Psychosomatic Medicine Faster Service from Outside North America
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
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
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Porcelli, P.
Right arrow Articles by Todarello, O.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Porcelli, P.
Right arrow Articles by Todarello, O.
Related Collections
Right arrow Gastrointestinal
Right arrow Depression
Right arrow Stress and Coping
Right arrow Personality
Psychosomatic Medicine 65:911-918 (2003)
© 2003 American Psychosomatic Society

ORIGINAL ARTICLES

Alexithymia as Predictor of Treatment Outcome in Patients with Functional Gastrointestinal Disorders

Piero Porcelli, PhD, R. Michael Bagby, PhD, Graeme J. Taylor, MD, FRCPC, Massimo De Carne, MD, Gioacchino Leandro, MD and Orlando Todarello, MD

From the Psychosomatic Unit (P.P.) and Department of Gastroenterology, IRCCS De Bellis Hospital (M.D.C., G.L.), Castellana Grotte, Italy; Department of Psychiatry, University of Toronto and Centre for Addiction and Mental Health-Clarke Site (R.M.B.), Toronto, Ontario, Canada; Department of Psychiatry, University of Toronto and Mount Sinai Hospital (G.J.T.), Toronto, Ontario, Canada; and Department of Psychiatry, University of Bari (O.T.), Italy

Address reprint requests to: Dr. Piero Porcelli, Unità di Psicosomatica, IRCCS Ospedale "De Bellis", Via Valente, 4, I-70013, Castellana Grotte (Bari), Italy. E-mail: porcellip{at}mail.media.it


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION>
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
OBJECTIVE: A previous study found a strong association between alexithymia and functional gastrointestinal disorders (FGID). The objective of this study was to investigate whether alexithymia might be a predictor of treatment outcome in patients with FGID.

METHODS: A group of FGID outpatients classified by the ‘Rome I’ criteria was divided into improved (N= 68) and unimproved (N= 44) groups on the basis of pre-established criteria after 6 months of treatment. Patients were administered the 20-item Toronto Alexithymia Scale, the Hospital Anxiety and Depression Scale, and the Gastrointestinal Symptom Rating Scale both before and after 6 months of treatment.

RESULTS: At the base-line assessment, compared with the improved patients, the unimproved patients had significantly higher levels of anxiety, depression, alexithymia, and gastrointestinal symptoms. Stability of alexithymia was demonstrated by significant correlations between base-line and follow-up TAS-20 scores in the entire sample. Moreover, hierarchical regression analyses showed that the stability of TAS-20 scores over the 6-month treatment period could not be accounted for by their associations with anxiety and depression scores. In logistic regression analyses, base-line alexithymia and depression emerged as significant predictors of treatment outcome. Relative to depression, however, alexithymia was the stronger predictor.

CONCLUSIONS: Alexithymia is a reliable and stable predictor of treatment outcome in FGID patients. Although further studies are needed, clinicians might improve treatment outcome by identifying patients with high alexithymia, and attempting to improve these patients’ skills for coping with emotionally stressful situations.

Key Words: alexithymia, • functional gastrointestinal disorders, • anxiety, • depression, • prediction, • treatment outcome,

Abbreviations: FGID = functional gastrointestinal disorders;; GSRS = Gastrointestinal Symptom Rating Scale;; HADS = Hospital Anxiety and Depression Scale;; HADS-A = Hospital Anxiety and Depression Scale, Anxiety subscale;; HADS-D = Hospital Anxiety and Depression Scale, Depression subscale;; TAS-20 = Twenty-item Toronto Alexithymia Scale.


   INTRODUCTION>
TOP
 ABSTRACT
 INTRODUCTION>
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Functional gastrointestinal disorders (FGID) occur commonly in the general population and comprise a high percentage of patients who present to primary care physicians or gastroenterologists (1–4). Although the pathophysiology of FGID is not fully understood, there is accumulating evidence that symptom formation involves an interaction among multiple factors that vary in importance from one patient to another. These factors include motility disturbances, altered thresholds of pain and other sensory input from the gut, gastrointestinal inflammation and infection, psychological distress, and personality disturbances (5–10). Personality traits and emotional states may not only induce effects on the physiology of the gut (6, 11), but also influence how the symptoms of FGID are experienced and acted on, and the outcome of treatment (9, 12–13).

One personality trait that has been associated with both illness behavior and emotion dysregulation is alexithymia (14, 15). As defined by Nemiah et al. (16), the salient features of the alexithymia construct are: (a) difficulty identifying feelings and distinguishing between feelings and the bodily sensations of emotional arousal; (b) difficulty describing feelings to others; (c) constricted imaginative processes; and (d) a stimulus-bound, externally orientated cognitive style. These characteristics are thought to reflect deficits in the cognitive processing and regulation of emotions (14, 15). In a preliminary study, a group of patients with FGID, classified by the ‘Rome I’ criteria (5), was found to be significantly more alexithymic than a group of patients with inflammatory bowel disease, and the two groups were significantly more alexithymic than a group of healthy adults (17). Moreover, 66% of the FGID patients scored in the high range (>=61) of the 20-item Toronto Alexithymia Scale (TAS-20). This study used a cross-sectional design and therefore provided no information about the stability of alexithymia or its impact on the course of FGID and patients’ responses to treatment.

