SEARCH

SEARCH BY CITATION

Keywords:

  • eating disorder;
  • maturity fear;
  • Parental Bonding Instrument;
  • paternal care;
  • self-injurious behavior

Abstract

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES

Aim:  The goal of the study was to investigate the correlations among parental bonding patterns, eating disorders (ED) and self-injurious behavior (SIB).

Methods:  The subjects were 80 patients with ED who were divided into two groups based on those that had (n = 25) and had not engaged in SIB at least once in the last month. The patients and 120 healthy control subjects were evaluated using the Eating Disorder Inventory and the Parental Bonding Instrument.

Results:  On the Parental Bonding Instrument, the ED + SIB group showed significantly lower paternal care compared with the ED/no SIB and control groups. On the Eating Disorder Inventory, the ED + SIB group showed higher body dissatisfaction, and increased maturity fear compared with the other two groups. The total scores and perfectionism and interoceptive awareness differed significantly among the three groups.

Conclusion:  ED + SIB patients had a more severe ED pathology than ED/no SIB patients, and also felt that they had received low paternal care during childhood. Therefore, paternal care may be linked to severe ED pathology.

EARLY ETIOLOGICAL INVESTIGATIONS suggested that the family played an important role in the development of eating disorders (ED), including anorexia nervosa (AN) and bulimia nervosa (BN). These studies focused on overprotection from mothers and indifference of fathers. Evaluations using the Parental Bonding Instrument (PBI) have produced uneven results, with patients with binge eating purging type AN having less care and greater overprotection from both parents compared to those with restricting type AN and normal controls;1 BN patients having greater maternal overprotection and AN patients having higher paternal overprotection;2 and binge eating patients having less care from both parents compared to AN patients, with no significant difference in overprotection.3

In the 1980s, an association between self-injurious behavior (SIB) and ED was noted in several studies.4–6 Svirko and Hawton found that the incidence of SIB in ED ranged from 25.4% to 55.2%.7 These patients are generally understood to be difficult to treat and to have more complicated family environments. In an early review, Simpson and Porter noted that SIB patients more often experience emotional distancing from parents and inconsistent parental warmth.8 Carroll et al. suggested that SIB patients are more likely to come from families characterized by divorce, neglect, or parental deprivation,9 and recent studies have demonstrated a close association between SIB and childhood sexual or physical abuse.10 Claes et al. found that the family environment of ED patients with SIB was less cohesive, expressive and socially orientated, and more conflictual and disorganized compared to that of ED patients without SIB.11

Previous studies of ED with SIB have had several limitations, one of which is the use of a system based on operational diagnoses. Some authors have proposed that SIB is an independent clinical entity that occurs in the context of ED,4,12 while others have suggested that self-injurious and abnormal eating behaviors are only partial symptoms in patients with borderline personality disorder.13 In recent years, the significance of diagnostic subtyping of ED according to DSM criteria has been questioned. Some studies have shown that patients are more distinctively classified if personality features,14 temperament, and character15 are taken into account. Therefore, placement of ED patients on an impulsivity–compulsivity spectrum has been proposed, especially as high rates of SIB have been described in these patients.16,17

Another problem is the ambiguous definition of self-injury. Self-injurious behavior is often synonymously used with terms such as self-mutilation or self-harm. The concept of self-mutilation refers to more severe forms of self-harm, such as eye enucleation or castration. Self-harm is a more general term that is often used to refer to suicide attempts and may also include indirect methods of self-damage, such as binge eating, alcohol or drug abuse, or overdose. Farverow named obesity, smoking and substance abuse as ‘indirect methods of self-mutilating behavior’, and suggested that these types of behavior are greatly influenced by unconscious factors, while SIB refers to behavior involving direct and deliberate damage of one's own body.18 Thus, the ‘official’ definition of terms shown above is clear, however many studies use them in an inaccurate manner. Therefore, it seems that these kinds of behavior should not be discussed collectively.

