Skip to main content
. 2020 Jan 5;11(2):227–236. doi: 10.1002/jrsm.1385

Table 1.

Trial characteristics by Physiotherapy Evidence Database (PEDro) scores of below 6 or 6 and above

All Trials PEDro Score <6 PEDro Score ≥6 P Value*
Number of trials 345 188 157
PEDro score 5.00 (4.00‐7.00) 5.00 (4.00‐5.00) 7.00 (6.00‐7.00) <.001
Year of publication 2002 (1998‐2004) 2000 (1997‐2004) 2003 (1999‐2005) <.001
Trial design .25
Parallel 322 (93%) 174 (93%) 148 (94%)
Factorial 15 (4.3%) 7 (3.7%) 8 (5.1%)
Crossover 7 (2.0%) 6 (3.2%) 1 (0.6%)
Cluster 1 (0.29%) 1 (0.53%) 0 (0.0%)
Multicentric study 85 (25%) 30 (16%) 55 (35%) <.001
Hypothesis .41
Superiority 317 (92%) 169 (90%) 148 (94%)
Equivalence 15 (4.3%) 11 (5.9%) 4 (2.5%)
Noninferiority 4 (1.2%) 2 (1.1%) 2 (1.3%)
Unclear/no comparison 9 (2.6%) 6 (3.2%) 3 (1.9%)
Placebo‐controlled 28 (8.1%) 9 (4.8%) 19 (12%) .017
Study population .20
Adult 218 (63%) 122 (65%) 96 (61%)
Geriatric 116 (34%) 57 (30%) 59 (38%)
Pediatric 5 (1.4%) 4 (2.1%) 1 (0.6%)
Unclear 6 (1.7%) 5 (2.7%) 1 (0.6%)
Interventions .031
Exercise 266 (77%) 154 (82%) 112 (71%)
Education 9 (2.6%) 5 (2.7%) 4 (2.5%)
Manual therapy 13 (3.8%) 4 (2.1%) 9 (5.7%)
Acupuncture 8 (2.3%) 1 (0.53%) 7 (4.5%)
Other 49 (14%) 24 (13%) 25 (16%)
Outcome source .036
Clinician assessment 128 (37%) 82 (44%) 46 (29%)
Self‐reported 180 (52%) 86 (46%) 94 (60%)
Laboratory data 15 (4.3%) 9 (4.8%) 6 (3.8%)
Administrative data 22 (6.4%) 11 (5.9%) 11 (7.0%)
Sample size
Randomized 72.0 (41.0‐148) 59.0 (32.0‐109) 104 (60.0‐191) <.001
Analyzeda 63.0 (37.0‐128) 49.0 (30.0‐95.0) 93.0 (54.0‐170) <.001
Fundingb
Industry 36 (10%) 17 (9.0%) 19 (12%) .86
Government 183 (53%) 79 (42%) 104 (66%) .002
Academic 41 (12%) 23 (12%) 18 (11%) .50
Foundation 96 (28%) 46 (24%) 50 (32%) .70
No funding 18 (5.2%) 12 (6.4%) 6 (3.8%) .22
Not declared 69 (20%) 50 (27%) 19 (12%) .001

Note. Numbers (%) or medians (interquartile range) are shown.

a

Unknown for 10 trials with a PEDro score <6 and 1 trial with a PEDro score ≥6.

b

Does not add up to 100% as multiple entries are possible.

*

P values obtained from Wilcoxon rank‐sum and Fisher exact tests for continuous and categorical variables, respectively.