EMPIRICAL ARTICLE
Bulimia Nervosa-Nonpurging Subtype: Closer to the Bulimia Nervosa-Purging Subtype or to Binge Eating Disorder? Jennifer Jordan, PhD1,2* Virginia V.W. McIntosh, PhD1,2 Janet D. Carter, PhD3 Sarah Rowe, PhD, MED4 Kathryn Taylor, Dip Clin Psyc, MSc1 Christopher M.A. Frampton, PhD1 Janice M. McKenzie, MB ChB, FRANZCP1 Janet Latner, PhD5 Peter R. Joyce, MD, PhD, DSc, FRSNZ, FRANZCP1
Abstract Objective: DSM-5 has dropped subtyping of bulimia nervosa (BN), opting to continue inclusion of the somewhat contentious diagnosis of BN-nonpurging subtype (BN-NP) within a broad BN category. Some contend however that BN-NP is more like binge eating disorder (BED) than BN-P. This study examines clinical characteristics, eating disorder symptomatology, and Axis I comorbidity in BN-NP, BN-P, and BED groups to establish whether BN-NP more closely resembles BN-P or BED. Method: Women with BN-P (n 5 29), BN-NP (n 5 29), and BED (n 5 54) were assessed at baseline in an outpatient psychotherapy trial for those with binge eating. Measures included the Structured Clinical Interviews for DSM-IV, Eating Disorder Examination, and Eating Disorder Inventory-2. Results: The BN-NP subtype had BMIs between those with BN-P and BED. Both BN subtypes had higher Restraint and Drive for Thinness scores than BED. Body Dissatisfaction was highest in BN-NP and predicted BN-NP compared to BN-P.
Introduction Bulimia nervosa (BN) is a serious condition with significant morbidity and mortality.1 DSM-IV2 introduced subtyping of BN to capture compensatory behaviors used to prevent weight gain following binge episodes. Uncertainty about the validity of the BN-NP category led DSM-53 to Accepted 5 October 2013 Additional Supporting Information may be found in the online version of this article. *Correspondence to: Dr J. Jordan; Psychological Medicine, University of Otago, Christchurch, PO Box 4345, Christchurch 8140, New Zealand. E-mail: [email protected] 1 Department of Psychological Medicine, University of Otago, Christchurch, New Zealand 2 Canterbury District Health Board, Christchurch, New Zealand 3 Psychology Department, University of Canterbury, Christchurch, New Zealand 4 Kings College London, London, United Kingdom 5 Psychology Department, University of Hawaii, Manoa, Hawaii Published online 26 November 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/eat.22218 C 2013 Wiley Periodicals, Inc. V
International Journal of Eating Disorders 47:3 231–238 2014
Higher Restraint and lower BMI predicted BN-NP relative to BED. BN-NP resembled BED with higher lifetime BMIs; and weight-loss clinic than eating disorder clinic attendances relative to the BN-P subtype. Psychiatric comorbidity was comparable except for higher lifetime cannabis use disorder in the BN-NP than BN-P subtype Discussion: These results suggest that BN-NP sits between BN-P and BED however the high distress driving inappropriate compensatory behaviors in BN-P requires specialist eating disorder treatment. These results support retaining the BN-NP group within the BN category. Further research is needed to determine whether there are meaningful differences C 2013 Wiley in outcome over follow-up. V Periodicals, Inc. Keywords: bulimia nervosa; binge eating disorder; DSM-5; bulimia nervosa nonpurging; subtype; classification (Int J Eat Disord 2014; 47:231–238)
drop subtyping but retain BN-NP and BN-P within a broad BN category. Questions remain however about whether the nonpurging subtype is more closely related to BN-P or to binge eating disorder (BED). The DSM-IV purging subtype included those with regular (twice weekly on average for 3 months) vomiting, laxatives, abuse of diuretics, thyroid, or insulin medication or enemas. The nonpurging subtype included regular compensatory behaviors such as fasting or excessive exercise. BN-P might be considered to be more severe than BN-NP due to the purging-related morbidity and mortality due to physical complications such as electrolyte imbalance causing cardiac abnormalities, or to suicide.1,4,5 Those with purging BN may also use nonpurging compensatory behaviors such as fasting or excessive exercise, the risks of which are less well known. Excessive exercise has been associated with menstrual abnormalities, osteoporosis, hypercortisolaemia, and hypothalamic 231
JORDAN ET AL.
