Table 2.

Details of included study arms for the estimation of effectiveness

Intervention	Study	Cumulative dose	Route	Duration	Day of assessment	Fraction with reduced relapse	Relapse severity (baseline EDSS)	Diagnostic criteria
High-dose methylprednisolone	Durelli 1986a [31]	7035 mgb	IV	15 days	15	10/11	5.8 (mean)	Poser 1983 [51]
	Milligan 1987 [32]	2500 mg	IV	5 days	28	10/13	4.0 (median)	McDonald 1977 [52]
	La Mantia 1994 [27]	5750 mg	IV	14 days	14	8/10	4.6 (mean)	McDonald 1977 [52]
	Barnes 1997 [33]	3000 mg	IV	3 days	28	13/38	6.0 (median)	Not stated
	Sellebjerg 1998 [34]	3676 mg	PO	15 days	21	14/26	4.5 (median)	Poser 1983 [51]
	Visser 2004 [35]	2500 mgc	IV	5 days	28	6/9	3.5 (median)	Not stated
	Ramo-Tello 2013 [36]	3000 mg	IV	3 days	28	15/23	4.0 (median)	McDonald 2005 [53]
	Ramo-Tello 2013 [36]	3750 mg	PO	3 days	28	15/22	3.0 (median)	McDonald 2005 [53]
Low-dose methylprednisolone	Milanese 1989 [26]	390 mg	IV	14 days	30d	3/10	4.9 (mean)	McDonald 1977 [52]
	La Mantia 1994 [27]	390 mg	IV	14 days	14	6/10	4.7 (mean)	McDonald 1977 [52]
	Barnes 1997 [33]	588 mg	PO	21 days	28	20/42	5.0 (median)	Not stated
Placebo	Miller 1961 [37]	-	IM	21 days	21	4/18e	Not stated	Not stated
	Rose 1970 [38]	-	IM	14 days	28	39.25/94f	5.2 (mean)	Rose 1968 [54]
	Durelli 1986a [31]	-	IV	15 days	15	4/10	5.9 (mean)	Poser 1983 [51]
	Milligan 1987 [32]	-	IV	5 days	28	2/9	4.0 (median)	McDonald 1977 [52]
	Sellebjerg 1998 [34]	-	PO	15 days	21	6/25	4.0 (median)	Poser 1983 [51]

IV intravenous, PO per oral, IM intramuscular

^a Only the 15 days controlled period of this trial is considered here

^b Based on a weight of 70 kg

^c Patients also received 2 % human albumin

^d This study lacked reported assessments in the 14–28 day interval

^e No EDSS measurement performed; the result refers to the fraction with an ‘undoubted response to treatment’

^f Conversion based on the assumption that a DSS of e.g. 4 is equally likely to correspond to EDSS 4.0 as EDSS 4.5