Eating Disorders Exam 3

ventral medial hypothalamus (VMH)

• helps you know when you’re full

  • lesions (destruction) cause overeating/elevated weight

  • stimulation causes starvation

lateral hypothalamus (LH)

• helps you know when you’re hungry

  • lesions (destruction) cause starvation

  • stimulation causes overeating/elevated weight

brain activity and neurotransmitters

• within brain cells (neurons), signals are carried electronically

• between neurons, signals are transmitted by chemicals (NTs)

  • serotonin (5-HT)

  • dopamine

  • norepinephrine

serotonin/EDs

• serotonin involved in regulation of appetite, mood, impulse control

  • decreases food intake/weight

• serotonin hypothesis of EDs

  • AN caused by over-function of serotonin

  • BN caused by under-function of serotonin

• studies:

  • studies DON’T support serotonin hypothesis for AN

    • evidence showed serotonin function is reduced in AN

  • studies DO support serotonin hypothesis for BN

    • higher binge frequency associated w/ lower cerebrospinal fluid serotonin levels

leptin

• reduces food intake

• genetic mutation makes mice unable to make leptin

  • overeat, low metabolism, weight gain, development of diabetes in adulthood (impacts of not being able to make leptin)

• impacts in EDs

  • BN patients have lower levels of leptin compared to controls

    • consequence of weight suppression

  • AN patients have lower levels of leptin compared to controls

    • consequence of low weight

ghrelin

• “hunger hormone”

  • results in increased food intake

• functions in EDs compared to controls

  • AN → elevated

  • BN → elevated

  • BED → mixed

  • PD → elevated

cholecystokinin (CCK)

• regulates gastric emptying and induces satiety

• function in EDs compared to controls

  • AN → mixed

  • BN → lower

  • BED → normal

  • PD → normal

glucagon-like peptide 1 (GLP-1)

• stimulates insulin secretion

• function in EDs compared to controls

  • AN → mixed

  • BN → lower

  • BED → normal

  • PD → normal

peptide tyrosine tyrosine (PYY)

• “satiety hormone”

• function in EDs compared to controls

  • AN → mixed

  • BN → nixed

  • BED → mixed

  • PD → elevated

twin studies

• MZ

• DZ

• results:

  • concordance (shared genes) for EDs is higher in MZ compared to DZ teens post-pubertally

    • one twin most likely to have ED after puberty if other twin has ED

  • twins are not at increased risk for EDs

steiger et al. (2011)

• if certain alleles/genotypes are most common in ED participants than controls, then this allele/genotype may increase risk for ED

• background

  • women w/ BN more likely to report histories of sexual/physical abuse

  • women w/ BN have higher cortisol levels, suggesting HPA-axis dysfunction

  • the C-allele (compared to G allele) of the Bcl 1 gene has been linked to higher cortisol response following a stressor via its influence on glucocorticoid receptors

• hypothesis:

  • C allele of the Bcl 1 gene represents a diathesis (vulnerability) and abuse represents a stress that causes BN

  • there should be a significant interaction between history of abuse/Bcl 1 gene in predicting presence vs. absence of BN

• IVs

  • either no history or only history

  • genotype (CC, GC, or GG)

    • GG

    • C allele carrier (CC or CG/GC)

• DVs

  • presence vs. absence of BN

    • 129 female patients w/ BN

    • 98 “normal-eating” women

• results

  • C allele

    • women w/ BN more likely to carry allele of Bcl 1 gene

  • abuse

    • women w/ BN report greater history of abuse compared to controls

  • odds ratio

    • individual affects of either a allele or abuse are eliminated in their interaction together

      • interaction between them is significant → they influence each other

• conclusions

  • women w/ BN report greater history of abuse compared to normal eaters

  • women w/ BN more likely to carry C allele of Bcl 1 gene

  • both effects eliminated when evaluating interaction

  • combination of diathesis (c allele)/stress (abuse) that particularly predicts risk for BN

    • having both together increases risk for BN, not them individually

consequences of starvation

• dry, brittle hair, skin, nails

• fine hairs grow on body → lanugo; tries to keep you warm due to the lack of fat on the body

• constipation

• amenorrhea

• osteoporosis/bone fractures

• impaired immune system

• muscle loss

• major organ failures

  • cardiovascular complications

    • arrhythmia, heart failure

purging complications

• electrolyte imbalance due to self-induced vomiting/diuretic/laxative mouse (can lead to heart failure)

