Psychiatric comorbidities among people with epilepsy: A population-based assessment in disadvantaged communities

Psychiatric disorders are frequent among people with epilepsy but often under-recognized. The diagnosis and treatment of these disorders in low- and low-middle-income countries (LMICs) are challenging.

This cross-sectional survey included people recruited during a community epilepsy screening program involving 59,509 individuals from poor communities in Ludhiana in Northwest India. Adults (age ≥18 years) with confirmed epilepsy on antiseizure medications were screened for depression and anxiety using the Neurological Disorders Depression Inventory for Epilepsy (NDDI-E) and Generalized Anxiety Disorder-7 (GAD-7) twice over two years of follow-up. They were later interviewed for symptoms using the Brief Psychiatric Rating Scale, which was then confirmed by assessments by an experienced psychiatrist.

Of the 240 people with confirmed epilepsy, 167 (70%) were adults, of whom, 116 (70%) eventually participated in the study. The NDDI-E with a cut-off of 15 identified depression in 14 (12%) of 116 people after one year of follow-up and 17 (15%) at two years. The GAD-7 using a cut-off of 6 identified 22 (19%) at one year and 32 (28%) with anxiety at two years. The area under the curves for NDDI-E was estimated as 0.62 (_95%CI, 0.51-0.73; SE: 0.06; p = 0.04) and for GAD-7 as 0.62 (_95%CI, 0.46-0.78; SE: 0.08; p = 0.12). Brief Psychiatric Rating Scale identified 63 (54%) people with psychiatric symptoms, for whom, a psychiatric diagnosis was confirmed in 60 (52%). A psychiatric diagnosis was associated with education below high school [Odds Ratio (OR): 2.59, _95%CI, 1.12-5.1; p = 0.03], later age of seizure onset (OR, 1.05, _95%CI: 1.0-1.10; p = 0.04), seizure frequency of at least one/year at enrolment (OR, 2.36, _95%CI: 1.0-5.58; p = 0.05) and the use of clobazam (OR, 5.09, _95%CI, 1.40-18.42; p = 0.01).

Depression and anxiety are common in people with epilepsy. Our findings underscore the low yields of screening instruments, NDDI-E and GAD-7, and comparatively better professionally-administered diagnostic assessments in resource-limited settings in LMICs. Moreover, previously established cut-offs do not apply to the community studied.