Abstracts

Rationale: Treatment resistant depression in the general population leads to increased healthcare utilization, risk of suicide, poor productivity and impaired psychosocial functioning. In epilepsy, it is unknown what clinical and demographic variables are associated with persistent symptoms despite antidepressant therapy, and practice habits such as antidepressant dosing have not been examined in real world epilepsy samples. The objective was to explore factors associated with anxiety and/or depression symptoms during ongoing antidepressant treatment in a large consecutive adult epilepsy clinic sample.

Methods: Anxiety and depression scores on the Generalized Anxiety Disorder-7 (GAD-7) and Neurological Disorders Depression Inventory-Epilepsy (NDDI-E) were collected prospectively on a tablet for dual research and clinical purposes at a tertiary epilepsy clinic visit. Consecutive adults with epilepsy who completed both instruments for the first time in a 16-month study period were included. Sociodemographic, epilepsy-related, health utilization (system no-show rate, neurology telephone/portal encounter count) and medication records were abstracted from the electronic health record. Antidepressant dosing levels were categorized as low versus medium/high using categories 1-2(low), 3&4(medium & high) of the dosing component of the Antidepressant Treatment History Form. The main analysis focused on the subset with ongoing antidepressant prescription at the time of anxiety and depression screener completion. Prior to data analyses on the primary outcome of persistent symptoms (GAD-7 ≥10 or NDDI-E ≥16), the factors to include were selected by overall rankings among individual author-ranked selections. Simple and multiple logistic regression were conducted.

Results: Of 563 adults with epilepsy, 154 had evidence of ongoing antidepressant prescription at the time of screening and 73/154 (42%) had a positive anxiety and/or depression screen. Table 1 shows sociodemographic and clinical characteristics of the sample. Among the factors selected a priori for multivariable modeling (Table 2), only the presence of at least one no-show visit to any specialty in the institution was significantly associated with persistent positive symptoms (adjusted OR 3.10, 95% CI 1.27-8.08). With low antidepressant dosing, the odds of positive screen increased twofold (adjusted OR 2.06, 95% CI 0.91-4.80). Other variables also yielded odds ratios in directions expected a priori, but were not statistically significant (Table 2).

Conclusions: Among the sociodemographic and clinical variables tested, presence of prior no-shows to any clinic visit in the health system was associated with positive anxiety or depression screen in the setting of existing antidepressant prescription. Further research in larger, longitudinal studies is indicated to replicate the findings and examine system factors and social determinants of health that may underlie the observed association between missed clinic visits and persistent anxiety or depression symptoms.

Funding: Please list any funding that was received in support of this abstract.: National Institutes of Health R25 NS088248; UL1 TR001420, U24 NS107197, 2KL2TR001421-05.