  • electrolyte imbalance is the effect of multiple purging methods

• erosion of dental enamel

• hypersensitive gag reflex

• ruptured esophagus/stomach

consequences of binge eating

• elevated weight

  • correlated cardiovascular, metabolic consequences

• compromised gastrointestinal function

  • enlarged gastric capacity

  • delayed gastric emptying

  • gastric emptying (rare but deadly)

inpatient treatment

• this is when the patient is admitted to the hospital

  • reserved for the medically severe cases/those at-risk of suicide

• most expensive

  • average cost of one month of this treatment → $30k

residential treatment

• full-time treatment → go to a center where everyone’s getting treatment

• resource intensive

  • average duration is about 83 days → average cost is $956/day, $79,348 (2006)

• this is mostly limited to individuals w/ resources to cover the costs

  • insurance and disposable

outpatient treatment

• this is when the patient sees a therapist/professional

• costs substantially less compared to the other methods

  • $4k for a full course of CBT; $200/hr.

  • about $1,689 for full-course CBT w/ Medicare/Medicaid reimbursement level

• have to consider the safety/health, efficacy

  • if someone’s at risk of death, this may not be the right treatment method for them

within-subjects designs

• participant serves as their own control → changes in the same person over time

  • assess the baseline symptom level

  • begin treatment → reassess symptom level

  • discontinue after effective component of treatment → reassess symptom level

• if the treatment is impacting the symptoms, then changes in the symptoms should be observed w/ changing presence and absence of the active treatment

• limits:

  • limited generalizability → what works for one may not work for others

  • once patient is recovered, they never have the opportunity to test whether a different treatment would have been better/worse

  • not ethical to withhold treatment that you know works

between-subjects designs

• comparing the outcomes between two groups → those who receive intervention vs. those that don’t receive intervention (or receive another one)

  • different people are the controls

  • IV → treatment condition or the group to which the patients belonged

  • DV → recovery, symptom reduction, something along those lines

randomized controlled trials (rct)

• type of between-subjects design

  • active treatment group or control group (random assignment)

    • observes outcomes after treatment → did the number of people who recovered differ between the groups?

random assignment

• increases the representativeness of those in a condition to the population that’s seeking treatment

• reduces the likelihood that differences before the treatment (ex: differences in symptom severity) influence comparison of outcomes between the conditions

empirically supported interventions

• evidence suggests that this is a good treatment

• produce superior outcomes compared to control conditions

  • no treatment/waitlist or placebo or alternative treatment

• also called evidence-based treatments

waitlist control

• between-subjects design (w/ treatment group)

  • compare people who receive the intervention to the people placed on a waiting list for the treatment

    • whether the treatment is better than doing nothing

placebo control

• between-subjects design (w/ treatment group)

  • compare people who receive intervention to people who receive fake treatment

    • mimics treatment pills (thinking you’re getting the treatment may actually help even if you’re not actually getting it)

alternative treatment control

• between-subjects design (w/ treatment group)

  • compare people who receive the treatment to people given a different kind of intervention

    • two treatments studied compared to just one

comparing treatment modes

• gowers et al. (2010)

• three types:

  • specialized inpatient treatment

  • specialized outpatient treatment

  • general (routine) treatment through a service in England

• acceptance/satisfaction

  • acceptance of randomization (treatment) greater for the outpatient care

  • both specialist treatments preferred over general outpatient treatment

• outcome

  • no differences in outcome across treatment modalities

    • didn’t matter what you received in terms of the final improvements

  • improvement observed over time (good, but slow)

• cost effectiveness

  • to achieve the same outcomes from cheapest to most expensive

    • specialist outpatient → general outpatient → inpatient

• conclusions

  • specialist outpatient treatment determined to be the superior treatment modality

    • it’s acceptable to patients and results in higher satisfaction

    • produces outcomes that are just as good as the other options

    • costs the least amount of money

CBT for BN

• Fairburn’s three stages

  • CBT Stage 1

  • CBT Stage 2

  • CBT Stage 3

Fairburn’s stage 1

• identify the triggers and tackle them

  • establish control over eating w/ behavioral techniques

    • self-monitoring of food intake/symptoms w/ diary

    • prescription of a regular pattern of eating

      • ex: every 2/3 hours → 3 meals + 2/3 snacks

    • stimulus control (identify triggers to binge eating + avoid/employ incompatible behaviors)

