There is a question I have been asking, in different forms, for thirty years of clinical practice. I first heard it explicitly from a client who came to see me after losing her mother. She had been on a waiting list for six weeks. By the time she sat in my office, she had already navigated the worst weeks of her grief alone, had made several decisions she would later regret, and had developed a habit of drinking two glasses of wine every evening to quiet the noise in her chest.
“What was I supposed to do in between?” she asked. Not unkindly. Not accusingly. Just honestly. What was I supposed to do while I was waiting? While I was between sessions? While the crisis was actually happening?
I did not have a good answer for her then. I have spent the years since trying to build one.
The 50-Minute Hour Is a Remarkable Tool With a Serious Limitation
The traditional therapy hour is one of the most powerful interventions in human history. A trained clinician, a private space, a relationship built on trust and therapeutic contract, the application of evidence-based frameworks to the specific texture of a specific life: there is nothing else quite like it. I believe in it completely. I have practiced it for thirty years and watched it transform people.
But it has a structural limitation that we do not always name clearly enough: it happens once a week, or once every two weeks, or once a month, for fifty minutes. And the rest of life happens in the other 10,030 minutes of the week.
The Tuesday afternoon panic attack does not schedule itself around the Thursday therapy appointment.
What Happens In Between Determines Almost Everything
Research on therapy outcomes consistently demonstrates that what clients do between sessions is among the strongest predictors of therapeutic progress. A landmark meta-analysis by Mausbach and colleagues, synthesizing 23 studies encompassing more than 2,000 subjects, found a significant relationship between between-session homework engagement and treatment outcomes, with effect sizes robust across depression, anxiety, and other target symptoms. A more recent 2024 review in the Journal of Clinical Psychology confirmed that between-session homework demonstrates efficacy for ultimate treatment outcomes and is likely effective for intermediate, session-to-session outcomes as well.
The implications are direct: what people do between sessions is not supplementary to therapy. It is, in a meaningful sense, where therapy lives. The application of coping strategies in real time, the use of journaling and reflection, the practice of emotional regulation skills in the actual moments when regulation is needed: these are the conditions under which therapeutic learning becomes therapeutic change.
The Difference Between Distraction and Support
Not everything marketed as mental wellness support is grounded in psychological science. There is a meaningful difference between an app that offers breathing exercises and mood-tracking checkboxes and an app that is built on clinical frameworks, developed by a licensed practitioner, and designed to meet people in the actual texture of their specific emotional experience.
The distinction matters enormously for communities who are already navigating mistrust of mental health systems. The APA’s 2024 health advisory on AI chatbots and wellness applications explicitly cautions that many general-purpose mental health apps are trained on vast, unvetted internet data rather than clinically validated information, and that sycophantic engagement patterns can reinforce confirmation bias and maladaptive beliefs rather than challenging them. When a person reaches for support in a vulnerable moment and receives something generic and culturally incompetent, the cost is not just unhelpfulness. It is the reinforcement of a belief that mental health tools were not built for them.
What Genuine Between-Session Support Looks Like
In my clinical practice, I have always thought carefully about what I send clients away with. Not homework in the punitive sense, but something to hold onto between sessions: a question to sit with, a journaling prompt, a specific strategy to try when a particular kind of moment arises. The goal is to extend the therapeutic relationship into the week, to give the client a resource that carries some of the clinical thinking into the moments when I am not in the room.
That is what psychologically grounded digital support can do at scale. Not replace the clinician. Not simulate the relationship. But offer evidence-based reflection, regulation, and psychoeducation that helps people navigate their inner lives in real time, and arrive at their next session having done something useful with the week rather than simply survived it.
The goal is to extend the therapeutic relationship into the week, to give the client a resource that carries some of the clinical thinking into the moments when I am not in the room.
Why This Matters Most for Underserved Communities
For communities that face the most significant barriers to accessing formal mental health care, the between-session gap is not just a therapeutic optimization problem. It is often the primary reality of their mental health experience. They may see a therapist once a month, if they are fortunate enough to have one. They may be on a waiting list. They may have decided, based on prior experience, that the available options were not designed for them.
For these communities, between-session support is not supplementary. It is often the only support available. The quality of that support, its cultural grounding, its clinical integrity, its accessibility, matters more, not less. That is the problem Epiphany was built to solve.
Kamilah M. Woodson, Ph.D., is a licensed clinical psychologist, tenured Full Professor at Howard University, and the Founder and CEO of Epiphany Psychological Solutions. Learn more at epiphanypsy.com.
References
American Psychological Association. (2024). Health advisory: Use of generative AI chatbots and wellness applications for mental health. https://www.apa.org/topics/artificial-intelligence-machine-learning/health-advisory-chatbots-wellness-apps
Mausbach, B. T., Moore, R., Roesch, S., Cardenas, V., & Patterson, T. L. (2010). The relationship between homework compliance and therapy outcomes: An updated meta-analysis. Cognitive Therapy and Research, 34(5), 429-438. https://doi.org/10.1007/s10608-010-9297-z
Peretz, G., Taylor, C. B., Ruzek, J. I., Jefroykin, S., & Sadeh-Sharvit, S. (2023). Machine learning model to predict assignment of therapy homework in behavioral treatments. JMIR Formative Research, 7, e45156. https://doi.org/10.2196/45156
Ryum, T. (2024). Homework as a driver of change in psychotherapy. Journal of Clinical Psychology. https://doi.org/10.1002/jclp.23627
Hill, C. E., & Norcross, J. C. (2023). Psychotherapy skills and methods that work. Oxford University Press.