To our knowledge, parental bonding patterns in patients with ED and SIB have not been examined. After consideration of the above issues, we investigated the clinical features of ED with SIB in Japanese patients from the perspective of parental bonding. We defined SIB as the ‘direct and deliberate damage of one's own body surface without suicidal intent’.19 Claes et al. developed the Self-Injury Questionnaire – Treatment Related (SIQ-TR) to assess active SIB in female ED patients, with a focus on the causes and functional consequences of SIB.20 In the SIQ-TR, SIB is limited to that occurring within the last month, with definition of five types of SIB that are frequently reported in ED: scratching, bruising, cutting, burning, and biting. Each subject was also free to list a type of self-injury (if any) that was not included among the five types. We adopted this definition of SIB in the current study.

METHODS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES

Subjects

The subjects of the study were 80 female ED patients who visited University Hospital, Kyoto Prefectural University of Medicine and were diagnosed based on DSM-IV criteria. Of these subjects, 25 (31.3%) had engaged in at least one type of SIB in the last month (ED + SIB group), and had a mean age of 24.3 years (standard deviation [SD] 5.6). The rest (55 subjects) had not engaged in SIB in the last month (ED/no SIB group), and had a mean age of 26.9 years (SD 7.9). The demographic and clinical characteristics of the subjects are shown in Tables 1 and 2. There was no significant difference in age, years of education, parental divorce, running away from home, seeing quarrels between parents, financial difficulty or physical violence between the ED + SIB and ED/no SIB groups, but there was a significant difference in age at onset (P < 0.05). The healthy control group consisted of 120 female university students in Kyoto, with a mean age of 19.5 years (SD 1.2). The criteria for SIB were scratching, bruising, cutting, biting, picking skin and pulling of hair, based on Claes et al.20 The frequencies of different types of SIB are shown in Table 3. Overall, the ‘arms, hands, and/or nails’ were the most frequently injured body parts. There were some differences in the frequency and degree of pain, but patients generally stated that they occasionally felt mild or moderate pain during the SIB. The study was approved by the Ethical Review Committee of Kyoto Prefectural University of Medicine.

Table 1.  Diagnoses of eating disorder (ED) patients with and without self-injurious behavior (SIB)
 ED + SIB n = 25ED/no SIB n = 55All ED n = 80
  1. AN-BP, anorexia nervosa, binge-eating/purging type; AN-R, anorexia nervosa, restricting type; BN-NP, bulimia nervosa, non-purging type; BN-P, bulimia nervosa, purging type; ED-NOS, eating disorder – not otherwise specified; ED/no SIB, did not engage in SIB within the last month; ED + SIB, engaged in SIB within the last month.

AN-R11920
AN-BP92029
BN-P111223
BN-NP123
ED-NOS325
Table 2.  Comparison of demographic and clinical characteristics of patients with ED
VariableED + SIB n = 25ED/no SIB n = 55Significance
  • Data shown as the mean ± standard deviation.

  • Data shown as the number of patients (with the percentage in parentheses).

  • ED + SIB, engaged in SIB within the last month; ED/no SIB, did not engage in SIB within the last month; NS, not significant.

Age at onset17.2 ± 3.019.1 ± 4.8<0.05
Current age24.3 ± 5.626.9 ± 7.9NS
Years of education14.3 ± 2.214.1 ± 2.0NS
Parental divorce2 (8%)3 (5.5%)NS
Ran away from home4 (16%)6 (10.9%)NS
Saw quarrels between parents9 (36%)19 (34%)NS
Financial difficulty2 (8%)2 (3.6%)NS
Physical violence4 (16%)5 (9.1%)NS
Table 3.  Frequencies (n) of different types of self-injurious behavior (SIB)
 Scratching (Nmax = 10) nBruising (Nmax = 8) nCutting (Nmax = 16) nBiting (Nmax = 8) nPicking Skin (Nmax = 3) nPulling hair (Nmax = 5) n
  1. Nmax: A total of 25 subjects engaged in at least one type of SIB. Subjects who engaged in two or more types of SIB are included in each type of SIB.

