Regression and Mixed Modeling Sample Clauses

Regression and Mixed Modeling. In addition to performing bivariate analyses, two linear regressions were performed using the keratometric astigmatism value at 1 year of age and the change of astigmatism between baseline and 1 year as the dependent variables. Backward elimination was used with a significance level of 0.10 to identify surgical factors that were significant predictors of keratometric astigmatism. A mixed model analysis was also performed, using subject-specific intercepts and slopes along with surgical factors and a variable for a patient’s age at each visit. Each subject had two time points: one at the baseline visit, and the other at the 1 year of age visit, using 365 days as the time point (exact ages at the 1 year visit were not available for this analysis). The mixed model takes the form: Ŷij = αi + bij tij + β1 X1 + ... + βk Xk + βt tij, where Ŷij = keratometric astigmatism in patient i at visit j; αi = Subject specific random effect on the intercept for subject i; bij = Subject specific random effect on the slope; tij = Age of patient i at visit j, given in months; βt is the population effect for age tij, and β1, ... βk are fixed effects for predictors X1, ..., Xk.