Eating Disorders in Children and Teens with Type 1 Diabetes 1984ongoing

1
Eating Disorders in Children and Teens with Type
1 Diabetes 1984-ongoing

• Denis Daneman
• University of Toronto
• And
• The Hospital for Sick Children

2
ED Classification

• Clinical/full-blown DSM-lV
• Anorexia nervosa
• Bulimia nervosa
• Eating Disorder Not Otherwise Specified (EDNOS)
• Subthreshold (not subclinical)
• Disturbed Eating Behavior that does not meet
  criteria for full-blown ED, but with clinical
  consequences (e.g. A1c, complications)

3
Working Model Rodin Daneman 1992
Individual, family, and societal factors
Diabetes-specific vulnerabilities Insulin-relate
d weight gain Nutritional counseling Poor
self-esteem
Eating Disorders Core Features Body
dissatisfaction Drive for thinness Dietary
restraint
Disordered eating attitudes and behavior Insulin
omission Binge eating Dieting
Diabetes-specific outcomes Poor metabolic
control high HbA1c Microvascular
complications, e.g., retinopathy
4
Predictions arising from our model

• Prevalence
• Natural history
• Associated with
• poorer control
• specific behavior, especially insulin omission
• early complications
• specific family issues
• Difficult to treat

5
Jones et al, BMJ 2000 DSM-IV diagnosable ED

• Controls 1098 (31)
• 49 (4)
• 0 (0) NS
• 5 (0.5) NS
• 44 (4.0)
• OR 2.4
• (1.5-3.7)

• DM 356
• DSM-IV 36 (10)
• AN 0 (0)
• BN 5 (1.4)
• NOS 31 (8.7)

6
Jones et al, 2000 Subthreshold Disorders

• DM 356
• 49 (14)
• DSM-IV ST
• 85 (24)

• Controls 1098
• 84 (8)
• OR 1.9
• (1.3-2.8)
• 134 (12)
• OR 2

7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10
(No Transcript)
11
Obstacles identified during initial assessment
12
HbA1c by Disordered Eating Status at Baseline and
Follow-up.
HbA1c ()
HbA1c for the highly disordered group was
significantly higher than the moderately and
non-disordered groups at baseline,
pdisordered groups was significantly higher than
the non-disordered group at follow-up, p(Rydall et al., 1997).
13

Common behaviors in girls with type 1 diabetes.
Percentage of Sample
Binge eating
Insulin omission
Self- induced vomiting
Laxative use
Dieting
McNemars test for change in prevalence, baseline
to follow-up p0.01 p0.003 p.06
(Rydall et al., 1997).
14
Age and Prevalence of Insulin Omission for
Weight Control.
Insulin Omission
Prevalence of Insulin Omission ()
12-18 years
16-22 years
9-13 years
1Colton et al., 2000 (n90) 1 prevalence of
insulin omission in pre-teen girls 2 Rydall et
al., 1997 (n91) 14 in adolescent girls
(baseline assessment) 3 Rydall et al., 1997
(n91) 34 in young adult women (four-year
follow-up of baseline sample).
15
(No Transcript)
16
Evolution of ED in teen girls with T1D

• In progress, study of natural history of ED in
  girls with T1D
• Baseline 101 9-13.9 yo with T1D 303 controls
• Follow-up of DM cohort for 5-8 years
• Demographics at Baseline
• Mean age 11.8 years
• Mean A1c 8.2
• Mean duration of T1D 4.7 years

17
DISTURBED EATING BEHAVIOUR (last month)


18
EATING DISORDERS T1D VS. SCHOOL GIRLS
sample
p .001

• No sign differences in
• Age
• A1c
• Duration of T1D
• Those with ED BMI
• those without

19
FIVE-YEAR FOLLOW-UP

• 13.3 of participants (13/98) met criteria for an
  ED
• 3 girls had bulimia nervosa
• 3 had ED-NOS
• 7 had a subthreshold ED
• 44.9 of participants were classified as
  overweight or obese

20
FIVE-YEAR FOLLOW-UP

• A1c not higher in girls with DEB
• (8.7 vs. 8.4 p 0.11)
• Trend for higher A1c in those with an ED
• (9.1 vs. 8.5 p 0.08)
• BMI higher in those with DEB
• (26.1 versus 23.5 p 0.001)

