CBASP@YoungAge

The researchers designed a modular, individualized psychotherapy program called CBASP@YoungAge for children and adolescents with depression and interpersonal problems with their primary caregivers. The intervention is an adaptation of CBASP, originally developed for chronically depressed adults with a history of childhood maltreatment. The key innovation is that CBASP@YoungAge is modular (therapists can pick and choose components based on the patient's age and needs), includes mandatory caregiver involvement, and targets both depressive symptoms and the quality of the parent-child relationship. The comparator is treatment-as-usual (TAU), which in Germany typically means standard cognitive-behavioral therapy or psychodynamic therapy without mandatory caregiver involvement. The primary outcomes are (1) feasibility (acceptability to therapists, patients, and caregivers; adherence to the modular approach; session attendance; dropout rates) and (2) change in depressive symptoms from pre- to post-treatment. Secondary outcomes include interpersonal behavior between child and caregiver, parenting behavior, and a global health index for both children and parents. Assessments occur at baseline, post-treatment, and 6-month follow-up.

The study aims to recruit 44 participants total (22 per treatment arm), aged 10–21 years, who meet DSM-5 criteria for a depressive disorder (confirmed by structured clinical interview, Kinder-DIPS). Inclusion requires a participating caregiver for the CBASP@YoungAge group (mandatory), sufficient German fluency, and an IQ of ≥80. Exclusion criteria include ongoing psychotherapy, acute suicidality, or a primary diagnosis requiring different treatment (e.g., traumatic disorder, schizophrenia, bipolar disorder). Comorbid disorders and ongoing psychopharmacotherapy are allowed to increase generalizability. Participants are recruited from three university-affiliated outpatient clinics in Germany: Philipps-University Marburg (intervention group), Ruhr-University Bochum, and Landau Psychotherapy Outpatient Clinic (control groups). Therapists are psychotherapy trainees. Assuming a 60% retention rate, the researchers expect 38 treatment completers with complete data.

**Depressive symptoms:** Assessed via structured clinical interview (Kinder-DIPS, parent and child versions) at baseline for diagnosis, plus self-report and clinician-rated scales at pre-, post-, and 6-month follow-up (specific instruments not fully detailed in the protocol excerpt, but standard measures for child depression include the Children's Depression Inventory or Beck Depression Inventory for adolescents).

**Feasibility:** Session-by-session quantitative and qualitative process evaluations completed by therapists, patients, and caregivers. Measures include whether the therapist used the modular approach, whether patients and caregivers understood the material, and attendance rates.

**Interpersonal behavior:** Assessed via observational or self-report measures of caregiver-child interaction (specific instruments not detailed in the excerpt).

**Parenting behavior:** Self-report or observational measures of parenting styles and behaviors.

**Global health index:** A composite measure of physical and mental health for both children and parents.

**Dropout rate:** Tracked throughout the study; expected ~40% based on comparable studies.

**Study design:** This is a quasi-experimental pilot feasibility trial (phase 1 to phase 2). It is non-randomized: participants are consecutively recruited from different outpatient clinics depending on which clinic they attend. The CBASP@YoungAge intervention group is at the Marburg clinic; the TAU control groups are at the Bochum and Landau clinics. This is a major design limitation because the groups are not randomly assigned — differences between clinics, therapists, and patient populations could confound results.

**Randomization:** None. This is a quasi-experimental design, meaning participants are assigned to groups based on which clinic they attend, not by random allocation. This cannot prove causality — any differences between groups could be due to pre-existing differences in the clinics, therapists, or patient populations rather than the intervention itself.

**Blinding:** Not mentioned. Given the nature of psychotherapy, blinding of therapists and patients is impossible. However, outcome assessors could theoretically be blinded to group assignment — the protocol does not specify whether this is done. Lack of blinding introduces risk of bias, especially for subjective outcomes like self-reported depression.

**Duration:** The treatment phase duration is not explicitly stated in the excerpt, but typical CBASP protocols run 16–24 sessions (4–6 months). Follow-up is at 6 months post-treatment. Total study duration per participant is approximately 10–12 months (baseline + treatment + 6-month follow-up).

**Washout:** Not applicable — this is not a crossover design.

**Statistical approach:** The protocol mentions sample size calculations guided by Hertzog (2008) for pilot studies, aiming for 22 per arm. They assume 60% retention. The primary analysis will likely compare pre-post changes within groups and between groups using appropriate parametric or non-parametric tests, but specific statistical methods are not detailed in the excerpt. As a pilot study, the focus is on feasibility outcomes (recruitment rates, dropout, acceptability) and estimating effect sizes for a future fully powered RCT, not on definitive hypothesis testing.