dysfunction related to insulin-induced hypoglycaemia, as well as depressive-like symptoms.6,7 BED has been promoted from the catch-all “eating disorders not otherwise specified” (EDNOS) category to full recognition in DSM-5. A DSM-IV diagnosis of BED requires regular objective binges, at least three of the following: eating much more rapidly than usual, eating until uncomfortably full, eating in the absence of hunger, eating alone because of embarrassment over how much is being eaten, feelings of disgust, depression or guilt after overeating; marked distress associated with binges and no regular compensatory behaviors.8 DSM-5 modified frequency of binge episodes and duration of disorder criteria from on average at least 2 days a week for 6 months, to on average once weekly for three months, matching the duration and revised frequency of binge eating for BN in DSM-5. Despite recognition of overlapping eating psychopathology and associated features, there is reasonable consensus regarding the validity and utility of BN-P and BED as discrete disorders. There is less consensus regarding BN-NP, with competing views on its validity, utility and boundaries with BN-P and BED (see9,10), and debate about whether it is really a variant of BN or of BED. Ramacciotti and colleagues concluded that although some distinct features of BN-NP exist, given the larger number of similarities with BED, there seemed little benefit in maintaining the BN-NP category.11 Hay and Fairburn noted difficulties discriminating BN-NP and BED in a “present state” assessment but found differences in 1-year outcome,12 concluding that BNNP sits between BN-P and BED in severity. Uncertainty regarding definitional and threshold issues related to dieting/fasting and excessive exercise may have limited the clinically utility of BNNP.13 Some may have been diagnosed with EDNOS or BED if nonpurging compensatory behaviors were not fully assessed.10 A review justifying changes in DSM-5 examined studies comparing BN-NP to either BN-P or BED.14 Overall, the number of BNNP participants was much lower than the BN-P or BED groups limiting power to detect differences. Although the authors considered that use of structured diagnostic assessments might increase prevalences, they concluded there was little evidence for the validity and utility of the BN-NP subtype, echoing earlier comments by Wilfley and colleagues.15 If the function of diagnostic criteria is to provide valid descriptions of specific conditions to facilitate communication, convey accurate information about aetiology, associated difficulties, course, and to inform treatment, then doubts about the validity n ~ ezof a diagnosis undermine these functions. Nu 232
Navarro and colleagues note that BN-NP has been relatively under-researched, with few direct comparisons with BN-P or BED addressing a wide range of demographic, clinical and psychiatric comorbidity variables which might characterise subgroups.9 Although there are some generally consistent findings, for example, of increased age and higher weights in BED than BN,9,16 there are also inconsistent findings. Further research is needed to clarify uncertainty regarding possible defining features of BN-NP compared with BN-P and BED. This study examines whether the DSM-IV BN-NP subtype relates more closely to the BN-P subtype (as in DSM-5), or to BED in clinical characteristics, eating disorder symptomatology and psychiatric comorbidity in a treatment seeking sample of women with regular binge eating.