    • psychoeducation on weight regulation, dieting, risks of purging

fairburn’s stage 2

• explore the distorted thoughts and poke holes in them

• reduce dieting/body image disturbance though a combination of behavioral/cognitive techniques, engaging in problem solving

Fairburn’s stage 3

• maintenance of progress/reduction of risk for future relapse

  • ex: recognize and challenge dichotomous thinking to prevent lapses from becoming relapses

• frequency of sessions transitions from initially being twice/week, once/week to twice/month, once/month, ending

disease-specific pathways model

• focuses on specific factors that increase risk for EDs within individuals

  • their presence increases the risk of an ED but the absence does nothing

  • ex: weight/shape concerns

  • prevention would seek to reduce body dissatisfaction

• success depends on the accurate identification of these risk factors

nonspecific vulnerability-stressor model

• focus on nonspecific risk factors within individuals that contribute to EDs and other problems (ex: depression, anxiety disorders, SUDs)

  • ex: increased negative affect (sadness, anxiety) is a nonspecific risk factor for EDs (and depression, anxiety disorders, SUDs)

  • prevention would seek to reduce stress

• these are the general risk factors that contribute to the etiology of many related problems

health promotion paradigm

• focuses on protective factors that increase health when present and do nothing when absent (their absence does not make you sick)

  • focuses on the individual and the community

  • ex: people wouldn’t resort to unhealthy weight control behaviors if they had healthy approaches

  • prevention would promote healthy eating and exercise patters at the individual and community level (like workplace incentives for joining a gym)

empowerment-relational model

• focuses on using the individual to change the community to help both the individual and society (based on feminist theory)

• employs girls/young women as agents of change

  • ex: recruit girls to participate in letter-writing campaigns to ad agencies to focus on what women can do rather than appearance; empowers these girls and reduces sociocultural risk factors

universal prevention programs

• includes everyone → works w/ the general population (like students in a school setting)

  • knowledge → improved significantly

  • attitudes → improved modestly in some

  • behaviors → largely unchanged

selective prevention programs

• subsets of the population → focuses on the individual at increased risk for EDs

  • knowledge → improved significantly

  • attitudes → improved significantly

  • behaviors → improved significantly for in some but not all studies

• more interactive programs more likely to reduce risks

indicated prevention programs

• focuses on those w/ emerging problems → have already begun reporting some eating problems

  • knowledge → improved significantly

  • attitudes → improved significantly

  • behaviors → improved significantly

• usually more effective than the universal programs

challenges for prevention

• need for greater understanding of risk factors for EDs for disease-specific pathways model

  • narrow focus on sociocultural risk factors w/ assumptions that these have the greatest influence/easiest to change

• constraints on duration/intensity of the programs

  • particularly true of universal programs where prevention takes time away from regular curriculum

• selective/indicated preventions also face statistical/methodological challenges

  • selective biases may reduce difference from control conditions

    • individuals volunteering for prevention study may be motivated to improve regardless of efficacy of prevention

    • “regression to the mean” may contribute to improvements regardless of condition

planet health

• a type of universal prevention program

• uses the health promotion paradigm

• employs longer duration/intensity than prior programs

  • school-based curriculum to prevent childhood elevated weight

• results

  • elevated weight

    • both boys/girls showed reduced TV viewing/increased fruit/veggie consumption

    • prevalence of elevated weight decreased by half in intervention schools compared to control schools in girls, but not boys

    • in girls, reduced TV time predicted decreases in rates of elevated weight

  • EDs

    • at follow-up, 6.2% of girls in the control schools vs. 2.8% of girls in intervention schools reported use of purging to control weight

    • reduced time watching TV predicted reduced purging at 3-year follow-up

regression to the mean

• when you test the same group at two time points; individuals w/ extreme scores (either very high or low) on the first test tend to produce scores closer to the average on the second test

• scores on a test/measure reflect two things:

  • actual levels you’re testing/measuring

  • random error (ex: misreading an item)

• actual levels may stay the same over time or change as an effect of intervention