**What this design can and cannot prove:**

**Can prove:** Feasibility — whether the intervention can be delivered as intended, whether patients and therapists find it acceptable, whether recruitment and retention are achievable. It can also provide preliminary estimates of effect sizes for depressive symptom reduction and interpersonal improvement.

**Cannot prove:** Efficacy or effectiveness. Without randomization, any observed differences between CBASP@YoungAge and TAU could be due to clinic effects, therapist effects, patient selection bias, or other confounds. The small sample size (44 total, ~38 completers) means the study is underpowered to detect anything but very large effects. The design cannot establish that CBASP@YoungAge causes improvement — only that it is associated with improvement in this specific, non-randomized sample.

**Major methodological weaknesses:**

1. **No randomization** — the biggest flaw. Groups are from different clinics, so clinic-level differences (therapist expertise, patient demographics, clinic culture) are confounded with treatment.

2. **Small sample size** — 22 per arm, with expected 40% dropout, leaves only ~13 completers per arm for analysis.

3. **No blinding** — risk of expectancy effects and biased reporting.

4. **Pilot design** — not powered for definitive conclusions about effectiveness.

5. **Retrospective trial registration** — registered after recruitment began (November 2020, but protocol published 2022), which raises concerns about selective reporting.

**Important: This is a study protocol — no results are reported yet.** The following are the planned analyses and expected outcomes based on the hypotheses stated in the protocol:

**Primary outcome 1 (Feasibility):** The researchers expect that CBASP@YoungAge will be acceptable to therapists, patients, and caregivers, with high session attendance, low dropout (though they assume 40% dropout based on comparable studies), and high therapist adherence to the modular approach.

**Primary outcome 2 (Depressive symptoms):** They expect a greater reduction in depressive symptoms from pre- to post-treatment in the CBASP@YoungAge group compared to TAU. No specific effect sizes are predicted.

**Secondary outcomes:** They expect improved interpersonal behavior between child and caregiver, improved parenting behavior, and improved global health index in the CBASP@YoungAge group compared to TAU.

**Caregiver involvement:** They expect significantly higher proportion of parent involvement in the intervention group than in the control group (by design — it's mandatory in CBASP@YoungAge).

**Long-term effects:** They expect stronger long-term alleviation of symptoms at 6-month follow-up in the CBASP@YoungAge group compared to TAU.

**Context from the background section:**

Meta-analyses of child/adolescent depression psychotherapy show overall effect size g = 0.36 at post-treatment and g = 0.21 at follow-up — these are small effects.

Effects drop significantly between post-treatment and follow-up assessments.

CBASP for adults with chronic depression is considered evidence-based, but has high relapse and dropout rates, especially among young adults.

No results are available yet. However, based on the background literature:

Typical psychotherapy for child/adolescent depression produces a small effect (g = 0.36), meaning the average treated child improves about one-third of a standard deviation more than the average untreated child. In practical terms, this might mean a drop of 4–6 points on a depression inventory (e.g., CDI total score 0–54).

The researchers hope CBASP@YoungAge will produce larger effects by targeting interpersonal problems with caregivers, which they believe is a key maintaining factor. If successful, effect sizes might be in the moderate range (g = 0.5–0.7), but this is speculative.

The high expected dropout rate (40%) suggests that even if the intervention works for completers, real-world effectiveness may be limited by poor retention.

**Acknowledged by authors:**

This is a pilot feasibility trial, not a definitive efficacy trial.

Non-randomized design limits causal inference.

Small sample size (44 total, ~38 completers expected).

High expected dropout rate (40%).

Therapists are trainees, which may limit treatment fidelity and effectiveness.

The study is conducted in German outpatient clinics, limiting generalizability to other healthcare systems and cultures.

**Critical reader observations:**

**No randomization** is the most serious limitation. The intervention and control groups are from different clinics, so any differences could be due to clinic effects (e.g., Marburg vs. Bochum vs. Landau have different patient populations, therapist training, and clinic cultures).

**No blinding** of patients, therapists, or assessors (apparently) introduces substantial risk of bias, especially for subjective outcomes like self-reported depression.

**Retrospective trial registration** (registered after recruitment began) is a red flag for potential selective outcome reporting.

**Broad age range (10–21)** is problematic because a 10-year-old and a 21-year-old have vastly different developmental stages, cognitive abilities, and relationship dynamics with caregivers. The modular approach attempts to address this, but pooling such a wide age range could obscure age-specific effects.

**IQ ≥ 80 cutoff** excludes children with intellectual disabilities, limiting generalizability to that population.

**Mandatory caregiver involvement** in the intervention group but not the control group means that any benefits could be due to caregiver involvement per se, not the specific CBASP components.