Method Participants Participants were assessed on entering a randomised controlled trial17 of three psychotherapies for binge eating: cognitive-behavior therapy (CBT),18 appetitefocussed CBT,19 and schema therapy.20 The trial had ethical approval from the Upper South A Regional Ethics Committee, New Zealand. Participants provided written informed consent. Participants were 112 women aged 16 years and older with a current primary DSM-IV diagnosis of BED (n 5 54), BN-P subtype (n 5 29) or BN-NP subtype (n 5 29). Exclusion criteria were current severe major depression, serious suicidal intent, current severe psychoactive substance dependence, cognitive impairment, bipolar I disorder, schizophrenia, severe physical illness, severe medical complications of BN or BED, on psychotropic medication, or a recent (past year) adequate trial of CBT or schema therapy. Recruitment included referrals from health professionals and self-referral. Measures Assessment data included demographic details (age, ethnicity, marital status, and education), and history of suicide attempts and self-harm. The Structured Clinical Interview for DSM-IV8 determined the presence of lifetime mood, anxiety, and substance-related disorders (SRD). Severity of current depressive symptoms was rated using the Montgomery and Asberg Depression Rating Scale (MADRS),21 and current (past month) psychosocial functioning using the Global Assessment of Functioning (GAF, Axis V of the DSM-IV8). Eating, mood, anxiety or substance abuse treatment history was International Journal of Eating Disorders 47:3 231–238 2014
BULIMIA NERVOSA NONPURGING SUBTYPE TABLE 1. groups
Demographic and clinical characteristics of bulimia nervosa (BN) subtypesa or binge eating disorder (BED) Effect sizes or Odds ratios (confidence intervals)
Mean (sd) or Median (range) or %
Age (years) Marital status Never married Education (years) Unemployed orsickness benefit GAFc MADRSd Prior suicide attempt Prior self-harm Any psychiatric or ED-clinic treatment Prior ED-clinic treatment Other psychiatric treatment Depression Anxiety Substancerelated Eating related self-help groupe Weight loss clinic
BN-Pa n 5 29
BN-NPb n 5 29
BEDc n 5 54
Statb
p
BN-P v BN-NP
BN-NP v BED
BN-P v BED
29.5 (12.5)
34.9 (10.8)
38.6 (12.6)
5.3
.006 ab, c
0.2 (0.1–0.7)
1.4 (0.4–4.4)
0.3 (0.1–0.7)
55%
69%
76%
3.8
.15
1.8 (0.6–5.3)
1.4 (0.5–3.9)
2.6 (1.0–6.7)
38% 10% 3%
52% 31% 14%
65% 7% 6%
5.6 9.0 2.7
.06 .01 b>c .26
1.8 (0.6–5.0) 3.9 (0.9–16.3) 4.5 (0.5–42.8)
1.7 (0.7–4.3) 0.2 (0.1–0.6) 0.4 (0.1–1.8)
3.0 (1.2–7.7) 0.7 (0.1–3.3) 1.7 (0.2–16.6)
41%
21%
35%
2.7
.26
0.4 (0.1–1.2)
2.0 (0.7–5.8)
0.8 (0.3–1.9)
35%
62%
69%
9.2
.01 a
Bulimia Nervosa-Nonpurging Subtype: Closer to the Bulimia Nervosa-Purging Subtype or to Binge Eating Disorder? Jennifer Jordan, PhD1,2* Virginia V.W. McIntosh, PhD1,2 Janet D. Carter, PhD3 Sarah Rowe, PhD, MED4 Kathryn Taylor, Dip Clin Psyc, MSc1 Christopher M.A. Frampton, PhD1 Janice M. McKenzie, MB ChB, FRANZCP1 Janet Latner, PhD5 Peter R. Joyce, MD, PhD, DSc, FRSNZ, FRANZCP1
Abstract Objective: DSM-5 has dropped subtyping of bulimia nervosa (BN), opting to continue inclusion of the somewhat contentious diagnosis of BN-nonpurging subtype (BN-NP) within a broad BN category. Some contend however that BN-NP is more like binge eating disorder (BED) than BN-P. This study examines clinical characteristics, eating disorder symptomatology, and Axis I comorbidity in BN-NP, BN-P, and BED groups to establish whether BN-NP more closely resembles BN-P or BED. Method: Women with BN-P (n 5 29), BN-NP (n 5 29), and BED (n 5 54) were assessed at baseline in an outpatient psychotherapy trial for those with binge eating. Measures included the Structured Clinical Interviews for DSM-IV, Eating Disorder Examination, and Eating Disorder Inventory-2. Results: The BN-NP subtype had BMIs between those with BN-P and BED. Both BN subtypes had higher Restraint and Drive for Thinness scores than BED. Body Dissatisfaction was highest in BN-NP and predicted BN-NP compared to BN-P.