• random error is random and thus unlikely to affect scores the same way on the second test, allowing the score to move closer to the mean (though rank order may be relatively stable over time)

efficacy

• demonstrated when intervention produces superior outcomes to the control conditions (no treatment/waitlist vs. placebo vs. alternative treatment) in RCT

  • this is done in a controlled setting

effectiveness

• demonstrated when intervention produces improvement in naturalistic setting that mirrors how it will be employed in the real world

primary prevention

• stopping an event before it occurs

  • hard to tackle this b/c it’s difficult to pinpoint when they may start

secondary prevention

• stopping emerging problems from developing into more serious ones

outcome

• how well the patients are doing at some point after they were initially encountered

  • where you end up at a given time

    • % recovered at 10-year follow-up

    • % ill at 10-year follow-up

course

• path the patients have taken between when they were first encountered and when the outcomes assessed

  • how you get there

    • changes in illness status assessed repeatedly over follow-up

Becker et al. (2008)

• background

  • cognitive dissonance (CD) based prevention programs

  • higher and lower risk groups compared to the active control group (media advocacy (MA))

• IVs

  • treatment condition (CD vs. MA)

  • risk status (high vs. low)

• DVs

  • internalization of thin ideal

  • body dissatisfaction

  • dietary restraint

  • bulimic pathology

• results summary

  • hypotheses only partially supported

    • CD did produce improvements (for both high/low); some effects diminished over time

    • effects of MA, when observed, did not diminish as much as expected

  • across 3 DVs, CD worked well in both high/low participants whereas MA only worked in high-risk participants

    • this could show that MA is only good for high-risk populations

wilksch et al. (2017)

• background

  • purely online intervention has key advantages (helps spread this a lot)

    • delivery does not require training of therapists/interventionists

    • no risk for loss of fidelity to program

    • no limit to number of participants

• objectives

  • test online version of Student Bodies w/out moderator (therapist/trained interventionist) and against an active condition

  • develop and test an online version of Media Smart

  • test both in general population community w/ full range of eating pathology

• results → acceptability

  • no differences in initiating/completing first nodule of MS-T vs. SB

  • among those who completed first module, significant more completed all 9 session in MS-T

    • only MS-T had weekly email reminders of the modules

  • primary results (EDE-Q global score):

    • no difference among those who initiated intervention/controls

    • follow-up measures → MS-T had lower EDE-Q scored than controls (good thing)

  • secondary results (weight concerns, depression, etc.):

    • MS-T pretty much better for everything over SB/controls

  • tertiary results (prevention/remission):

    • prevention → no significant differences

    • remission → MS-T significantly less likely to keep DE, but not significantly different from SB

• discussion

  • ITT analyses produced no differences across conditions on primary outcome (not ideal)

    • didn’t eliminate people w/ higher symptoms b/c this mimics how interventions would be used in the real world

  • engagement was lower than in prior intervention studies

    • completely online test to fully scalable intervention

    • engagement even lower in SB than MS-T, suggesting value of automated email reminders

  • where differences were found, they favored MS-T

mortality

• type of outcome domain

  • % dead

  • standardized mortality ratio (SMR) = actual mortality rate/expected mortality rate

• AN associated w/ one of the highest risks of premature death of any premature death of any disorder

• more recent evaluation of global burden of mental illness/SUDs suggests that Eds are 5th place for premature death (whiteford et al., 2013)

  • number could be due to the fact that non-Western areas also included

  • cause of death determined by what was on the death certificate

recovery

• type of outcome domain

  • % w/ no symptoms (full remission)

  • % w/ symptoms below threshold for diagnosis (partial remission)

  • % still ill (non-recovery, current ED)

relapse

• type of outcome domain

  • among those who achieved remission, % who returned to full-threshold diagnosis

crossover

• type of outcome domain

  • % who migrate from one ED diagnosis to another one

prognostic factors

• type of outcome domain

  • variables that predict any of the other outcome variables

    • mortality

    • recovery

    • relapse

    • crossover

crow et al. (2009)

• background

  • several studies have supported increased risk of death in AN, including increased risk of death by suicide

  • fewer studies completed on BN, but small number of studies of not support increased mortality

  • no systematic studies on EDNOS

• purpose: assess mortality in large sample of AN, BN, EDNOS w/ follow-up durations of 8 to 25 years

