OBJECTIVE: To evaluate the relationship between insomnia and suicidality (SI) in Chinese patients with first or recurrent stroke. DESIGN: Cross-sectional survey. SETTING: Acute stroke unit of a general hospital. PARTICIPANTS: Patients (N=787) with acute ischemic stroke admitted to the acute stroke unit of a university-affiliated regional hospital in Hong Kong. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: Suicidality (SI) was assessed with the Geriatric Mental State Examination at 3 months after subjects' index stroke. Insomnia symptoms were evaluated with a standard insomnia questionnaire. The association between insomnia symptoms and SI was examined and adjusted for potential confounders, including age, sex, marital status, previous stroke, depression, fatigue, Mini-Mental State Examination score, and neurologic deficits measured with the National Institutes of Health Stroke Scale. RESULTS: Eighty-seven patients (11.1%) were found to have SI (SI group). Frequent awakening was significantly more common in the SI group than in the non-SI group and remained a significant predictor of SI in forward logistic regression analysis (odds ratio, 1.7) after adjusting for possible confounders. CONCLUSIONS: These findings should alert clinicians to the potential danger of insomnia and the importance of its early identification and treatment in stroke patients.
OBJECTIVE: To evaluate the relationship between insomnia and suicidality (SI) in Chinese patients with first or recurrent stroke. DESIGN: Cross-sectional survey. SETTING:Acute stroke unit of a general hospital. PARTICIPANTS: Patients (N=787) with acute ischemic stroke admitted to the acute stroke unit of a university-affiliated regional hospital in Hong Kong. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: Suicidality (SI) was assessed with the Geriatric Mental State Examination at 3 months after subjects' index stroke. Insomnia symptoms were evaluated with a standard insomnia questionnaire. The association between insomnia symptoms and SI was examined and adjusted for potential confounders, including age, sex, marital status, previous stroke, depression, fatigue, Mini-Mental State Examination score, and neurologic deficits measured with the National Institutes of Health Stroke Scale. RESULTS: Eighty-seven patients (11.1%) were found to have SI (SI group). Frequent awakening was significantly more common in the SI group than in the non-SI group and remained a significant predictor of SI in forward logistic regression analysis (odds ratio, 1.7) after adjusting for possible confounders. CONCLUSIONS: These findings should alert clinicians to the potential danger of insomnia and the importance of its early identification and treatment in strokepatients.