| | Temperament and character of Japanese eating disorder patients☆☆☆Abstract Previous studies consistently reported a link between bulimic behaviors and high Novelty Seeking (NS), as measured by the Temperament and Character Inventory (TCI). However, it remains unclear whether this relationship is universal and occurs in different environments. Subjects of the present study consisted of 66 patients with anorexia nervosa restricting type (AN-R), 59 patients with anorexia nervosa binge eating/ purging type (AN-BP), 101 patients with bulimia nervosa purging type (BN), and 75 controls. NS score in AN-R patients was significantly lower than that in controls, although NS in BN patients did not differ from that in controls. The temperament of AN-R patients seems to be universal, even in different environments. Conversely, something other than temperament might be important in the development of bulimia nervosa in Japan. Copyright 2003, Elsevier Science (USA). All rights reserved.
Recently, the contribution of temperament to the pathogenesis of personality disorders as well as psychiatric disorders has received attention.1 One of the influential theories regarding temperament is Cloninger's temperament and character theory,1 which can be measured by the Tridimensional Personality Questionnaire (TPQ) or Temperament and Character Inventory (TCI).1, 2 These tests postulate the existence of four heritable temperamental traits called Novelty Seeking (NS), Harm Avoidance (HA), Reward Dependence (RD), and Persistence (P), and three character dimensions called Self-Directedness (SD), Cooperativeness (CO), and Self-Transcendence (ST).
The TCI has been translated into several languages,3, 4, 5 the results suggest that this temperament and character model is universal, even in different cultures. However, some studies have pointed out cross-cultural differences in temperament. For example, a French population might have lower NS and higher HA compared with those of a US population.4 Moreover, whether the relationship between temperament and psychiatric disorder is universal remains unclear.
A number of previous studies found a significant relationship between temperament and eating disorders, especially bulimic behaviors. Bulimic patients were consistently reported to have significantly higher NS scores compared with patients with restricting anorexia nervosa6, 7 and controls.8, 9 Conversely, in anorexic patients, decreased NS was consistently reported.6, 10, 11 Whether these findings are universal even in very different cultures remains to be established.
We hypothesized that bulimic behaviors are linked with high NS scores, while anorexic behaviors are linked with low NS among Japanese eating disorder patients.
Method  Subjects Subjects consisted of 73 patients with anorexia nervosa restricting type (AN-R), 68 patients with anorexia nervosa binge eating/purging type (AN-BP), and 113 patients with bulimia nervosa purging type (BN). All subjects were diagnosed by T.N or N.K. according to DSM-IV criteria based on a personal interview. The TCI3 and Beck Depression Scale12 were administered on the second or third visit. The total patient number was reduced to 226 (66 AN-R, 59 AN-BP, and 101 BN) after excluding patients who gave incomplete responses on the above questionnaires. All patients were female outpatients at Osaka City University Hospital, and all gave informed consent prior to participating in the study. Controls were female students attending a nursing school in the Osaka area along with medical staff such as nurses, receptionists, and secretaries at a psychiatric hospital other than Osaka City University Hospital. In the psychiatry class, 70 students were asked to anonymously fill out inventories on a voluntary basis. For controls, one question, “Have you ever had any psychiatric or eating disorder histories?” was added at the end of the questionnaire. Although 65 students (80%) responded, three students were excluded from the control group because of past (two) and current (one) depressive episodes. One male student was also excluded. Sixteen medical staff also volunteered to fill out the above questionnaires. Fourteen staff responded. None of the staff admitted any psychiatric histories. Medical staff were significantly older than the students (26.6 ± 1.3 v 20.0 ± 1.3, t = 6.1, P < .001); however, there were no significant differences in body mass index (20.0 ± 1.4 v 20.2 ± 1.6) or any subscale scores of the TCI (NS: 21.9 ± 5.7 v 23.0 ± 5.2; HA: 19.6 ± 5.1 v 20.8 ± 6.3; RD: 17.1 ± 3.4 v 17.8 ± 2.9; P: 3.5 ± 2.3 v 3.3 ± 1.9, SD: 25.6 ± 9.3 v 22.6 ± 6.4; CO: 28.0 ± 5.2 v 27.2 ± 4.8; ST: 11.1 ± 4.8 v 11.8 ± 5.1) even by independent t test. Therefore, the two groups were collapsed for subsequent analyses. Statistics Independent t test, one-way analysis of variance (ANOVA), and multiple analysis of covariance (MANCOVA) were used where appropriate. All procedures were performed using SPSS for Windows software (Chicago, IL). Results Demographics are presented in Table 1.
