For Clinicians

The Therapy Homework Problem: Why Patients Don't Practice (And How to Fix It)

Between 20-50% of therapy patients don't complete homework. Here's what the research says about why — and how digital tools can close the compliance gap.

8 min readFor Therapists

You assign a thought record. Your patient nods. They seem motivated. And then next week, when you ask about it, you get a familiar answer: "I meant to, but I didn't get to it."

You're not alone. Between-session homework non-adherence is one of the most persistent problems in evidence-based psychotherapy — and one of the least discussed outside of clinical training.

This post is for therapists. It reviews the research on homework compliance, examines why patients don't follow through, and explores what the evidence says about improving adherence — including how digital tools fit into the picture.

The numbers

The compliance data has been remarkably consistent over decades. A 2017 review in JMIR Mental Health summarized the literature: surveys of practitioners suggest rates of non-adherence in adult clients of approximately 20% to 50%, while adherence rates in adolescents have been reported at approximately 50%.

That means roughly half your patients are not completing their between-session exercises in a given week. For adolescents, the number is even more stark.

And this matters clinically, not just aspirationally. A meta-analysis of 23 studies (2,183 participants) found a significant positive relationship between homework compliance and treatment outcomes across conditions — anxiety, depression, and substance use (r = .26). The relationship held regardless of therapy type, specific exercises assigned, or measurement method.

A 2016 meta-analysis in Behavior Therapy of 17 CBT studies (2,312 clients) found that both quantity and quality of homework predict outcomes, with quality showing dramatically stronger effects at follow-up (g = 1.07 vs. g = 0.51). In clinical terms: patients who engage deeply with even a few exercises benefit more long-term than those who check boxes on many.

The most recent 2024 review estimates the causal effect of between-session homework at a medium effect size (d = .53). This is substantial — roughly the difference between a treatment that clearly works and one with marginal benefit.

Why patients don't do it

A 2023 systematic review of 25 studies examined the factors that drive homework completion. The barriers were overwhelmingly practical, not motivational:

Forgetting. The most cited reason. By mid-week, the specific instructions from session have faded. The worksheet is in a bag somewhere. The rationale made sense on Tuesday but is fuzzy by Friday.

Uncertainty. "Am I doing this right?" Without guidance in the moment, patients second-guess the exercise and abandon it rather than risk doing it wrong.

No support at the point of need. The moment when a thought record would be most useful — when the patient is actually experiencing distorted thinking — is also the moment when they're least equipped to remember the steps, find the worksheet, and complete it without help.

Life. Work deadlines, children, exhaustion, social obligations. Homework doesn't compete with other therapy; it competes with everything else in a patient's life.

Shame about non-completion. Once a patient misses a week, the avoidance compounds. They feel guilty, which makes them less likely to bring it up, which makes the therapist less likely to review it, which reduces its perceived importance.

What the research says therapists can do

The same systematic review identified therapist behaviors that significantly improve homework adherence:

Present a convincing rationale. Patients who understand why a specific exercise matters are more likely to do it. Generic instructions ("try to notice your thoughts") are less effective than connected rationales ("this thought record will help us see whether the pattern we noticed in session shows up during the week").

Design collaboratively. Homework assigned unilaterally has lower compliance than homework co-created with the patient. When patients choose the frequency, timing, or difficulty level, they feel more ownership.

Review it consistently. If you don't ask about homework at the start of each session, patients quickly learn it's optional. Consistent review signals that between-session work is a core part of treatment, not an afterthought.

Be flexible. Rigid expectations backfire. If a patient did 2 of 5 thought records, explore what made the other 3 hard rather than treating it as failure.

The Beck Institute has noted that the field is shifting from measuring strict "adherence" toward genuine engagement — quality of practice matters more than checking boxes.

Where digital tools fit

A systematic review of mobile apps for treatment adherence found that digital tools using reminders, gamification, progress tracking, and guided exercises show promise for improving compliance. The JMIR Mental Health review identified six essential features for homework support apps:

  1. Therapy congruency — the app aligns with the modality you're using (CBT, DBT, ACT, etc.)
  2. Fostering learning — clear, concise instructions and psychoeducation
  3. Guiding therapy — individualized feedback and homework review between sessions
  4. Building connections — strengthening the therapeutic alliance
  5. Emphasis on completion — graphical progress tracking and automated reminders
  6. Population specificity — adapted for the target population

Research on digital micro-interventions supports the approach of brief, daily, guided exercises as an effective delivery mechanism. The insight: patients are more likely to complete a 2-minute guided exercise on their phone than a 20-minute paper worksheet.