In assessing the stability of a personality trait, such as alexithymia, it is important to distinguish between absolute stability and relative stability. Absolute stability refers to the extent to which personality test scores in an identified group of individuals, as a whole, change over time, including during the course of treatment. Relative stability refers to the degree to which the relative differences on personality test scores among individuals in a group remain the same over time, even in the context of acute symptom change (18). Although several investigators have concluded that alexithymia is unstable on the basis of a failure to demonstrate absolute stability, Luminet et al. (19) have argued that relative stability, even in the absence of absolute stability, is sufficient to establish alexithymia as a vulnerability factor for certain medical and psychiatric disorders, and as a moderator of their response to treatment. The aim of the present study was to assess the stability of alexithymia in patients with FGID and to investigate the extent to which alexithymia can predict the response to the treatment strategies that are usually administered to FGID patients.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION>
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Participants and Procedures
The participants were 130 consecutive patients (44 men and 86 women) referred for their first consultation at the FGID outpatient facility of a gastrointestinal tertiary care hospital in southern Italy, between January 1997 and December 1998. All patients satisfied the ‘Rome I’ symptom-based diagnostic criteria for one or more of the FGID subgroups (functional dyspepsia, irritable bowel syndrome, or functional abdominal pain) (5), and for all patients, biochemical, ultrasonographic, and endoscopic examinations revealed no evidence of any structural organic gastrointestinal disease. The mean age of the group was 37.9 years (SD = 14.4) and the mean education was 9.5 years (SD = 3.8). All of the patients were willing to participate in the study and gave written informed consent.

At the time of their initial evaluation (base-line) all patients completed self-report psychological scales and were rated by their gastrointestinal physician on a gastrointestinal symptom questionnaire (see "Measures" section). Patients were treated according to their symptoms case by case with combination forms of gastrointestinal medications (usually antisecretory, prokinetic, and antispasmodic drugs), diet modifications (usually including high-fiber intake), psychotropic medications (usually anxiolytic and antidepressant drugs, in low doses), and sessions of psychological counseling or brief psychotherapy. After 6 months of treatment (follow-up), the patients were re-administered the same questionnaires.

Measures
Alexithymia was assessed with the self-report 20-item Toronto Alexithymia Scale (TAS-20) (20, 21), which is composed of 20 items rated on a 5-point Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree). Total scores range from 20 to 100. The reliability and validity of the scale are well established (14, 20, 21). A validated Italian version of the TAS-20 (22) was used in this study.

Psychological distress was assessed with an Italian translation of the Hospital Anxiety and Depression Scale (HADS) (23), which is a 14-item self-report scale providing two separate subscale scores for anxiety (HADS-A; 7 items) and depression (HADS-D; 7 items). The HADS is specifically designed for assessing physically ill patients and has demonstrated reliability and validity (24, 25) . The Italian translation of the HADS was developed several years ago by the first author (P.P.) who used the method of translation-backtranslation by bilingual colleagues to establish crosslanguage equivalence with the English version of the scale. Minor changes were made to the wording of a few of the items after preliminary testing of the translated scale on groups of patients and hospital staff with a wide range of educational backgrounds. This Italian version of the HADS has since been used for both clinical and research purposes (17, 26).

Gastrointestinal symptoms were evaluated by means of a disease-specific observer-rated questionnaire, the Gastrointestinal Symptom Rating Scale (GSRS) (27). The questionnaire includes 15 symptoms rated on a 7-point Likert scale ranging from 0 (no symptoms) to 3 (pronounced symptoms), including half points. The total score represents overall severity of gastrointestinal symptomatology, while three subscale scores correspond to dyspeptic (epigastric and abdominal pain, squeezing sensations in the epigastrium, acid regurgitation, heartburn, and nausea), digestive (borborygmus, abdominal distension, eructation, and increased flatus), and intestinal (decreased passage of stools, increased passage of stools, loose stools, hard stools, urgent need of defecation, and feeling of incomplete evacuation) syndromes. The GSRS is well validated and widely used in gastrointestinal research (28). For this study an Italian translation of the GSRS was developed following a procedure similar to that used in developing the Italian translation of the HADS. After evaluating the translation and backtranslation of the GSRS, the first author discussed the item content with a group of Italian gastroenterologists who were to administer the scale. Before conducting the study, preliminary testing of the scale was done with small groups of patients and hospital staff representing a range of educational backgrounds.

Outcome Criteria
A major problem with prospectively designed studies of FGID is the lack of consensus on how to measure outcome. However, a panel of experts (29) has recently provided helpful guidelines including the following recommendations that are applicable to this study: (a) the main results of the study should be based on the primary outcome measure constituted mainly by an assessment that integrates the key symptoms of FGID and is based mainly on patients’ reports of symptoms; and (b) the study protocol should include a priori the definition of a responder, namely the change in the outcome measure that is considered clinically meaningful. In the present study, the GSRS total score was used to categorize patients into improved and unimproved outcome groups. Adopting the recommended guidelines, we defined two criteria that patients had to satisfy to be included in the improved group. The first criterion was determined by the change of overall gastrointestinal symptoms, expressed as the proportion of change from base-line to follow-up and calculated as follows:

(GSRS total score at base-line minus the GSRS total score at follow-up)/GSRS total score at base-line) x 100.

