This pilot randomized control trial investigates whether a psychosocial intervention targeting parent-adolescent conflict is feasible and acceptable, in preparation for a future trial that will test whether the intervention can improve treatment outcomes for youth diagnosed with depression.

Parent involvement in psychosocial interventions for youth diagnosed with internalizing disorders has been cited as a contributor to better outcomes for these youth, and parent components have been included in several evidence-based interventions for adolescent mental health problems. Additionally, youth experiencing mood and behaviour changes that come about as a result of a depressive episode can lead to frustrated/overprotective reactions from parents, which can contribute to worse outcomes for these adolescents. Interventions focused on addressing parent-adolescent conflict and reducing negative interactions can potentially support recovery from depression and reduce episode relapse. However, there is very little research on including parent components in psychosocial interventions specifically aimed at youth depression. This randomized control trial aims to fill this gap by randomizing parents of youth receiving treatment for depression into either 1) an 8-week virtual or in-person group parent intervention, or 2) usual care.

Participants will be parents or caregivers (for simplicity, the term "parent" is used to refer to parents or caregivers) of adolescents aged 13-18 with depression who are referred for outpatient services at a mental health research hospital in Toronto, Canada. Their adolescents will also participate in the study in order to provide multi-informant data; however, should adolescents decline to participate in the study, their parent will still be able to take part.

Parents will complete a combination of questionnaire and observationally coded measures at pre- and posttreatment, along with questionnaire measures at mid-treatment and 3 months follow-up. Adolescents will only complete questionnaire measures for all four time points.

Inclusion Criteria:

• parent/caregiver of an adolescent aged 13-18 years
• the adolescent is referred for outpatient services at CAMH
• adolescent has significant depressive symptoms, having either: a) a self-reported score of 23 or higher on the Mood and Feelings Questionnaire; or b) a depression diagnosis based on the Kiddie-Schedule for Affective Disorders and Schizophrenia from the last 3 months
• parent or youth endorse problematic levels of conflict in their relationship, having either: a) a total score of 9 or greater on the adolescent version, or a total score of 11 or greater on the parent version, of the 20-item Conflict Behavior Questionnaire; or b) a rating of 2 or higher on item 2 or 3 (Item 2 "getting along with your mother/mother figure;" item 3 "getting along with your father/father figure") of the youth-report version of the Columbia Impairment Scale
• both parent and adolescent speak, read, and write English at a Grade 6 level or above

Exclusion Criteria:

• the adolescent's primary diagnosis is not depression, based on information in their health record
• the adolescent has been diagnosed with bipolar disorder (I or II), schizophrenia, moderate-severe eating disorder, moderate-severe substance use disorder, or an intellectual disability (as the clinical and family needs of these adolescents are not likely to be sufficiently addressed by a general group for parents of adolescents with depression), or requires immediate hospitalization for suicide risk
• the parent/caregiver does not consent to participate
• (specific only to the virtual format of the study) the parent is not comfortable communicating via email or using WebEx for virtual appointments (as the use of WebEx and email is required for study visits and for participation in the parent intervention)

Sharing IPD has not currently been approved by our Research Ethics Board and would also require a data sharing agreement.

Research personnel involved in data collection will be blind to participants' treatment condition (single blind). In addition, coders for the interaction task and speech sample will be blind to treatment condition (intervention versus control) and measurement time point (baseline vs. posttreatment).