When the same set of data is analyzed in multiple parallel ways using different models, groupings, or summary measures, the meaning of statistical tests becomes distorted by the multiple-comparisons problem.
Kaplan Meier curves for OS, using the RFS and RFoT model risk groupings, are presented in Figures 2D and E, respectively.
Populations (internal branches) were modeled using groupings of subpopulations (areas, SAMOVA groups) appropriate to each hypothesis, and we scaled population-lineage sizes (internal branch widths) to their respective proportions of overall Ne, so they summed to overall Ne at all time points [ 90].
The multidimensionality of OC symptoms was also supported by a confirmatory study performed in a large sample of OCD adult patients [ 25] that tested different models of symptom groupings and found that only the model composed of four factors fitted adequately.
So it is impossible to imagine now the simple "two groupings" model of the past being restored.
Our 4-strain model utilizes four groupings: (1) serogroup B, (2) serogroup C, (3) serogroups A, W-135, X, Y, 29e, and 'ungroupable' (UG) serogroups, and (4) Neisseria lactamica.
A more robust critique of the sub-grouping perspective requires consideration of the predictions that a sub-grouping model makes, and investigation of data related to these predictions.
Sensitivity analysis was undertaken to test the alternative assumptions, such as free or fixed knot of spline quadratic models, different age groupings and truncating criteria of PHC visits.
The sub-grouping model would predict that subjects in multimodal care have a greater likelihood of receiving their appropriate intervention, thus decreasing the chance of 'washout'washout
Supporters of the sub-grouping model contend that the small effect sizes seen in clinical trials result from the large positive effects seen in some patients being 'washed out' by the smaller or negative effects seen in others [ 32].
