Eating Disorders are a debilitating and serious mental illness. This illness is associated with medical complications, psychological and social impairment. Families of people with an eating disorder also report that they lack resources and have many unmet needs. Families often have insufficient information regarding the eating disorder, available treatment options and strategies for supporting the person with the illness. The purpose of this study is to evaluate the effectiveness of two different family interventions for people with eating disorders and their family members (parents or partners) receiving treatment either in the inpatient or day treatment Eating Disorder Program at the Toronto General Hospital, University Health Network. Family supportive counseling consists of people with eating disorders and their family members meeting with a family therapist. Multi-family group therapy involves eight to ten families who meet as a group with two therapists. The investigators are conducting a study to assess the differences between these two different family interventions. This study will help us identify who benefits the most from participating in family supportive counseling or multi-family therapy. The investigators are also evaluating which intervention is more effective at helping the person with the eating disorder overcome their illness while helping their family members learn how to support the recovery process. Both family therapy interventions are delivered by experienced family therapy clinicians who work in either the Inpatient or Day Treatment Eating Disorder Program at Toronto General Hospital.

Background: Bulimia Nervosa (BN) is a serious and persistent mental illness with a high rate of mortality and morbidity. Research demonstrates that individuals with BN have high drop-out rates, multiple re-admissions to tertiary care programs and high relapse rates all of which may contribute to the chronicity of this illness. A systematic review of randomized controlled trials for eating disorders reveals dropout rates ranging from 0-43%. Individuals with BN are at the highest risk for relapse during the first three months following the completion of inpatient treatment for BN. A higher BMI at completion of an intensive eating disorder program and weight maintenance immediately following treatment are associated with lower relapse rates at six and twelve months. Women with chronic BN have a poorer quality of life than women in the general population in a variety of domains. Individuals with BN have access to a smaller support network outside of their nuclear family compared with normal controls. They are often confronted with negative stereotypes and stigmatizing attitudes from the general public who often perceive affected individuals as having control over their eating behaviours. Fear of being stigmatized and negatively stereotyped, a poor quality of life, a smaller social support network and lower expectations of receiving appropriate support, may contribute to and heighten the reliance of individuals with BN on their immediate families for emotional and instrumental support. The over-reliance on families for support may contribute to duress in families who may be willing to assist the person afflicted with BN but lack the sufficient knowledge and skills to effectively intervene to assist with the significant medical and psychological impairment that results from this illness.

Caregivers of individuals with eating disorders experience significant distress poor quality of life, and intense feelings of self-blame and shame as a result of their caregiving role. Caregivers often feel responsible for causing a loved one to develop the illness, and these feelings are often reinforced by professionals. In fact, the general public is more likely to blame families for causing BN compared with other mental illnesses such as schizophrenia. Poor family functioning in families of individuals with BN is predicted by problematic eating symptoms, conflict about how best to support the affected person, stigma and lack of social support. Concerns and fears that their loved one will be discriminated against or labeled negatively have been shown to result in family members withdrawing from their own social support network. Social support has been shown to mitigate caregiver distress and burden and poor health in carers of individuals with mental illnesses. Hence, an intervention is required to increase social support for individuals and their family members while providing educational information that challenges stigma, self-blame and shame.

Caring for an adult with chronic BN creates many challenges for the carer, including psychological duress, social isolation, stigma and poor family functioning. In turn, increased family conflict may exacerbate illness behaviours in the person with BN. Due to the emaciated appearance of the person with BN and the life threatening nature of this illness, family members may become overly protective and avoid discussions that may be distressing to the affected individual. Anxiety and depression may further heighten caregiver distress and intensify their propensity to over-protect the affected individual. Family members may also respond to severe symptoms by accommodating to the illness while taking on responsibility for the affected person in ways that are not age-appropriate. Negative feelings may be increased in the carers when their efforts to assist and motivate the person into treatment are met with denial about the seriousness of the illness and ambivalence to engage in treatment. These responses may elicit criticism from carers, especially if they perceive the eating symptoms as volitional. Emotional over-involvement, accommodation and criticism can have negative influences on the person with BN including heightened distress, decreased willingness to seek help and poorer treatment outcomes. An intervention is urgently needed to address the burden that this illness imposes on families and improve the quality of life of the carer, while teaching communication and problem solving skills that facilitate recovery from BN.

In the last decade, Multi-family group therapy (MFT) has been recognized internationally as an innovative approach to working with families of BN adolescents and has been shown to contribute to positive outcomes. MFT has also been widely employed and shown to be effective in families of adults affected with other mental illnesses including substance abuse, bipolar disorder and schizophrenia. To date, MFT has not been investigated in families of adults with BN. Given the evidence that demonstrates the efficacy of MFT in adolescent BN and other mental illnesses in adults, the proposed research study seeks to assess the efficacy of MFT for adults with BN and their families.

Overview of Proposed Study: In this study, the investigators propose to investigate MFT, which has never before been studied in adults with BN and their carers. The overarching objective will be to conduct a randomized controlled trial comparing MFT with treatment as usual (TAU) in patients receiving intensive treatment for BN and their carers.

Hypotheses and Research Questions: This proposed randomized controlled trial aims to provide pilot data regarding the efficacy of Multi-Family Therapy (MFT) in improving treatment outcomes for patients with Bulimia Nervosa (BN) and in reducing caregiver distress. Compared with patients not participating in MFT, the investigators hypothesize that patients with BN and their carers participating in MFT will have: 1) lower drop out rates, 2) higher likelihood of sustaining a body mass index (BMI) of 18.5 or higher at three months post-treatment, 3) greater improvement in general psychological health for carers from pre-treatment to three months post-treatment, and 4) reductions in the adverse impact of caregiving, expressed emotion and accommodation of eating behaviours at three months post-treatment. Finally, the investigators expect that descriptive data will show that MFT is a feasible, acceptable option for patients with BN and their carers.

Inclusion Criteria:

• any patient attending the eating disorders program at University Health Network, Toronto General Hospital and their family members over the age of 16 (siblings, parents, partners).

Exclusion Criteria:

• current family violence.

There is no plan or REB approval to share the IPD data. IPD will be kept confidential and destroyed as per UHN data policy when the study is concluded.

For the entirety of the study (2011-2014.

Lacked random assignment to MFT or SFT. Sample size of pilot was small and subsample of individuals were smaller in SFT than MFT. We may not have had statistical power to determine all differences in MFT and SFT groups. Sample not generalized.