to verbally describe feelings, although little data exist. We administered the ... administered the Toronto Alexithymia Scale (TAS) as part of the evaluation process. .... pression or other psychiatric disorders before and after recovery are needed.
Alexithymia in the Eating Disorders
Carolyn E. Cochrane Timothy D. Brewerton Diane B. Wilson Elizabeth L. Hodges (Accepted 23 September 1992)
Eating disorder patients appear to have high degrees of alexithymia, a diminished capability to verbally describe feelings, although little data exist. We administered the Toronto Alexithymia Scale (TAS) to 114 females with DSM-III-R defined eating disorders. Patients, regardless of subtype, scored significantly higher than 370 college-aged females. TAS scores were significantly correlated to self-ratings of affective symptoms, but not weight or binge-purge frequency. © 1993 by John Wiley & Sons, Inc.
Alexithymia is defined as a diminished capability to verbally describe and identify feeling states, as well as restricted imaginal capacities (Nemiah & Sifneos, 1970; Sifneos, 1973). Several clinicians have observed that bulimic and anorexic patients have high degrees of alexithymia (Garner & Garfinkle, 1982; Anderson, 1983). Researchers have used different tools to measure alexithymia in these patients and have reported varying conclusions. Two studies reported a high degree of alexithymia in patients with anorexia nervosa (AN) (Bourke, 1985; Taylor, Bagby & Parker, 1991) and one study was equivocal (Pierloot, Houben, & Acke, 1988). Patients with bulimia nervosa (BN) and obese binge eaters who do not purge (EDNOS) have not been studied. Although the application of the alexithymia concept to eating disorders (ED) is of recent origin, the concept has been previously applied to several psychiatric conditions, including suicidal behavior (Lester, 1991), substance abuse disorders (Haviland, Hendryx, Cummings, Shaw, & MacMurray, 1991; Taylor, Parker, & Bagby, 1990), and psychological factors affecting physical conditions (Rubino, Sonnino, Stefanato, Pezzarossa, & Ciami, 1989; Fernandez, Sriram, Rajikumar, & Chandrasekar, 1989). The following study was therefore undertaken to refute or substantiate the hypothesis that ED patients display high levels of alexithymia.
Carolyn E. Cochrane, Ph.D., R.N., C.S., is Instructor of Psychiatry, Assistant Professor in Nursing, and Associate Clinical Director of Eating Disorders Program at the Medical University of South Carolina. Timothy D. Brewerton, M.D., is Associate Professor in Psychiatry and Director of Eating Disorders Program at the Medical University of South Carolina. Diane B. Wilson, R.D., M.S., is Doctoral Candidate, School of Public Health, University of South Carolina, Columbia, South Carolina. Elizabeth L. Hodges, A.C.S.W., L.I.S.W., is Instructor in Psychiatry at the Medical University of South Carolina. Address reprint requests to Dr. Cochrane at Institute of Psychiatry Eating Disorders, Medical University of South Carolina, Charleston, South Carolina 29425-0742. International journal of Eating Disorders, Vol. 14, No. 2, 219-222 (1993) (c) 1993 by )ohn Wiley & Sons, Inc.
CCC 0276-3478/93/020219-04
^^°
Cochrane et al.
SUBJECTS AND METHODS The subjects included 114 female ED patients who were consecutively evaluated for treatment in the Medical University of South Carolina Eating Disorders Program and who met DSM-III-R (American Psychiatric Association, 1987) criteria for either AN (N = 19), BN (N = 52), both AN and BN {N = 18), or eating disorder not otherwise specified (EDNOS) (N = 25). These latter patients would best be described as obese or overweight patients who binge but not purge and would meet proposed criteria for binge eating disorder (BED) (American Psychiatric Association, 1991). All subjects were administered the Toronto Alexithymia Scale (TAS) as part of the evaluation process. Demographic and clinical features including age, percent of average body weight (%ABW) according to standard tables (Society of Actuaries, 1980), and frequency of binge eating and vomiting are shown in Table 1. For the measurement of alexithymia we used the TAS (Taylor, Bagby, Parker, Ryan, & Citron, 1988a; Taylor, Bagby, Ryal, Parker, & Doody, 1988b; Taylor et al., 1990), which is a 26-item self-report likert scale measuring: (1) difficulty in identifying and describing feelings; (2) difficulty in distinguishing between feelings and bodily sensations; (3) concrete and reality-based thinking; and (4) reduced daydreaming. The instrument possesses internal consistency, good test-retest reliability, construct and criterion validity, and a stable and replicable factor structure theoretically congruent with the alexithymia construct (Taylor, 1982; Taylor et al., 1988b; Bagby, Taylor, Parker, & Loiselle, 1990; Bagby, Taylor, & Ryan, 1986). We categorized subjects as alexithymic, that is, if they scored 74 or more on the TAS (Taylor & Bagby, 1988). For comparison purposes, we used published norms for 370 female college students (mean age = 23.8 years) (Taylor, Ryan, & Bagby, 1985), which was not significantly different from the age of our patients, except for the EDNOS group (p < .05). We also administered the Diagnostic Survey for Eating Disorders (DSED); (Johnson, 1985), a multi-item self-report survey that records various clinical features of eating disorders, for example, weight, binge-vomit frequency. We used data derived from the DSED for analysis of several affective symptoms including: (1) depression; (2) anxiety; (3) crying episodes; (4) irritability; (5) fatigue; (6) difficulty falling asleep; and (7) difficulty getting up in the morning.
