Background Psychiatric comorbidities are among the most disabling non-seizure manifestations of epilepsy yet remain underrecognised in routine care, particularly in low- and middle-income settings. Understanding their clinical spectrum and correlates is vital for comprehensive epilepsy management. Purpose To determine the prevalence and spectrum of psychiatric disorders in adults with epilepsy and evaluate their associations with seizure characteristics and antiseizure medication (ASM) use. Methods A cross-sectional study was conducted on 98 adults (aged 18–65 years) attending a tertiary epilepsy clinic between January 2024 and February 2025. Psychiatric evaluation was performed using the Patient Health Questionnaire (PHQ-SADS), PHQ-9, the Generalised Anxiety Disorder-7 (GAD-7), the Yale–Brown Obsessive–Compulsive Scale (Y-BOCS), the Brief Psychiatric Rating Scale (BPRS), PHQ-15 and the Columbia Suicide Severity Rating Scale (C-SSRS). Associations between psychiatric symptoms and seizure-related variables (frequency, duration, aura, injury) and the ASM profile were analysed using Pearson correlation and ANOVA. Results Overall, 59.2% of patients exhibited at least one psychiatric disorder. Depression (38.1%) and anxiety (19.1%) were the most common, followed by somatic symptom disorder (10.2%), suicidal ideation/behaviour (8.1%), obsessive–compulsive disorder (OCD) (7.2%) and psychosis (6.1%). Higher seizure frequency and longer epilepsy duration correlated significantly with depressive and anxiety symptom severity ( p < .05). Aura and seizure-related injuries were associated with elevated affective scores. Polytherapy and older ASMs (phenytoin, carbamazepine) increased depressive risk, whereas newer agents showed better tolerability. Conclusion Psychiatric comorbidities are highly prevalent among adults with epilepsy, strongly influenced by seizure burden rather than demographic variables. Routine mental health screening and integrated neuropsychiatric care should be integral to epilepsy management, particularly in resource-limited settings.
Shanthi et al. (Mon,) studied this question.