# Intimate disclosure in online‐only friendships predicts adolescents' depressive symptoms

## TL;DR

Adolescents who have online-only friendships may experience higher depressive symptoms, especially if they engage in intimate disclosures within these friendships.

## Contribution

This study reveals that intimate disclosure in online-only friendships predicts increased depressive symptoms in adolescents.

## Key findings

- The presence of online-only friendships correlates with depressive symptoms at baseline.
- Higher levels of intimate disclosure in online-only friendships predict increased depressive symptoms later.
- Depressive symptoms do not significantly predict future levels of intimate disclosure.

## Abstract

Adolescents frequently use social media to form online‐only friendships (OOFs; i.e., friendships that are exclusive to digital platforms), raising questions about their emotional impact. While OOFs may provide social support and increase well‐being, they could also displace in‐person relationships or increase exposure to online risks. This study examined longitudinal associations between OOFs, intimate disclosure within these friendships, and depressive symptoms among adolescents. A total of 1089 teens aged 13–19 (M = 15.65, SD = 1.19; 581 girls, 492 boys, and 16 another gender identity) completed online surveys as part of a longitudinal study investigating adolescent development at two timepoints: October 2019 (T1) and February 2020 (T2). Teens reported whether they had any OOFs, and, if so, levels of intimate disclosure within their OOFs, as well as depressive symptoms. There was a significant correlation between OOF status and depressive symptoms at T1 (r = .13, p < .01). However, T1 OOF status was not significantly associated with depressive symptoms at T2 (unstandardized b = 0.02, p = .64, IRR = 1.02), or vice versa (OR = 1.02, p = .055, CI = 1.00–1.05), controlling for age, gender, and baseline levels of the outcome variable. Higher T1 intimate disclosure with OOFs predicted higher T2 depressive symptoms (unstandardized b = 0.088, p = .011, IRR = 1.09), while T1 depressive symptoms did not significantly predict T2 intimate disclosure (unstandardized b = 0.012, p = .069, CI = −0.001–0.03), controlling for age, gender and baseline levels of the outcome variable. These results suggest that while the sole presence of OOFs is associated with depressive symptoms, intimate disclosure within these friendships may predict increases in depressive symptoms.