An online therapy program helped students. Half its users spent under twenty minutes.

Randomised controlled trial (RCT). 6,205 people. 6-month access to the program; follow-ups at 6 weeks, 6 months and 2 years (recruited Oct 2019 - Nov 2021, data collection completed Oct 2023).

A trial. Chance decided who did what.

Who: Undergraduates aged 18+ at 26 US colleges and universities who screened (self-report) at clinical level (4,538) or high risk (1,667) for generalized anxiety, social anxiety, panic, depression and/or eating disorders (not anorexia), not in psychotherapy in the past month, owning a smartphone; mean age 20.2, 26.5% male; 39,194 students answered the screen

• Had a disorder, 6 mo-4.9 pts56% vs 61%
• Had a disorder, 2 yrs-3.8 pts55.5% vs 59.3%
• Got any help, 6 mo74 vs 30% of students

More to know, from our read of the paper:

• Disorder status is based on validated self-report questionnaires with clinical cut-offs, not clinician interviews: say 'by their own answers' / 'self-reported'
• The comparison is referral to campus care, not in-person therapy; both groups also got screening and feedback, and both groups improved over time
• At 6 weeks the benefit was only significant in students with clinical-level symptoms, not in the high-risk group; by 6 months and 2 years the full sample benefited
• Symptom-score effects are small (between-group d about 0.06-0.19). Panic symptoms: no difference at any time. Generalized anxiety: only at 6 weeks. Eating-disorder and social-anxiety symptoms: not at 2 years. Depression: at all three times
• Engagement was low: among the 2,308 who used the program, median 4 visits and median 18.0 minutes in total (mean 98.8); on average 2.4 modules of 6-8

In this randomized trial at 26 US colleges, 6,205 students with signs of anxiety, depression or an eating disorder (by their own questionnaire answers) got either a coached online CBT program or a referral to campus care. After 6 months, 56% vs 61% still had at least one disorder; after 2 years, 55.5% vs 59.3%. 74% started the program, while 30% of the referral group reported therapy or new medication. Half of the program's users spent 18 minutes or less in it, and panic symptoms did not change more than in the other group.

We read the abstract and the full text of this paper and checked every number on this page against it (16 facts, checked on 2026-10-29). The study design is named above. Cohort and other observational studies are described as linked, not proven. Found an error? Email theis@onestudyaday.com.