Injured body parts      
 Head040005
 Arms, hands, nails9315810
 Torso, belly, buttocks000010
 Legs, feet, toes111010
 Breasts, genitals000000
Frequency of pain      
 Never417432
 Occasional457403
 Often212000
 Always010000
Degree of pain      
 None521422
 Mild226313
 Moderate217100
 Strong121000
 Very strong011000

Measurements

Self-reporting questionnaires were used in the study. The Eating Disorder Inventory (EDI) was used to evaluate the severity of ED. The EDI was developed by Garner et al. to assess the psychological and behavioral traits common to AN and BN.21 It consists of 64 items scored on a 6-point scale: seven items on ‘Drive for Thinness’, seven on ‘Bulimia’, nine on ‘Body Dissatisfaction’, 10 on ‘Ineffectiveness’, six on ‘Perfectionism’, seven on ‘Interpersonal Distrust’, 10 on ‘Interoceptive Awareness’, and eight on ‘Maturity Fear’. The reliability and validity of the Japanese version of the EDI has been confirmed, and this questionnaire is widely used for evaluation of AN, BN and normal female subjects.22,23

The Parental Bonding Instrument (PBI) was used to measure the parental bonding pattern. The PBI was developed and validated by Parker et al. to assess parental rearing attitudes, based on the memory of subjects regarding their parents in their first 16 years.24 The PBI score reflects the perceived parental rearing attitudes and can change from before to after treatment.1 The PBI consists of 25 items, including 12 on ‘Care’ (parental warmth and affection) and 13 on ‘Protection’ (parental promotion of psychological autonomy versus psychological control). A subject scores their father's and mother's attitudes separately using a 4-point scale; thus, ‘Paternal Care’, ‘Paternal Protection’, ‘Maternal Care’, and ‘Maternal Protection’ are scored. Care scores range from 0 to 36, with higher scores indicating accepting and affectionate attitudes, and lower scores indicating neglectful and rejecting attitudes. Protection scores range from 0 to 39, with higher scores indicating overprotective and overly interfering attitudes, and lower scores indicating encouragement of the child's independence. Kitamura and Suzuki translated the PBI into Japanese, back-translated it, and validated the translated scale in Japanese subjects by obtaining results similar to those in the original PBI validation studies.25

PBI quadrants using the established cut-off score are created from two factors (care and overprotection) that form the x- and y-axes in the quadrant.26 These four quadrants indicate Affectionless control (I), Affectionate constraint (II), Neglectful parenting (III), and Optimal parenting (IV) (Fig. 1). The dichotomizing care and overprotection scores are 27.0 and 13.5, respectively, for mothers, and 24.0 and 12.5, respectively, for fathers.27

image

Figure 1. Quadrants in the parental bonding instrument.

Download figure to PowerPoint

At the time the EDI and PBI were administered, the life history and details of the SIB were established through a set of original questions that we developed: (i) ‘Did your parents divorce before you were 15 years old?’ (ii) ‘Did you run away from home before you were 15 years old?’ (iii) ‘Have you often seen quarrels between your parents?’ (iv) ‘Was your family in financial difficulty before you were 15 years old?’ (v) ‘Did your parents use violence against you before you were 15 years old?’ (vi) ‘Did you engage in self-injurious behavior (scratching, bruising, cutting, burning, biting, pulling hair, picking skin) within the last month?’ (vii) ‘Which body parts did you mostly injure?’ (viii) ‘How often did you feel pain during this act?’ and (ix) ‘To what degree did you feel pain during this act?’

Statistical analysis

An unpaired t-test was performed for comparison of age, age at onset, and years of education between the ED + SIB and ED/no SIB groups. Fisher's exact test was performed for comparison of categorical variables, such as parent divorce between the two groups. anova was performed for comparison of continuous variables, such as PBI and EDI scores among the ED + SIB, ED/no SIB, and control groups. For comparisons with a significant difference, a Bonferroni post-hoc test or Dunnett's T3 test (when homogeneity could not be shown) was performed to establish where the differences lay. Fisher's exact test was also performed for comparison of the PBI quadrants (Fig. 1). P < 0.05 was taken to indicate significance in all tests, and all P-values were two-tailed. All analyses were performed using spss (version 17.0) for Windows (spss, Chicago, IL, USA).

