collab w/ the wonderful @sgoldbergphd.bsky.social @lorihilt.bsky.social @hadarfisher.bsky.social @carolineswords.bsky.social and Dan Bolt

Conventional self-report and EMA measures of rumination provide distinct and clinically meaningful information. When deciding to use EMA in intervention studies, researchers should carefully consider the psychometric properties of their measures and the precise construct they intend to capture

We found larger changes in rumination for the conventional self-report measures relative to EMA. Notably, change in both self-report and EMA rumination accounted for unique variance in overall depressive symptom improvement, demonstrating incremental predictive validity. 5/n

Conventional self-report and EMA rumination were only modestly correlated, particularly w/ regards to change over time, which may be due to the lower reliability of change scores and that they tap different “selves” (e.g., experiencing vs. remembering self) 4/n

While the reliability of both conventional self-report rumination (at baseline and post-intervention) and mean EMA rumination scores were high, we found a substantial decline in reliability when assessing change over time, especially for EMA. 3/n

We analyzed data from 4 app-based meditation trials (N=412) to assess the reliability, validity, and sensitivity to change of conventional retrospective self-report vs. EMA measures of improvement in rumination. 2/n

Thanks, Charly. in general it was teens with lower levels of symptoms (e.g., fewer emotional difficulties (SDQ), lower depressive symptoms (CESD), and less executive function difficulties (BRIEF)). Details in the paper

Preprint now available here: osf.io/preprints/ps...

Coauthors @willemkuyken.bsky.social @hinzeverena.bsky.social and bluesky-less Drs. Ren, Dalgeish, Ford, Greenberg, and Montero-Marin (hope not missing any bluesky accounts)

Focusing only on individual-level interventions risks overlooking broader influences. Future precision prevention should integrate multilevel data—capturing both individual and contextual drivers of outcomes.
6/n

Youth mental health ≠ just individual factors.
School climate, poverty & inequities matter too.
Prevention must move beyond one-size-fits-all to integrate both personal + systemic influences.
5/n

Machine learning can identify which adolescents are more likely to benefit from SBMT vs. TAU.

This may point to a path toward precision prevention in schools. Next step: prospective trials testing whether algorithm-guided intervention assignment improves outcomes. 4/n