Only the trajectory of depression symptoms that demonstrated the highest mean scores and variability over time predicted suicide attempt in high-risk young adults, according to a study analyzing the most important clinical predictors of suicide attempt.
“Despite years of mental health research, our ability to predict suicidal behavior is only slightly better than chance,” Nadine M. Melhem, PhD, from the department of psychiatry, University of Pittsburgh School of Medicine, told Healio Psychiatry. “While psychiatric diagnoses are quite useful, diagnoses alone don’t do a great job because they are labels that often don’t change. We wanted to develop a predictive model that would identify symptoms that can change over time and signals risk for suicidal behavior.”
In their longitudinal study, researchers examined which clinicals trajectories — depression symptoms, hopelessness, impulsivity, aggression, impulsive aggression and irritability—best predicted suicide attempt in the offspring of parents (or probands) with mood disorders.
Probands and their children were interviewed at baseline and at yearly follow-ups for 12 years. The investigators evaluated lifetime and current psychiatric disorders and suicidal behavior as well as participant-reported questionnaires to determine severity of symptoms. For model evaluation, Melhem and colleagues used 10-fold cross-validation. They also used machine learning regression approaches to determine a subset of predictors with the strongest effects.
Of the 663 offspring, 71 (10.7%) had suicide attempts over the study period. Analysis revealed that the trajectory of depression symptoms with the and variability over time (OR = 4.72; 95% CI, 1.47-15.21).
Other predictors of suicide attempt included:
- younger age (OR = 0.82; 95%CI, 0.74-0.9);
- lifetime history of unipolar disorder (OR = 4.71; 95% CI, 1.63-13.58);
- lifetime history of bipolar disorder (OR = 3.4; 95% CI, 0.96-12.04);
- history of childhood abuse (OR = 2.98; 95% CI, 1.4-6.38); and
- proband actual attempt (OR = 2.24; 95% CI, 1.06-4.75).
After computing a risk score using these statistically significant predictors endorsed from the models, Melhem and colleagues found that a score of or higher resulted in the highest sensitivity (87.3%) and moderate specificity (63%; area under the curve = 0.8) for suicide attempt. Although the positive predictive value was low, rose as the prevalence of attempt increased, according to the study.
“Our prediction risk score is based on variables that are already or should be collected as part of the clinical assessment,” Melhem told Healio Psychiatry. “Clinicians must pay particular attention to the severity of current and past depressive symptoms and try to reduce their severity and fluctuations to decrease suicide risk.” – by Savannah Demko
Disclosure: Melhem reported support from the American Foundation for Suicide Prevention, Brain and Behavior Research Foundation and NIMH.