Table 1. Demograpbic clinical features and Toronto Alexithymia Scale scores in patients with anorexia (AN), bulimia (BN), both disorders, or EDNOS patients with eating disorder not otherwise specified (EDNOS) Diagnosis Sample size (N) Age (years)* %ABW" Binge frequency Vomiting frequency TAS scores % Alexithymic *p < .05, ANOVA. **p < .01, ANOVA.
AN
BN
AN + BN
EDNOS 25 38.6+11.8 163.6+36.0 5.5±8.2
19
52
18
28.1±9.3 78.4±9.3
28.2±12.1 106±20.7% 9.5+11.7 9.9±21.1 74.9+11.2
27.3±7.1 80.2±6.4 7.1+6.5 8.1±6.9 78.5+11.1
56
61
— —
72.4±12.2 63
74.1+8.4 64
Alexithymia in Eating
221
RESULTS There were no significant differences in TAS scores among the various subtypes of ED patients using an analysis of variance (ANOVA) (see Table 1). However, the scores of the total group of patients with ED were significantly higher (74.8 ± 10.8) than those of 370 female college students previously reported by Taylor et al. (1985) (61.1 ± 11.3, p < .01 non-paired t test). TAS scores were significantly correlated with DSED self-ratings of depression (rho = .50, p < .0001), anxiety (rho = .29, p < .0005), crying (rho = .38, p < .0001), irritability (rho = .29, p < .005), difficulty falling asleep (rho = .24, p < .02), and the total score. TAS scores were not significantly correlated with age, %ABW, or frequency of bingeeating or vomiting. The percentage of patients from each diagnostic group who were considered alexithymic (Taylor, et al., 1985) did not differ by diagnosis (chi-square) (Table 1).
DISCUSSION Our results confirm that levels of alexithymia as measured by the TAS are significantly higher in ED patients, regardless of subtype, compared to a college sample. Although patients with both AN and BN had the highest scores and patients with AN alone the lowest scores, these differences were not statistically significant. Nor was the percentage of patients in each diagnostic group who scored in the alexithymic range. Nevertheless, the majority of ED patients scored in the alexithymic range. Taylor et al. (1991) has reported a much higher rate of alexithymia (77%) in a large (48) sample of AN patients. By comparison, in our sample of AN only 63% were alexithymic. TAS scores were highly correlated to several self-reported ratings of affective symptoms. This is compatible with other reports linking TAS scores with measures of depression and anxiety in college students and alcoholics (Haviland et al., 1991; Hendryx, Haviland, & Shaw, 1991; Taylor, 1990a). However, to what extent alexithymia is a state- versus traitrelated phenomenon remains unclear. Studies of alexithymia in patients with major depression or other psychiatric disorders before and after recovery are needed. There was no apparent relationship to measures of ED pathology, such as weight or frequency of binge-eating or vomiting. Further studies of the relationship of alexithymia to EDs, depression, and anxiety are warranted. Theoretically, the degree of alexithymia may have important implications for prognosis and response to treatment, but this remains to be demonstrated. Clinically, alexithymia provides very useful information for treating ED patients. Measurement of the degree of alexithymia may help to determine more specifically the type of psychotherapeutic modality to be prescribed. Patients with EDs have been reported to respond to a variety of treatments, including pharmacotherapy, cognitive-behavioral therapy (CBT), interpersonal psychotherapy, etc. Individuals with low alexithymia scores may be expected to be more able to discuss feelings readily. Those with greater degrees of alexithymia may be less able to identify feelings and more likely to channel feelings into bodily sensations such as body image distortion. Those patients with higher levels of alexithymia may be more appropriate candidates for CBT. These highly alexithymic individuals will need help in identifying and tolerating feeling states following stabilization of their eating patterns (Taylor, 1977, 1984; Warner, 1986).
222
Cochrane et al.
Swiller (1988) has suggested that highly alexithymic individuals will need both individual and group therapy. The highly alexithymic individual may learn about expression of feelings in a non threatening way during individual session. In group therapy, the patient can practice expressing affect in a supportive environment.
