Incorporating Data From Text Messages and Email Into Psychotherapy: Randomized Controlled Trial.

```mdx

title: "Adding Text Message Data to Psychotherapy: What a Randomized Trial Found"

description: "A randomized controlled trial tests whether reviewing patients' text messages during psychotherapy improves depression, anxiety, quality of life, or the therapeutic relationship."

date: "2025"

studyType: "Randomized controlled trial"

sampleSize: 101

journal: "JMIR Mental Health"

pubmedId: "42896155"

TL;DR: In a randomized trial of 101 adults in psychotherapy for depression or anxiety, reviewing patients' text messages before sessions did not significantly improve depression or anxiety scores or the therapeutic alliance, though it did change what clinicians did and produced a modest pain-related gain on one quality-of-life measure. This is a single small trial, so the findings are preliminary.

What the study found

Digital communication now generates a large amount of data about people's daily lives, and researchers have asked whether clinicians can use it to guide care. Most prior work on this topic has focused on digital phenotyping, which uses passive sensor data such as phone usage or movement. Far less attention has gone to whether the content of personal messages, such as texts and emails, improves psychotherapy outcomes. This trial, called the Electronic Media and its Impact on Psychotherapy (EMAP) study, addressed that gap directly.

Researchers at an academic psychiatric hospital randomized adults and older adults receiving outpatient or partial hospitalization psychotherapy for primary depression or anxiety to one of two conditions. In the electronic media-enhanced therapy (EME) group (n=52), research staff reviewed participants' text messages before sessions using personalized mood-related key terms and presented aggregated findings on a HIPAA-compliant clinician dashboard. The treatment-as-usual (TAU) group (n=49) received standard care without any message review. Outcomes included change scores on the Patient Health Questionnaire-9 (PHQ-9) for depression, the Generalized Anxiety Disorder-7 (GAD-7) for anxiety, the 36-item Short Form Health Survey (SF-36) for health-related quality of life, and the Working Alliance Inventory-Short Revised (WAI-SR) for therapeutic alliance.

On the primary clinical questions, the results were largely null. The trial found no statistically significant between-group differences in anxiety or in working alliance. The abstract reports that integrating personalized text message data did not significantly improve depression either. Among the subset of participants with SF-36 follow-up data (n=65), the EME group showed greater improvement on the pain subscale than the TAU group. This is a single subscale finding in a subsample, and it should not be read as evidence of broad quality-of-life benefit. The authors also found that the alliance was neither improved nor worsened by the intervention, which means the approach did not appear to damage the therapeutic relationship despite its privacy implications.

The more consistent signal concerned clinical decision-making rather than outcomes. Within the EME group, a greater frequency of access to the collateral information was associated with larger treatment-related changes. Access to additional collateral information was linked to more clinical actions, but it was not associated with better distal outcomes such as symptom scores. In other words, clinicians who had more information did act on it, but acting on it did not translate into measurable symptom improvement over the short treatment period studied. The authors describe these effects as nuanced and context-dependent, and that characterization fits the data. The moderation analyses are exploratory, and the study was not powered to detect small effects across all outcomes.

What this means for you

This trial does not support the claim that reviewing text messages makes psychotherapy more effective for depression or anxiety. Its primary symptom outcomes were not improved, so clinicians and patients should not expect this approach to change symptom trajectories on the basis of this evidence alone.

The findings do show that additional personal data changes how clinicians behave. That is worth weighing, because clinical actions that do not improve outcomes can still carry costs, including time, attention, and the possibility of misreading context from informal messages. Patients considering whether to share digital communication with a therapist should weigh privacy, consent, and the intended use of that information.

Because psychotherapy outcomes depend heavily on the therapeutic relationship, it is reassuring that the alliance was not harmed. Still, the trial was short term and small, and its findings may not generalize to other clinics, populations, or longer treatment courses. Replication in larger, longer trials would be needed before any recommendation could be made.

Symptom improvement in psychotherapy is driven by well-established factors. Our habits and social connection resources cover approaches that have a stronger research base for everyday mental wellbeing, though they are not substitutes for professional care.

Study details

Sources

1. Electronic Media and its Impact on Psychotherapy (EMAP) study. Incorporating Data From Text Messages and Email Into Psychotherapy: Randomized Controlled Trial. JMIR Mental Health. PubMed. https://pubmed.ncbi.nlm.nih.gov/42896155/

```