Suicide is amongst the top three leading causes of mortality in the United States for individuals aged 15 to 44 years (National Institute of Mental Health, 2022) and accounts for over 700,000 annual deaths globally (World Health Organization, 2021). Despite increased efforts over the last twenty years to understand its etiology, identify factors moderating risk, and evaluate effectiveness of available interventions, suicide rates in the United States have increased 30 % since 2000 (Garnett et al., 2022).

Suicidal crises, regardless of time since the original plan, usually have rapid onset with only a few minutes for intervention before a life-threatening attempt is initiated (Deisenhammer et al., 2009). Adding to this challenge, 60 % of suicides are completed on the first attempt (Bostwick et al., 2016). Established pharmacological and behavioral treatments for suicidal crises generally take weeks to have noticeable effects (Harmer et al., 2009; Levy et al., 2020) and for individuals hospitalized for suicidal crises, risk of suicide peaks in the months following discharge (Qin and Nordentoft, 2005). Hence, there is a critical need for interventions that both reduce the risk of attempt and have lasting effects.

The U.S. Food and Drug Administration (FDA) approved the N-methyl D-aspartate (NMDA) receptor antagonist esketamine nasal spray in 2020 for major depressive disorder (MDD) with suicidal thoughts or actions. Ketamine, the racemic generic version of esketamine used off-label for psychiatric indications, can have rapid antidepressant effects beginning at 40 min and peaking at 1 day after a single dose (Hochschild et al., 2021; Kishimoto et al., 2016). Additionally, a course of repeated administrations of ketamine shows persistent benefits following the acute treatment period (Phillips et al., 2019). However, meta-analyses have found heterogenous effects of ketamine on suicidal ideation (Reinstatler and Youssef, 2015; Wilkinson et al., 2018; Witt et al., 2020). Further, a significant proportion of individuals do not respond adequately to ketamine (Ballard et al., 2018; Niciu et al., 2014; Rong et al., 2018). Given the time intensive and monetary costs of off-label ketamine treatment, it is particularly important to identify markers of optimal response. Unfortunately, only a few reliable predictors of ketamine treatment response (i.e., greater treatment resistance, higher body mass index (BMI) and family history of alcohol use disorder) have emerged in preliminary studies. Further, these findings do not necessarily generalize to suicidal symptoms (Meshkat et al., 2021; Niciu et al., 2014; Price et al., 2022; Rong et al., 2018).

The aim of the current study was to extend our previous work on real-world patients’ antidepressant response patterns after repeated intravenous (IV) ketamine infusions (O'Brien et al., 2023, 2021). We focused on potential gaps between ketamine's reported efficacy in treating suicidality in controlled clinical research trial settings and its effectiveness for the much larger, and complex group of patients seeking this treatment in community-based clinical practices. We used a hypothesis-free, data-driven latent-class analysis to evaluate the trajectories of changes in symptoms of suicidality. Analyses were conducted to (i) find specific groups of patients showing similar trajectories in suicidality during ketamine treatment and (ii) examine specificity of demographic and clinical characteristics relative to response trajectory groups. In addition to evaluating the trajectories of overall symptom severity, we evaluated domains that have been demonstrated to be highly relevant to the experience of suicidal thoughts and behaviors including (i) pessimism, helplessness, perceived lack of social support and despair; (ii) impulsivity; and (iii) active suicidal ideation.