Abstract TP179: Key Factors Associated with Major Depression in a National Sample of Stroke Survivors
BACKGROUND: Depression is a strong emerging independent risk factor for vascular brain injury. Up to a third of stroke survivors are depressed at any time, putting them at risk for sub-optimal recovery and recurrent vascular events. Identifying factors associated with depression after stroke in a nationally representative sample may provide insights into avenues for implementing potential population-based interventions to improve stroke outcomes.
METHODS: We assessed factors associated with depression among adults (≥20 years) with self-reported stroke who participated in NHANES 2005-2010. Depression was identified using a Patient Health Questionnaire-8 (PHQ-8) score ≥10. The following potential predictors of depression were assessed using a random forest model: age, sex, race/ethnicity, income to poverty index ratio, education level, marital status, body mass index (BMI), smoking, history of congestive heart failure, myocardial infarction (MI), coronary artery disease, angina pectoris, diabetes, emphysema, chronic bronchitis, liver disease, and cancer. All estimates were weighted to obtain nationally representative estimates.
RESULTS: Among stroke survivors in NHANES 2005-2010 (n=678), 546 completed the PHQ-8 survey. Of those who completed the survey, 17% had depression, corresponding to 872,237 stroke survivors in the US with depression. Compared to stroke survivors without depression, depressed individuals were younger (55.0 vs. 65.8 years), more likely to have an income to poverty ratio <200% (70% vs. 44%) and more likely to smoke (50% vs. 19%; all p<0.05). The following 10 factors, in decreasing order of importance were informative for predicting depression after stroke: age, education level, BMI, race, smoking, female sex, marital status, history of cancer, and history of MI. The model predicted co-morbid depression among stroke survivors with 70% specificity and 64% sensitivity (overall accuracy 69%).
CONCLUSIONS: In this nationally representative sample, 10 key factors predicted depression after stroke. Knowledge of these socio-demographic and medical factors could be used to highlight US population subsets which may benefit from more focused depression screening and intervention to enhance stroke outcomes.
- © 2012 by American Heart Association, Inc.