Kayser 2017.

Study characteristics
Methods	Study design: open‐label, parallel, randomised Total duration of study: 4 to 6 weeks Number of study centres and locations: 1, University Hospital, Bonn, Germany. Study setting: no information Withdrawals: 5 patients discontinued the study for different reasons: 3 because of MST device defect, 1 because of cognitive adverse effect during ECT, and 1 because of worsening of preexisting coronary heart disease during ECT Dates of study: February to June 2011
Participants	Number: 10 for MST (8 MDD, 2 BPII); 10 for ECT (9 MDD, 1 BPI) Mean age (SD), age range: 45 (14) for MST; 55 (12) for ECT; no information on range Gender: 3 females for MST; 4 females for ECT Course of disease: no information Severity of condition: HAMD‐28 26.1 (4) for MST; 28.4 (4) for ECT Diagnostic criteria MDD, BPI, and BPII according to DSM‐IV TRD defined as failure of 2 different antidepressants (given > 5 weeks at maximum recommended or tolerated dose) during current depressive episode according to Thase and Rush stage 2 definition Inclusion criteria TRD 18 to 65 years old Exclusion criteria Pregnancy Younger than 18 years Other psychiatric, cognitive, or neurological disorder At high risk for anaesthesia Magnetisable material in the head, cardiac pacemaker, vagus nerve stimulator, or any medical pump Co‐morbid conditions: no information
Interventions	MST (coil placement, frequency, dose, number of sessions) Twin coil containing 2 individual round coils positioned over Cz according to the international 10‐20 system 100 Hz 6 times seizure threshold up to 800 pulses in a train 8 to 12 sessions ECT (electrode placement, pulse width, dose, number of sessions) Right unilateral for 9 participants; bifrontotemporal for 1 0.5 ms 6× seizure threshold for right unilateral; 3× seizure threshold for bifrontotemporal 8 to 12 sessions Concomitant medications: psychotropic medication was stable for a minimum of 4 weeks before MST/ECT treatments and remained unchanged during the study Concomitant psychosocial interventions: no information Excluded medications: no information
Outcomes	Primary and secondary outcomes specified and collected Specified Recovery time Collected HAMD Seizure features (polyspike wave duration and polyspike wave amplitude in tonic phase, slow wave duration and slow wave amplitude in clonic phase, postictal suppression in termination phase, and regularity and stereotypy of global pattern) Time points reported Before and after all treatments for HAMD Average of all treatments for seizure features Outcome data reported in a usable way: yes
Notes	Funded in part by MagVenture A/S
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Protocol described the study as a case‐control trial. However, paper claims "the patients were randomized to ECT or MST using a randomized block design, with a block size of 5 patients"
Allocation concealment (selection bias)	Unclear risk	No information
Blinding of participants and personnel (performance bias) All outcomes	High risk	Described as open‐label
Blinding of outcome assessment (detection bias) All outcomes	High risk	Described as open‐label
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Attrition rates were similar (3 for MST vs 2 for ECT), however, for different reasons (MST device defect vs adverse effects). As‐treated analysis was done
Selective reporting (reporting bias)	High risk	Protocol specified recovery time as the only outcome. However, study reported HAMD and seizure features
Other bias	High risk	The study is funded in part by a manufacturer of MST devices