Introduction Bulimia nervosa (BN) is a serious condition with significant morbidity and mortality.1 DSM-IV2 introduced subtyping of BN to capture compensatory behaviors used to prevent weight gain following binge episodes. Uncertainty about the validity of the BN-NP category led DSM-53 to Accepted 5 October 2013 Additional Supporting Information may be found in the online version of this article. *Correspondence to: Dr J. Jordan; Psychological Medicine, University of Otago, Christchurch, PO Box 4345, Christchurch 8140, New Zealand. E-mail: [email protected] 1 Department of Psychological Medicine, University of Otago, Christchurch, New Zealand 2 Canterbury District Health Board, Christchurch, New Zealand 3 Psychology Department, University of Canterbury, Christchurch, New Zealand 4 Kings College London, London, United Kingdom 5 Psychology Department, University of Hawaii, Manoa, Hawaii Published online 26 November 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/eat.22218 C 2013 Wiley Periodicals, Inc. V
International Journal of Eating Disorders 47:3 231–238 2014
Higher Restraint and lower BMI predicted BN-NP relative to BED. BN-NP resembled BED with higher lifetime BMIs; and weight-loss clinic than eating disorder clinic attendances relative to the BN-P subtype. Psychiatric comorbidity was comparable except for higher lifetime cannabis use disorder in the BN-NP than BN-P subtype Discussion: These results suggest that BN-NP sits between BN-P and BED however the high distress driving inappropriate compensatory behaviors in BN-P requires specialist eating disorder treatment. These results support retaining the BN-NP group within the BN category. Further research is needed to determine whether there are meaningful differences C 2013 Wiley in outcome over follow-up. V Periodicals, Inc. Keywords: bulimia nervosa; binge eating disorder; DSM-5; bulimia nervosa nonpurging; subtype; classification (Int J Eat Disord 2014; 47:231–238)
drop subtyping but retain BN-NP and BN-P within a broad BN category. Questions remain however about whether the nonpurging subtype is more closely related to BN-P or to binge eating disorder (BED). The DSM-IV purging subtype included those with regular (twice weekly on average for 3 months) vomiting, laxatives, abuse of diuretics, thyroid, or insulin medication or enemas. The nonpurging subtype included regular compensatory behaviors such as fasting or excessive exercise. BN-P might be considered to be more severe than BN-NP due to the purging-related morbidity and mortality due to physical complications such as electrolyte imbalance causing cardiac abnormalities, or to suicide.1,4,5 Those with purging BN may also use nonpurging compensatory behaviors such as fasting or excessive exercise, the risks of which are less well known. Excessive exercise has been associated with menstrual abnormalities, osteoporosis, hypercortisolaemia, and hypothalamic 231
JORDAN ET AL.
dysfunction related to insulin-induced hypoglycaemia, as well as depressive-like symptoms.6,7 BED has been promoted from the catch-all “eating disorders not otherwise specified” (EDNOS) category to full recognition in DSM-5. A DSM-IV diagnosis of BED requires regular objective binges, at least three of the following: eating much more rapidly than usual, eating until uncomfortably full, eating in the absence of hunger, eating alone because of embarrassment over how much is being eaten, feelings of disgust, depression or guilt after overeating; marked distress associated with binges and no regular compensatory behaviors.8 DSM-5 modified frequency of binge episodes and duration of disorder criteria from on average at least 2 days a week for 6 months, to on average once weekly for three months, matching the duration and revised frequency of binge eating for BN in DSM-5. Despite recognition of overlapping eating psychopathology and associated features, there is reasonable consensus regarding the validity and utility of BN-P and BED as discrete disorders. There is less consensus regarding BN-NP, with competing views on its validity, utility and boundaries with BN-P and BED (see9,10), and debate about whether it is really a variant of BN or of BED. Ramacciotti and colleagues concluded that although some distinct features of BN-NP exist, given the larger number of similarities with BED, there seemed little benefit in maintaining the BN-NP category.11 Hay and Fairburn noted difficulties discriminating BN-NP and BED in a “present state” assessment but found differences in 1-year outcome,12 concluding that BNNP sits between BN-P and BED in severity. Uncertainty regarding definitional and threshold issues related to dieting/fasting and excessive exercise may have limited the clinically utility of BNNP.13 Some may have been diagnosed with EDNOS or BED if nonpurging compensatory behaviors were not fully assessed.10 A review justifying changes in DSM-5 examined studies comparing BN-NP to either BN-P or BED.14 Overall, the number of BNNP participants was much lower than the BN-P or BED groups limiting power to detect differences. Although the authors considered that use of structured diagnostic assessments might increase prevalences, they concluded there was little evidence for the validity and utility of the BN-NP subtype, echoing earlier comments by Wilfley and colleagues.15 If the function of diagnostic criteria is to provide valid descriptions of specific conditions to facilitate communication, convey accurate information about aetiology, associated difficulties, course, and to inform treatment, then doubts about the validity n ~ ezof a diagnosis undermine these functions. Nu 232
Navarro and colleagues note that BN-NP has been relatively under-researched, with few direct comparisons with BN-P or BED addressing a wide range of demographic, clinical and psychiatric comorbidity variables which might characterise subgroups.9 Although there are some generally consistent findings, for example, of increased age and higher weights in BED than BN,9,16 there are also inconsistent findings. Further research is needed to clarify uncertainty regarding possible defining features of BN-NP compared with BN-P and BED. This study examines whether the DSM-IV BN-NP subtype relates more closely to the BN-P subtype (as in DSM-5), or to BED in clinical characteristics, eating disorder symptomatology and psychiatric comorbidity in a treatment seeking sample of women with regular binge eating.
