Abstracts

Rationale: Patterns of suicide timing have been studied in the general population but not in people with epilepsy. We investigate suicide timing in data from the U.S. National Violent Death Reporting System (NVDRS). Methods: NVDRS is a population-based, active surveillance system that monitors US violent deaths including suicides to better understand and guide their prevention. In 18 states from 2003-2014, NVDRS collected information on 103,892 suicides (ICD-10 X60–X84, Y87.0 as the underlying cause of death) in persons 10 years old or older. Among these suicide decedents, epilepsy was identified from ICD-10 codes (G40, G41, P90, R56) in the first ten multiple conditions recorded on the death certificate or from epilepsy-related terms ("epilepsy", "seizure", "convulsion", "drop attack", "falling out spell", and "staring spell") in narratives about the suicide incidents. We calculated suicide mortality rates by age, sex, and race/ethnicity with respect to suicide timing by hour of day, day of week, week of month, and month of year. In suicide decedents with (n=1,310) and without (n=102,582) epilepsy, we calculated subgroup ratios of these rates relative to reference times within each group. We compared trends in these ratios for both groups by the above time intervals. Results: In people with epilepsy, suicides occurred significantly more often during the morning, afternoon, and evening hours than at night in all age groups, men, and white non-Hispanics, a pattern like that in people without epilepsy (p<0.05). Suicides in people with epilepsy occurred significantly more often on Mondays than Sundays in those 45 years old or older and men, a pattern like that in people without epilepsy (p<0.05). Suicides in people with epilepsy did not differ statistically significantly relative to reference times by week of month or month of year.  However, suicides in people without epilepsy showed decreased ratios in the third through fifth weeks of a month in those 45 years or older, men, and non-Hispanic whites and increased ratios from the spring through the fall in those of all ages, both sexes, and all race-ethnicities (p<0.05). Conclusions: Except for hours within a day and on Mondays, suicides in people with epilepsy did not exhibit the monthly or the weekly timing patterns of suicides as seen in people without epilepsy. Why most suicide timing patterns seen in decedents without epilepsy were not seen in decedents with epilepsy deserves further investigation. These data on timing and demographics of suicide among people with epilepsy can help to inform prevention programs.  A comprehensive public health strategy to prevent suicide includes approaches designed to impact individual, relationship, community and societal risk and protective factors, not only at peak times but all the time. Funding: N/A