Abstract W P335: Variation and Trends in the Documentation of National Institutes of Health Stroke Scale (NIHSS) Among GWTG-Stroke Hospitals

Abstract

Background: Despite its clinical importance NIHSS is often not documented in routine practice and in stroke registries. We describe trends in NIHSS documentation by GWTG-Stroke hospitals, the relationship between hospital documentation rates and reported NIHSS, and identify hospital- and patient-level factors associated with documentation. Methods: We analyzed NIHSS documentation in 1,159,981 acute ischemic stroke patients admitted to 1,682 GWTG-Stroke hospitals between 2003-2012. We used multivariable logistic regression models to identify hospital- and patient-level predictors of NIHSS documentation. Because of rapid increases in NIHSS documentation in recent years, multivariable analyses were restricted to the 2011-2012 period. Results: The overall NIHSS documentation rate was 51% and the mean NIHSS score was 6.8. Between 2003 -2012 mean hospital-level NIHSS documentation increased from 27% to 70%. The mean hospital-level NIHSS documentation rates and NIHSS score were inversely related (r = -0.228, p <0.0001), but this relationship was driven primarily by hospitals with documentation rates of<20% (Figure). Multivariable analysis indicated that NIHSS documentation was impacted by arrival mode, last known well-to-arrival time, and was higher at primary stroke centers and hospitals with larger case volumes (Table). Conclusion: Evidence of selection bias in reporting NIHSS was limited to hospitals with low documentation rates. Documentation of NIHSS was influenced by both hospital and patient-level factors and was higher in patients who were candidates for thrombolysis treatment. Documentation of NIHSS has improved dramatically in recent years.