Objectives: Telestroke is the application of telemedicine to stroke care. We estimated the effect of participation in a telestroke network on in-hospital mortality in the state of Georgia, and explored its impact on mitigating the difference in mortality for patients admitted in nighttime compared to those admitted in daytime.
Methods: We selected patients with ischemic stroke from 15 non-teaching hospitals in the Georgia’s Paul Coverdell Acute Stroke Registry from 2005 to 2016. We applied a quasi-experimental study design by classifying patients from 4 hospitals that participated in a telestroke network in 2009 as the treatment group, and patients from 11 hospitals that were not covered by the telestroke network as the comparison group. All selected hospitals are located in non-Metropolitan Areas. We compared mortality between treatment and comparison groups in 2005 - 2008 (pre-participation period for treatment group) and in 2009 - 2016 (post-participation period for treatment group), and estimated difference in in-hospital mortality attributable to participation in a telestroke network by applying a difference-in-differences approach, while adjusting for patients’ age, sex, race/ethnicity, insurance coverage, arrival mode, ambulatory status prior to the current stroke, stroke severity, medical history of atrial fibrillation/flutter and hospital admission time.
Results: The mortality among ischemic stroke patients decreased in all selected hospitals over the last decade. Participation in a telestroke network significantly decreased in-hospital mortality by 3.2% (p-value= 0.003). There was a positive association between nighttime admission and in-hospital mortality in the entire patient sample. After controlling for the effect of participation in a telestroke network, the nighttime effect on mortality still remained significant (odds ratio=1.25, 95% confidence interval: 1.10 - 1.42).
Conclusions: Acute ischemic stroke patients admitted in hospitals participating in a telestroke program had a more pronounced reduction in in-hospital mortality. However, telestroke coverage did not alter the effect of nighttime admission on in-hospital mortality.