| | |  | AN-R (n = 66) | AN-BP (n = 59) | BN (n = 101) | Controls (n = 75) | F(P) | | |
|---|
| Age, yr | 22.2 (5.6) | 25.4 (5.5)*†‡ | 23.0 (4.8) | 21.2 (3.3) | 8.7 (<.001) | |
| Onset, yr | 19.4 (4.6) | 20.0 (5.0) | 18.6 (4.0) | NA | 1.8 (.17) | |
| Duration, yr | 2.8 (3.2) | 5.4 (4.3)† | 4.3 (3.8)a | NA | 7.5 (.001) | |
| Height, cm | 158.1 (4.9) | 157.7 (5.7) | 159.2 (5.1) | 158.6 (4.3) | 1.3 (.28) | |
| Weight, kg | 36.5 (7.5)*‡ | 37.7 (7.2)*‡ | 52.1 (8.2) | 50.8 (4.8) | 100 (<.001) | |
| BMI | 14.6 (2.9)*‡ | 15.2 (2.8)*‡ | 20.5 (2.9) | 20.1 (1.6) | 108 (<.001) | |
| Minimum BMI | 13.0 (1.9)*‡ | 13.4 (2.2)*‡ | 16.8 (2.7)* | 19.0 (1.2) | 125 (<.001) | |
| Beck Depression Scale | 18.3 (11.1)* | 24.6 (11.0)*† | 25.7 (10.6)* | 9.9 (6.1) | 42 (<.001) | |
| | | | | |
Patients with AN-BP were significantly older than subjects in the other groups, although there were no significant differences in age of onset among the three patient groups. Duration of illness in AN-R patients was significantly shorter than in AN-BP and BN patients. AN-BP and BN patients had significantly higher Beck Depression Scale scores than AN-R patients and controls. Mean TCI scale scores are presented in Table 2.
| | | | AN-R (n = 66) | AN-BP (n = 59) | BN (n = 101) | Controls (n = 75) | F(P) | | |
|---|
| Novelty seeking | 17.6 (4.7)* | 20.2 (5.7) | 21.6 (5.6) | 22.8 (5.5) | 12 (<.001) | |
| Harm avoidance | 24.2 (6.5) | 23.7 (5.4) | 24.3 (7.6) | 20.6 (6.1) | 1.6 (.18) | |
| Reward dependence | 16.1 (3.3)* | 15.7 (3.3)* | 15.0 (3.6)* | 17.6 (3.0) | 6.3 (<.001) | |
| Persistence | 4.8 (2.1)* | 4.8 (2.0)* | 4.4 (2.1) | 3.3 (1.9) | 4.3 (.006) | |
| Self-directedness | 21.7 (7.8) | 18.7 (7.0) | 16.2 (7.1)* | 23.2 (7.0) | 3.8 (.01) | |
| Cooperativeness | 25.7 (6.2) | 25.4 (7.0) | 23.6 (7.1) | 27.4 (4.8) | 1.1 (.33) | |
| Self-transcendence | 9.1 (4.7)* | 9.7 (5.2)* | 10.7 (4.8) | 11.6 (5.0) | 3.4 (.02) | |
| | | | | |
Beck Depression Scale score and current age were entered as covariates when comparing the means among four groups. HA score in all three groups (AN-R, AN-BP, and BN) tended to be higher than that in the controls. Similarly, all three patient groups showed significantly lower RD than controls. Anorexic patients regardless of the presence or absence of bulimia (AN-R and AN-BP) had significantly higher P and lower ST than controls. Only AN-R patients showed significantly lower NS than controls. Conversely, BN patients had significantly lower SD than controls. Discussion We found significantly lower NS scores and slightly higher HA in AN-R patients than in controls. In the most recent large study, Klump et al.11 also found significantly higher HA and P, and significantly lower NS, RD, CO, SD, and ST in AN-R patients than those in controls. We also found significantly higher P, slightly higher HA, and significantly lower NS, RD, and ST in AN-R patients than those in controls. Therefore, both the study by Klump et al.11 our study similarly found a particular temperament significantly more prevalent among patients with anorexia nervosa restricting type. Our result indicates that these temperament and character features of restricting anorexic patients are similar even in cultures very different from Western culture. We found that eating disorder patients regardless of subtype had significantly lower RD scores than controls even after controlling for depressive symptoms, as previous studies found.7, 8, 11 Lower RD is related to social withdrawal, detachment, and coldness in social attitudes.1, 2 Therefore, lower RD might be related to common interpersonal trends in eating disorder patients such as interpersonal