The mean GSRS total score change was 53% and the median score change was 68% (range: -100% to +100%). Because a positive change of two thirds could be reasonably considered as a good clinical improvement of symptoms, we considered improved those patients who showed >=68% (median value) of symptom change. By itself, however, this criterion is not sufficient to define a responder because overall symptoms at base-line may vary greatly among patients. A second criterion was therefore chosen, which was a low level of symptoms at follow-up. The mean GSRS total score at follow-up was 5.8 (SD = 6.1) and the median was 3 (range: 0–25; higher score represents higher level of symptoms). We considered improved those patients who obtained a GSRS total score <=3 at follow-up. Thus, using both criteria, the improved group included patients with at least 68% of GSRS total score change from base-line to follow-up, and a GSRS total score of <=3 at follow-up.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION>
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Eighteen patients (4 men and 14 women) did not present for the 6-month follow-up evaluation, two of whom had moved from the catchment area. No significant differences were found between the patients who did not participate in the follow-up and those who completed the study for age, gender, education, illness duration, types of FGID, and base-line anxiety (HADS-A), alexithymia (TAS-20), and gastrointestinal symptoms (GSRS, both the total as well as the three subscale scores). Those who did not complete the study had lower base-line depression (as measured by HADS-D scores) than those who completed the study (mean = 6.7, SD = 4.8 vs. mean = 10.6, SD = 5.7, respectively) [t(128)= 2.76, p< .01].

On the basis of the two outcome criteria defined previously, the 112 patients who completed the study were divided into improved and unimproved groups. The improved group was comprised of 68 patients (25 men and 43 women; mean age = 37.3 years, SD = 14.7; mean education = 10.0 years, SD = 3.9). The unimproved group was comprised of 44 patients (15 men and 29 women; mean age = 37.7 years, SD = 14.8; mean education = 9.2 years, SD = 3.4). Gender, age, and education were not significantly different between the improved and unimproved groups. In the improved group, 25 patients (36.8%) fulfilled the ‘Rome I’ criteria (5) for functional dyspepsia, 25 patients (36.8%) for irritable bowel syndrome, 8 patients (11.7%) for both functional dyspepsia and irritable bowel syndrome, and 10 patients (14.7%) for functional abdominal pain. In the unimproved group, 19 patients (43.2%) fulfilled the ‘Rome I’ criteria for functional dyspepsia, 4 patients (9.1%) for irritable bowel syndrome, 14 patients (31.8%) for both functional dyspepsia and irritable bowel syndrome, and 7 patients (15.9%) for functional abdominal pain. The proportion of subgroups of FGID patients who were classified as either improved or unimproved was significantly different, {chi}2 (3)= 13.68 (p< .003). Examination of individual cell sizes for the subgroups of the FGID patients who were improved or unimproved suggested that the difference was attributable to a higher proportion of patients with both functional dyspepsia and irritable bowel syndrome in the unimproved group, and a significantly higher proportion of patients with irritable bowel syndrome alone in the improved group. Because of the small number of subjects and the high overlap of symptoms that is generally found across FGID subgroups (5, 30, 31), all FGID patients were collapsed into the two outcome-related groups.

Mean Scores and Correlations among the Measures
Table 1 displays the means and standard deviations for the TAS-20, HADS-A, HADS-D, and GSRS total scores for the total sample and for the improved and the unimproved patients at base-line and at follow-up. At base-line, compared with the improved patients, the unimproved patients displayed significantly higher scores on the HADS-A [t (111)= 1.98, p=.04], HADS-D [t (111) = 4.39, p< .001], TAS-20 [t (111) = 7.89, p< .001], and GSRS [t (111) = 3.62, p= .005]. These results suggest that higher levels of anxiety, depression, alexithymia, and symptom severity may all serve as predictors of treatment outcome. They may also suggest, however, that these constructs are part of the illness. Based on the empirically established cut-off score of >=61 on the TAS-20 (14), 56% of the total sample of FGID patients scored in the high alexithymia range before treatment.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Means, standard deviations, t-tests, effect sizes, and test-retest correlations for the TAS-20, HADS anxiety and depression, and GSRS total score at baseline and follow-up
 
Table 2 shows the correlations among the measures both at base-line and at follow-up. Although only the TAS-20 correlated significantly with the GSRS total score at base-line, the HADS-A and HADS-D as well as the TAS-20 correlated significantly with the GSRS total score at follow-up.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Correlations among the measures at baseline and follow-up
 
Stability
We first examined the change in mean scores from base-line to follow-up (ie, absolute stability) of the TAS-20, HADS-A, HADS-D, and GSRS total scores for the total sample, and for the improved and unimproved groups. These results are displayed in Table 1. As would be expected, there were statistically significant changes in the improved patients for all of these variables. It is interesting to note, however, that the magnitude of change (calculated with Cohen’s d) for the TAS-20 scores was substantially smaller than it was for the HADS-A, HADS-D, and GSRS total scores. For the unimproved patients, also as would be expected, there were no significant changes in HADS-A, HADS-D, and GSRS total scores and the magnitudes of these changes were substantially smaller than they were in the improved group of patients. Only TAS-20 scores changed significantly in the unimproved group; however, the magnitude of change was very close to that in the improved group, suggesting that any lack of absolute stability for TAS-20 scores is probably not associated with change in symptom scores.

To assess relative stability of the TAS-20, we calculated test-retest reliabilities. We correlated base-line and follow-up scores for the entire sample; the reliability coefficient (Pearson’s correlation) was r= 0.76, p< .001, suggesting significant relative stability. Examination of a scatterplot of correlations between base-line and follow-up TAS-20 scores revealed that the association between base-line and follow-up scores for those who initially scored low on the TAS-20 covaried in a similar fashion to those who initially scored high on the TAS-20.