RESULTS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES

EDI and PBI scores for the ED + SIB, ED/no SIB and control groups are shown in Table 4. On the EDI, ‘Total Score’ (P < 0.001), ‘Perfectionism’ (P < 0.001) and ‘Interoceptive Awareness’ (P < 0.001) differed among the three groups. Bonferroni post-hoc tests showed that the ‘Total Score’ was significantly higher for the ED + SIB group compared to the ED/no SIB (P < 0.005) and control (P < 0.001) groups, and for the ED/no SIB group compared to the control group (P < 0.005). The ‘Perfectionism’ score was significantly higher for the ED + SIB group compared to the ED/no SIB (P < 0.05) and control (P < 0.001) groups, and for the ED/no SIB group compared to the control group (P < 0.001). The ‘Interoceptive Awareness’ score was significantly higher for the ED + SIB group compared to the ED/no SIB (P < 0.01) and control (P < 0.001) groups, and for the ED/no SIB group compared to the control group (P < 0.001). The ‘Body Dissatisfaction’ (P < 0.01) and ‘Maturity Fear’ (P < 0.05) scores also differed among the three groups. Bonferroni post-hoc tests showed that the ‘Body Dissatisfaction’ score was significantly higher in the ED + SIB group compared to the ED/no SIB (P < 0.01) and control (P < 0.05) groups. The ‘Maturity Fear’ score was significantly higher in the ED + SIB group compared to the ED/no SIB (P < 0.05) and control (P < 0.05) groups, but did not differ between the ED/no SIB and control groups.

Table 4.  Comparisons of scores on the EDI and PBI
 ED + SIB (A) n = 25ED/no SIB (B) n = 55Control (C) n = 120Fd.f.PPost-hoc tests
  1. EDI, Eating Disorder Inventory; ED/no SIB, did not engage in SIB within the last month; ED + SIB, engaged in SIB within the last month; NS, not significant; PBI, Parental Bonding Instrument; Post-hoc test, Bonferroni post-hoc test, except for ‘Maternal Care’, which was evaluated using Dunnett's T3 test.

EDI       
 Total score113.9 ± 36.885.6 ± 36.963.5 ± 35.922.72,197<0.001A > B > C
 Perfectionism9.7 ± 4.17.0 ± 4.64.3 ± 4.120.62,197<0.001A > B > C
 Interoceptive awareness19.8 ± 8.314.1 ± 7.77.7 ± 7.7312,197<0.001A > B > C
 Body dissatisfaction19.2 ± 7.514.1 ± 6.315.1 ± 6.85.12,197<0.001A > B,C
 Maturity fear11.0 ± 5.37.7 ± 5.47.7 ± 5.24.42,197<0.05A > B,C
PBI       
 Paternal care17.6 ± 9.222.9 ± 9.023.7 ± 8.45.42,195<0.005A < B,C
 Paternal protection14.8 ± 8.511.4 ± 7.912.7 ± 7.11.72,195NSNS
 Maternal care24.4 ± 8.224.0 ± 10.929.0 ± 6.88.22,197<0.001A,B < C
 Maternal protection16.2 ± 9.912.5 ± 8.611.3 ± 7.03.92,197NSNS

On the PBI, the ‘Paternal Care’ (P < 0.005) and ‘Maternal Care’ (P < 0.001) scores differed among the three groups. A Bonferroni post-hoc test showed that the ‘Paternal Care’ score was significantly lower in the ED + SIB group compared to the ED/no SIB (P < 0.05) and control (P < 0.005) group, but did not differ between the ED/no SIB and control groups. Dunnett's T3 test showed that the ‘Maternal Care’ score was significantly higher in the control group compared to the ED + SIB (P < 0.05) and ED/no SIB (P < 0.001) groups, but did not differ between the ED + SIB and ED/no SIB groups.

Results for the PBI quadrants (Fig. 1) are shown in Table 5. A Fisher's exact test indicated a difference between the ED + SIB and ED/no SIB groups for paternal bonding (P < 0.05), but not for maternal bonding.

Table 5.  Scores for quadrants on the Parental Bonding Instrument (PBI)
ParentPBI quadrantED + SIBED/no SIB
n%n%
  • Fisher's exact test: significant difference for ED + SIB versus ED/no SIB, P < 0.05.

  • ED/no SIB, did not engage in SIB within the last month; ED + SIB, engaged in SIB within the last month.

