Table 1.

Risk of bias assessment of RCTs of all interventions for low-back pain and neck pain (total of 131 RCTs).

Quality components	N studies	%
Adequate method of randomization	57	43.5%
Adequate method of allocation concealment	41	31.3%
Similarity at baseline regarding the most important prognostic indicators	89	67.9%
Appropriate patient blinding to the intervention	30	22.9%
Appropriate care provider blinding to the intervention	4	3.1%
Appropriate outcome assessor blinding to the intervention	66	50.4%
Similar or no cointerventions between-groups	40	30.5%
Acceptable compliance in all groups	53	40.5%
Described and acceptable drop-out rates	99	75.6%
Similarity of timing of the outcome assessment in all groups	118	90.1%
Inclusion of an intention-to-treat analysis	57	45.5%
Absence of selective outcome reporting	78	59.5%
Absence of other potential bias	7	5.3%
Total risk of bias scores (max 13); median (IQR)	6	4–7