Another procedure that can be conducted to assess the relative stability of personality traits, especially in treatment outcome studies, which typically involve substantial change in symptom severity, is hierarchical regression. For example, Santor et al. (18) argued that stability is demonstrated when the personality test scores at the beginning of treatment, when patients are acutely ill, can predict test scores on the same measure of personality at treatment completion, when symptoms have diminished, even after controlling for symptom severity at both the beginning and end of treatment. To this end, we constructed regression models in which TAS-20 scores at follow-up served as the dependent (criterion) variable and base-line and follow-up HADS-D or HADS-A scores and TAS-20 scores at base-line served as the independent (predictor) variable. In the first set of analyses, which used the entire sample, base-line TAS-20 scores were a significant predictor of follow-up TAS-20 scores even after controlling for base-line and follow-up HADS-D scores, R2chg= 0.23, p< .001, and base-line and follow-up HADS-A scores, R2chg= 0.35, p< .001. A similar pattern of results emerged in the improved group, when controlling for base-line and follow-up HADS-D scores, R2chg= 0.27, p< .001, and base-line and follow-up HADS-A scores, R2chg= 0.32, p< .001; and also in the unimproved group, when controlling for base-line and follow-up HADS-D scores, R2chg= 0.22, p< .001, and base-line and follow-up HADS-A scores, R2chg= 0.28, p< .001. These results suggest that any association between anxiety and depression scores and TAS-20 scores cannot account for the stability of TAS-20 scores across the treatment interval. Thus, TAS-20 scores can be seen as a reliable and stable variable for predicting treatment outcome.

Predicting Treatment Outcome
Having established the relative stability of alexithymia, as measured by the TAS-20, we next sought to determine if alexithymia can predict treatment outcome. A series of logistic and linear regression analyses were performed. For the logistic regressions, treatment outcome (improved/unimproved) served as the dependent (criterion) variable and base-line TAS-20, HADS-A, HADS-D, and GSRS scores served as the independent (predictor) variables. In the subsequent linear regressions, percentage of change in GSRS total scores from base-line to follow-up served as the criterion (dependent) variable. For the logistic and linear regressions we computed standardized scores for the TAS-20, HADS-A, HADS-D, and GSRS.

In the first set of logistic regression analyses, we entered each of the independent variables as single predictors to determine how well each variable alone predicted treatment outcome. As shown in Table 3, TAS-20 scores proved to be the strongest predictor of treatment outcome, more than doubling the Cox and Snell R2 values and odds ratio estimates for HADS-A, HADS-D, and GSRS total scores. As evident from Table 2, however, the TAS-20 is significantly correlated with these other independent variables.


View this table:
[in this window]
[in a new window]
  		TABLE 3. Results from individual variable logistic regressions predicting treatment outcome
 
Therefore, a set of hierarchical regression analyses were performed to determine if TAS-20 base-line scores were a unique predictor of treatment outcome, independent of the variance shared with these other variables. HADS-A, HADS-D, and GSRS base-line scores were entered first into the regression model as a block (Step 1), followed by the entry of TAS-20 base-line scores (Step 2). The results from this analysis are displayed in Table 4. The model that included the block containing base-line HADS-A, HADS-D, and GSRS scores (Step 1) produced a statistically significant fit, {chi}2(3) = 28.97, p< .001, Cox and Snell R2= 0.23, accurately predicting 84% of the improved and 66% of the unimproved patients (overall correct classification rate = 77%). The addition of base-line TAS-20 scores (Step 2) increased significantly the overall fit of the model, {chi}2dif (1) = 32.51, p< .001. The overall fit of the model containing all four variables was significant, {chi}2 (4)= 61.48, p< .001; the addition of TAS-20 scores increased the Cox and Snell R2 to 0.42 (from 0.23), and increased the accurate prediction of the improved patients to 85% and unimproved patients to 82%; the overall correct classification rate improved to 83%.


View this table:
[in this window]
[in a new window]
  		TABLE 4. Results from hierarchical logistic regressions predicting treatment outcome
 
In a second hierarchical logistic regression analysis, the entry order of the variables was reversed, with base-line TAS-20 scores entered first into the model, followed by the block entry of base-line HADS-A, HADS-D, and GSRS scores. This analysis, when compared with the previous regression equation, allows one to gauge the relative predictive capacity of the block of HADS-A, HADS-D, and GSRS scores with TAS-20 scores in the prediction of treatment outcome. The model that included only the TAS-20 scores (Step 1) produced a statistically significant fit, {chi}2 (1)= 53.64, p< .001, Cox and Snell R2= 0.38, accurately predicting 74% of the improved and 89% of the unimproved patients (overall correct classification rate = 80%). In Step 2, the addition of the block containing HADS-A, HADS-D, and GSRS increased the overall fit of the model, {chi}2dif (1) = 7.84, p< .05. It is important to note, however, that the R2 value for alexithymia as a single predictor was 0.38 compared with an R2 value of 0.23 for the symptom variables measured by the HADS-A, HADS-D, and GSRS. Thus, whereas the addition of TAS-20 scores increased the R2 value from 0.23 to 0.42 when added to the block entry of HADS-A, HADS-D, and GSRS, adding these latter variables only increased the R2 values from 0.38 to 0.42.