FatherI Affectionless control1354.2%1018.5%
II Affectionate constraint312.5%814.8%
III Neglectful parenting625.0%1935.2%
IV Optimal parenting28.3%1731.5%
MotherI Affectionless control1144.0%1629.1%
II Affectionate constraint312.0%59.1%
III Neglectful parenting520.0%1221.8%
IV Optimal parenting624.0%2240.0%

DISCUSSION

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES

The results of the study indicated that patients with ED and SIB had a more severe ED pathology than those with ED without SIB. This result is similar to the association of higher impulsivity with severity of ED symptoms found by Favaro.28 The EDI score for ineffectiveness in BN patients with multi-impulsive types of behavior (three or more) has been reported to be higher than that in BN patients with two or fewer impulsive behavior types.29 Ross et al. also found that high-school students who engage in non-suicidal self-injury (NSSI) display significantly more eating pathologies than their non-NSSI peers.30

The results of this study differ from other studies focusing on recent self-injuries in the last month. On the EDI, body dissatisfaction and maturity fear were significantly higher in these patients than in ED patients without SIB and controls. As far as we are aware, the EDI has not been used previously to compare ED with and without SIB. It is possible that our results indicate that patients with ED and SIB loathe their body and reject becoming adults through self-injury. These may be some aspects of the functions of SIB. Patients with ED and SIB, and other kinds of psychiatric patients with SIB, are understood not to care about their bodies and to replace negative affection with psychical pain. Our results are consistent with this view. The ED + SIB and ED/no SIB groups both had lower interoceptive awareness, but this was significantly lower in the ED + SIB group. This low interoceptive awareness may contribute to other psychopathology, such as a twilight or numb state. This may explain the different degrees of the ‘painless’ state shown in Table 3. The high perfectionism in patients with ED and SIB also suggests that compulsivity might also be included in SIB. This result is consistent with previous findings that patients with BN and SIB are significantly more likely to have comorbid obsessive–compulsive disorder compared to BN patients without SIB.31

The PBI score showed that patients with ED and SIB perceived that they received low paternal care, with more than half of these patients feeling that their father exhibited ‘Affectionless control’. We also investigated the difference in parental care scores between restricting type AN patients and others because of the difference in the number of patients with restricting type AN who did and did not engage in SIB. However, there was no difference in parental care scores between these subgroups, which indicates that differences in ED subtypes are unlikely to influence the paternal care score. It is also of note that maternal care scores for ED patients with and without SIB were lower than for controls, whereas the paternal care score did not differ significantly between ED patients without SIB and controls. This suggests that ED patients without SIB mainly have conflict with their mother, whereas ED patients with SIB also have conflict with their father. These results support the hypothesis that the family environment of ED patients with SIB is more complicated than that of ED patients without SIB.

ED studies have focused on family or parental bonding as the cause of the disease, but the role of the family in treatment is more important. Our results indicate the need for participation of a patient's father in this treatment. ED patients are generally thought to have conflict with their mother, with their father needing to contain the complicated mother–child relationship or promote separation when they are deadlocked.32 In treatment, ED patients tend to focus on food or weight. It is often hard to make progress with the treatment, as the treatment relationship is likely to be superficial and patients readily drop out. Our results suggest that development of insights into the paternal relationship may provide effective treatment for patients with ED and SIB. We suggest that therapy for patients with ED should vary depending on the presence or absence of SIB, and that further assessment of this therapy is required in patients with active SIB.

This is the first study in Japan to examine ED patients with SIB. Previous Japanese studies have examined psychological traits in patients with impulsive behavior or in those who made suicide attempts,29,33 but not in patients with ED and SIB. Furthermore, most previous Japanese studies on SIB have had important limitations. First, the timeframe of the SIB has not been established, as the studies have used questions such as ‘Have you ever committed self-harm behavior?’ The PBI score also changes from before to after treatment;1,34 therefore, assessment of the score for ED patients with recent SIB is required. A second problem is the ambiguity of the definition of SIB. Since previous research has shown that SIB differs from other general self-harm or suicide attempts,18,20 we specified both the timing and manner of the SIB in this study.

