Mental Health Services Utilization among Suicidal Patients: Comparing the Impact of Co-Occurring Opioid or Other Substance Use Disorders

Abstract

Background: Prior literature establishes bidirectional associations between suicide and substance use disorders (SUDs), particularly opioid use disorder (OUD). However, the context of mental health (MH) utilisation of services remains under-investigated. This analysis examined patterns of MH services utilisation in patients with SUDs and suicidality, identified associated risk factors and evaluated the impact of patient engagement on subsequent MH outcomes. Method(s): Electronic health records (EHRs) derived from seven health systems across New York City between 2010 and 2019 were analysed. Suicidality was identified as any International Classification of Disease (ICD)-9/10 diagnosis of suicide attempt, suicidal ideation or self-harm injury. SUDs were identified as any opioid, cannabis, cocaine, hallucinogen, inhalant, sedative/hypnotic/anxiolytic, amphetamine or other substance abuse or dependence. Quasi-Poisson regression adjusted for age, gender and chronic disease was used to model associations between OUD exposure and the frequency of encounters and to estimate the relative risk (RR) of significant covariates. Result(s): A total of 6977 adults with suicidality and any comorbid SUD were selected, including 2203 (31.6%) with a diagnosis of OUD and 4774 (68.4%) without a diagnosis of OUD. Most patients were male (54.8%) and aged between 25 and 64 years (79.3%). Many (61.3%) had more than three chronic diseases, including depression (80.8%), hypertension (60.6%), anaemia (43.0%) and hyperlipidaemia (41.9%). Compared to patients with other SUDs, those with OUD had higher odds of self-harm injury [odds ratio (OR): 1.26 [95% confidence interval (CI): 1.13- 1.41]), depressive disorders [1.47 (1.29-1.67)], anxiety disorders (1.65 [1.48-1.84]), psychotic disorders (1.23 [1.11-1.37]), personality disorders (1.30 [1.16-1.48]) and post-traumatic stress disorder (1.37 [1.20-1.57]). Patients with OUD were more likely to utilise all-cause outpatient (RR: 1.16), emergency department (ED) (RR: 1.43) and inpatient (RR: 1.60) services (p < 0.001). Among OUD patients, males were less likely to have outpatient visits (RR: 0.79) and inpatient hospitalizations (RR: 0.88), and older age was protective against ED admissions (RR range: 0.62-0.71). In addition, individuals with OUD were more likely than those with other SUDs to have SUD--related encounters, as well as suicide-related ED admissions and inpatient hospitalizations (p < 0.0001). Those who had more MH outpatient visits were less likely to have suicide-related ED admissions (RR: 0.85); however, this association was weaker among younger or male patients with comorbid OUD. Conclusion(s): Among suicidal adults with comorbid SUDs, those with a diagnosis of OUD were more likely to utilise MH services and have psychiatric comorbidity. Males and older adults were less likely to utilise services. These findings provide a first look at utilisation among this vulnerable population based on real-world EHRs and may inform future interventions simultaneously aimed at suicide and substance abuse prevention.