• hypothesis

  • mortality rates would be high in AN

  • elevated mortality rates would be found in BN (due to longer duration of follow-up)

  • only modest elevation in mortality would be found for EDNOS

• measures of mortality

  • crude mortality rate (CMR)

  • standardized mortality ratio (SMR)

• results

  • hypothesis only partially supported

  • CMR for AN slightly lower than observed in previous studies; SMR not elevated significantly

  • CMR for BN higher than observed in previous studies; SMR significantly elevated for both all causes/suicide

  • CMR for EDNOS not lower for AN/BN; SMR significantly elevated for both all causes/suicide

crude mortality rate (cmr)

• % of patients who died by follow-up

standardized mortality rate (smr)

• ratio of observed deaths in patients to the expected number of deaths in the demographically matched sample

• size of sample determines confidence interval around SMR, which determines where it differs significantly from 1

mortality in AN

• franko et al. (2013)

• 9-year follow-up

  • 11.6

• 22-year follow-up

  • 4.37

• what this tells us:

  • mortality goes down over the years (factoring aging death)

mortality in BN

• franko et al. (2013)

• 9-year follow-up

  • 1.3

• 22-year follow-up

  • 2.33

• what this tells us:

  • mortality slightly goes up

remission in AN

• franko et al. (2013)

• 9-year follow-up

  • 31.4%

• 22-year follow-up

  • 62.8%

• what this tells us:

  • still has recovery over 2 decades after treatment

  • tells patients that recovery is still possible even if they think they’re losing hope/there’s no reason to continue

• positive linear relationship in recovery

remission in BN

• franko et al. (2013)

• 9-year follow-up

  • 68.2%

• 22-year follow-up

  • 68.2%

• what this tells us:

  • recovery pretty much stays the same; there’s not much difference

• positive linear relation up until 20 years, in which it then becomes to level off

franko et al. (2018)

• long-term recovery

• methods:

  • participants

    • 176 women originally recruited for a longitudinal study of AN/BN over 20 years previously completed interviews at follow-up

      • 40 w/ baseline AN-R diagnosis

      • 60 w/ baseline AN-BP diagnosis

      • 76 w/ baseline BP diagnosis

  • interview: longitudinal interval follow-up evaluation EDs version

    • wave I: 1987-1996 → assessed every 6-12 months

    • wave II: 2011-2013 → assessed once and asked to recall over the past 12 months

• psychiatric status rating (PSR) score for ED made weekly based on retrospective recall

  • 5/6 = meets diagnostic criteria

  • 3/4 = subthreshold criteria

  • 1/2 = remission/minimal symptoms

• results

  • for AN-R (vs. recovered) at wave II

    • intake diagnosis of AN-R vs. BN

    • intake diagnosis of MDD

  • for AN-BP (vs. recovered) at wave II

    • older age

    • lower BMI

    • higher % of weeks meeting diagnostic criteria for AN/BN

  • for BN (vs. recovered) at wave II

    • higher % of week meeting diagnostic criteria for BN

• discussion

  • for those w/ AN-R, treatments to address both ED symptoms and depression may be important

  • for those w/ AN-BP/BN, features of the disorder predict the presence of the disorder at long-term follow-up

agras et al. (1974)

• positive reinforcement can work when it comes to caloric intake

  • caloric intake can increase when being positively reinforced

  • this effect can still last even if you don’t actually give them the reinforcer, as long as they have done it before

diathesis

a genetic or biological vulnerability

• ex: C allele of there Bcl 1 gene

Eisler et al. (2016)

• MFT-AN vs. FT-AN

  • MFT-AN a more intensive version of FT

• hypotheses

  • primary: MFT-AN will be more effective tan FT-AN in restoring healthy nutritional state

  • secondary: satisfaction will be higher in MFT-AN compared to FT-AN

• methods

  • RCT comparing efficacy of MFT-AN and FT-AN

• results

  • 58% of FT vs. 76% of MFT in the good/intermediate category after follow-up (12 months)

  • other secondary outcomes:

    • mostly no differences (but there was a lot of missing data for secondary outcomes)

• discussion

  • MFT-AN was superior to FT-AN

    • caveat: participants spent more time in MFT-AN than FT-AN

  • both treatments had high acceptability in terms of treatment acceptance (only one drop-out after randomization)

  • few hospitalizations during treatment (35) or entire study (7%)