The current study also has several important limitations. First, we were unable to investigate ED subtypes and the SIB type (impulsive or compulsive) because of the small number of subjects. In BN patients, Favaro and Santonasstaso17 classified SIB into ‘impulsive SIB’, which included skin cutting, burning, suicide attempts, substance/alcohol abuse, and laxative/diuretics abuse; and ‘compulsive SIB’, which included hair pulling, severe nail biting, and self-induced vomiting. Impulsive SIB is repetitive but egodystonic, with patients trying to resist the urge for the behavior. On the other hand, compulsive SIB is more episodic, but gratifying and egosyntonic, and is often triggered by external events and involves little resistance on the part of the patient. The difference between the two types of SIB depends on personality characteristics. Furthermore, Maggini suggested that personality characteristics in ED with SIB differed for the two types of SIB.35 Likewise, the PBI is linked to the psychological defense style in ED.36 Therefore, perception of parental bonding may also differ for the two types of SIB. Thus, the possible influence of ED subtypes on SIB remains to be examined in a further study in a larger population that is categorized based on ED subtype and the type of SIB. Second, we did not consider comorbid psychiatric states in this study. A previous study showed that depression is not correlated with PBI score,37 but further research on the correlation between comorbid psychiatric states and PBI scores is required.

ACKNOWLEDGMENTS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES

The authors wish to thank the women who participated in this research. We are also grateful to Drs Kiwamu Yasuda, Tatsuo Ando, Hiroyuki Moritoki, Yuki Mizuhara and Hisayo Kudo for their support of this work.

