Study	Reason for exclusion
Burke‐Ramirez 2001	Number of participants in each treatment arm not reported and no indication of baseline pain intensity for any treated participants
Cady 1994	First dose of subcutaneous sumatriptan not randomised, only for subsequent doses of oral sumatriptan for recurrence (from 2 to 24 h after initial dosing) were patients randomised to either sumatriptan or placebo
Cull 2001	All participants initially treat with sumatriptan at the onset of migraine headache, and are only randomised to either sumatriptan or placebo to treat any subsequent recurrence that occurred between 1 and 24 h after the first dose was administered
Ensink 1991	2 studies: Study 1 ‐ Baseline pain intensity of treated participants not reported and at least 50% of participants in each treatment arm took a second dose of study medication at 30 minutes. No useable efficacy data at 1 or 2 h and no adverse event data reported. Study 2 ‐ Data reported in Mathew 1992
Friedman 2005	Only comparator (intravenous metoclopramide 20 mg) was not self administrable. No placebo group.
Friedman 2006	Only comparator (intramuscular combination of trimethobenzamide 200 mg + diphenhydramine 25 mg) was not self administrable. No placebo group.
Gonzalez‐Espinosa 1997	Only comparator (intramuscular dihydroergotamine 1 mg) was not self administrable. No placebo group. In addition, blinding of study medication is uncertain (study does not appear to use double‐dummy technique) and the baseline pain intensity of treated participants is not reported
Melchart 2003	Non‐standard pain scale (50‐point categorical scale) and use of an additional dose of sumatriptan by the majority of participants at unknown, variable time point (any time after initial dosing if participants developed a full migraine attack: ˜60% used 2nd dose) meaning no useable efficacy or safety data
Pradel 2006	Not subcutaneous route of administration
Russell 1995	Data reported in Russell 1994
S2BM04	All participants initially treated with oral sumatriptan 100 mg; only those failing to respond to this initial treatment were subsequently randomised to receive either subcutaneous sumatriptan 4 mg or placebo
Solbach 1993	Subgroup analysis of data reported in Cady 1991 for menstruation‐associated migraine. No additional data reported.

H: hour