Voet 2014.

Methods	Evaluator‐blind, parallel‐group RCT
Participants	Sample size Intervention group: 20 adults with FSHD type 1, control group: 24 adults with FSHD type 1 Inclusion criteria Age ≥ 18 years, severe fatigue (CIS‐fatigue > 35), able to walk independently (ankle‐foot orthoses and canes are accepted), able to exercise on a bicycle ergometer, able to complete either type of intervention Exclusion criteria Cognitive impairment, insufficient mastery of the Dutch language, neurologic or orthopedic comorbidity interfering with the interventions or possibly influencing outcomes, pregnancy, use of psychotropic drugs (except simple sleeping medication), severe cardiopulmonary disease (chest pain, arrhythmia, pacemaker, cardiac surgery, severe exertional dyspnea, emphysema), epileptic seizures, poorly regulated diabetes mellitus or hypertension, clinical depression, as diagnosed with the Beck Depression Inventory for Primary Care. Baseline demographics Mean age participants control group: 52 years, mean age participants training group: 59 years. FSHD clinical score participants control group: 3.0, Ricci score training group: 3.0. Female/male ratio control group: 17/17, female/male ratio training group: 12/18. Duration of illness participants control group: 16.7 years, duration of illness participants training group: 13.0 years
Interventions	Aerobic exercise vs usual care (and a 3rd group with cognitive behaviour therapy) Type of training and exercise. Cycling exercises on an ergometer Intensity Resistance was increased until an increase of 50%‐65% in heart rate reserve was achieved. Frequency 3 times/week, twice a week at home, once a week at a rehabilitation centre Setting Once a week at the department of rehabilitation at one of 6 the participating centres (St Maartenskliniek Nijmegen, de Hoogstraat Utrecht, de Vogellanden Zwolle, het Roessingh Enschede, St Franciscus ziekenhuis Roosendaal, Reade Amsterdam, the Netherlands) Duration Session: 30 min, with additional warming‐up and cooling‐down periods of 5 and 3 min, respectively. Programme: 16 weeks Supervision Supervised by a physiotherapist
Outcomes	Primary The subscale fatigue of the checklist individual strength Secondary Isometric strength for the quadriceps as MVIC, aerobic exercise tolerance (VO2 peak and distance walked in a 6‐min walk test), physical activity using actometers and a questionnaire, sleep quality, pain intensity, social participation restrictions Time‐points measured Before and after the intervention period of 16 weeks and after 12 weeks of follow‐up
Dates	Participants were enrolled from January 2009 through February 2012
Funding/ declarations of interest	N Voet received grants from Prinses Beatrix SpierFonds (PBF) (The Dutch Public Fund for Neuromuscular Disorders), the Netherlands Organisation for Health Research and Development (ID: ZonMW 89000003), and global FSH. G Bleijenberg, J Hendriks, I de Groot, and G Padberg report no disclosures relevant to the manuscript. B van Engelen was research director of the European Neuromuscular Centre (ENMC), received grants from global FSH, the Netherlands Organisation for Health Research and Development, Prinses Beatrix SpierFonds (PBF) (The Dutch Public Fund for Neuromuscular Disorders), and the Dutch FSHD Foundation. S. Geurts received grants from Prinses Beatrix SpierFonds (PBF) (The Dutch Public Fund for Neuromuscular Disorders), the Netherlands Organisation for Health Research and Development (ID: ZonMW 89000003), and global FSH. Go to Neurology.org for full disclosures.
Notes	Outcomes were presented for the 3 treatment groups (i.e. aerobic exercise, cognitive behaviour therapy and usual care). Effect sizes were presented by group as well
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "An independent research assistant allocated each participant to 16 weeks of AET [aerobic exercise training], CBT [cognitive behavioural therapy], or UC [usual care] using a computer‐generated randomisation block list". The block sizes varied randomly in order to prevent predictability of the allocation process.
Allocation concealment (selection bias)	Low risk	Quote: "An independent research assistant allocated each participant to 16 weeks of AET [aerobic exercise training], CBT [cognitive behavioural therapy], or UC [usual care], using a computer‐generated randomisation block list".
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not blinded for group assignment, as it is impossible to blind exercise training compared to no exercise training.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "All measurements were performed (...) by 2 blinded physical therapists"
Incomplete outcome data (attrition bias) All outcomes	Low risk	Comment: 11 participants in the aerobic exercise group (39%) did not achieve the level of acceptable adherence. One participant in the aerobic exercise group withdrew due to time constraints. However, analysis was by ITT. Comment: in the usual care group, 4 participants withdrew just before the 2nd randomisation because they thought that the intervention would be too time‐consuming.
Selective reporting (reporting bias)	Low risk	Comment: no evidence found for selective reporting
Other bias	Low risk	Comment: no risk of bias from other sources detected