Renee Colsch

Date of Award

January 2019

Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Glenda Lindseth


Background/Purpose. Stroke is the leading cause of disability in the United States. Women who have experienced a stroke have greater disability than men. Thrombolytic agents decrease adverse side effects of stroke by dissolving blood clots. Yet, women have 8% higher odds against being treated with a thrombolytic agent. Also, about 17% of stroke cases occur in-hospital. Therefore, the purpose of this study was to investigate the effects and associated variables of having an in-hospital stroke alert activation on outcomes in women admitted to the hospital for a separate condition.

Methods. Guided by the Model for Nursing Effectiveness Research, a retrospective observational study of 149 women participants was completed for a 4 year period. Study measures based on empirical evidence included the primary independent variable of in-hospital stroke alert, and confounding variables (patient characteristics, clinical conditions, and context of care) that are conceptually related to the primary outcome of thrombolytic therapy and secondary outcome of discharge status. Analysis included regression models and propensity score matching to isolate the treatment (in-hospital stroke alert) and outcome (thrombolytic therapy) while controlling the effects of other influential variables.

Results. In-hospital stroke alert was activated in 46 of 149 or 30.9% women and 15 of 149 or 10.1% of women received thrombolytic therapy. In-hospital stroke alert was significant (p < .001) for women receiving thrombolytic therapy and significant to a home discharge status (p = .014). Age (p < .001), marital status (p = .067), ethnicity (p < .001), common (p = <.001) and unique symptoms (p = .012), stroke risk factors were present (p <.001), comorbid conditions

were present (p <.001), Time Last Known Well (the time that the patient was without stroke symptoms) (p = .041), diagnostic imaging (p <.001) were all significantly related to in-hospital stroke alert.

Discussion/Conclusions. Results from this study suggest that younger married women from non-Caucasian ethnic groups and women with risk factors or comorbid conditions are all at higher levels of late stroke symptom detection and no in-hospital stroke alert activation. Improved stroke detection in women with attention to barriers may improve in-hospital stroke alert activation and early treatment.