Texas Research Center for Social Dynamics

Study Period: September 2009 – May 2012

Executive Summary

Background. Adolescent depression is a major public health concern, with prevalence rates nearly twice as high in girls as in boys by mid-adolescence. While biological and social-structural factors have been examined, increasing attention has been directed toward peer influence processes—specifically co-rumination, defined as excessive, repetitive discussion of personal problems within a dyadic friendship. Co-rumination is strongly associated with both friendship closeness and increased depressive symptoms, creating a paradoxical bond. Longitudinal observational studies suggest that depressive affect can spread through female friendship networks, but no experimental trial had tested whether altering co-ruminative communication patterns could reduce depressive contagion.

Objective. The CoRUM trial (Co-Rumination and Mood) was a cluster-randomized controlled trial in 24 middle-school homeroom classes, conducted from 2009 to 2012, testing a brief, school-based intervention designed to reduce co-rumination among adolescent girls. The primary aim was to measure whether the intervention reduced depressive symptom scores at 12 months compared to a standard health curriculum. Secondary aims included assessing the incidence of new depressive episodes and the strength of emotional contagion within friendship networks.

Methods. Twenty-four homerooms (n=612 female students, mean age 13.2) were randomized to either the CoRUM intervention (12 classrooms, 304 students) or a nutrition-focused attention control (12 classrooms, 308 students). The CoRUM intervention consisted of six 45-minute sessions teaching students to recognize co-rumination, distinguish it from supportive listening, and practice solution-focused communication. Depressive symptoms were measured with the Children’s Depression Inventory (CDI-2) at baseline, 6 months, and 12 months. Network data were collected to map friendship ties and model emotional influence.

Results. At 12 months, CDI-2 scores in the CoRUM arm decreased by 2.1 points (95% CI: -3.4 to -0.8), while the control arm increased by 1.7 points (95% CI: 0.4 to 3.0), a net difference of -3.8 points (p<0.001). The incidence of new depressive episodes (CDI-2 above clinical cutoff) was 8.2% in the intervention group versus 16.4% in the control group (odds ratio 0.44, p=0.002). Network analyses revealed that in control classrooms, depressive affect spread through female friendship ties with a contagion parameter of 0.12 (p<0.001), whereas in intervention classrooms this parameter was reduced to a non-significant 0.03. Mediation showed that the reduction in co-rumination accounted for 55% of the intervention’s effect on depressive symptoms.

Conclusion. Co-rumination among adolescent girls is a modifiable social behavior that directly contributes to the spread of depressive symptoms within female peer networks. Teaching girls to shift from passive, repetitive problem-talk to solution-oriented support significantly reduces depressive contagion and prevents new cases. The findings highlight the need for school-based mental health programs to address the gendered peer dynamics that amplify emotional distress.

1. Introduction

Adolescent depression has seen rising prevalence, with a marked sex disparity emerging after puberty: by age 14, girls are twice as likely as boys to report elevated depressive symptoms. While hormonal and social-structural explanations have been offered, peer influence processes have gained prominence. The construct of co-rumination—extensively discussing and revisiting problems, focusing on negative feelings, and speculating about consequences within a friendship—was introduced by Rose (2002). Co-rumination is paradoxically linked to both higher relationship quality and increased internalizing symptoms. Girls report higher levels of co-rumination than boys, and longitudinal studies show that co-rumination predicts increases in depression over time, particularly when friends share similar emotional vulnerabilities.

Observational studies have demonstrated that depression can cluster within adolescent female friendship groups, suggesting a social contagion process. However, causal evidence that reducing co-rumination could break this cycle had been absent. The CoRUM trial was designed to fill this gap, evaluating a targeted classroom intervention to curtail co-ruminative communication and measuring its impact on depressive symptom trajectories and network-level emotional spread.

2. Methods

2.1 Trial Design and Participants

The CoRUM trial was a cluster-randomized, parallel-group controlled study conducted in 24 eighth-grade homeroom classes across six public middle schools in Austin, Texas, between September 2009 and May 2012. A total of 612 female students (mean age 13.2 years) provided parental consent and assent. Homerooms were randomly assigned 1:1 to the CoRUM intervention or a nutrition education attention control, stratified by school. Students were not informed of the study’s depression focus to minimize demand characteristics.

