Obsessive-Compulsive Disorder (OCD) is a psychiatric disorder characterized by recurrent, intrusive, and unwanted thoughts, images, or urges (obsessions) and repetitive actions, behaviors, or mental rituals (compulsions). The obsessions and compulsions often severely impair an individual's daily functionality without treatment (Skoog and Skoog, 1999). OCD is also associated with considerable morbidity and mortality, e.g., increased risk of suicide (Albert et al., 2019; Fernández de la Cruz et al., 2022). In fact, prior to the advancements in evidence-based treatments, OCD was listed by the World Health Organization as one of the top ten conditions associated with the greatest financial loss and decrease in quality of life in the 1990's (Bobes et al., 2001). Since then, the development and dissemination of effective treatments for OCD has decreased the morbidity substantially for many patients (Hirschtritt et al., 2017).

The current first-line treatment for OCD includes a combination of selective serotonin reuptake inhibitors (SSRIs) and cognitive behavioral therapy (CBT) (Bloch et al., 2013). However, even with evidence-based treatments, 40–60% of individuals still experience OCD symptoms that can significantly impair their quality of life (Bloch et al., 2013; Garnaat et al., 2015). Over the last couple of decades, there has been a considerable number of randomized clinical trials (RCTs) evaluating the efficacy of novel treatments for adults with OCD, including psychotherapeutic (Andersson et al., 2012, 2015; Lundström et al., 2022) and pharmacological interventions, such as antipsychotics (Bloch et al., 2006), N-methyl-D-aspartate (NMDA) receptor modulators (Pasquini and Biondi, 2006; Poyurovsky et al., 2005; Stewart et al., 2010), opioid agonists (Koran et al., 2005; Shapira et al., 1997), and ketamine (Rodriguez et al., 2013), and invasive/non-invasive neuromodulation, such as deep brain stimulation (DBS) (Abelson et al., 2005; Denys et al., 2010) repeated transcranial magnetic stimulation (rTMS) (Lusicic et al., 2018; Rehn et al., 2018), and transcranial direct current stimulation (tDCS) (Fineberg et al., 2023; Pinto et al., 2023; Silva et al., 2021).

In RCTs of adult OCD, the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS) (Goodman et al., 1989a), a 10-item clinician-rated scale that measures the severity of obsessive-compulsive symptoms by characterizing the time, interference, distress, resistance, and control related to the OCD symptomatology, is typically used as the primary outcome to determine the comparative efficacy of interventions. In the Y-BOCS, each item is scored in a 5-point Likert scale ranging from 0 (no symptoms) to 4 (extreme symptoms). Separate scores for obsessions (obsession severity scale) and compulsions (compulsion severity scale), each ranging from 0 (no symptoms) to 20 (extreme symptoms), are calculated and can be summed to provide a total score. The Y-BOCS has sound psychometric properties, including high internal consistency (Cronbach's α = 0.89) and interrater reliability (intraclass correlation coefficient = 0.98) (Goodman et al., 1989b; Storch et al., 2010). However, the precise clinical meaning of average changes in Y-BOCS scores during treatment remains unclear. While global measures of clinical judgment (very much/much improved; very much/much ill) may be less informative than Y-BOCS scores regarding specific OCD symptomatology, from a clinical perspective the dichotomized response and remission categories based on the CGI-I and CGI-S, respectively, may be at least as meaningful as previously discussed for schizophrenia (Leucht et al., 2005, 2006, 2019) and depression (Leucht et al., 2017). Dichotomous outcomes such as a treatment remission and response may be more important because they are more clinically interpretable than numeric change scores in the Y-BOCS. However, response and remission have been traditionally reported as secondary outcomes in OCD RCTs with varying definitions including different Y-BOCS percentage improvement cutoffs or auxiliary scales, e.g., Clinical Global Impressions-Improvement/Severity (CGI-I/S) (Busner and Targum, 2007), two global measures of improvement/severity that have been widely adopted in clinical research across psychiatric disorders, including OCD, depression and schizophrenia (Busner and Targum, 2007). The lack of consensus operational definitions of response and remission in OCD RCTs has impaired the standardization and comparability of OCD RCTs, e.g., in meta-analyses, creating difficulties for communication in the field (Kühne et al., 2020).

In response to these varied definitions, a web-based Delphi survey of OCD experts from around the world was conducted to determine consensus definitions of response and remission in OCD RCTs (Mataix-Cols et al., 2016). For response, the consensus determined an operational definition of a reduction of at least 35% in Y-BOCS score after treatment plus a CGI-I score of 1 (“very much improved”) or 2 (“much improved”) lasting for at least one week. For remission, the consensus determined an operational definition of a Y-BOCS score of at most 12 posttreatment plus a CGI-S score of 1 (“normal, not at all ill”) or 2 (“borderline mentally ill”) lasting for at least one week (Mataix-Cols et al., 2016).

Nevertheless, data from empirical studies have been conflicting regarding whether the Y-BOCS percent reductions and absolute endpoint raw scores suggested by the expert consensus provide the optimal thresholds for response and remission, respectively. Several previous studies have examined the optimal thresholds in the Y-BOCS that correspond to response and remission using a signal detection analytic approach with the CGI-I and CGI-S scales as the reference standards, respectively (Farris et al., 2013; Lewin et al., 2011; Tolin et al., 2005). The previous studies have suggested an optimal threshold between 30% and 45% reduction in Y-BOCS and a Y-BOCS endpoint score between 12 and 14 for treatment response and remission, respectively (Farris et al., 2013; Lewin et al., 2011; Tolin et al., 2005). However, these studies have been limited by only including a few hundred individuals at most. More recently, in children and adolescents, a meta-analysis of RCTs demonstrated a promising approach to investigate this problem by aggregating data across studies, therefore increasing sample sizes and statistical power. That study indicated that the thresholds proposed by the consensus study had the best discriminatory abilities for response and remission as defined by a CGI-I ≤ 2 and CGI-S ≤ 2, respectively (Farhat et al., 2022). However, at present a similar study has not been conducted for adult OCD.

To empirically validate the consensus Y-BOCS definitions, we conducted a large-scale individual participant data meta-analysis of adult OCD RCTs and used a novel multiple thresholds linear mixed effects model to find the optimal Y-BOCS threshold for response and remission defined as a CGI-I ≤ 2 and CGI-S ≤ 2, respectively.