Table 3

Multiple regression model for outcome at long term follow-up derived with stepwise model selection procedure: fitted regression model including predictors with p≤0.5 to enter; p<0.1 to stay in the model

	Estimate	SE	p Value
Physical outcome^* model fit (logistic regression): R²=0.51
Baseline variables
Intercept	–5.3	0.91	<0.0001
EDSS at baseline	1.20	0.22	<0.0001
MRI T2 BOD at baseline (cm²)	0.05	0.02	0.001
Gender	0.93	0.47	0.045
On-RCT variables
Actual EDSS change from baseline	0.86	0.21	<0.0001
Annualised relapse rate	0.52	0.23	0.025
Cognitive outcome^† model fit (linear regression): R²=0.49
Baseline variables
Intercept	–11.2	3.98	0.006
EDSS at baseline	–0.99	0.25	<0.0001
Premorbid IQ	0.12	0.035	0.0007
MRI T2 BOD at baseline (cm²)	–0.05	0.02	0.018
Third ventricular width at baseline (mm)	–0.41	0.16	0.014
On-RCT variables
Actual EDSS change from baseline	–0.67	0.24	0.007
Change, third ventricular width (mm)	–0.87	0.33	0.009

↵* Physical outcome=either SPMS or EDSS=6.
↵† Cognitive outcome=Cognitive Performance Index (see text).
BOD, burden of disease; EDSS, expanded disability status scale score; LTF, long term follow-up; RCT, randomised controlled trial; SPMS, secondary progressive multiple sclerosis.