The above analyses examined treatment outcome categorically with patients classified as either improved or unimproved. Since patients with more severe FGID symptoms were also more likely to be psychologically distressed, there was a potential to increase artifactually the association between the psychological measures and categorically defined outcome. In addition, given that one criterion of a treatment responder was that the GSRS total score be no greater than 3, it is possible that patients with milder gastrointestinal symptoms at the beginning of treatment were more likely to be categorized as improved at the end of treatment. We therefore performed a series of linear multiple regression analyses using percent change in symptoms as the outcome measure.

Using the same modeling procedure as we used with the logistic regression analyses, we entered the HADS-A, HADS-D, and TAS-20 base-line scores as single predictors to determine how well each of these independent variables alone predicted treatment outcome. As percentage change in GSRS total scores from base-line to follow-up was used exclusively to construct the criterion variable (ie, dependent variable), GSRS scores could not be used as a predictor variable. As shown in Table 5, the results are similar to those obtained using the categorical outcome measure in the logistic regression, with TAS-20 scores emerging as the best predictor of treatment outcome compared with the HADS-A and HADS-D. We next performed a series of hierarchical regression analyses to determine if TAS-20 scores were a unique predictor of percentage change in GSRS scores from base-line to follow-up. HADS-A and HADS-D base-line scores were entered first into the regression model as a block (Step 1), followed by the entry of TAS-20 base-line scores (Step 2). The results from this analysis are displayed in Table 6. Again, even after controlling for base-line HADS-A and HADS-D scores, base-line TAS-20 score proved to be a significant predictor, adding a significant increase in incremental variance in the prediction of percentage change in GSRS scores.


View this table:
[in this window]
[in a new window]
  		TABLE 5. Results from individual linear regressions predicting percentage change in GSRS total scores from baseline to follow-up
 

View this table:
[in this window]
[in a new window]
  		TABLE 6. Results from hierarchical linear regressions predicting percentage change in GSRS total scores from baseline to follow-up
 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION>
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The results of this study indicate that FGID patients who did not respond to treatment were more alexithymic, more depressed, and had more pronounced gastrointestinal symptoms both before and after a period of 6 months of treatment than FGID patients who did respond to treatment. Although the findings suggested that all of these variables might serve as base-line predictors of treatment outcome, alexithymia clearly emerged as the most powerful predictor of both recovery status and overall reduction in GSRS symptoms at follow-up. More impressively, alexithymia, assessed at base-line, was able to predict recovery status and reduction in symptoms even after controlling for base-line gastrointestinal symptoms, depression, and anxiety symptoms. Moreover, evidence that alexithymia can be considered a reliable and stable predictor of treatment outcome was provided by significant correlations between base-line and follow-up TAS-20 scores, and by the results of hierarchical regression analyses showing that the stability of TAS-20 scores over the 6-month treatment period could not be accounted for by their associations with anxiety and depression scores. The main contribution of TAS-20 scores was to improve the prediction of patients who did not respond to treatment compared with patients who responded to treatment.

The finding that 56% of the FGID patients scored in the high range of the TAS-20 is lower than the finding of 66% in a previous study of FGID patients (17), but this rate of high alexithymia is similar to rates that have been reported in studies of patients with panic disorder, posttraumatic stress disorder, essential hypertension, somatoform pain disorder, substance use disorders, and eating disorders (for reviews see 14,15). Despite the strong associations between alexithymia and these various disorders, however, only a few studies have examined the role of alexithymia in predicting treatment outcome. Kosten et al. (32) studied a group of Vietnam combat veterans with posttraumatic stress disorder who entered a double-blind, placebo-controlled pharmacological trial of 8 weeks. Alexithymia was measured with the Alexithymia Provoked Response Questionnaire (33) and was found to be stable over the 8 week treatment period and to significantly predict treatment outcome. In another study, Bach and Bach (34) investigated a group of psychiatric inpatients who had presented with functional somatic symptoms and received diagnoses of anxiety or somatoform disorders. The patients were treated with cognitive and behavioral therapies for a minimum of 8 weeks and reassessed 2 years later. Alexithymia, measured by the 26-item version of the Toronto Alexithymia Scale (35), was found to be a significant predictor of persistent somatization and was independent of sociodemographic variables, other psychopathology, and illness severity.

There are several potential pathways by which alexithymia might influence symptom severity and treatment outcome for FGID patients. These include a limited ability of high alexithymia individuals to cope adaptively with stressful situations (36, 37), which may contribute to high levels of psychological distress and a possible sustained arousal of the physiological component of emotion response systems (15, 38). In addition, high alexithymia individuals may be prone to functional somatic symptoms because of a tendency to amplify, focus on, and misinterpret the somatic sensations that accompany states of emotional arousal as well as other normal bodily sensations (38, 39). There is evidence that FGID patients have a low pain threshold in the perception of visceral stimuli, including in bodily sites beyond the symptom-related region of the gastrointestinal tract (40). Furthermore, the bidirectional ‘brain-gut’ axis allows somatic symptoms to be generated not only by motility disturbances and alterations in sensory inputs from the gut, but also by unregulated states of emotional arousal and by mutual interactions between these and other behavioral factors, such as abnormal illness behavior and health care-seeking behavior (41, 42). Consequently, patients with high alexithymia may experience more severe somatic symptoms and respond poorly to treatment because of the difficulty in cognitively processing emotional and somatic stimuli.