distrust.13, 14 The need to keep others at a distance in the eating disorder patients has long been described.15 However, NS and HA scores in BN patients did not differ from those in controls when depressive symptoms were controlled as covariates, although previous studies consistently reported a significant relationship between NS and bulimic symptomatology.6, 7, 8, 9 These findings can lead to several speculations. First, results might be confounded by differences in comorbidity such as substance use disorders. We previously found differences in prevalences of alcoholism or other substance use disorders between Japanese bulimic patients and those in Western countries.16 Therefore, high prevalence of substance use disorders or other kinds of axis I and II disorders in Western countries might contribute to higher the NS scores in bulimic patients shown in previous studies. Second, the Japanese temperament may be essentially different from that of patients in Western countries. Long alleles of DRD4 exon III polymorphism are indeed related to high novelty seeking in the Japanese population.17 However, long alleles in the Japanese population were reported to be less prevalent than those in Western countries as a racial difference.18 The Japanese version of the TCI is not exactly the same as the original version despite all efforts toward careful translation. Therefore, we cannot definitively establish whether NS in Japanese differs from that in Westerners. However, our results suggested that high NS is not a necessary trait for the development of bulimic behaviors. The fact that individuals with normal NS scores developed bulimia nervosa in Japan, suggested the importance of environmental factors in the pathogenesis of bulimic symptoms in a culture much different from Western ones. Otherwise, extremely low NS, suggesting a rigid, stoic, and reserved temperament,1, 2 is universal in restricting anorexic patients. A limitation of this study is that all patients were currently ill and the effects of starvation or bulimic behaviors were not controlled. Only follow-up study (after recovery) or family study would overcome this limitation. A second limitation is that the control group was too small to represent the general Japanese population. Despite these limitations, our results suggest that factors other than temperament may be more important in the pathogenesis of bulimia nervosa. During the last decade, Japanese and Western societies have both faced rapid technological evolution, changes in values and the family system (to a nuclear family). Although genetics and the heritability of eating disorders have recently received attention, the current study suggested that factors other than heritability related to temperament are important in the pathogenesis of bulimia. Further study to clarify these issues is warranted. References 1.
1
Cloninger CR, Svrakic DM, Przybeck TR.
A psychobiological model of temperament and character.
Arch Gen Psychiatry. 1993;50:975–990. 2.
2
Svrakic DM, Whitehead C, Przybeck TR, Cloninger CR.
Differential diagnosis of personality disorders by the seven-factor model of temperament and character.
Arch Gen Psychiatry. 1993;50:991–999. 3.
3
Kijima N, Tanaka E, Suzuki N, Higuchi H, Kitamura T.
Reliability and validity of the Japanese version of the Temperament and Character Inventory.
Psychol Rep. 2000;86:1050–1058. MEDLINE |
CrossRef
4.
4
Pelissolo A, Lepine JP.
Normative data and factor structure of the Temperature and Character Inventory (TCI) in the French version.
Psychiatry Res. 2000;94:67–76. Abstract | Full Text |
Full-Text PDF (97 KB)
|
CrossRef
5.
