial in optimizing pharmacotherapy adherence in patients with mood disorders. A multidimensional approach incorporating the use of several of these modalities, in addition to pharmacotherapy, was recently shown to be effective in improving the physical complaints and mental well-being of primary care patients with multiple medially unexplained symptoms (69). Theoretically, antidepressants with the lowest side-effect profiles and nonpharmacological interventions also may play important roles in patient subgroups with functional gastrointestinal disorder at greatest risk of premature medication discontinuation. Nonpharmacological treatments (behavioral/psychological interventions), however, will likely prove to be of greatest importance in this population (63), as use of newer antidepressants with seemingly cleaner side-effect profiles in psychiatric samples disappointingly has not translated into substantially improved medication adherence patterns (70,71).


   NOTES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
REFERENCES
 
Received for publication May 2, 2006; revision received November 6, 2006.

Supported in part by National Institutes of Health Grants DK 59364 and DK 63202 (P.J.L.) from the United States Public Health Service and a grant from the Sidney R. Baer, Jr., Foundation (P.J.L., R.E.C.).

DOI:10.1097/PSY.0b013e318031391d


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 

  1. Clouse RE. Antidepressants for functional gastrointestinal syndromes. Dig Dis Sci 1994;39:2352–63.[CrossRef][Medline]
  2. Mertz HR. Irritable bowel syndrome. N Engl J Med 2003;349:2136–46.[Free Full Text]
  3. Whitehead WE, Levy RL, Von Korff M, Feld AD, Palsson OS, Turner M, Drossman DA. The usual medical care for irritable bowel syndrome. Aliment Pharmacol Ther 2004;20:1305–15.[CrossRef][Medline]
  4. Talley NJ. Antidepressants in IBS: are we deluding ourselves? Am J Gastroenterol 2004;99:921–3.[CrossRef][Medline]
  5. Lembo A. Irritable bowel syndrome medications side effects survey. J Clin Gastroenterol 2004;38:776–81.[CrossRef][Medline]
  6. Drossman DA, Toner BB, Whitehead WE, Diamant NE, Dalton CB, Duncan S, Emmott S, Proffitt V, Akman D, Frusciante K, Le T, Meyer K, Bradshaw B, Mikula K, Morris CB, Blackman CJ, Hu Y, Jia H, Li JZ, Koch GG, Bangdiwala SI. Cognitive-behavioral therapy versus education and desipramine versus placebo for moderate to severe functional bowel disorders. Gastroenterology 2003;125:19–31.[CrossRef][Medline]
  7. Clouse RE, Lustman PJ, Geisman RA, Alpers DH. Antidepressant therapy in 138 patients with irritable bowel syndrome: a five-year clinical experience. Aliment Pharmacol Ther 1994;8:409–16.[Medline]
  8. Clouse RE. Antidepressants for irritable bowel syndrome. Gut 2003;52:598–9.[Free Full Text]
  9. Thiwan SM, Dalton CB, Morris CB, Diamant NE, Toner BB, Whitehead WE, Hu Y, Bangdiwala SI, Drossman D. Factors predicting symptoms of "side effects" when using tricyclic antidepressants. Gastroenterology 2005;128:A66.[CrossRef]
  10. Lancaster-Smith MJ, Prout BJ, Pinto T, Anderson JA, Schiff AA. Influence of drug treatment on the irritable bowel syndrome and its interaction with psychoneurotic morbidity. Acta Psychiatr Scand 1982;66:33–41.[Medline]
  11. Drossman D TN, Thompson W, Whitehead W, Corazzari E, eds. Rome II: functional gastrointestinal disorders. McLean, VA: Degnon Assoc., Inc.; 2000.
  12. Whitehead WE, Palsson O, Jones KR. Systematic review of the comorbidity of irritable bowel syndrome with other disorders: what are the causes and implications? Gastroenterology 2002;122:1140–56.[Medline]
  13. Lydiard RB, Fossey MD, Marsh W, Ballenger JC. Prevalence of psychiatric disorders in patients with irritable bowel syndrome. Psychosomatics 1993;34:229–34.[Abstract/Free Full Text]
  14. Assoc AP. Diagnostic and statistical manual of mental disorders. Washington, DC: APA; 1994.
  15. Kroenke K, Spitzer RL, Williams JB. The PHQ-15: validity of a new measure for evaluating the severity of somatic symptoms. Psychosom Med 2002;64:258–66.[Abstract/Free Full Text]
  16. Prakash C, Clouse RE. Long-term outcome from tricyclic antidepressant treatment of functional chest pain. Dig Dis Sci 1999;44:2373–9.[CrossRef][Medline]
  17. Prakash C, Lustman PJ, Freedland KE, Clouse RE. Tricyclic antidepressants for functional nausea and vomiting: clinical outcome in 37 patients. Dig Dis Sci 1998;43:1951–6.[CrossRef][Medline]
  18. Sawhney MS, Prakash C, Lustman PJ, Clouse RE. Tricyclic antidepressants for chronic vomiting in diabetic patients. Dig Dis Sci. 2006 Dec 30; Epub ahead of print.
