Predicting Non-Suicidal Self-Injury in Young Adults with and without Borderline Personality Disorder: a Multilevel Approach Combining Ecological Momentary Assessment and Self-Report Measures

A Machine Learning Approach for Predicting Non-Suicidal Self-Injury in Young Adults

“…The database used in this work had been used before in [16], in which a multilevel mixed-effect regression analysis was performed. This analysis showed that momentary frustration alone directly predicted NSSI.…”

“…The data used in our work are the same used in [16,35] and consist of 64 young adults (ages between 18 and 33 years), divided into three groups. The first two groups consist of participants with NSSI (≥5 NSSI events in the previous 12 months): (i) a subclinical group of university students recruited from the city of Igualada in Barcelona, Spain (STD group; N = 19), and (ii) a clinical group of BPD patients (BPD group; N = 22).…”

“…All the par-ticipants completed self-report measures of borderline pathology, emotion dysregulation, de-centredness and negative emotional symptoms, and used the Sinjur application (EMA instrument) at least three times a day for 15 days to capture negative affect and NSSI in daily life. We refer the reader to [16] for further details on this matter.…”

“…The Sinjur app was developed for the purpose of this kind of study, and was used to collect the data first analysed in [16]. We were not involved in its design, nor in the data collection step.…”

“…This, in turn, avoids the bias introduced by trying to recall motives and mood retrospectively, and ecological validity is maintained [11]. However, there are few published studies using this technique to determine the motives that triggered an episode of NSSI [14][15][16]. Previous published work indicates that an increase in negative emotions is one of the common events prior to an episode of NSSI [17,18].…”