Feeling Worse Doesn’t Mean Seeking Help: Severity Fails to Predict Care Booking on Saudi Mental Health App

On a commercial digital mental health platform in Saudi Arabia, users who reported the most severe symptoms at their first self-assessment were no more likely to book a session with a professional within the following 90 days than those reporting minimal symptoms. That is the central finding of a retrospective cohort study of 5,524 unique users of the Talbinah platform, published as a STROBE-compliant analysis of linked screening and booking records. The assumption that symptom severity drives help-seeking underpins much of digital mental health design: platforms routinely deliver self-administered questionnaires expecting that users scoring in the severe range will be nudged, or will naturally move, toward professional care.

Yet that assumption is rarely tested with participant-level data linking a screening score to an actual booking, and it has almost never been examined in an Arabic-speaking Gulf population. The researchers analyzed 10,100 first assessments spanning four symptom domains—depression, generalized anxiety, panic, and obsessive-compulsive symptoms—contributed by 5,524 users. Records were linked to demographic information and session-booking data through a pseudonymous, SHA-256 hashed user identifier.

The primary outcome was any professional session booked within 90 days of the assessment. Depression severity was banded using established PHQ-9 cut-offs and anxiety using GAD-7 cut-offs, while panic and obsessive-compulsive measures were banded by percentage of scale maximum, since no validated severity thresholds exist for them. The user base was young, with a mean age of 26.7 years, and predominantly female at 78.1 percent.

Overall, 16.6 percent of first assessments were followed by a booking within 90 days. In multivariable logistic regression models adjusted for age, sex, and device type, no severity band on any of the four scales predicted higher odds of booking. Comparing severe with none-or-minimal symptoms, the adjusted odds ratios were 0.71 for depression, 0.76 for generalized anxiety, 0.83 for panic, and 1.01 for obsessive-compulsive symptoms, with confidence intervals that all included or approached the null.

One contrast did reach statistical significance, and it pointed the opposite way: moderate depression was associated with lower odds of booking than minimal symptoms, with an adjusted odds ratio of 0.66. What did predict booking? Demographic and access-related attributes, not symptom load.

The null severity–booking association proved robust. It held across 30-, 90-, and 180-day follow-up windows, persisted when the outcome was restricted to attended sessions rather than mere bookings, and remained when percentage-of-maximum banding was applied to all four scales. A nominal association for moderate anxiety appeared under percentage banding but did not survive when the validated GAD-7 cut-offs were used.

The authors flag that estimates for the panic and obsessive-compulsive scales, which lack established severity thresholds, should be treated as exploratory. The study's limitations follow from its design. As a retrospective analysis of a single commercial platform's users, it captures people who had already downloaded an app and completed a self-assessment, a self-selected and digitally literate group.

Even with those caveats, the implication is uncomfortable for the industry: self-screening on its own appears insufficient as a pathway into care.

Subject of Research: Whether symptom severity at first digital self-assessment predicts subsequent booking of professional mental health care

Article Title: Symptom severity at first digital self-assessment is not associated with greater subsequent booking of professional care on a Saudi mental health platform: a cohort study of 5,524 users

Article References: Alkhamees, A. A. (2026). Symptom severity at first digital self-assessment is not associated with greater subsequent booking of professional care on a Saudi mental health platform: a cohort study of 5,524 users. Frontiers in Psychiatry, 17, 1928345. https://doi.org/10.3389/fpsyt.2026.1928345

Image Credits: AI Generated

DOI: 10.3389/fpsyt.2026.1928345

Keywords: digital mental health, symptom severity, help-seeking, PHQ-9, GAD-7, Saudi Arabia, cohort study, depression, anxiety, screening, care access, telehealth