REFERENCES American Psychiatric Association. (1987). Diagnosis and statistical manual of mental disorders (3rd ed., revised). Washington, DC: Author. American Psychiatric Association. (1991). DSM-IV option book. Washington, DC: Author. Anderson, A. E. (1983). Anorexia nervosa and bulimia: A spectrum of eating disorders. Journal of Adolescent Health Care, 4, 15-21. Bagby, R. M., Taylor, G. J., Parker, J. D., & Louiselle, C. (1990). Cross-validation of the factor structure of the Toronto Alexithymia Scale. Journal of Psychosomatic Research, 34, 47-51. Bagby, R. M., Taylor, G. J., & Ryan, D. (1986). Toronto Alexithymia Scale: Relationship with personality and psychotherapy measures. Psychotherapy and Psychosomatics, 45, 207-215. Bourke, M. P. (1985). Symbolic functioning in anorexia nervosa. Journal of Psychiatric Research, 19, 273-278. Fernandez, A., Sriram, T., Rajikumar, S., & Chandrasekar, A. (1989). Alexithymic characteristics in rheumatoid arthritis: A controlled study. Psychosomatics and Psychotherapeutics, 51, 45-50. Gamer, D., & Garfinkle, P. (1982). Perceptive and cogrutive disturbances. Anorexia nervosa: A multidimensional perspective (pp. 123-163). New York: Brunner/Mazel. Haviland, M. G., Hendryx, M. S., Cummings, M. A., Shaw, D. G., & MacMurray, J. P. (1991). Multidimensionality and state dependency of alexithymia in recently sober alcoholics. Journal of Nervous and Mental Disease, 179, 284-290. Hendryx, M. S., Haviland, M. G., & Shaw D. G. (1991). Dimensions of alexithymia and their relationship to anxiety and depression. Journal of Personality Assessment, 56, 227-237. Johnson, C. (1985). Initial consultation for patients with bulimia and anorexia nervosa. In D. M. Gamer & P. Garfinkel (Eds.), Handbook of psychotherapy for anorexia nervosa and bulimia (pp. 19-51). New York: Guilford. Lester, D. (1991), Depression and suicide preoccupation. Perceptual and Motor Skills, 72, 1058. Neimiah, J. C , & Sifneos, P. E. (1970). Affect and fantasy in patients with psychosomatic disorders. In Hillow (Ed.), Modern trends in psychosomatic medicine (Vol. 2, pp. 26-34). London: Butterworths. Pierloot, R. A., Houben, M. E., & Acke G. (1988). Are anorexia nervosa patients alexithymic? Acta Psychiatrica Belgica, 88, 222-232. Rubino, I., Sonnino, A., Stefanato, C , Pezzarossa, B., & Ciami, N. (1989). Separation-individuation, aggression and alexithymia in psoriasis. Acta Dermato-Venereologica, 146, 87-90. Sifneos, P. (1973). The presence of alexithymic characteristics in psychosomatic patients. Psychotherapeutics and Psychosomatics, 22, 253-263. Society of Actuaries and Association of Life Insurance Medical Directors of America. (1980). Build Study, 1979. Chicago, IL: Author. Swiller, H. I. (1988). Alexithymia: Treatment utilizing combined individual and group psychotherapy. International Journal of Group Psychotherapy, 38, 47-61. Taylor, G., Bagby, R. M., & Parker, J. D. (1991). The alexithymia construct: A potential paradigm for psychosomatic medicine. Psychosomatics, 32, 153-164. Taylor, G., Ryan, D., & Bagby, R. (1985). Toward the development of a new self-report alexithymia scale. Psychotherapeutics and Psychosomatics, 44, 191-199. Taylor, G. J. (1977). Alexithymia and countertransference. Psychotherapeutics and Psychosomatics, 28, 141-147. Taylor, G. J. (1984). Psychotherapy with the boring patient. Canadian Journal of Psychiatry, 29, 217-222. Taylor, G. ]., & Bagby, R. M. (1988). Measurement of alexithymia: Recommendations for clinical practice and future research. Psychiatric Clinics of North America, 11, 351—366. Taylor, G. J., Bagby, R. M., Parker, J. D. A., Ryna, D. P., & Citron, K. (1988a). Factorial validity of the Toronto Alexithymia Scale with a clinical sample. Psychosomatic Medicine, 50, 205-206. Taylor, G. J., Bagby, R. M., Ryal, D. P., Parker, J. D., & Doody, K. F. (1988b). Criterion validity of the Toronto Alexithymia Scale. Psychosomatic Medicine, 50, 500-509. Taylor, G. J., Parker, D. A., & Bagby, R. M. (1990). A preliminary investigation of alexythymia with psychoactive substance abuse. American Journal of Psychiatry, 147, 1228-1230. Warner, H. (1986). Alexythymia: Clinical and therapeutic aspects. Psychotherapeutics and Psychosomatics, 46, 96-104.