  19. Lesbros-Pantoflickova D, Michetti P, Fried M, Beglinger C, Blum AL. Meta-analysis: the treatment of irritable bowel syndrome. Aliment Pharmacol Ther 2004;20:1253–69.[CrossRef][Medline]
  20. Clouse RE, Lustman PJ. Use of psychopharmacological agents for functional gastrointestinal disorders. Gut 2005;54:1332–41.[Free Full Text]
  21. Jackson JL, O'Malley PG, Tomkins G, Balden E, Santoro J, Kroenke K. Treatment of functional gastrointestinal disorders with antidepressant medications: a meta-analysis. Am J Med 2000;108:65–72.[CrossRef][Medline]
  22. Clouse RE, Prakash C, Anderson RJ. Antidepressants for functional gastrointestinal symptoms and syndromes: a meta-analysis. Gastroenterology 2001;120:A642.
  23. Woodman CL, Breen K, Noyes R Jr, Moss C, Fagerholm R, Yagla SJ, Summers R. The relationship between irritable bowel syndrome and psychiatric illness. A family study. Psychosomatics 1998;39:45–54.[Abstract/Free Full Text]
  24. Walker EA, Gelfand AN, Gelfand MD, Katon WJ. Psychiatric diagnoses, sexual and physical victimization, and disability in patients with irritable bowel syndrome or inflammatory bowel disease. Psychol Med 1995;25:1259–67.[Medline]
  25. Walker EA, Roy-Byrne PP, Katon WJ, Li L, Amos D, Jiranek G. Psychiatric illness and irritable bowel syndrome: a comparison with inflammatory bowel disease. Am J Psychiatry 1990;147:1656–61.[Abstract/Free Full Text]
  26. Clouse RE. Managing functional bowel disorders from the top down: lessons from a well-designed treatment trial. Gastroenterology 2003;125:249–53.[CrossRef]
  27. Demyttenaere K, Enzlin P, Dewe W, Boulanger B, De Bie J, De Troyer W, Mesters P. Compliance with antidepressants in a primary care setting, 1: Beyond lack of efficacy and adverse events. J Clin Psychiatry 2001;62(Suppl 22):30–3.[Medline]
  28. Bull SA, Hu XH, Hunkeler EM, Lee JY, Ming EE, Markson LE, Fireman B. Discontinuation of use and switching of antidepressants: influence of patient-physician communication. Jama 2002;288:1403–9.[Abstract/Free Full Text]
  29. Al-Chaer E, Kawasaki M, Paricha P. A new model of chronic visceral hypersensitivity in adult rats induced by colon irritation during postnatal development. Gastroenterology 2000;119:1276–85.[CrossRef][Medline]
  30. Lingam R, Scott J. Treatment non-adherence in affective disorders. Acta Psychiatr Scand 2002;105:164–72.[CrossRef][Medline]
  31. Maddox JC, Levi M, Thompson C. The compliance with antidepressants in general practice. J Psychopharmacol 1994;8:48–53.[Abstract/Free Full Text]
  32. Kobak KA, Taylor L, Katzelnick DJ, Olson N, Clagnaz P, Henk HJ. Antidepressant medication management and health plan employer data information set (HEDIS) criteria: reasons for nonadherence. J Clin Psychiatry 2002;63:727–32.[Medline]
  33. Papakostas GI, Petersen T, Denninger J, Sonawalla SB, Mahal Y, Alpert JE, Nierenberg AA, Fava M. Somatic symptoms in treatment-resistant depression. Psychiatry Res 2003;118:39–45.[CrossRef][Medline]
  34. Agosti V, Quitkin FM, Stewart JW, McGrath PJ. Somatization as a predictor of medication discontinuation due to adverse events. Int Clin Psychopharmacol 2002;17:311–4.[CrossRef][Medline]
  35. Keeley R, Smith M, Miller J. Somatoform symptoms and treatment nonadherence in depressed family medicine outpatients. Arch Fam Med 2000;9:46–54.[Abstract/Free Full Text]
  36. North CS, Downs D, Clouse RE, Alrakawi A, Dokucu ME, Cox J, Spitznagel EL, Alpers DH. The presentation of irritable bowel syndrome in the context of somatization disorder. Clin Gastroenterol Hepatol 2004;2:787–95.[CrossRef][Medline]
