Applications of survival analysis and learning curves methods in neurosurgical stroke data and simulations to account for provider heterogeneity.

For the Cox proportional hazard regression model for time to complication (Table 3 ) as per prior analyses [ 1 ], we found that EVD device type remained as a significant predictor in either univariate ( p < 0.008) or multivariate ( p < 0.010) Cox model analyses while pre-operative MRS was initially significant in the univariate analysis ( p < 0.014) but became borderline significant in the multivariate analysis ( p < 0.060).

also in 11,409 other papers