Ref (type)	Population	Outcome, Interventions	Results and statistical analysis	Effect size	Favours
Clinical scores
[34] RCT Crossover design	24 people with ET (23 completed), clinical diagnosis	Clinical score (0–4) x segments 2 weeks 22/23 (96%) with propranolol 5/23 (22%) with placebo	P <0.001	Effect size not calculated	propranolol
[35] RCT Crossover design	11 people with ET (10 completed), clinical diagnosis	Clinical score 6 weeks' treatment with propranolol with placebo Absolute results not reported	P <0.003	Effect size not calculated	propranolol
[36] RCT Crossover design	9 people with ET (7 completed), clinical diagnosis	Clinical score with propranolol with placebo Absolute results not reported	P <0.01	Effect size not calculated	propranolol
[11] RCT Crossover design	16 people with ET (16 completed), clinical diagnosis	Clinical score with propranolol 240 mg/day with placebo Absolute results reported graphically	P <0.05 (propranolol 240 mg/day v placebo)	Effect size not calculated	propranolol 240 mg/day
[11] RCT Crossover design	16 people with ET (16 completed), clinical diagnosis	Clinical score with propranolol 120 mg/day with placebo Absolute results reported graphically	Unclear; no report of a significant difference (propranolol 120 mg/day v placebo)
[37] RCT Crossover design 5-armed trial	23 people with ET (15 completed), clinical diagnosis	Clinical score (0–5 each side, maximum 10) with propranolol 80 mg 3 times/day with propranolol LA 160 mg/day with propranolol LA 240 mg/day with propranolol LA 320 mg/day with placebo Absolute results reported graphically	P <0.05 (all doses of propranolol except 160 mg v placebo)	Effect size not calculated	propranolol (80 mg 3 times/day or LA 240 mg/day or LA 320 mg/day)
[12] RCT Crossover design 4-armed trial	9 people with ET (9 completed), clinical diagnosis	Objective tremor score (0–25) 6.6 with propranolol 10.7 with placebo	P <0.01 (propranolol v placebo)	Effect size not calculated	propranolol
[29] RCT Crossover design 3-armed trial	17 people with ET (12 completed), clinical diagnosis	Clinical score (0–10): mean change from baseline –2.58 with propranolol –1.08 with placebo	P <0.01 (propranolol v placebo)	Effect size not calculated	propranolol
[24] RCT Crossover design 3-armed trial	16 people with ET (16 completed), clinical diagnosis	Tremor Clinical Rating Scale score ( P1 + P2, 0–78) with propranolol with placebo	Mean difference (propranolol v placebo): –4.95 P <0.01	Effect size not calculated	propranolol
Accelerometry
[34] RCT Crossover design	24 people with ET (23 completed), clinical diagnosis	Accelerometry (frequency) 2 weeks with propranolol with placebo	Reported as unchanged
[34] RCT Crossover design	24 people with ET (23 completed), clinical diagnosis	Accelerometry (amplitude) 2 weeks with propranolol with placebo	P <0.001	Effect size not calculated	propranolol
[36] RCT Crossover design	9 people with ET (7 completed), clinical diagnosis	Accelerometry with propranolol with placebo	P <0.01	Effect size not calculated	propranolol
[17] RCT Crossover design	16 people with ET (16 completed), clinical diagnosis	Accelerometry (amplitude) with propranolol 120 and 240 mg/day with placebo Absolute results reported graphically	P <0.01 (propranolol 240 mg/day v placebo)	Effect size not calculated	propranolol 240 mg/day
[11] RCT Crossover design	16 people with ET (16 completed), clinical diagnosis	Accelerometry (amplitude) with propranolol 120 mg/day with placebo Absolute results reported graphically	P reported as not significant (propranolol 120 mg/day v placebo)		Not significant
[11] RCT Crossover design	16 people with ET (16 completed), clinical diagnosis	Accelerometry (amplitude) with propranolol 240 mg/day with placebo Absolute results reported graphically	P <0.02 (propranolol 240 mg/day v placebo)	Effect size not calculated	propranolol 240 mg/day
[37] RCT Crossover design 5-armed trial	23 people with ET (15 completed), clinical diagnosis	Accelerometry (magnitude) with propranolol 80 mg 3 times/day with propranolol LA 160 mg/day with propranolol LA 240 mg/day with propranolol LA 320 mg/day with placebo Absolute results reported graphically	P <0.02 (all propranolol dosing regimens combined v placebo)	Effect size not calculated	propranolol (all included dosing regimens combined v placebo)
[29] RCT Crossover design 3-armed trial	17 people with ET (12 completed), clinical diagnosis	Accelerometry (frequency: mean change from baseline) –0.20 with propranolol +0.11 with placebo	P value reported as not significant		Not significant
