0945 Sleep Health and Suicide Ideation in the US: Data from the National Sleep Foundation Sleep in America Poll

Abstract Introduction Previous studies have shown that poor sleep is generally associated with suicide ideation. But few studies have examined this relationship in the general population across many dimensions of sleep health, and it is not clear which sleep variables are most salient. Methods Data were collected as part of the National Sleep Foundation Sleep In America Poll. The population-based sample (N=1,042) provided data on a range of sleep questions, as well as each item of the PHQ9 depression scale. For the present analyses, PHQ item 9 (suicide ideation) was examined as a binary outcome, with population-weighted regression analyses adjusted for age, sex, race/ethnicity, education, income, and PHQ2 depression score. Independent variables included sleep satisfaction, feeling refreshed, daytime energy, difficulty falling asleep, frequent nighttime awakenings, difficulty resuming sleep, satisfaction with weekday and weekend sleep duration, ability to feel relaxed, overall sleep quality, weekday and weekend bedtime, waketime, time in bed, sleep duration, perceived sleep need, and days/week (0-7) of: feeling well-rested, having trouble falling or staying asleep, sleep impacts functioning, daytime sleepiness, and sleep medication use. Bonferonni correction (0.05/26) was used. Results Population-weighted analyses adjusted for age, sex, race/ethnicity, education, income, and PHQ2 depression score showed that elevated likelihood of suicide ideation was associated (all p 0.0019) with days/week: feeling well rested (OR=0.80/day), trouble falling asleep (OR=1.26/day), trouble staying asleep (OR=1.18/day), and sleep impacts functioning (OR=1.46/day). Weekday sleep duration (OR=0.73/hr), weekend sleep duration (OR=0.75/hr), “Very Dissatisfied” with sleep (OR=6.92), “Not energized at all” (OR=7.46), “A great deal” of trouble falling asleep (OR=6.91), “Very often” waking during the night (OR=5.75), “Very dissatisfied” with weekday (OR=6.95) and weekend (OR=7.81) sleep duration, and “Very difficult” to relax before bed (OR=10.04) were also associated. Stepwise analyses revealed that the variables that explain the most unique variance are (in order) days/week that sleep impacts functioning (R-squared 0.089) and satisfaction with weekend sleep duration (R-squared 0.012). Conclusion Many indicators of sleep health were associated with suicide ideation, implicating sleep duration, quality, efficiency, and daytime effects of sleep. Efforts might ideally focus on the daytime impacts of sleep as well as nighttime experiences. Support (if any)