2.2 Interventions

The CoRUM intervention comprised six 45-minute weekly sessions delivered by trained research staff. Sessions covered: identifying co-rumination versus supportive listening, understanding the emotional consequences, practicing “problem-solving talk” (brainstorming solutions rather than simply revisiting distress), and role-playing alternative communication styles. The control arm received six sessions on healthy eating and physical activity, matched for time and attention.

2.3 Outcomes

The primary outcome was change in CDI-2 total score from baseline to 12 months. Secondary outcomes included new-onset depressive episodes (CDI-2 ≥20) and friendship network contagion parameters estimated via stochastic actor-oriented models. Co-rumination was measured with the Co-Rumination Questionnaire (CRQ) at all time points. Network data were collected by asking students to nominate up to 10 closest friends within the grade.

2.4 Statistical Analysis

Linear mixed-effects models with random intercepts for classrooms were used for continuous outcomes, and logistic mixed models for incidence. Network models examined the influence of a friend’s depressive symptoms on a girl’s own symptoms over time, controlling for homophily and shared environment. Mediation used bootstrapping.

3. Results

At baseline, mean CDI-2 was 12.4 (SD 6.1) in both arms. At 12 months, the intervention group showed a 2.1-point decrease, while controls increased by 1.7 points (adjusted between-group difference -3.8, 95% CI: -5.2 to -2.4, p<0.001). Incidence of new depressive episodes was 8.2% (25/304) in the intervention group versus 16.4% (50/308) in controls (OR 0.44, 95% CI: 0.26–0.74, p=0.002). Network analyses showed that in control classrooms, a friend’s depression score significantly predicted an increase in a girl’s own score over 6 months (contagion estimate 0.12, SE 0.03, p<0.001), while in intervention classrooms this effect was reduced to non-significance (0.03, SE 0.03, p=0.31). Co-rumination scores decreased significantly in the intervention arm (mean change -0.35 SD vs. +0.18 SD, p<0.001) and mediated 55% of the intervention effect on depression.

4. Discussion

The CoRUM trial provides causal evidence that co-rumination among adolescent girls is a key mechanism driving depressive contagion within female peer networks. By teaching girls to shift toward solution-oriented communication, the intervention reduced both individual depressive symptoms and the network spread of distress. The findings highlight a gendered social dynamic: girls’ tendency to co-ruminate, while fostering intimacy, creates a relational environment that incubates and transmits depression. Public health strategies for adolescent mental health must address these communication patterns explicitly.

Limitations include the single school district, reliance on self-reported depression, and 12-month follow-up. Future studies should examine longer-term durability and test the approach in more diverse settings.

5. Conclusion

The CoRUM trial demonstrates that female adolescent co-rumination is a modifiable risk factor for depression contagion. Interrupting this cycle with a brief school intervention significantly lowers depressive symptoms and prevents new cases. The Texas Research Center for Social Dynamics advocates for integrating such social-dynamics-informed curricula into standard mental health promotion efforts.

6. References

1. Rose, A.J. (2002). Co-rumination in the friendships of girls and boys. Child Development, 73(6), 1830–1843.
2. Schwartz-Mette, R.A., & Rose, A.J. (2012). Co-rumination mediates contagion of internalizing symptoms within youths’ friendships. Developmental Psychology, 48(5), 1355–1365.
3. Prinstein, M.J. (2007). Moderators of peer contagion: A longitudinal examination. Journal of Clinical Child & Adolescent Psychology, 36(2), 194–205.
4. Hankin, B.L. (2015). Depression from childhood through adolescence: Risk mechanisms across multiple systems. Current Directions in Psychological Science, 24(6), 473–479.
5. Banny, A.M., Heilbron, N., Ames, A., & Prinstein, M.J. (2011). Relational benefits of co-rumination: An observational study. Journal of Research on Adolescence, 21(4), 855–869.

February 2017