Taylor 2013.

Methods	Design: randomised controlled trial Country of origin: Australia Intervention(s): resistance training Unit of allocation: individual
Participants	Number of participants: 49 randomised; exercise group: n = 24; control group: n = 25 (demographic data reported on n = 48 as 1 person withdrew from the exercise group after allocation but before the start of training) Age: mean (SD) 18 years 1 month (1 year 11 months);exercise group: mean (SD) = 18 years 2 months (1 year 11 months); control group: mean (SD) = 18 years 7 months (2 years 11 months) Sex (male/female): 26/22; exercise group: 13/10; control group: 13/12 Ethnicity: not stated GMFCS level: level II (n = 29) and level III (n = 19); exercise group: level II (n = 13) and level III (n = 10); control group: level II (n = 16) and level III (n = 9) Type of motor abnormality: spastic CP Anatomical distribution of CP: diplegia Inclusion criteria: diagnosis of spastic diplegia, aged 14‐22 years, GMFCS level II or III, able to follow simple instructions Exclusion criteria: participated in strength training in previous 6 months, single event multi‐level surgery in last 2 years, contractures > 20 degrees at hips and knees
Interventions	Aim of the intervention: to improve mobility Type of exercise programme: resistance training Exercise mode: weights machine; 4‐6 individualised exercises to target deficits identified during gait analysis. Targeted muscles were the knee extensors (7 people), the plantarflexors (4 people), the hip extensors (3 people), the hip abductors (2 people) and generalised extensors represented by the leg press (7 people) Comparator: usual recreation and physiotherapy provided it did not include progressive resistance training Setting: community gym; individually or in pairs Intervention provider: physiotherapist Duration of programme: 12 weeks (24 sessions) Exercise dose: 2 sessions per week, 3 sets of 10‐12 repetitions to fatigue (i.e. 60%‐80% 1 RM), at least 5 on Borg Rating of Percieved Exertion scale Tailoring of intervention to individual: when the participant was able to complete 3 sets of 12 repetitions of an exercise the weight to be lifted was increased Fidelity to prescribed intervention: participants kept logbooks detailing exercise, weight lifted, number of repetitions, sets and details of injuries. Mean (SD) sessions performed 21.9 (2.4). The mean (SD) rating of perceived exertion at the end of each session was 6.9 (1.1). Participants increased their training load of exercises for targeted muscles from session 3 to session 24 by a mean (SD) of 183% (23%). Monitoring of adverse events: short‐term muscle soreness reported by most participants but resolved in a few days. 1 participant reported minor calf strain, and 1 participant reported minor discomfort to plantar fascia; the programme was adjusted but participants did not miss sessions.
Outcomes	Assessment time points: baseline (week 0), postintervention (week 12), 12 weeks postintervention (week 24) Primary outcome: mobility Mobility was measured with the 6‐min walk test. Participants were instructed to walk as far as they could in 6 min and the distance (in m) was recorded Secondary outcomes: mobility‐related function, gross motor function, muscle performance Mobility‐related function was assessed with self‐selected walking speed over 10 m (reported in m/s) Mobility‐related function was assessed with a timed stairs test. The time participants took to walk up and down 3 stairs was recorded in seconds Gross motor function was assessed with dimensions D and E of the GMFM‐66. Possible range of scores 0 to 39 for dimension D and 0 to 72 for dimension E. Scores are presented as a percentage of total possible score for each domain. Higher score indicates better gross motor function. Gait was assessed using the Gait Profile Score, measured using motion analysis, which provides an overall measure of gait kinematic deviation from normal in degrees. Participant‐rated mobility was measured using the Functional Mobility Scale, which describes the level of assistance that participants require to cover different distances and environments. The score at each distance (5 m, 50 m, and 500 m) was reported on a scale of 1‐6. Higher score indicates better mobility. Participant‐rated mobility was measured using the Functional Assessment Questionnaire, a 10‐level report of the level that best describes typical walking ability. Higher score indicates better mobility. Muscle strength was measured using 1 RM of a leg press and a reverse leg press (reported in kg). Isometric muscle strength of the targeted muscles for each participant was measured using a hand‐held dynamometer. The percentage increase in muscle strength was reported. Physical activity was assessed using the ActivPAL activity monitor. Participants were instructed to wear the monitor for 7 consecutive days; number of steps per day and time spent in sitting and lying (reported in hours per day) were reported.
Notes	Source of funding: National Health and Medical Research Council of Australia Potential conflicts of interest: the authors report no conflict of interest; none perceived
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "[a] separate randomisation procedure was prepared for each stratum (GMFCS levels II and III) using permuted blocks. An independent researcher generated a block allocation sequence for each stratum by drawing pieces of paper from a sealed container and then sealing assignments in sequentially numbered opaque envelopes. The research coordinator allocated participants after enrolment and baseline testing".
Allocation concealment (selection bias)	Low risk	Quote: "[a] separate randomisation procedure was prepared for each stratum (GMFCS levels II and III) using permuted blocks. An independent researcher generated a block allocation sequence for each stratum by drawing pieces of paper from a sealed container and then sealing assignments in sequentially numbered opaque envelopes. The research coordinator allocated participants after enrolment and baseline testing".
Blinding of participants and personnel (performance bias) All outcomes	High risk	Comment: given the nature of the intervention participants were not blinded to treatment allocation
Blinding of outcome assessment (detection bias) All outcomes	High risk	Quote: "[a]ssessments were completed in a hospital gait laboratory by an assessor blinded to group allocation". Quote: "[i]n addition, two participant‐rated mobility outcomes were assessed: the Functional Mobility Scale, which describes the level of assistance that children with CP required to cover different distances and environments; and the Functional Assessment Questionnaire, a 10‐level report of the level that best describes typical walking ability". Comment: both self‐reported and objective outcome measures were used. As participants were not blinded to group allocation at least 1 outcome is at high risk of bias.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "[t]he intention‐to‐treat principle was applied with available data of all participants who were allocated and commenced their programme included in analyses". Quote: "[o]ne participant withdrew from the intervention group after allocation but before the start of training because surgery was scheduled unexpectedly". Comment: low rate of missing data (2%) and intention‐to‐treat analysis undertaken
Selective reporting (reporting bias)	Unclear risk	Comment: a protocol is not available for this study and therefore unable to determine if all expected outcomes are reported; no convincing text provided to indicate that published report includes all expected outcomes.
Other bias	Low risk	Comment: no other sources of bias identified