The 2025 Dartmouth Therabot trial adds another dimension: AI-guided CBT conversations produced clinically significant depression reduction over 8 weeks. When patients had a conversational guide walking them through exercises daily, they actually did the work.

What this means for your practice

The homework compliance problem isn't a patient problem. It's a delivery problem. The therapeutic techniques work. The evidence for between-session practice is strong. But the traditional delivery mechanism — verbal instructions and paper worksheets — doesn't match how modern patients live.

This doesn't mean abandoning homework. It means rethinking how it's delivered, tracked, and reviewed.

Specifically:

Make it easier to start. A 10-second mood check-in has a lower barrier than a full thought record. Once the habit is established, depth can follow.

Provide guidance at the point of need. A patient struggling with distorted thinking at 9 p.m. on a Tuesday needs a guided walkthrough, not a blank worksheet.

Track what happens between sessions. If you knew before each session what your patient practiced, what was hard, and how their mood shifted, you could use session time more efficiently.

Close the loop. When patients know their therapist will see what they practiced, the accountability changes. It's not surveillance — it's connection.

How BridgeCalm supports therapists

BridgeCalm's therapist portal was designed around these research findings:

Pre-session briefs — 24 hours before each session, you receive a 1-page summary: what your patient practiced, how their mood tracked, what they struggled with, and any flagged items. Session time shifts from catch-up to progress.

Exercise assignment in 60 seconds — browse a library of evidence-based exercises across CBT, DBT, ACT, and mindfulness categories. Assign an exercise with a personalized note. Jan (the AI wellness companion) delivers it to your patient and guides them through it.

Outcomes at a glance — PHQ-9 and GAD-7 tracking with baseline comparisons across your full caseload. Spot who's improving and who needs more support without digging through notes.

Patient self-reports — all data is labeled as patient self-report. Jan doesn't interpret, diagnose, or make clinical judgments. She guides exercises, tracks what patients report, and sends you the data. Clinical judgment stays where it belongs — with you.

Under 3 minutes per patient per week — the portal is designed to add insight without adding workload.

All data is HIPAA-compliant, end-to-end encrypted, and never sold. Business Associate Agreements are available for all therapist-tier subscriptions.

[Learn more about the therapist portal →]

Sources

  • Tang, W. & Kreindler, D. (2017). "Supporting Homework Compliance in CBT: Essential Features of Mobile Apps." JMIR Mental Health. PMC5481663
  • Mausbach, B.T., et al. (2010). "Homework Compliance and Therapy Outcomes: An Updated Meta-Analysis." Cognitive Therapy and Research. PMC2939342
  • Kazantzis, N., et al. (2016). "Quantity and Quality of Homework Compliance." Behavior Therapy. PubMed 27816086
  • Ryum, T. & Kazantzis, N. (2024). "Between-Session Homework in Clinical Training and Practice." Clinical Psychology in Europe. PMC11303922
  • Karyotaki, E., et al. (2023). "Integrating between-session homework in psychotherapy: Systematic review." PubMed 37104804
  • Beck Institute. "What is the Status of 'Homework' in CBT, 50 Years On?" beckinstitute.org
  • Ahmed, I., et al. (2018). "Mobile Apps for Increasing Treatment Adherence: Systematic Review." PMC6604503
  • Aguilera, A. (2020). "Digital Micro Interventions for Behavioral and Mental Health Gains." JMIR. PMC7661243
  • Dartmouth College. (2025). "First therapy chatbot trial yields mental health benefits." Dartmouth News

Built for therapists who want better between-session data

Pre-session briefs, PHQ-9/GAD-7 tracking, homework assignment, and outcomes at a glance — under 3 minutes per patient per week.

Explore the Therapist Portal

If you or someone you know is in crisis

Help is available 24/7. Call or text 988 (Suicide & Crisis Lifeline) or text HOME to 741741 (Crisis Text Line). BridgeCalm is a wellness tool, not a crisis service.

Keep Reading