REFERENCES

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES
  • 1
    Takahashi S. A study on appraisals of parental rearing attitude in eating disorders as measured by Parental Bonding Instrument (PBI). Jpn J. Clin. Psychiatry 1994; 23: 10351046 (in Japanese).
  • 2
    Hadano M. Family relationships of patients with eating disorders examined by the Parental Bonding Instrument and Family Relationship Inventory. Jpn J. Psychosom. Med. 1998; 38: 511522 (in Japanese).
  • 3
    Yamaguchi N. Relationship between parenting and severity in eating disorder: measured by the Parental Bonding Instrument. Jpn J. Clin. Psychiatry 1999; 28: 11191126 (in Japanese).
  • 4
    Pattison EM, Kahan J. The deliberate self-harm syndrome. Am. J. Psychiatry 1983; 140: 867872.
  • 5
    Lacey JH, Evans CD. The impulsivist: a multi-impulsive personality disorder. Br. J. Addict. 1986; 81: 641649.
  • 6
    Favazza AR, DeRosear L, Conterio K. Self-mutilation and eating disorders. Suicide Life Threat. Behav. 1989; 19: 352361.
  • 7
    Svirko E, Hawton K. Self-injurious behavior and eating disorders: the extent and nature of the association. Suicide Life Threat. Behav. 2007; 37: 409421.
  • 8
    Simpson CA, Porter GL. Self-mutilation in children and adolescents. Bull. Menninger Clin. 1981; 45: 428438.
  • 9
    Carroll J, Schaffer C, Spensley J, Abramowitz SI. Family experiences of self-mutilating patients. Am. J. Psychiatry 1980; 137: 852853.
  • 10
    Matsumoto T, Azekawa T, Yamaguchi A, Asami T, Iseki E. Habitual self-mutilation in Japan. Psychiatry Clin. Neurosci. 2004; 58: 191198.
  • 11
    Claes L, Vandereycken W, Vertommen H. Family environment of eating disordered patients with and without self-injurious behaviors. Eur. Psychiatry 2004; 19: 494498.
  • 12
    Favazza AR, Conterio K. Female habitual self-mutilators. Acta Psychiatr. Scand. 1989; 79: 283289.
  • 13
    Gunderson JG, Zanarini MC. Current overview of the borderline diagnosis. J. Clin. Psychiatry 1987; 48 (Suppl.): 514.
  • 14
    Westen D, Harnden-Fischer J. Personality profiles in eating disorders: rethinking the distinction between axis I and axis II. Am. J. Psychiatry 2001; 158: 547562.
  • 15
    Klump KL, Bulik CM, Pollice C et al. Temperament and character in women with anorexia nervosa. J. Nerv. Ment. Dis. 2000; 188: 559567.
  • 16
    Claes L, Vandereycken W, Vertommen H. Impulsive and compulsive traits in eating disordered patients compared with controls. Pers. Individ. Dif. 2002; 32: 707714.
  • 17
    Favaro A, Santonastaso P. Impulsive and compulsive self-injurious behavior in bulimia nervosa: prevalence and psychological correlates. J. Nerv. Ment. Dis. 1998; 186: 157165.
  • 18
    Farberow NL. The Many Faces of Suicide: Indirect Self-Destructive Behavior. McGraw-Hill, New York, 1980.
  • 19
    Favazza AR. The coming of age of self-mutilation. J. Nerv. Ment. Dis. 1998; 186: 259268.
  • 20
    Claes L, Vandereycken W. The Self-Injury Questionnaire-Treatment Related (SIQ-TR): Construction, Reliability, and Validity in A Sample of Female Eating Disorder Patients. Nova Science Publishers, New York, 2007; 111139.
  • 21
    Garner D, Olmstead M, Polivy J. Development and validation of a multi-dimensional eating disorder inventory for anorexia nervosa and bulimia. Int. J. Eat. Disord. 1983; 2: 1534.
  • 22
    Nagata T, Kiriike N, Matsunaga H, Iketani T, Yoshida M, Yamagami S. Clinical trial of the Eating Disorder Inventory (EDI) in patients with eating disorders. Jpn J. Clin. Psychiatry 1994; 23: 897903 (in Japanese).
  • 23
    Yamagami T, Moriguchi Y, Okawa A. Psychological characteristics associated with eatng disorders: comparison among eating disorder subtypes and healthy females. Jpn J. Clin. Psychiatry 2004; 33: 931938 (in Japanese).
  • 24
    Parker G, Tupling M, Brown C. A parental bonding instrument. Br. J. Med. Psychol. 1979; 52: 110.
  • 25
    Kitamura T, Suzuki T. A validation study of the Parental Bonding Instrument in a Japanese population. Jpn J. Psychiatry Neurol. 1993; 47: 2936 (in Japanese).
  • 26
    Parker G. Parental ‘affectionless control’ as an antecedent to adult depression. A risk factor delineated. Arch. Gen. Psychiatry 1983; 40: 956960.
  • 27
    Parker G. Parental characteristics in relation to depressive disorders. Br. J. Psychiatry 1979; 134: 138147.
  • 28
    Favaro A, Zanetti T, Tenconi E et al. The relationship between temperament and impulsive behaviors in eating disordered subjects. Eat. Disord. 2005; 13: 6170.
  • 29
    Nagata T, Kawarada Y, Kiriike N, Iketani T. Multi-impulsivity of Japanese patients with eating disorders: primary and secondary impulsivity. Psychiatry Res. 2000; 94: 239250.
  • 30
    Ross S, Heath NL, Toste JR. Non-suicidal self-injury and eating pathology in high school students. Am. J. Orthopsychiatry 2009; 79: 8392.
  • 31
    Anderson CB, Carter FA, McIntosh VV, Joyce PR, Bulik CM. Self-harm and suicide attempts in individuals with bulimia nervosa. Eat. Disord. 2002; 10: 227243.
  • 32
    Matsuki K. Therapeutic Technique for Eating Disorder. Kongo Shuppan, Tokyo, 1997 (in Japanese).
  • 33
    Yamaguchi N, Kobayashi J, Tachikawa H et al. Adverse parenting as a risk factor for suicide attempts among eating disorder patients: measurement by the Parental Bonding Instrument. Jpn J. Psychosom. Med. 2000; 40: 2532 (in Japanese).
  • 34
    Takahashi S. Parental Bonding Instrument (PBI) and eating disorders. Jpn J. Seisinka-Shindangaku 1999; 10: 417427 (in Japanese).
  • 35
    Maggini C, Ampolline P, Marchesi C, Gariboldi S, Cloninger CR. Relationship between tridimensional personality questionnaire dimensions and DSM-III-R personality traits in Italian adolescents. Compr. Psychiatry 2000; 41: 426431.
  • 36
    Steiger H, Van der Feen J, Goldstein C, Leichner P. Defense styles and parental bonding in eating disordered women. Int. J. Eat. Disord. 1989; 8: 131140.
  • 37
    Pole R, Waller DA, Stewart SM, Feigenbaum LP. Parental caring versus overprotection in bulimia. Int. J. Eat. Disord. 1988; 7: 601606.