Another pathway whereby alexithymia might contribute to a persistence of FGID symptoms is through an association with unhealthy behaviors such as poor nutritional consumption (eg, a high-fat diet or eating foods high in sugar content), poor eating behavior (eg, fast eating and bingeing), alcohol and drug use, and a sedentary lifestyle (43). For example, Tang et al. (44) found that several psychological, behavioral, and cognitive aspects of eating disorders were present among patients with irritable bowel syndrome. Crowell et al. (45) studied FGID in obese binge eaters, obese nonbinge eaters, and nonobese binge eaters and found that functional symptoms of the digestive tract were dependent on binge eating behavior and not on body weight. Porcelli et al. (46) found that lifetime eating disorders were significantly more prevalent in patients with FGID than in patients with gallstones. Moreover, as noted earlier, alexithymia is associated strongly with both eating disorders and alcohol abuse; these disorders have been conceptualized as maladaptive efforts to self-regulate distressing affects (14). Future studies might evaluate the influence of eating disorders and alcohol abuse on treatment outcome in patients with FGID.

In the present study, treatment outcome was also predicted by depression. This finding is consistent with previous studies showing that higher psychological distress and psychiatric comorbidity may be associated with persistent and severe FGID symptoms (42, 47) . Moreover, whereas both the improved and unimproved groups of patients showed only modest reductions in alexithymia, the improved patients had much larger reductions in depression and anxiety than the unimproved patients. It is possible that a reduction in depression and anxiety results in fewer FGID symptoms, or that an improvement in FGID relieves associated depression and anxiety.

There are several limitations to this study that should caution against generalizing from the results. First, the FGID patients were recruited from a tertiary care clinic and may therefore represent the most severe groups of FGID patients in which a high level of alexithymia may be more prevalent. FGID patients referred to tertiary care clinics have been found to have more psychological distress, psychiatric disorders, and abnormal illness behavior than FGID patients referred to primary and secondary care settings (5, 48). In our particular sample, however, the severity of gastrointestinal symptoms and level of psychological distress were not very high, and 60% responded to treatment, although no comparative data are available on the epidemiology of FGID in the region where the study was performed. There is some evidence also that alexithymia may be linked more with increased illness behavior than with severity of illness. Lumley and Norman (49), for example, in a study of college students, found that the alexithymic difficulty in identifying feelings was associated with increased use of outpatient treatment, even after controlling for depression and somatic complaints.

A second limitation is that because the study was not intended as a treatment controlled investigation, patients underwent different forms of treatment (including gastrointestinal and psychotropic medications, psychological counseling, and brief psychotherapy) that were combined in various ways on a case by case basis. Moreover, during the 6-month treatment period, the combinations of therapies sometimes changed for individual patients, in accordance with changes in their symptomatology and overall health. This therapeutic approach to FGID is quite usual in clinical practice, and the study therefore could not be regarded as a controlled investigation of one specific therapy.

Third, the methodology of the study was limited by a failure to control for abnormal eating behavior, alcohol use, coffee consumption, and smoking. Since these behaviors may aggravate gastrointestinal symptoms, they should be controlled for in future studies.

Fourth, the interval of 6 months between base-line and follow-up assessments was quite short and precludes generalizing to the prediction of treatment outcomes over longer periods. In addition, follow-up outcomes were assessed with a single observation and may not be reliable because FGID symptoms tend to fluctuate. Further studies with longer follow-up intervals and multiple outcome assessments are needed to establish more conclusively that alexithymia is stable in the context of symptom change.

Finally, the small numbers of patients in some of the FGID subgroups precluded examining whether there were any statistically significant differences in outcome for each of the subgroups. Nonetheless, the higher number of patients with both functional dyspepsia and irritable bowel in the unimproved group raises the possibility that treatment outcome might be influenced by the number of coexisting FGID subgroups. In the present study, however, the severity of gastrointestinal symptoms was not a significant predictor of treatment outcome.

In conclusion, the findings from the present study suggest that alexithymia is a stable personality trait in FGID patients, and like depression, it may be a useful predictor of treatment outcome. However, because the patients in this study were recruited from a tertiary care hospital and the medical and psychological treatments were tailored to each patient’s needs, further studies should evaluate the association between alexithymia and treatment outcome in FGID patients from different medical settings and in randomized, controlled treatment trials. Nonetheless, because no treatment has been shown to be definitely effective for FGID, and multicomponent therapeutic strategies are usually employed, the present findings may have important clinical implications. In addition to treating comorbid anxiety and depression, clinicians might improve treatment outcome for FGID patients by identifying those patients who are high in alexithymia, and attempting to improve these patients’ skills for coping with emotionally stressful situations.

