Ansari 2006.

Study characteristics
Methods	RCT conducted in outpatient physiotherapy clinic (secondary care setting)
Participants	15 participants were randomised, who were aged 18 to 65 years with non‐radiating, non‐specific low back pain lasting for more than 3 months. Exclusion criteria were: abnormal neurological status; concomitant severe disease; psychiatric illness; current psychotherapy; pathological lumbosacral X‐rays (except for minor degenerative changes); rheumatic inflammatory disease; planned hospitalisation; addiction to any kind of substance; and any contraindication to ultrasound therapy.
Interventions	Intervention (I) group (n = 5) received 1 MHz continuous ultrasound, at 1.5 W/cm² for 10 sessions, 3 days per week. Duration of ultrasound application was calculated according to the formula: total treatment time = planned average local exposure time x (tissue area/effective radiation area of applicator). Control (C) group (n = 5) received placebo ultrasound for 10 sessions, 3 days per week.
Outcomes	Mean (SD) pre‐ and post‐treatment scores on Functional Rating Index were: (I) 56.5 (20.35), 34.5 (13.5); and (C) 46.95 (14.38), 39.9 (16.5). Mean (SD) pre‐ and post‐treatment degrees of flexion range of motion were: (I) 117.4 (2.5), 128.6 (14.3); and (C) 103.4 (13.39), 109.2 (10.6). Mean (SD) pre‐ and post‐treatment degrees of extension range of motion were: (I) 23.8 (4.15), 30 (6.4); and (C) 27.2 (3.03), 29 (4.2). No between‐group difference was seen for H‐reflex parameters (electroneurophysiological evaluation) or in lumbar spine lateral flexion (left and right) range of motion.
Notes	No conflict of interest declared with regard to commercial funding.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method for determining randomisation sequence not reported in text.
Allocation concealment (selection bias)	Unclear risk	Not reported in text
Blinding of participants and personnel (performance bias) Participants	Low risk	Participants blinded to group allocation.
Blinding of participants and personnel (performance bias) Care providers	High risk	Care providers not blinded.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Primary outcome measure is self‐reported, participants blind to group allocation.
Incomplete outcome data (attrition bias) Dropout rate described	High risk	Dropout rate described, 30% of sample dropped out.
Incomplete outcome data (attrition bias) Intention‐to‐treat	High risk	Participants who dropped out were excluded from analysis.
Selective reporting (reporting bias)	Unclear risk	No trial pre‐registration or published protocol was available.
Similar groups	High risk	Large differences in gender, age, BMI between groups
Co‐interventions	Low risk	Participants advised not to commence new treatments.
Compliance	High risk	High proportion of dropouts due to noncompliance.
Timing of outcome measures	Low risk	Similar timing of outcome assessment (post‐treatment) for both groups.