**No active control** — TAU is a weak comparator because it varies widely between clinics and therapists. A more rigorous design would compare CBASP@YoungAge to an equally structured, equally intensive alternative psychotherapy.

**Industry funding:** Not mentioned, but the study is funded by German academic institutions, so no obvious industry bias.

For someone running their own n=1 experiment (e.g., a parent trying to improve their child's depression or a young person working on their own mood and relationships):

### What to test

**Intervention:** A structured approach that combines (a) identifying and modifying negative interpersonal patterns with a key caregiver, (b) practicing new interpersonal behaviors in real-time during sessions, and (c) involving the caregiver as a co-participant in the therapeutic process. The core CBASP technique is "Situational Analysis" — a structured method for analyzing a specific interpersonal situation, identifying the patient's interpretation and behavior, and practicing a more effective response.

**Dose:** The protocol suggests 16–24 sessions (4–6 months of weekly therapy). For a self-experiment, you could test a simplified version: 8–12 weekly sessions of structured interpersonal work with a caregiver, using a workbook or guided self-help format.

### Minimum meaningful duration

**4 months** (16 weekly sessions) is the minimum to see meaningful change in depressive symptoms and interpersonal patterns, based on the CBASP literature. Shorter durations (e.g., 8 weeks) may show initial improvement but are unlikely to produce lasting change in entrenched relational patterns.

**6-month follow-up** is critical because depression treatment effects often decay after treatment ends. A positive result would be sustained improvement at 6 months post-treatment.

### What to measure

**Primary metric:** Depressive symptoms — use a validated self-report scale like the Patient Health Questionnaire-9 (PHQ-9, 0–27, lower = better) for adolescents/adults, or the Children's Depression Inventory (CDI, 0–54, lower = better) for children aged 7–17. Measure weekly during treatment and monthly during follow-up.

**Secondary metric:** Interpersonal behavior with the caregiver — use the Conflict Behavior Questionnaire (CBQ, 0–20, lower = less conflict) or the Inventory of Parent and Peer Attachment (IPPA, higher = better attachment). Measure at baseline, post-treatment, and follow-up.

**Process metric:** Session-by-session ratings of the quality of the caregiver-child interaction during "homework" practice — e.g., "On a scale of 0–10, how well did we communicate this week?" and "How connected did I feel to my caregiver this week?"

**Feasibility metric:** Track session attendance, homework completion, and whether the caregiver attended. Dropout is a key outcome — if you can't stick with it, the intervention isn't feasible for you.

### Key confounds to control for

**Medication changes:** If the child is on antidepressants, note any dose changes during the study period. These could independently affect mood.

**Life events:** Major stressors (school change, family conflict, loss) or positive events (new friendship, achievement) could confound results. Keep a brief daily log of major events.

**Other therapy:** If the child is receiving other psychological support (school counselor, other therapist), this is a confound. Ideally, no other concurrent therapy during the experiment.

**Caregiver's own mental health:** Depressed or anxious caregivers may struggle to participate effectively. Measure caregiver depression (e.g., PHQ-9) at baseline and post-treatment.

**Developmental stage:** A 10-year-old and a 17-year-old will respond very differently to the same intervention. If running this as a self-experiment with a child, note their age and cognitive level.

**Expectancy effects:** Both the child and caregiver may expect improvement because they're trying something new. Use a no-treatment baseline period (e.g., 4 weeks of monitoring before starting the intervention) to establish a pre-treatment trajectory.

### What a positive result would look like

**Clinically meaningful improvement:** A drop of ≥5 points on the PHQ-9 (for adolescents/adults) or ≥8 points on the CDI (for children) from pre- to post-treatment. This is roughly equivalent to moving from moderate depression (PHQ-9 = 15) to mild depression (PHQ-9 = 10).

**Sustained improvement:** The improvement is maintained at 6-month follow-up, not just at post-treatment. This is important because many depression treatments show decay over time.

**Interpersonal improvement:** A drop of ≥5 points on the CBQ (less conflict) or an increase of ≥10 points on the IPPA (better attachment) from pre- to post-treatment.

**Feasibility:** The child and caregiver attend ≥80% of sessions, complete ≥70% of homework assignments, and rate the intervention as "moderately" to "very" helpful on a session-by-session basis.

**Caveat:** In an n=1 experiment, you cannot prove causality — you can only observe whether improvement coincides with the intervention. A positive result is suggestive but not definitive. To strengthen inference, use a multiple-baseline design (e.g., start the intervention after a 4-week baseline, then after 8 weeks, then after 12 weeks in different family members or different problem areas) or an ABAB design (intervention, withdrawal, intervention) if ethically feasible.