Received for publication August 15, 2002.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION>
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Drossman DA, Li Z, Andruzzi E, Temple RD, Talley NJ, Thompson WG, Whitehead WE, Janssens J, Funch-Jensen P, Corazziari E, Richter JE, Koch GG. U.S. householder survey of functional gastrointestinal disorders: prevalence, sociodemography and health impact. Dig Dis Sci 1993; 38: 1569–80.[CrossRef][Medline]
  2. Talley NJ, Zinstmeister AR, Schleck CD, Melton III LJ. Dyspepsia and dyspepsia subgroups: a population-based study. Gastroenterology 1992; 102: 1259–68.[Medline]
  3. Mitchell CM, Drossman DA. Survey of the AGA membership relating to patients with functional gastrointestinal disorders. Gastroenterology 1987; 92: 1282–84.[Medline]
  4. Thompson WG, Heaton KW, Smyth GT, Smyth C. Irritable bowel syndrome in general practice: prevalence, characteristics, and referral. Gut 2000; 46: 78–82.[Abstract/Free Full Text]
  5. Drossman DA, ed. The Functional Gastrointestinal Disorders: Diagnosis, Pathophysiology, and Treatment. Boston: Little, Brown and Company; 1994.
  6. Ringel Y, Drossman DA. From gut to brain and back. A new perspective into functional gastrointestinal disorders. J Psychosom Res 1999; 47: 205–10.[CrossRef][Medline]
  7. Kellow JE, Delvaux M, Azpiroz F, Camilleri M, Quigley EMM, Thompson DG. Principles of applied neurogastroenterology: physiology/motility-sensation. Gut 1999; 45 (Suppl II): II17–II24.[Abstract/Free Full Text]
  8. Talley NJ, Spiller R. Irritable bowel syndrome: a little understood organic bowel disease? Lancet 2002; 360: 555–64.[CrossRef][Medline]
  9. Drossman DA, Creed FH, Olden KW, Svedlund J, Toner BB, Whitehead WE. Psychosocial aspects of the functional gastrointestinal disorders. Gut 1999; 45 (Suppl. II): II25–II30.[Abstract/Free Full Text]
  10. Osterberg E, Blomquist L, Krakau I, Weinryb RM, Asberg M, Hultcrantz R. A population study on irritable bowel syndrome and mental health. Scand J Gastroenterol 2000; 35: 264–68.[CrossRef][Medline]
  11. Wood JD, Alpers DH, Andrews PLR. Fundamentals of neurogastroenterology. Gut 1999; 45 (Suppl II): II6–II16.[Abstract/Free Full Text]
  12. Whitehead WE, Bosmajian L, Zonderman AB, Costa PT, Schuster MM. Symptoms of psychologic distress associated with irritable bowel syndrome. Gastroenterology 1988; 95: 709–14.[Medline]
  13. Bennett EJ, Palmer K, Badcock C-A, Tennant CC, Kellow JE. Functional gastrointestinal disorders: psychological, social, and somatic features. Gut 1998; 42: 414–20.[Abstract/Free Full Text]
  14. Taylor GJ, Bagby RM, Parker JDA. Disorders of Affect Regulation. Alexithymia in Medical and Psychiatric Illness. Cambridge: Cambridge University Press; 1997.
  15. Taylor GJ. Recent developments in alexithymia theory and research. Can J Psychiatry 2000; 45: 134–42.[Medline]
  16. Nemiah JC, Freyberger H, Sifneos PE. Alexithymia: a view of the psychosomatic process. In: Hill OW, ed. Modern Trends in Psychosomatic Medicine.Vol 3. London: Butterworths; 1976: 430–39.
  17. Porcelli P, Taylor GJ, Bagby RM, De Carne M. Alexithymia and functional gastrointestinal disorders. A comparison with inflammatory bowel disease. Psychother Psychosom 1999; 68: 263–69.[CrossRef][Medline]
  18. Santor DA, Bagby RM, Joffe RT. Evaluating stability and change in personality and depression. J Pers Soc Psychol 1997; 73: 1354–62.[CrossRef][Medline]
  19. Luminet O, Bagby RM, Taylor GJ. An evaluation of the absolute and relative stability of alexithymia in patients with major depression. Psychother Psychosom 2001; 70: 254–60.[CrossRef][Medline]
  20. Bagby RM, Parker JDA, Taylor GJ. The twenty-item Toronto Alexithymia Scale. I. Item selection and cross-validation of the factor structure. J Psychosom Res 1994; 38: 23–32.[CrossRef][Medline]
  21. Bagby RM, Parker JDA, Taylor GJ. The twenty-item Toronto Alexithymia Scale. II. Convergent, discriminant, and concurrent validity. J Psychosom Res 1994; 38: 33–40.[CrossRef][Medline]
  22. Bressi C, Taylor GJ, Parker JDA, Bressi G, Brambilla V, Aguglia E, Allegranti I, Bongiorno A, Giberti F, Bucca M, Todarello O, Callegari C, Vender S, Gala C, Invernizzi G. Cross validation of the factor structure of the 20-item Toronto Alexithymia Scale: An Italian multicenter study. J Psychosom Res 1996; 41: 551–59.[CrossRef][Medline]
  23. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand 1983; 67: 361–70.[Medline]
  24. Hermann C. International experiences with the Hospital Anxiety and Depression Scale. A review of validation data and clinical results. J Psychosom Res 1997; 42: 17–41.[CrossRef][Medline]
  25. Naifeh KH. Psychometric testing in functional GI disorders. In: Olden KW, ed. Handbook of Functional Gastrointestinal Disorders. New York: Marcel Dekker; 1996: 79–126.
  26. Porcelli P, Leoci C, Guerra V, Taylor GJ, Bagby RM. A longitudinal study of alexithymia and psychological distress in inflammatory bowel disease. J Psychosom Res 1996; 41: 569–73.[CrossRef][Medline]
  27. Svedlund J, Sjödin I, Dotevall G. GSRS. A clinical rating scale for gastrointestinal symptoms in patients with irritable bowel syndrome and peptic ulcer disease. Dig Dis Sci 1988; 33: 129–34.[CrossRef][Medline]