  37. Miller AR, North CS, Clouse RE, Wetzel RD, Spitznagel EL, Alpers DH. The association of irritable bowel syndrome and somatization disorder. Ann Clin Psychiatry 2001;13:25–30.[CrossRef][Medline]
  38. Brown WH, Chey WD, Elta GH. Number of responses on a review of systems questionnaire predicts the diagnosis of functional gastrointestinal disorders. J Clin Gastroenterol 2003;36:222–7.[CrossRef][Medline]
  39. Gwee KA, Leong YL, Graham C, McKendrick MW, Collins SM, Walters SJ, Underwood JE, Read NW. The role of psychological and biological factors in postinfective gut dysfunction. Gut 1999;44:400–6.[Abstract/Free Full Text]
  40. Whitehead WE, Palsson OS. Is rectal pain sensitivity a biological marker for irritable bowel syndrome: psychological influences on pain perception. Gastroenterology 1998;115:1263–71.[CrossRef][Medline]
  41. Vandvik PO, Wilhelmsen I, Ihlebaek C, Farup PG. Comorbidity of irritable bowel syndrome in general practice: a striking feature with clinical implications. Aliment Pharmacol Ther 2004;20:1195–203.[CrossRef][Medline]
  42. Smith RC, Gardiner JC, Lyles JS, Sirbu C, Dwamena FC, Hodges A, Collins C, Lein C, Given CW, Given B, Goddeeris J. Exploration of DSM-IV criteria in primary care patients with medically unexplained symptoms. Psychosom Med 2005;67:123–9.[Abstract/Free Full Text]
  43. Blanchard EB, Keefer L, Galovski TE, Taylor AE, Turner SM. Gender differences in psychological distress among patients with irritable bowel syndrome. J Psychosom Res 2001;50:271–5.[CrossRef][Medline]
  44. Lee OY, Mayer EA, Schmulson M, Chang L, Naliboff B. Gender-related differences in IBS symptoms. Am J Gastroenterol 2001;96:2184–93.[CrossRef][Medline]
  45. Munakata J, Naliboff B, Harraf F, Kodner A, Lembo T, Chang L, Silverman DH, Mayer EA. Repetitive sigmoid stimulation induces rectal hyperalgesia in patients with irritable bowel syndrome. Gastroenterology 1997;112:55–63.[CrossRef][Medline]
  46. Naliboff BD, Berman S, Chang L, Derbyshire SW, Suyenobu B, Vogt BA, Mandelkern M, Mayer EA. Sex-related differences in IBS patients: central processing of visceral stimuli. Gastroenterology 2003;124:1738–47.[CrossRef][Medline]
  47. Berman S, Munakata J, Naliboff BD, Chang L, Mandelkern M, Silverman D, Kovalik E, Mayer EA. Gender differences in regional brain response to visceral pressure in IBS patients. Eur J Pain 2000;4:157–72.[CrossRef][Medline]
  48. Tillisch K, Mayer EA, Labus JS, Stains J, Chang L, Naliboff BD. Sex specific alterations in autonomic function among patients with irritable bowel syndrome. Gut 2005;54:1396–401.[Abstract/Free Full Text]
  49. Camilleri M, Mayer EA, Drossman DA, Heath A, Dukes GE, McSorley D, Kong S, Mangel AW, Northcutt AR. Improvement in pain and bowel function in female irritable bowel patients with alosetron, a 5-HT3 receptor antagonist. Aliment Pharmacol Ther 1999;13:1149–59.[CrossRef][Medline]
  50. Brown FW, Golding JM, Smith GR Jr. Psychiatric comorbidity in primary care somatization disorder. Psychosom Med 1990;52:445–51.[Abstract/Free Full Text]
  51. Simon GE, VonKorff M, Piccinelli M, Fullerton C, Ormel J. An international study of the relation between somatic symptoms and depression. N Engl J Med 1999;341:1329–35.[Abstract/Free Full Text]
  52. Kroenke K, Spitzer RL, Williams JB, Linzer M, Hahn SR, deGruy FV, 3rd, Brody D. Physical symptoms in primary care. Predictors of psychiatric disorders and functional impairment. Arch Fam Med 1994;3:774–9.[Abstract/Free Full Text]
  53. Kroenke K, Jackson JL, Chamberlin J. Depressive and anxiety disorders in patients presenting with physical complaints: clinical predictors and outcome. Am J Med 1997;103:339–47.[CrossRef][Medline]
