Prospective Relations between Life Stress, Emotional Clarity, and Suicidal Ideation in an Adolescent Clinical Sample

ABSTRACT

Objective

Although life stress has been linked to adolescent suicidal ideation, most past research has been cross-sectional, and potential processes characterizing this relation remain unclear. One possibility may be a lack of emotional clarity. Informed by stress generation, the current study examined prospective relations between episodic life stress, lack of emotional clarity, and suicidal ideation in an adolescent clinical sample.

Methods

The sample consisted of 180 youths (M_age = 14.89; SD = 1.35; 71.7% female; 78.9% White; 43.0% sexual minority) recruited from a psychiatric inpatient facility. Suicidal ideation severity was assessed at baseline and 18-month follow-up. Lack of emotional clarity and life stress were assessed at baseline, as well as 6-, and 12-month follow-ups. Two random-intercepts cross-lagged panel models were created to estimate within-person relations for variables of interest.

Results

At the within-person level, lack of emotional clarity at baseline predicted greater 6-month impact of interpersonal dependent stressors (b = 0.29, p = .012, 95% CI [0.07, 0.52]), which subsequently predicted a greater 12-month lack of emotional clarity (b = 0.41, p = .005, 95% CI [0.12, 0.70]). Next, a 12-month lack of emotional clarity but not interpersonal dependent stress, predicted greater 18-month suicidal ideation (b = 0.81, p = .006, 95% CI [0.23, 1.30]; R² = .24, p < .001). No significant relations were found for the lack of emotional clarity and independent stress.

Conclusions

Results support the stress generation hypothesis and suggest that future research should be conducted evaluating whether bolstering youth’s understanding of their emotional experiences may reduce subsequent suicidal ideation.

Acknowledgments

Preparation of this manuscript was supported in part by the National Institute of Mental Health of the National Institutes of Health under Award Numbers RF1MH120830, R01MH101138 R01MH115905, R01MH124899 and R21MH130767 and the American Foundation for Suicide Prevention (PDF-0-10-252) to RTL. RTL currently serves as a consultant to Relmada Therapeutics. The content is solely the responsibility of the authors and does not necessarily represent the official views of the funding agencies or Relmada Therapeutics.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Supplementary Material

Supplemental material for this article can be accessed online at https://doi.org/10.1080/15374416.2024.2344735.

Notes

¹ Age was considered as a potential covariate when models were initially created, though age was unrelated to all variables of interest at the bivariate level. Thus, age was not included in the final models for parsimony. In sensitivity analyses where lifetime suicide attempt history (i.e., presence versus absence) was included as a covariate in both models, the pattern of results remained unchanged. Results without lifetime suicide attempt history are presented here for parsimony.

² Consistent with prior studies using life stress interviews, the impact of independent stress was aggregated across all domains (i.e., interpersonal, or otherwise) in the current study. Nevertheless, sensitivity analyses were conducted using interpersonal independent stress only. An equivalent pattern of results emerged, so results with the impact of all independent stressors is presented here.

³ Given significant temporal fluctuations in life stress, ER difficulties, and SI (see Czyz et al., 2022; López et al., 2022; Weber et al., 2022) as well as the specific mediation hypotheses for the present study, 12-month within-person components of emotional clarity and life stressors, instead of their between-level components, were used to predict 18-month SI.

⁴ Orth et al. (2022) suggest that standardized regression weights (i.e., $\mathit{\beta }$) of .02, .05, and .11 for cross-lagged effects in RI-CLPMs correspond to small, medium, and large effect sizes, respectively. Thus, all significant cross-lagged paths demonstrated large effect sizes in this model (see Table 2 for corresponding $\mathit{\beta }\mathrm{s}$).

Additional information

Funding

This work was supported by the National Institute of Mental Health [NIMH; grant numbers R01MH101138, R01MH115905, RF1MH120830, R21MH130767, and R01MH124899] and the American Foundation for Suicide Prevention [AFSP; PDF-0-10-252].