So it seems unrealistic and unreasonable to use bootstrap resampling methods for marginal models that can only utilise a balanced data set, with equally spaced QoL assessments.
On the other hand, Poisson, MZIP and MPois-Pois models, which did not directly account for overdispersion, had poor fits and gave standard errors of regression coefficients for the marginal mean model that were too small.
The main difference to the marginal model is that a player cannot be at risk for the later injury until a prior event occurs.
The marginal model indicated that there was a difference in the average effect of intervention on the first four falls (Wald χ2 test = 32.2, df = 3, p < 0.0001).
Rate ratios based on the marginal model indicated that, on average, the transition rate from zero falls at the start of treatment to one, two, three and four falls were 32%, 58 %, 70 and 80%and80%, respectivelowern the interespectivelyup than the control.
Treatment effects for these ordinal outcome scales were estimated via a marginal ordinal regression model that allowed for the covariates and for time as a within person effect, with the participant as the clustering variable.
Finally, for Income & Health data, the model with GB1 marginals outperforms the BG truncated exponential marginals model in 1980 , 1985and 1995 while in the other four years the BG truncated exponential marginals is the model that provides the best fit.
Thus, a study using marginal structural models that adjust for time dependent confounding would be clinically relevant.
Because blood draws were not always conducted at exactly 12 or 24 months, we also examined the association of age at blood draw and Th1, Th2, and CD4 status using longitudinal marginal regression models that allowed for analysis of repeated measures.
