Cognitive control impairment in suicide behaviors: what do we know? A systematic review and meta-analysis of Stroop in suicide behaviors.

Abstract

Background: Suicidal behavior results from a complex interplay between stressful events and vulnerability factors, including cognitive deficits. Poorer performance on the Stroop task, a measure of cognitive control, has been associated with suicidal behavior in numerous studies. The objective was to conduct an updated systematic review of the literature on the Stroop task as a neuropsychological test of vulnerability to suicidal acts in patients with mood and other psychiatric disorders, while also looking at how the type (classic versus emotional) or the version (paper or computerized) of the Stroop task, as well as the characteristics of the patient (clinical population, age, sex) moderated the Stroop effect.

Methods: A search on Medline, Embase, PsycInfo databases, and article references was performed. 53 studies (6781 participants) met the selection criteria. Interference time and errors of the Stroop Test were assessed in at least 3 studies to be analyzed. Moderators, such as the type (classic versus emotional) of the Stroop task and the characteristics of the patient (clinical population, age, sex) were also assessed.

Results: Interference time on Stroop performance was lower in suicide attempters than in patient controls (g = 0.20; 95%CI [0.10-0.30]) and healthy controls (g = 0.79; 95 % CI [0.29-1.29]), with patient controls scoring lower than healthy controls (g = -0.63; 95%CI [-1.01-0.25]). This was moderated by age and having a mood disorder. In terms of interference errors, suicide attempters performed worse than healthy controls (g = 0.57; 95%CI [0.01-1.15]) but did not perform differently from patient controls (g = 0.20; 95 % CI [-0.06-0.45]). Patient controls also did not score differently than healthy controls (g = -0.18; 95 % CI [-0.54-0.18]). There was a significant moderation effect for the type (i.e., original Stroop task) and version (i.e., paper format) of the Stroop task, and for some characteristics of the patient (i.e., older patients and having a mood disorder).

Conclusions: Cognitive control impairment was associated with a history of suicidal behavior in patients, especially in older populations and those with mood disorders, however this result was moderated by outcome measure (interference time vs. errors), the type (i.e., original Stroop task) and the version (i.e., paper format) of the Stroop task. Cognitive control processes may be an important factor of suicidal vulnerability. Choosing the right neurocognitive test in the right population to detect suicide vulnerability is important direction for future research.