Method Participants Participants were assessed on entering a randomised controlled trial17 of three psychotherapies for binge eating: cognitive-behavior therapy (CBT),18 appetitefocussed CBT,19 and schema therapy.20 The trial had ethical approval from the Upper South A Regional Ethics Committee, New Zealand. Participants provided written informed consent. Participants were 112 women aged 16 years and older with a current primary DSM-IV diagnosis of BED (n 5 54), BN-P subtype (n 5 29) or BN-NP subtype (n 5 29). Exclusion criteria were current severe major depression, serious suicidal intent, current severe psychoactive substance dependence, cognitive impairment, bipolar I disorder, schizophrenia, severe physical illness, severe medical complications of BN or BED, on psychotropic medication, or a recent (past year) adequate trial of CBT or schema therapy. Recruitment included referrals from health professionals and self-referral. Measures Assessment data included demographic details (age, ethnicity, marital status, and education), and history of suicide attempts and self-harm. The Structured Clinical Interview for DSM-IV8 determined the presence of lifetime mood, anxiety, and substance-related disorders (SRD). Severity of current depressive symptoms was rated using the Montgomery and Asberg Depression Rating Scale (MADRS),21 and current (past month) psychosocial functioning using the Global Assessment of Functioning (GAF, Axis V of the DSM-IV8). Eating, mood, anxiety or substance abuse treatment history was International Journal of Eating Disorders 47:3 231–238 2014
BULIMIA NERVOSA NONPURGING SUBTYPE TABLE 1. groups
Demographic and clinical characteristics of bulimia nervosa (BN) subtypesa or binge eating disorder (BED) Effect sizes or Odds ratios (confidence intervals)
Mean (sd) or Median (range) or %
Age (years) Marital status Never married Education (years) Unemployed orsickness benefit GAFc MADRSd Prior suicide attempt Prior self-harm Any psychiatric or ED-clinic treatment Prior ED-clinic treatment Other psychiatric treatment Depression Anxiety Substancerelated Eating related self-help groupe Weight loss clinic
BN-Pa n 5 29
BN-NPb n 5 29
BEDc n 5 54
Statb
p
BN-P v BN-NP
BN-NP v BED
BN-P v BED
29.5 (12.5)
34.9 (10.8)
38.6 (12.6)
5.3
.006 ab, c
0.2 (0.1–0.7)
1.4 (0.4–4.4)
0.3 (0.1–0.7)
55%
69%
76%
3.8
.15
1.8 (0.6–5.3)
1.4 (0.5–3.9)
2.6 (1.0–6.7)
38% 10% 3%
52% 31% 14%
65% 7% 6%
5.6 9.0 2.7
.06 .01 b>c .26
1.8 (0.6–5.0) 3.9 (0.9–16.3) 4.5 (0.5–42.8)
1.7 (0.7–4.3) 0.2 (0.1–0.6) 0.4 (0.1–1.8)
3.0 (1.2–7.7) 0.7 (0.1–3.3) 1.7 (0.2–16.6)
41%
21%
35%
2.7
.26
0.4 (0.1–1.2)
2.0 (0.7–5.8)
0.8 (0.3–1.9)
35%
62%
69%
9.2
.01 a
