Internet CBT for pediatric OCD maintains, or even increases, effectiveness over 12 months

A follow-on analysis found tentative evidence that, unlike in-person CBT, the online intervention could become more effective over the course of a year.
By Dave Muoio
02:38 pm

New data suggest that an Internet-delivered cognitive behavioral therapy (ICBT) intervention not only remained effective after 12 months among adolescents with obsessive-compulsive disorder, but seemed to have an even greater impact on their symptoms with time.

Published last week in npj digital medicine, the analysis updated a previously reported three-month study with one-year data from the same group of participants. Although the researchers noted that the study's naturalistic follow-up approach prevents any claims of causality, they said that the information raises new questions regarding the therapeutic mechanisms of ICBT and how long it should be offered before switching to a different treatment option.

"Although tentative, these results indicate that the treatment effect of ICBT for pediatric OCD may accrue in an atypical manner over a longer period of time," researchers from the Karolinska Institutet and Stockholm Health Care Services wrote. "If confirmed, these results have important implications both for researchers and clinicians."


The average age of the 61 adolescents who received ICBT was about 14-and-a-half. Eleven (18%) were receiving selective serotonin reuptake inhibitor (SSRI) medication throughout the course of the study and 14 (23%) had previously received at least five sessions of in-person CBT prior to the study.

Fifty-three of these participants completed a 12-month follow-up. Among these, the researchers saw mean scores on an OCD symptom scale (the CY-BOCS) significantly improve from pre-ICBT baseline to three months (p < .0001), post-ICBT to three months (p < .001) and from three months to 12 months (p < .0001).

While responder rates increased 33% from pre-ICBT baseline to immediately post-ICBT, the minor bump to 39% by three-month follow-up outlined in the original investigation was not a significant change. However, the researchers found a significant increase in responses between the three-month and 12-month check-ins, with the latter having increased to a rate of 73% (p < .001). The researchers saw a similar nine-month jump in remitters (18% to 40%; p < .01).

SSRI treatments did not have any impact on these scores across the study period, nor did the receipt of additional face-to-face CBT. The researchers also noted that they were able to develop a machine learning algorithm to predict patients' response to ICBT "with good to excellent accuracy," suggesting that such an approach "could be useful when exploring the question of patient selection for ICBT."


The researchers conducted a 12-month follow-up of their prior investigation of adolescents between 12 and 17 years with OCD. To be included, they needed to have daily access to the Internet, an OCD diagnosis, a total CY-BOCS score of 16 or more, and a parent able to participate. Those who had certain other diagnoses, such as autism spectrum disorder or bipolar disorder, were excluded.

Evaluations at specific time points were administered in person by trained psychologists using the CY-BOCS tool. The researchers interviewed the child and a primary parent at these time points. Those who did not respond to the treatment at the three-month follow-up were offered a referral to standard treatments.

The intervention itself comprised various means of psychoeducation, exposure and response prevention, cognitive strategies, and relapse-prevention materials, delivered via online portal and designed for presentation over 12 weeks. A separate account was also provided to parents that included a smaller quantity of family and caregiver information. Both the child and larger family could also receive written or telephone support from a licensed psychologist.

A response was defined by the researchers as CY-BOCS scores decreasing by at least 35% and an accompanying CGI-I rating indicating "very much improved" or "much improved." Remission was a CY-BOCS score of 12 or lower, and a CGI-S rating of "not at all ill" or "borderline mentally ill."


The researchers described a fair amount of data investigating the effectiveness of ICBT interventions for OCD, but little information regarding longer-term follow-up among a younger cohort. Still, the data that is available suggests that the treatment is feasible and acceptable among these patients, and supports the idea that treatment gains may improve with continued use.

"This is normally not to be expected in standard face-to-face CBT, where the treatment effects obtained at post-treatment tend to be maintained without changes occurring during the follow-up," they wrote.

Across the digital health industry, CBT is the cornerstone of many digital tools targeting behavioral health or wellness. These include Big Health's Sleepio and Daylight, a mental health program offered by Omada Health and digital OCD identification and management platform NOCD, to name a few.


"This study indicates that the therapeutic gains of ICBT for pediatric OCD are not only maintained over time, but that further improvements can be expected up to one year after the end of treatment. Well-controlled replications are warranted, as well as studies that explore the causes of the observed delayed treatment effect of ICBT," the researchers wrote.


The latest news in digital health delivered daily to your inbox.

Thank you for subscribing!
Error! Something went wrong!