Tsukayama 2002.

Study characteristics
Methods	Two‐arm RCT Number analyzed/randomized: 19/20 Statistical analysis: analyzed as assigned group; repeated ANOVA; power analysis not conducted Funding source: grant from the Foundation for Training and Licensure Examination in Anma‐Massage‐Acupressure, Acupuncture and Moxibustion and the Tsukuba College of Technology Ethical approval and informed consents obtained
Participants	Participant recruitment: by leaflets in Tsukuba, Japan Setting: at Tsukuba College of Technology Clinic, Japan Inclusion criteria: 1) LBP without sciatica; 2) at least a 2‐week history of LBP; (3) > 20 years old Exclusion criteria: 1) radiculopathy or neuropathy in the lower extremity; 2) fracture, tumor, infection, or internal disease; 3) other general health problems; 4) other conflicting or ongoing treatments Mean age: 44.9 years Gender (female): 84% Pain duration (mean): 8.3 years Pain intensity (mean ± SD): not reported
Interventions	1) GROUP 1: electroacupuncture (EA, semi‐standardized) Acupuncture points: 8 points in total; BL23 and BL26 most frequently used Depth: the average insertion depth was approximately 20 mm De Qi: 15 min stimulation with a frequency of 1Hz Sessions: 4 sessions (2/week for 2 weeks) Acupuncturist experience: conservative physical therapy currently practiced in Japan After electroacupuncture treatment at each session, the press tack needles (Seirin Jr®: Seirin Kasei Co. Ltd., Japan) were inserted at four of the eight points chosen, left in situ for several days, and then removed. 2) GROUP 2: TENS Gel type disposable electrodes of size 20 x 30 mm were used for eight points. Electrostimulation was applied and muscle contraction was observed. After each session, a poultice containing methyl salicylic acid, menthol, and antihistamine was prescribed for home application between treatments to the low back region. Co‐intervention: not reported Duration of treatment: 2 weeks Duration of follow‐up: immediately post‐treatment
Outcomes	1) Pain intensity: VAS (0 to 100) 2) Pain‐related disability: subjective symptoms and restriction of daily activities items of Japanese Orthopaedic Association (JOA) score, higher values = better Assessment time: immediately post‐treatment Costs: NR Adverse effects: 1) EA group: transient aggravation of LBP (n = 1), discomfort due to press tack needles (n = 1), pain on needle insertion (n = 1) and small subcutaneous bleeding (10 mm in diameter, n = 1); 2) TENS group: transient aggravation of back pain (n = 1), transient fatigue (n = 1), itching with electrode (n = 1)
Notes	Conclusion: "In the present preliminary study, although some placebo effect may be included, EA showed more effectiveness than TENS in short‐term treatment of LBP." Language: English Mean and SD data of VAS and JOA score were obtained from Figures. For results, see comparison 4.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer software‐generated random numbers
Allocation concealment (selection bias)	Low risk	Sealed envelopes were used.
Blinding of participants (performance bias) All outcomes	Unclear risk	Not reported
Blinding of personnel /care providers (performance bias) All outcomes	High risk	Not possible
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Blinded assessor was mentioned; otherwise outcomes seemed to be participant‐reported.
Incomplete outcome data (attrition bias) All outcomes	Low risk	There was a 5% dropout rate, and acceptable withdrawal reasons.
Intention‐to‐treat‐analysis (attrition bias)	Low risk	Subjects with available data were analyzed according to the assigned group.
Selective reporting (reporting bias)	Low risk	All outcomes were reported.
Group similarity at baseline (selection bias)	Unclear risk	The risk factors were similar between the two groups except for VAS (pain).
Co‐interventions (performance bias)	Unclear risk	TENS group also used a poultice to the low back region.
Compliance bias (performance bias)	Low risk	Comparable compliance between groups
Timing of outcome assessments (detection bias)	Low risk	Same time
Other bias	Low risk	Not identified