  54. Kroenke K, Rosmalen JG. Symptoms, syndromes, and the value of psychiatric diagnostics in patients who have functional somatic disorders. Med Clin North Am 2006;90:603–26.[CrossRef][Medline]
  55. Katon W, Lin E, Von Korff M, Russo J, Lipscomb P, Bush T. Somatization: a spectrum of severity. Am J Psychiatry 1991;148:34–40.[Abstract/Free Full Text]
  56. Katon W, Kleinman A, Rosen G. Depression and somatization: a review. Part I. Am J Med 1982;72:127–35.[CrossRef][Medline]
  57. Bridges KW, Goldberg DP. Somatic presentation of DSM III psychiatric disorders in primary care. J Psychosom Res 1985;29:563–9.[CrossRef][Medline]
  58. Kirmayer LJ, Robbins JM, Dworkind M, Yaffe MJ. Somatization and the recognition of depression and anxiety in primary care. Am J Psychiatry 1993;150:734–41.[Abstract/Free Full Text]
  59. Sawhney M, Prakash C, Lustman PJ, Clouse RE. Tricyclic antidepressants for persistent or recurrent vomiting in diabetic patients. Gastroenterology 2001;120:A243.
  60. Seltzer A, Roncari I, Garfinkel P. Effect of patient education on medication compliance. Can J Psychiatry 1980;25:638–45.[Medline]
  61. Lin EH, Von Korff M, Ludman EJ, Rutter C, Bush TM, Simon GE, Unutzer J, Walker E, Katon WJ. Enhancing adherence to prevent depression relapse in primary care. Gen Hosp Psychiatry 2003;25:303–10.[CrossRef][Medline]
  62. Hoffman L, Enders J, Luo J, Segal R, Pippins J, Kimberlin C. Impact of an antidepressant management program on medication adherence. Am J Manag Care 2003;9:70–80.[Medline]
  63. Peveler R, George C, Kinmonth AL, Campbell M, Thompson C. Effect of antidepressant drug counselling and information leaflets on adherence to drug treatment in primary care: randomised controlled trial. BMJ 1999;319:612–5.[Abstract/Free Full Text]
  64. Lin EH, Von Korff M, Katon W, Bush T, Simon GE, Walker E, Robinson P. The role of the primary care physician in patients' adherence to antidepressant therapy. Med Care 1995;33:67–74.[Medline]
  65. Bull SA, Hunkeler EM, Lee JY, Rowland CR, Williamson TE, Schwab JR, Hurt SW. Discontinuing or switching selective serotonin-reuptake inhibitors. Ann Pharmacother 2002;36:578–84.[Abstract]
  66. Katon W, Von Korff M, Lin E, Simon G, Walker E, Unutzer J, Bush T, Russo J, Ludman E. Stepped collaborative care for primary care patients with persistent symptoms of depression: a randomized trial. Arch Gen Psychiatry 1999;56:1109–15.[Abstract/Free Full Text]
  67. Katon W, Von Korff M, Lin E, Walker E, Simon GE, Bush T, Robinson P, Russo J. Collaborative management to achieve treatment guidelines. Impact on depression in primary care. JAMA 1995;273:1026–31.[Abstract/Free Full Text]
  68. Paykel ES. Cognitive therapy in relapse prevention in depression. Int J Neuropsychopharmacol 2006;1–6.
  69. Smith RC, Lyles JS, Gardiner JC, Sirbu C, Hodges A, Collins C, Dwamena FC, Lein C, William Given C, Given B, Goddeeris J. Primary care clinicians treat patients with medically unexplained symptoms: a randomized controlled trial. J Gen Intern Med 2006;21:671–7.[CrossRef][Medline]
  70. Demyttenaere K, Haddad P. Compliance with antidepressant therapy and antidepressant discontinuation symptoms. Acta Psychiatr Scand 2000;403(Suppl):50–6.
  71. Hotopf M, Hardy R, Lewis G. Discontinuation rates of SSRIs and tricyclic antidepressants: a meta-analysis and investigation of heterogeneity. Br J Psychiatry 1997;170:120–7.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sayuk, G. S.
Right arrow Articles by Clouse, R. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sayuk, G. S.
Right arrow Articles by Clouse, R. E.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS