Ali, Alisha PhD; Toner, Brenda B. PhD; Stuckless, Noreen PhD; Gallop, Ruth PhD; Diamant, Nicholas E. MD; Gould, Michael I. MD; Vidins, Eva I. MD
BDI = Beck Depression Inventory, FBD = functional boweldisorder, GI = gastrointestinal, IBD = inflammatory boweldisease, IBS = irritable bowel syndrome.
A history of abuse has been identified as a risk factor for women in the development of a number of psychiatric disorders, including depression, anxiety, somatoform disorders, borderline personality disorder, and dissociative disorders (1–5). Studies examining the relationship between sexual or physical abuse and health status report that women who have experienced abuse are significantly more likely than nonabused women to present to physicians with complaints of pelvic pain, headaches, back pain, fatigue, and joint pain (6, 7). Women with sexual or physical abuse histories have also been shown to have had more lifetime surgeries, hospitalizations, and physician visits than their nonabused counterparts (8). All of these findings point to the importance of investigating the impact of abuse on women’s physical and mental health.
A number of recent investigations have examined the relationship between a history of physical or sexual abuse and FBDs (8–10). IBS, the most common FBD, is a chronic GI disorder, characterized by symptoms of abdominal pain, bloating, and altered bowel function (11, 12), that affects mostly women (12). IBS is described as a functional disorder because its physical symptoms persist in the absence of any identifiable structural or biochemical abnormalities (13). Functional somatic symptoms have been found to be significantly more prevalent in women than in men (14), and somatic reactions without known organic causes have been found to correlate with the presence of history of sexual abuse in female patients (15).
Despite some evidence linking physical abuse and sexual abuse to IBS (8–10), few studies have examined the association between emotional abuse and IBS (16). Emotional abuse includes various forms of psychological maltreatment, trauma, and nonphysical aggression (17, 18). Recent evidence suggests a relationship between emotional abuse and functional GI disorders (19, 20); however, no studies to date have used a standardized questionnaire measure to assess emotional abuse, and no studies have addressed the question of whether emotional abuse is associated with IBS over and above an association of physical and sexual abuse to IBS. Because emotional abuse can coexist with physical and/or sexual abuse within the same abusive relationship (17), we were interested in determining whether the experience of emotional abuse is associated with IBS beyond this syndrome’s association with a history of physical and/or sexual abuse. This is especially crucial given that the link between emotional abuse and FBDs is not yet well understood. Accordingly, this study was designed to empirically assess the presence of emotional abuse and two psychosocial constructs in a sample of women diagnosed with IBS.
The first psychosocial variable investigated in this study was self-silencing. Self-silencing is a cognitive self-schema described by Jack (21). Adoption of this schema is an attempt to create and maintain safe, intimate relationships by silencing certain thoughts, feelings, and actions, a silencing that, in turn, can precipitate an overall self-negation through progressive devaluation of one’s own thoughts and beliefs. Self-silencing has been found to correlate significantly with depressive symptomatology in studies of various populations of women and men (22, 23).
The other psychosocial construct assessed in this study was self-blame. Self-blame reflects a psychosocial mechanism of self-criticism and low self-evaluation in which the individual accepts personal responsibility for negative events (24, 25). This construct has been found to be associated with bulimic symptomatology, depression, and history of abuse (26–29). The decision to examine self-silencing and self-blame in the context of IBS was based on previous literature linking psychosocial variables such as unassertiveness and social desirability to this syndrome (30, 31).
This study therefore investigated the presence of emotional abuse, self-blame, and self-silencing in a sample of women diagnosed with IBS, using a sample of female patients diagnosed with IBD as a comparison group. IBD consists primarily of ulcerative colitis and Crohn’s disease (32). Unlike IBS, IBD is characterized as an organic disorder. IBD patients were included as the comparison group in this study for a number of reasons. One is that IBS and IBD have some salient GI symptoms in common, including abdominal pain and altered bowel function (8), which allowed us to control for the major physiological aspects of IBS. Also, previous studies of history of abuse in women with IBS have used women with IBD as an organic GI disorder comparison group (10), and our aim with this study was primarily to extend the existing literature by investigating emotional abuse and specific psychosocial constructs. A final reason for our use of patients with IBD as a comparison group is that IBS and IBD have been found to differ on certain psychosocial dimensions, including depression, anxiety, and differential help-seeking behavior (31), which in part led us to examine the presence of certain psychosocial variables that may also differentiate these two patient groups. Each patient’s diagnosis of current active symptoms of either IBS or IBD had been determined by her GI specialist before participating in the study.
Because this study was cross-sectional and not longitudinal, we were not able to assess any hypotheses aimed at investigating whether certain psychosocial variables actually play a direct causal role in the onset of bowel symptoms. Rather, our objectives were to explore a possible association between IBS and emotional abuse, self-silencing, and self-blame and to generate hypotheses for subsequent research in this area.
Patient Samples and Procedures
Fifty women participated in the study, 25 of whom had IBS and 25 of whom had IBD. Participants were recruited over a period of 26 weeks in the following manner. We selected five gastroenterologists associated with three university teaching hospitals to introduce the study to their consecutive IBS and IBD patients in their clinics. Although we have a large referral base among teaching hospitals associated with the university, we selected only these five physicians to ensure that we could have ongoing contact and communication during the recruitment phase. The principal investigator met with each physician to describe the recruitment procedure and to reinforce the need for introducing the study to consecutive patients (eg, not only patients with a history of abuse). During these meetings, we also ensured that these specialists were very familiar with diagnostic criteria for the study, as outlined in Table 1 (12), and emphasized that a consistent procedure should be followed for introducing the study to patients.
The gastroenterologists introduced the study to each patient by providing 1) a brief verbal description of the study (ie, “A study is being conducted investigating irritable bowel syndrome and inflammatory bowel disease”) and 2) a copy of a form entitled “Willingness to be Contacted Regarding Irritable Bowel Syndrome and Inflammatory Bowel Disease Study”; this form described what would be required of them should they choose to participate in the study and also informed the patient that information on stressful and traumatic events would be gathered as part of the study. The form also informed patients that their decision to either participate in the study or decline participation would in no way influence their clinical care. During the recruitment phase of the study, the principal investigator met with the referring physicians to ensure adherence to the recruitment protocol. The physicians informed us that all patients who were informed of the study had completed the “Willingness to be Contacted” form.
Patients completed the contact form with their name and phone number and were then contacted by the principal investigator, who provided them with more information about their participation. None of the patients who completed the contact form refused participation in the study when contacted. All participants met privately and individually with the investigator. After signing a consent form and being given a copy of the consent form to keep, participants were administered the study measures.
Emotional abuse was assessed using a psychometrically validated measure, the Abusive Behavior Inventory (18). This questionnaire includes a self-contained 21-item subscale to assess emotional abuse in adulthood. Items within the emotional abuse scale describe such abusive or denigrating behaviors as being verbally threatened, being personally insulted or put down, being denied personal or economic independence, or being deliberately humiliated or degraded in public. On the basis of previous literature on history of abuse in general (33, 34) and on history of abuse in GI patients in particular (35), the cutoff age for the distinction between childhood and adulthood with respect to the measurement of abuse was 14 years of age. On the Abusive Behavior Inventory scale, participants indicate how often (ranging from “never” to “very frequently” on a five-point Likert scale) the abuse experience described in each item has occurred. A total score for emotional abuse is derived by summing the numbers endorsed on the Emotional Abuse scale of the Abusive Behavior Inventory. Thus, a maximum score of 5 is possible for each of the 21 items on the emotional abuse scale, giving a total maximum score of 105. Shepard and Campbell (18) report good reliability for this measure, with Cronbach’s α values ranging from 0.70 to 0.92, as well as good criterion-related validity, good construct validity (both convergent and discriminant), and good factor validity.
Self-blame was assessed through a series of validated self-blame scenarios (25). Three scenarios were presented in random order to each participant. Each scenario describes a different negative event in which the participant was asked to imagine herself (eg, a car accident on an icy road). After this fictitious event is described, a series of questions that assess the degree of blame the respondent would attribute to herself in each situation is asked (eg, “Given what happened, how much do you blame yourself for your actions?”). These questions are rated by participants on a six-point Likert scale. There are six questions in total, and, following the method outlined by Janoff-Bulman (25), a total self-blame score is derived for each participant by summing the responses to each item to provide a score out of a maximum of 36. Janoff-Bulman (25) reports good reliability for this measure, with coefficient α values ranging from 0.62 to 0.74.
It is important to note that, for the purposes of this study, we were interested in assessing a pervasive form of self-blame. This is to be contrasted with other studies of abuse that assess the degree of self-blame reported for a specific abuse incident. There are two reasons for this. First, we anticipated that this study would include individuals with a history of abuse as well as individuals without a history of abuse; thus, self-blame, if operationalized on the basis of specific abuse events, would not be applicable to all participants. Second, we were interested in self-blame as a specific psychosocial construct that has never before been measured in an IBS population and that may be associated with IBS. Thus, we operationalized self-blame as it has been defined in other patient and control populations, where it has been shown to be a useful and relevant construct (25).
The construct of self-silencing was measured with the Silencing the Self Scale (21), which was developed and validated specifically to assess this construct. The Silencing the Self Scale consists of 31 statements describing behavior and beliefs about oneself in relationship to others. Participants rate how strongly they agree with each statement on a five-point scale ranging from “strongly disagree” to “strongly agree.” Items reflect the components of the self-silencing construct (eg, “I don’t speak my feelings in an intimate relationship when I know they will cause disagreement” and “Caring means putting the other person’s needs in front of my own”). High construct validity has been reported for this scale, as has high reliability, with a Cronbach’s α coefficient of 0.91 (23). Using Jack’s (21) recommended scoring method, a total self-silencing score is derived by summing the responses on the 31-item questionnaire, giving a total score out of a maximum of 155.
The BDI is a 21-item self-report measure that assesses current depressive symptomatology and that has established validity and reliability (36). We assessed depressive symptoms in the sample because IBS patients have been found to report elevated levels of depressive symptoms (30, 31) and because depression has also been found to be associated with a history of abuse (37), self-blame (25, 27), and self-silencing (23), making it necessary for us to statistically control for levels of depressive symptoms should these associations be found in our sample.
Physical and sexual abuse.
Physical and/or sexual abuse were assessed to determine whether emotional abuse is associated with IBS beyond the association between physical/sexual abuse and IBS. Accordingly, the same age criterion (ie, 14 years) was applied to the operationalization of adulthood abuse for physical/sexual abuse as was applied for emotional abuse. We based our physical/sexual abuse assessment on questions previously used with a female GI population (8). The questions asked about whether anyone has ever 1) sexually touched the participant; 2) forced the participant to sexually touch them; 3) forced the participant into intercourse; or 4) hit, kicked, or beaten the participant.
Continuously distributed total scores for the variables of emotional abuse, self-blame, self-silencing, and depressive symptoms were calculated on the basis of the recommended scoring for each measure. Although the physical/sexual abuse questions have previously been used to derive categorical variables of abuse (8), we derived a continuously distributed variable as a composite of physical and/or sexual abuse in adulthood that could function as a control variable in our analyses (38). This variable was calculated by summing the number of positive responses to each of these abuse questions to derive a total score for each participant. All three study variables (emotional abuse, self-blame, and self-silencing) and each of the two control variables (depressive symptoms and physical/sexual abuse) were examined for normality and outliers using the explore function of SPSS version 7.5. No transformation of variables was necessary, and there were no missing data.
The Pearson correlation method was applied to assess bivariate correlations among the three study variables for the IBS patient sample and for the IBD patient sample. Because the study variables were found to be significantly correlated with each other (see Results) and because we wanted to accommodate the use of control variables, we used a Roy-Bargman stepdown F analysis. This analysis allows for investigation of each study variable in turn, with the control variables tested first in a univariate analysis of variance and with higher-priority study variables treated as covariates. Our a priori ordering of study variables was as follows: 1) emotional abuse, 2) self-blame, and 3) self-silencing. These variables were analyzed in the stepdown F analysis after controlling for physical/sexual abuse. Depressive symptom level was not correlated with any of the study variables; therefore, it was not necessary to use depressive symptom scores as a control variable in this analysis (39, 40).
Demographic Characteristics of Samples
The mean age of IBS patients was 36.1 years (SD = 8.62 years), with ages ranging from 21 to 55 years. The mean age of IBD patients was 35.8 years (SD = 7.77 years), with ages ranging from 19 to 53 years. There was no significant difference between the two groups on mean age. The mean age of initial symptom onset for the IBS and IBD groups was identical at 20.4 years. All participants were referred by their gastroenterologists and had current active symptoms based on international diagnostic criteria for IBS (12) or IBD (41) for at least 3 months before participation in the study. Among the IBD patients, 18 had a diagnosis of Crohn’s disease, and 7 had a diagnosis of ulcerative colitis.
The two patient groups did not differ significantly on measures of childhood family income, current personal income, current combined household income, education level, ethnic background, religion, and name of referring gastroenterologist. All IBS patients were white, and 23 of the 25 IBD patients were white; the other two IBD patients were Asian. The majority of participants in both groups had at least partial university education. The only demographic variable on which the two groups differed significantly was occupation, with slightly more IBS patients being “major professionals” (ie, executives or proprietors of large business) and more IBD patients being “lesser professionals” (ie, managers or proprietors of medium business) (χ2 = 5.71, p < .02). We did not consider this to constitute a major demographic difference between the two samples, especially since the groups did not differ significantly on income or education level.
IBS patients had higher levels of depressive symptomatology than IBD patients. Using total scores on the BDI, it was found that the mean level of depression for IBS patients (mean = 15.6, SD = 13.7) was significantly greater than the mean level of depression for IBD patients (mean = 9.20, SD = 6.57) (t (48) = 2.12, p < .05). The mean score for the IBS sample falls within the mild to moderate range for BDI scores, and the mean score for the IBD patients falls within the normal range of scores (36). This finding is consistent with previous findings of moderate levels of depressive symptoms in IBS patients (30, 31). However, unlike in some previous samples drawn primarily from nonmedical populations (22, 25, 27, 37), depressive symptom level was not significantly correlated with any of the study variables (emotional abuse, self-blame, or self-silencing) in our sample.
Physical and Sexual Abuse
For purposes of comparison with previous findings, we calculated the percentages of participants responding “Yes” to the questions assessing physical and sexual abuse. Our results were similar to those previously reported using the same abuse questions (8). Specifically, 44% of the IBS patients and 18% of the IBD patients reported being sexually touched; 22% of the IBS patients and 10% of the IBD patients reported being forced to sexually touch another; 34% of the IBS patients and 14% of the IBD patients reported forced intercourse; and 18% of the IBS patients and 10% of the IBD patients reported being repeatedly hit, kicked, or beaten.
Emotional Abuse, Self-Blame, and Self-Silencing by Diagnosis
The three study variables were first assessed for their intercorrelations. Table 2 presents the Pearson correlation coefficients among these variables for the IBS and IBD patients; these coefficients indicate that all study variables were significantly correlated with each other in each of the two groups but that none correlated significantly with symptoms of depression as measured by the BDI.
Results of the Roy-Bargman stepdown F analysis are summarized in Table 3. With use of the Wilks’ criterion, the study variables in the analysis, including the physical/sexual abuse variable, were multivariately affected by diagnosis (ie, IBS vs. IBD) (F (4,45) = 9.49, p < .001). As shown in Table 3, all study variables were univariately significant. A univariate contribution to the diagnosis variable was made by the physical/sexual abuse variable (F (1,48) = 21.57, p < .001); significantly higher physical/sexual abuse scores were reported by IBS patients (mean = 2.24, SD = 1.85) than by IBD patients (mean = 0.360, SD = 0.810). After the pattern of differences measured by physical/sexual abuse was entered, a difference remained on emotional abuse (stepdown F (1,47) = 6.27, p < .02); significantly higher emotional abuse scores were reported by IBS patients (mean = 39.5, SD = 12.4) than by IBD patients (mean = 29.4, SD = 6.94). After the pattern of differences measured by physical/sexual abuse and emotional abuse were entered, a difference also remained on self-blame (stepdown F (1,46) = 5.62, p < .05); significantly higher self-blame scores were reported by IBS patients (mean = 24.6, SD = 7.92) than by IBD patients (mean = 14.2, SD = 5.47). Although univariate comparison revealed that significantly higher self-silencing scores were reported by IBS patients (mean = 77.2, SD = 20.5) than by IBD patients (mean = 64.7, SD = 13.8) (univariate F (1,48) = 6.35, p < .02), this difference was already accounted for in the stepdown analysis by higher-priority study variables. Comparison with mean scores from female general population samples on the psychological variables shows that scores for our IBD sample are similar to previously reported norms but that scores for our IBS sample are elevated (see Table 4). Sufficient information was not available to directly analyze the statistical significance of the difference on self-blame scores with the previous sample. However, on the self-silencing scores, our IBS sample is significantly higher than the norm (t (127) = 2.90, p < .01), and our IBD sample is not statistically different from the norm (t (127) = 0.012, NS).
The present study investigated emotional abuse, self-blame, and self-silencing in female patients with IBS relative to a comparison group of female patients with IBD. The IBS patients reported significantly higher levels of emotional abuse, self-blame, and self-silencing than did the IBD patients, although the statistical difference on self-silencing was not significant after controlling for differences on the other psychosocial variables. We also found that these three variables were strongly intercorrelated. Finally, we found emotional abuse to be associated with IBS even when we controlled for physical/sexual abuse. We were somewhat surprised to find that depressive symptom level did not correlate significantly with emotional abuse, self-blame, or self-silencing in either the IBS or IBD sample. Although additional research is required to more fully explore this lack of association of these variables with depression, we suggest that these results may indicate that the psychosocial constructs explored in this study can function independently of an association with depressive symptoms; therefore, these factors seem to be relevant to our understanding of IBS and are most likely not simply an expression of depression.
The intercorrelation between emotional abuse, self-blame, and self-silencing suggests that women who experience emotional abuse may be more likely to develop response patterns of inhibiting self-expression and taking responsibility for negative events. Such response patterns may increase their levels of stress, a factor that has been shown to be important in the exacerbation of IBS symptoms (11, 42, 43). Future investigations should further examine this relationship to develop a more comprehensive conceptualization of the interplay between trauma and stress in the experience of IBS.
There are some limitations of the present study that must be considered in the interpretation of our findings. For instance, it should be noted that we cannot infer causality among the variables in this study, and we are thus unable to conclude definitively that any form of abuse precedes IBS symptoms. Only a controlled longitudinal investigation can address this issue directly; such an investigation could establish the timing of abusive events and their temporal relation to symptom onset and could use a case-referent design. Also, we focused on emotional abuse occurring in adulthood in this study; however, there is also a need to investigate childhood emotional abuse in the context of functional somatic symptoms. Similarly, although we used a psychometrically rigorous quantitative measure to examine emotional abuse, we suggest that future studies could integrate a qualitative approach to further refine this construct. Accordingly, our team is currently developing and validating a semistructured interview to further our conceptualization of emotional abuse in women.
This study also has limitations related to the participants who made up the samples. As with the majority of studies examining psychosocial aspects of FBDs (42), this investigation included only patients seen in tertiary care settings. This limitation is confounded by the previously reported finding that the psychological variables associated with IBS, including elevated levels of depression and anxiety, may be specific to patients who are seen in GI clinics and may therefore be associated with certain patterns of healthcare seeking rather than with IBS per se (42). There is also evidence that most standardized personality variables are not significantly higher in IBS patients than in patients with organic bowel disorders (44). Therefore, psychosocial factors, such as those investigated in this study, should be further examined in samples drawn from a broader recruitment base to include patients in primary care settings as well as individuals with IBS symptoms who do not seek medical intervention. Such studies should also include an analysis of the role of trauma and gender-related variables in men with IBS. Finally, although we devoted much effort to ensuring that the physicians referring patients to the study were familiar with the diagnostic criteria, we could not monitor each physician-patient interaction to ensure that the recruitment procedure was consistent across all physicians and across all patients.
We suggest that future research should investigate other psychosocial constructs, such as social support and self-efficacy, as variables in IBS populations and should examine these constructs in male as well as female patient samples. Additionally, there is a need to explore both self-silencing and self-blame in male participants. Finally, future studies should use structured psychiatric screening tools to examine possible interrelations between psychosocial variables and psychiatric symptoms.
1. Beitchman JH, Zucker KJ, Hood JE, daCosta GA, Akman D, Cassavia E. A review of the long-term effects of child sexual abuse. Child Abuse Negl 1992; 16:101–18.
2. Browne A, Finkelhor D. Initial and long-term effects: a review of the research. In: Finkelhor D, editor. A sourcebook on child sexual abuse. Beverly Hills, CA: Sage Publications; 1986. pp. 143–79.
3. Greenwald E, Leitenberg H, Cado S, Tarran M. Childhood sexual abuse: long-term effects on psychological and sexual functioning in a nonclinical and nonstudent sample of adult women. Child Abuse Negl 1990; 14:503–13.
4. Stein JA, Golding JM, Diegal JM. Long-term psychological sequelae of child abuse: the Los Angeles Epidemiological Catchment Area Study. In: Wyatt GE, Powell GJ, editors. Lasting effects of child sexual abuse. Newbury Park (CA): Sage Publications; 1988. p. 135–54.
5. Walker EA, Gelfand AN, Gelfand MD, Koss MP, Katon WJ. Medical and psychiatric symptoms in women gastroenterology clinic patients with histories of sexual victimization. Gen Hosp Psychiatry 1995; 97:108–18.
6. Domino JV, Haber JD. Prior physical and sexual abuse in women with chronic headache: clinical correlates. Headache 1987; 27:310–4.
7. Harrop-Griffiths J, Katon WJ, Walker E, Holm H, Russo J, Hickok L. The association between chronic pelvic pain, psychiatric diagnosis, and childhood sexual abuse. Obstet Gynecol 1988; 71:589–94.
8. Drossman DA, Leserman J, Nachman G, Li ZM, Gluck H, Toomey TC, Mitchell CM. Sexual and physical abuse in women with functional or organic gastrointestinal disorders. Ann Intern Med 1990; 113:828–33.
9. Longstreth GF, Wolfe-Tsadik G. Irritable bowel-type symptoms in HMO examinees: prevalence, demographics and clinical correlates. Dig Dis Sci 1993; 38:1581–9.
10. Walker EA, Katon WJ, Roy-Burne PP, Jemelka RP, Russo J. Histories of sexual victimization in patients with irritable bowel syndrome or inflammatory bowel disease. Am J Psychiatry 1993; 150:1502–6.
11. Drossman DA. Irritable bowel syndrome. Gastroenterologist 1994; 2:315–26.
12. Thompson WG, Dotevall G, Drossman DA, Heaton KW, Kruis W. Irritable bowel syndrome: guidelines for diagnosis. Gastroenterol Int 1989; 2:92–5.
13. Kirmayer LJ, Robbins JM. Functional somatic syndromes. In: Kirmayer LJ, Robbins JM, editors. Current concepts of somatization: research and clinical perspectives. Washington DC: American Psychiatric Press; 1991. p. 79–105.
14. Moldofsky H, Lue F. Disordered sleep, pain, fatigue and gastrointestinal symptoms in fibromyalgia, chronic fatigue and irritable bowel syndromes. In: Mayer EA, Raybould HE, editors. Pain research and clinical management. Vol 9: Basic and clinical aspects of chronic abdominal pain. Amsterdam: Elsevier Science Publications; 1993. p. 249–56.
15. Rimsza ME, Berg RA, Locke C. Sexual abuse: somatic and emotional reactions. Child Abuse Negl 1988; 12:201–8.
16. Talley NJ. Abuse and functional gastrointestinal disorders: what is the link and should we care? Gastroenterology 1996; 110:1301–4.
17. Geffner R, Rossman BBR. Emotional abuse: an emerging field of research and intervention. J Emotional Abuse 1998; 1:1–5.
18. Shepard MF, Campbell JA. The Abusive Behavior Inventory: a measure of psychological and physical abuse. J Interpers Violence 1992; 7:291–305.
19. Talley NJ, Fett SL, Zinsmeister AR, Melton LJ. Gastrointestinal symptoms and self-reported abuse: a population-based study. Gastroenterology 1994; 107:1040–9.
20. Talley NJ, Fett SL, Zinsmeister AR. Self-reported abuse and gastrointestinal disease in outpatients: association with irritable bowel-type symptoms. Am J Gastroenterol 1995; 90:366–71.
21. Jack DC. Silencing the self: women and depression. New York: Harper Perennial; 1991.
22. Jack DC, Dill D. The Silencing the Self Scale: schemas of intimacy associated with depression in women. Psychol Women Q 1992; 16:97–106.
23. Thompson JM. Silencing the self: depressive symptomology and close relationships. Psychol Women Q 1995; 19:337–53.
24. Beck AT. Depression: clinical experiments and theoretical aspects. New York: Harper & Row; 1967.
25. Janoff-Bulman R. Characterological versus behavioral self-blame: inquiries into depression and rape. J Pers Soc Psychol 1979; 37:1798–809.
26. Kaner A, Bulik CM, Sullivan PF. Abuse in adult relationships of bulimic women. J Interpers Violence 1993; 8:52–63.
27. Moscarello R. Victims of violence: aspects of the “victim-to-patient” process in women. Can J Psychiatry 1992; 37:497–502.
28. Root MPP, Fallon P. Treating the victimized bulimic. J Interpers Violence 1989; 4:90–100.
29. Wyatt GE, Newcomb M. Internal and external mediators of women’s sexual abuse in childhood. J Consult Clin Psychol 1990; 58:758–67.
30. Toner BB, Koyama E, Garfinkel PE, Jeejeebhoy KN, DiGasbarro I. Social desirability and irritable bowel syndrome. Int J Psychiatry Med 1992; 22:99–103.
31. Toner BB, Garfinkel PE, Jeejeebhoy KN, Scher HB, Shulman D, DiGasbarro I. Self-schema in irritable bowel syndrome. Psychosom Med 1990; 52:149–55.
32. Walker EA, Roy-Burne 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.
33. Finkelhor D, editor. A sourcebook on child sexual abuse. Beverly Hills, CA: Sage; 1986.
34. Russell DEH. The secret trauma: incest in the lives of girls and women. New York: Basic Books; 1986.
35. Leserman J, Drossman DA, Li Z, Toomey TC, Nachman G, Glogau L. Sexual and physical abuse history in gastroenterology practice: how types of abuse impact health status. Psychosom Med 1996; 58:4–15.
36. Beck AT, Ward CH, Mendelson M, Mock J, Erbauch J. An inventory for measuring depression. Arch Gen Psychiatry 1961; 4:561–71.
37. Cascardi M, O’Leary KD. Depressive symptomatology, self-esteem, and self-blame in battered women. J Fam Violence 1992; 7:249–59.
38. Tabachnik BG, Fidell LS. Using mulitvariate statistics. New York: Harper & Row; 1989.
39. Bock RD. Multivariate statistical methods in behavioral research. New York: McGraw-Hill; 1975.
40. Bock RD, Haggard EA. The use of multivariate analysis of variance in behavioral research. In: Whitla DK, editors. Handbook of measurement and assessment in behavioral sciences. Reading (MA): Addison-Wesley; 1968. pp. 100–42.
41. Sleisenger MH, Fordtran JS, editors. Gastrointestinal disease: pathophysiology, diagnosis, management. Philadelphia: WB Saunders; 1989.
42. Drossman DA, Creed FH, Fava GA, Olden KW, Patrick DL, Toner BB, Whitehead WE. Psychosocial aspects of the functional gastrointestinal disorders. Gastroenterol Int 1995; 8:47–90.
43. Toner BB. Cognitive-behavioral treatment of functional somatic syndromes: integrating gender issues. Cognit Behav Pract 1994; 1:157–78.
44. Talley NJ, Phillips SF, Bruce B, Twomey CK, Zinsmeister AR, Melton LJ. Relation among personality and symptoms in nonulcer dyspepsia and the irritable bowel syndrome. Gastroenterology 1990; 99:327–33.