Real-Life Outcome in Multiple Sclerosis in the Czech Republic

Table 2

Marginal structural Cox model of the risk of relapses by treatment.

(a)

Analysis of Maximum Likelihood Estimates
with Sandwich Variance Estimate, IPTW weighted
Parameter		DF	Parameter Estimate	Standard Error	StdErr Ratio	Chi-Square	Pr > ChiSq	Hazard Ratio
Treatment group	1	1	-0.26179	0.15279	1.180	2.9357	0.0866	0.770
Treatment group	2	1	-0.05780	0.09597	1.390	0.3627	0.5470	0.944
Treatment group	3	1	-0.38346	0.14685	1.351	6.8183	0.0090	0.682

(b)

Analysis of Maximum Likelihood Estimates
with Sandwich Variance Estimate, Unweighted
Parameter		DF	Parameter Estimate	Standard Error		StdErr Ratio	Chi-Square	Pr > ChiSq	Hazard Ratio
Treatment group	1	1	0.22118	0.13351	1.162	2.7443	0.0976	1.248
Treatment group	2	1	0.30767	0.09959	1.237	9.5443	0.0020	1.360
Treatment group	3	1	-0.07314	0.14171	1.237	0.2664	0.6058	0.929