[29] RCT Crossover design 3-armed trial	17 people with ET (12 completed), clinical diagnosis	Accelerometry (amplitude: mean change from baseline) –87.60 with propranolol –66.90 with placebo	P <0.01	Effect size not calculated	propranolol
[24] RCT Crossover design 3-armed trial	16 people with ET (16 completed), clinical diagnosis	Accelerometry (tremor magnitude) with propranolol with placebo	Reported as not significant P value not reported		Not significant
[24] RCT Crossover design 3-armed trial	16 people with ET (16 completed), clinical diagnosis	Accelerometry (tremor frequency with propranolol with placebo	Reported as not significant P value not reported		Not significant
[13] RCT Crossover design 3-armed trial	24 people with ET (24 completed), clinical diagnosis	Accelerometry (frequency) 9.1 with propranolol for 5 to 7 days 9.4 with placebo for 5 to 7 days	Reported as not significant P value not reported		Not significant
[13] RCT Crossover design 3-armed trial	24 people with ET (24 completed), clinical diagnosis	Accelerometry (maximum amplitude) 71 with propranolol for 5 to 7 days 128 with placebo for 5 to 7 days	P <0.05 (propranolol v placebo)	Effect size not calculated	propranolol
[14] RCT Crossover design 3-armed trial	24 people with ET (24 completed), clinical diagnosis	Accelerometry (tremor intensity, a summated value of the acceleration of postural tremor for 40 seconds) 42.9 with propranolol 4.9 with placebo	P <0.01 (propranolol v placebo)	Effect size not calculated	propranolol
Self-rating
[34] RCT Crossover design	24 people with ET (23 completed), clinical diagnosis	Self-rating with propranolol with placebo
[35] RCT	11 people with ET (10 completed), clinical diagnosis	Self-rating 6 weeks' treatment 5/5 (100%) with propranolol 1/5 (20%) with placebo	P value not reported
[36] RCT Crossover design	9 people with ET (7 completed), clinical diagnosis	Number of people improved >2 assessments 12 with propranolol 2 with placebo	P <0.05	Effect size not calculated	propranolol
[11] RCT Crossover design	16 people with ET (16 completed), clinical diagnosis	Self-rating (0–5) with propranolol 120 and 240 mg/day with placebo Absolute results reported graphically	P <0.01 (propranolol 120 mg/day or 240 mg/day v placebo)	Effect size not calculated	propranolol
[37] RCT Crossover design 5-armed trial	23 people with ET (15 completed), clinical diagnosis	Self-rating (0–5) with propranolol 80 mg 3 times/day with propranolol LA 160 mg/day with propranolol LA 240 mg/day with propranolol LA 320 mg/day with placebo Absolute results reported graphically	P <0.05 (all doses of propranolol v placebo)	Effect size not calculated	propranolol
[12] RCT Crossover design 4-armed trial	9 people with ET (9 completed), clinical diagnosis	Mean subjective tremor score (0–10) 4.4 with propranolol 6.2 with placebo	P = 0.1 (propranolol v placebo)		Not significant
[29] RCT Crossover design 3-armed trial	17 people with ET (12 completed), clinical diagnosis	Self-rating (0–10): mean change from baseline –4.50 with propranolol –2.04 with placebo	P <0.01 (propranolol v placebo)	Effect size not calculated	propranolol
[24] RCT Crossover design 3-armed trial	16 people with ET (16 completed), clinical diagnosis	Self-rating: subjective Disability Scale (25–100) with propranolol with placebo	Mean difference in Subjective Disability Scale score (propranolol v placebo) –4.48 P = 0.11		Not significant
Performance tests
[35] RCT	11 people with ET (10 completed), clinical diagnosis	Performance tests: pegboard test 6 weeks' treatment +2.9 with propranolol –2.1 with placebo	P = 0.06		Not significant
[11] RCT Crossover design	16 people with ET (16 completed), clinical diagnosis	Performance tests (hand-writing, drawing geometrical figures, and tracing an Archimedes' spiral, scores 0–5) with propranolol 120 and 240 mg/day with placebo Absolute results reported graphically	P <0.01 (propranolol 120 mg v placebo)	Effect size not calculated	propranolol 120 mg
[37] RCT Crossover design 5-armed trial	23 people with ET (15 completed), clinical diagnosis	Performance test scores (copy a short sentence and trace inside an Archimedes' spiral, 0–5) with propranolol 80 mg three times/day with propranolol LA 160 mg/day with propranolol LA 240 mg/day with propranolol LA 320 mg/day with placebo Absolute results reported graphically	P <0.01 (propranolol and propranolol LA 320 mg v placebo)	Effect size not calculated	propranolol (80 mg/day three times/day or LA 320 mg/day)
[29] RCT Crossover design 3-armed trial	17 people with ET (12 completed), clinical diagnosis	Performance tests: mean change from baseline in pegboard test (time to complete in seconds) –8.58 with propranolol –6.17 with placebo	P value reported as not significant		Not significant