21
ED POINT PREVALENCE CUMULATIVE PREVALENCE BY AGE
22
FIVE-YEAR FOLLOW-UP

• Higher BMI and DEB were strongly associated,
  which presents a management dilemma
• Both dietary restraint and higher weight are risk
  factors for the development of ED and their
  negative health consequences

23

• PREDICTION OF THE ONSET OF DISTURBED EATING
  BEHAVIOUR IN ADOLESCENT GIRLS WITH TYPE 1 DIABETES

24
LOGISTIC REGRESSION MODEL WITH BACKWARD STEPWISE
REGRESSION

• Dietary Restraint
• Weight Shape Concern Physical Appearance
• Self-Worth
• Depression
• X2 43.254, df 5, p
• McFaddens R2 0.416

25
If the model is correct, then the prevalence of
complications should be more common in ED
Percentage of Sample
MicroAlbuminuria
(Rydall et al., NEJM 1997).
26
(No Transcript)
27
Is family dysfunction more common in ED than
nonED DM

• To investigate if and how eating disturbances in
  girls with type 1 DM are associated with
• Mothers weight and shape concerns
• Mother-daughter relationships
• Adolescent self-concept

28
Method

• TEENS (N88)
• Age 14.9 yrs. ( 2.2)
• Weight 58.9 kg (12.7)
• BMI 22.4 kg/m2 (3.7)
• Age of Diabetes Onset 7.9 yrs ( 4.0)
• Illness Duration 7.1yrs (3.9)
• HbA1c 8.9 ( 1.6)

• MOTHERS (N88)
• Age 43.7 yrs ( 5.5)
• Weight 69.3 (13.7)
• BMI 25.9 (4.9)
• Middle Class
• Completed 1-2 years of college, university, or
  specialized training

29
Perceived Relationships With Mothers

• Multivariate Group effect
• F(6, 160 ) 3.97,
• p .001
• Highly Mildly Disturbed girls report
  more impaired relations with mothers on all
  dimensions compared
• to Non-Disturbed girls
• (p .01)

Communic. Trust Alienation
30
(No Transcript)
31
Mothers Eating and Weight Loss Behaviors

• Multivariate Group effect F (10, 138)
  2.12, p .03
• Mothers of Highly Mildly Disturbed girls are
  more weight dissatisfied (p .01) and are more
  likely to exercise for weight control (p
  .02), diet (p .05), and binge eat (p .02).

Satisfaction Diet Exercise
Binge
32
Prevention and Treatment in DM and ED

• Prevention not reported
• Treatment
• CBT - Peveler and Fairburn 1989
• Fluoxetine - case report - 1990
• Psychoeducation - Olmsted 2000

33
Evidence-based conclusionsModel validation

• Eating disorders are more common in adolescent
  and young adult females with diabetes (Level 1)
• When present they are associated with
• high frequency of insulin omission (Level 1)
• worse metabolic control (Level 2)
• earlier onset of complications (Level 1)
• family dysfunction (Level 2)
• They are (more) difficult to treat (Level 4)

34
Approach to ED in DM

• Awareness of the association
• Ask the right questions
• If suspect fullblown ED - refer
• If subthreshold - clinic-based intervention
• Complication surveillance

35
Asking the right questions

• Red flags
• Dealing with reluctance to disclose
• Their stories
• Partnering with patients
• Regaining control
• Treatment options

36
Red Flags

• Persistently high A1c
• Frequent DKA, illnesses
• Distress re weight
• Widely fluctuating b.g.s
• Skipping meals
• Binging feeling hungry all the time
• Skipping dosing/underdosing

37
Initial response to high A1c

• Raise the dose
• Labeled insulin resistent
• Problem insulin avoidant

38
Disclosure is very difficult

• Shame
• Feel like failures
• Failed
• Their families
• Their providers
• Themselves
• Important to be nonjudgemental and supportive

39
In their words.
40
A start.

• Information can be helpful
• Unfortunately something many young people
  struggle with
• Insulin omission drives hunger
• Losing control over eating behavior
• Information for parents
• Families are angry, blaming
• They feel like failures too

41
Regaining control

• A step at a time
• Steps forward, steps back
• Treatment options
• Partner with existing ED programs
• Requires collaboration
• Groups
• Conventional treatment
• Medication/Consult