  28. Wiklund I. Aspects of quality of life in gastrointestinal disease: some methodological issues. Scand J Gastroenterol 1993; 30 Suppl 208: 129–32.
  29. Veldhuyzen van Zanten SJO, Talley NJ, Bytzer P, Klein KB, Whorwell PJ, Zinsmeister AR. Design of treatment trials for functional gastrointestinal disorders. Gut 1999; 45 (Suppl II): 69–77.
  30. Agreus L, Svardsudd K, Nyren O, Tibblin G. Irritable bowel syndrome and dyspepsia in the general population: overlap and lack of stability over time. Gastroenterology 1995; 109: 671–80.[CrossRef][Medline]
  31. Holtmann G, Goebell H, Talley NJ. Functional dyspepsia and irritable bowel syndrome: is there a common pathophysiological basis? Am J Gastroenterol 1997; 92: 954–59.[Medline]
  32. Kosten TR, Krystal JH, Giller EL, Dan E. Alexithymia as a predictor of treatment response in post-traumatic stress disorder. J Traum Stress 1992; 5: 41–50.
  33. Krystal JH, Giller EL, Cicchetti DV. Assessment of alexithymia in post-traumatic stress disorder and somatic illness: introduction of a reliable measure. Psychosom Med 1986; 48: 84–94.[Abstract/Free Full Text]
  34. Bach M, Bach D. Predictive value of alexithymia: a prospective study in somatizing patients. Psychother Psychosom 1995; 64: 43–8.[Medline]
  35. Taylor GJ, Ryan DP, Bagby RM. Toward the development of a new self-report alexithymia scale. Psychother Psychosom 1985; 44: 191–99.[Medline]
  36. Parker JDA, Taylor GJ, Bagby RM. Alexithymia: relationship with ego defense and coping styles. Compr Psychiatry 1998; 39: 91–8.[CrossRef][Medline]
  37. Parker JDA, Taylor GJ, Bagby RM. The relationship between emotional intelligence and alexithymia. Pers Indiv Diff 2001; 30: 107–15.[CrossRef]
  38. Lumley MA, Stettner L, Wehmer F. How are alexithymia and physical illness linked? A review and critique of pathways. J Psychosom Res 1996; 41: 505–18.[CrossRef][Medline]
  39. Nakao M, Barsky AJ, Kumano H, Kuboki T. Relationship between somatosensory amplification and alexithymia in a Japanese psychosomatic clinic. Psychosomatics 2002; 43: 55–60.[Abstract/Free Full Text]
  40. Trimble KC, Farouk R, Pryde A, Douglas S, Heading RC. Heightened visceral sensation in functional gastrointestinal disease is nonsite-specific. Evidence for a generalized disorder of gut sensitivity. Dig Dis Sci 1995; 40: 1607–13.[CrossRef][Medline]
  41. Herschbach P, Henrich G, von Rad M. Psychological factors in functional gastrointestinal disorders: characteristics of the disorder or of the illness behavior? Psychosom Med 1999; 61: 148–53.[Abstract/Free Full Text]
  42. Drossman DA, Whitehead WE, Toner BB, Hu YJ, Bangdiwala SI, Jia H. What determines severity among patients with painful functional bowel disorders? Am J Gastroenterol 2000; 95: 974–80.[CrossRef][Medline]
  43. Helmers KF, Mente A. Alexithymia and health behaviors in healthy male volunteers. J Psychosom Res 1999; 47: 635–45.[CrossRef][Medline]
  44. Tang TN, Toner BB, Stuckless N, Dion KL, Kaplan AS, Ali A. Features of eating disorders in patients with irritable bowel syndrome. J Psychosom Res 1998; 45: 171–78.[CrossRef][Medline]
  45. Crowell MD, Cheskin LJ, Musial F. Prevalence of gastrointestinal symptoms in obese and normal weight binge eaters. Am J Gastroenterol 1994; 89: 387–91.[Medline]
  46. Porcelli P, Leandro G, De Carne M. Functional gastrointestinal disorders and eating disorders. Scand J Gastroenterol 1998; 33: 577–82.[CrossRef][Medline]
  47. Camilleri M. Management of the irritable bowel syndrome. Gastroenterology 2001; 120: 652–68.[CrossRef][Medline]
  48. Drossman DA, Thompson WG. Irritable bowel syndrome: a graduated, multi-component treatment approach. Ann Intern Med 1992; 116: 1009–16.
  49. Lumley MA, Norman S. Alexithymia and health care utilization. Psychosom Med 1996; 58: 197–202.[Abstract]



This article has been cited by other articles:


Home page
 Ann Rheum DisHome page
H van Middendorp, R Geenen, M J Sorbi, L J P van Doornen, and J W J Bijlsma
Emotion regulation predicts change of perceived health in patients with rheumatoid arthritis
Ann Rheum Dis, July 1, 2005; 64(7): 1071 - 1074.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. PsychiatryHome page
M. SPERANZA, M. CORCOS, O. GUILBAUD, G. LOAS, and P. JEAMMET
Alexithymia, Personality, and Psychopathology
Am J Psychiatry, May 1, 2005; 162(5): 1029 - 1030.
[Full Text] [PDF]

Home page
 PsychosomaticsHome page
M. P. Jones, A. Schettler, K. Olden, and M. D. Crowell
Alexithymia and Somatosensory Amplification in Functional Dyspepsia
Psychosomatics, December 1, 2004; 45(6): 508 - 516.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
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
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Porcelli, P.
Right arrow Articles by Todarello, O.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Porcelli, P.
Right arrow Articles by Todarello, O.
Related Collections
Right arrow Gastrointestinal
Right arrow Depression
Right arrow Stress and Coping
Right arrow Personality

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS