Introduction

Mental and substance use disorders (M/SUDs) inflict the highest costs to society of all diseases, affecting production levels, legal systems, law enforcement, and continued damage to users (Whiteford, Ferrari, Degenhardt, Feigin, & Vos, 2015). Key populations (KPs), most vulnerable to HIV exposure such as men who have sex with men (MSM), people who use drugs (PWUD), sex workers, and transgender individuals are disproportionately burdened due to stigma-induced psychosocial stress (WHO, 2014). While behavioural interventions support abstinence during early recovery, 80-95% of people relapse within 12 months (Brandon et al., 2007). Relapse is especially problematic in KPs because it exacerbates mental disorders, risk behaviours, and the spreading of HIV to other groups (Gupta, Kumar, & Garg, 2013). The reason many behavioural interventions fail in natural contexts is because patients have difficulty enacting them outside of controlled clinical settings in which they were first introduced (Boyer, Smelson, Fletcher, Ziedonis, & Picard, 2010). As a result, patients with M/SUDs neither have the ability to detect biological and affective changes that trigger drug cravings or modify their behaviours to decrease health risk (Boyer et al., 2010).
Current limitations in the diagnosis and treatment of M/SUDs has prompted a clinical shift toward personality-informed approaches (Trull & Widiger, 2013). Within psychiatry, growing consensus argues that dimensional trait models can more effectively diagnose M/SUDs (Krueger & Markon, 2014; Suzuki, Samuel, Pahlen, & Krueger, 2015). This consensus draws on clear evidence that key personality traits can act as endophenotypes for M/SUDs (Belcher, Volkow, Moeller, & Ferré, 2014). Specifically, high negative emotionality/neuroticism (NEM/N) increases, whereas high positive emotionality/extraversion (PEM/E) and constraint (CON) decreases vulnerability (Belcher et al., 2014). Researchers in mobile health (mHealth) also argue that mobile sensing technology can disrupt and change habitual behaviours via self-monitoring and feedback (Hermsen, Frost, Renes, & Kerkhof, 2016). While the long-term impact of such disruptions remain unclear, mobile applications (apps) offer clinicians unprecedented potential to increase user engagement whilst continuously self-monitoring disease outcomes (Jardine, Fisher, & Carrick, 2015). The fact that mobile apps can detect many of the affective changes that trigger drug craving and risk behaviours suggest their particular efficacy in treating M/SUDs (Boyer et al., 2010; Donker et al., 2013; Litvin, Abrantes, & Brown, 2013).
What these two converging research domains share above an emphasis on personality-informed intervention is the undying assumption that personality traits can no longer be defined as fixed dimensions (Ferguson, 2010). Rather, traits are highly dynamic qualities that morph considerably in response to spatio-temporal changes (Chapman, Hampson, & Clarkin, 2014). Health-protective personality changes are thus not only achievable but indeed desirable and beneficial (Hampson, Goldberg, Vogt, & Dubanoski, 2006). Findings confirm the timely nature of personality-informed diagnosis (Suzuki et al., 2015; Trull & Widiger, 2013) and intervention to meet continuous support demands in people with M/SUD (Donker et al., 2013; Litvin et al., 2013). The following section examines how such intervention may be achieved.