5
Gutierrez F, Torrens M, Boget T, Martin-Santos R, Sangorrin J, Perez G, et al.
Psychometric properties of the Temperament and Character Inventory (TCI) questionnaire in a Spanish psychiatric population.
Acta Psychiatr Scand. 2001;103:143–147.
CrossRef
6.
6
Kleifield EI, Sunday S, Hurt S, Halmi KA.
The Tridimensional Personality Questionnaire: an exploration of personality traits in eating disorders.
J Psychiatr Res. 1994;28:413–423. MEDLINE |
CrossRef
7.
7
Bulik CM, Sullivan PF, Weltzin TE, Kaye WH.
Temperament in eating disorders.
Int J Eating Disord. 1995;17:251–261. 8.
8
Waller DA, Gullion CM, Petty F, Hardy BW, Murdock MV, Rush AJ.
Tridimensional Personality Questionnaire and serotonin in bulimia nervosa.
Psychiatry Res. 1993;48:9–15. MEDLINE |
CrossRef
9.
9
Brewerton TD, Hand LD, Bishop ER.
The Tridimensional Personality Questionaire in eating disorder patients.
Int J Eating Disord. 1993;14:213–218. 10.
10
Casper RC.
Personality features of women with good outcome from restricting anorexia nervosa.
Psychosom Med. 1990;52:156–170. MEDLINE 11.
11
Klump KL, Bulik CM, Pollice C, Halmi KA, Fichter MM, Berrettini WH, et al.
Temperament and character in women with anorexia nervosa.
J Nerv Ment Dis. 2000;188:559–567. MEDLINE |
CrossRef
12.
12
Beck AT, Epstein N, Brown G, Stter RA.
An inventory for measuring clinical anxiety: Psychometric properties.
J Consult Clin Psychol. 1988;56:893–897.
CrossRef
13.
13
Lilenfeld LR, Stein D, Bulik CM, Strober M, Plotnicov K, Pollice C, et al.
Personality traits among currently eating disordered, recovered and never ill first-degree female relatives of bulimic and control women.
Psychol Med. 2000;30:1399–1410. MEDLINE |
CrossRef
14.
14
Garner DM, Olmstead MP, Polivy J.
Development and validation of a multidimensional eating disorder inventory for anorexia nervosa and bulimia.
Int J Eating Disord. 1983;2:15–34. 15.
15
Goodsitt A.
Eating disorders: a self-psychological perspective.
In:
Garner DM, Garfinkel PE editor.
Handbook of Treatment for Eating Disorders. Ed 2. New York, NY: Guilford; 1997;p. 205–228. 16.
16
Nagata T, Kawarada Y, Ohshima J, Iketani T, Kiriike N.
Drug use disorders in Japanese eating disorder patients.
Psychiatry Res. 2002;109:181–191. Abstract | Full Text |
Full-Text PDF (95 KB)
|
CrossRef
17.
17
Ono Y, Manki H, Yoshimura K, Muramatsu T, Mizushima H, Higuchi S, et al.
Association between dopamine D4 receptor (D4DR) exon III polymorphism and novelty seeking in Japanese subjects.
Am J Med Genet. 1997;74:501–503. MEDLINE |
CrossRef
18.
18
Tomitaka M, Tomitaka S, Otuka Y, Kim K, Matuki H, Sakamoto K, et al.
Association between novelty seeking and dopamine receptor D4 (DRD4) exon III polymorphism in Japanese subjects.
Am J Med Genet. 1999;88:469–471. MEDLINE |
CrossRef
Department of Neuropsychiatry, Osaka City University Medical School, Osaka, Japan ☆ Address reprint requests to Toshihiko Nagata, M.D., Department of Neuropsychiatry, Osaka City University Medical School, 1-4-3 Asahimachi, Abenoku, Osaka 545-8585, Japan. ☆☆ 0010-440X/03/4402-0010$30.00/0 PII: S0010-440X(03)00043-9 doi:10.1053/comp.2003.50023 © 2003 Published by Elsevier Inc. | |
|
FULL TEXT