Table 1.

Characteristics of included studies [ordered by study ID]

Her 2010
Methods	Study design: RCT. Number randomized (total and per group): 33 participants in total; 16 participants (28 glaucoma eyes) in the auricular acupressure group and 17 participants (32 glaucoma eyes) in the sham group Number analyzed (total and per group): 33 participants in total; 16 participants (28 glaucoma eyes) in the auricular acupressure group and 17 participants (32 glaucoma eyes) in the sham group Exclusions and loss to follow-up: 0. Study follow-up: 8 weeks.
Participants	Country: Taiwan. Age (mean ± SD, range), years: 73.6 ± 9.1 in the acupressure group, 76.3 ± 10.5 in the sham group Gender: 3/16 in the acupressure group and 4/17 in the sham group are female Inclusion criteria: (1) age 30 years or older and (2) history of unilateral or bilateral glaucoma or ocular hypertension that has been treated for at least 1 year Exclusion criteria: (1) laser trabeculoplasty in the past; (2) filtering surgery for glaucoma in the past; (3) persistent ocular inflammation within the past 1 year; (4) start of or adjustment to the use of any systemic medication that affects the IOP within 3 months; (5) end-stage glaucoma with impending phthisis bulbi; (6) no detectable IOP because of poor psychological condition
Interventions	Treatment or intervention 1: auricular acupressure. Control or intervention 2: sham group. General procedures (e.g., pre-op or post-op tests, medications, procedures): “Patients in the acupressure group received ‘stimulator tapping’ and massage of the auricular points (kidney, liver, and eye) over the right ear at our first course of outpatient clinic. Tapping stimulation was administered using a 1-mm alloy ball (Magrain; Sakamura, Kyoto, Japan) designed for acupressure, which was applied to the surface of the auricular acupoints. Subsequently, over a 4-week period, patients regularly massaged their ears using thumb and index finger (squeezing each side of the ear with inward pressure from opposing finger and thumb) twice per day for 9 minutes each time (3 minutes/each point). A regular follow-up was performed and the side for tapping, left ear or right ear, was alternated weekly. After 4 weeks of follow-up, tapping of the acupoints was discontinued and no further massages were given. The investigators re-evaluated IOP, uncorrected visual acuity, and best corrected visual acuity at the 8-week follow-up conducted at the clinic. Patients in the sham group received stimulator tapping but no massage at the sham auricular points (wrist, shoulder, and jaw). All other treatment modalities were the same as those for the acupressure group.”
Outcomes	Primary outcome(s): IOP and visual acuity with or without glasses were measured Secondary outcome(s): N/A. Measurements taken, specify intervals at which outcomes assessed: “IOP was measured with a Goldmann applanation tonometer (Haag-Streit AG, Koeniz, Switzerland). Visual acuity was examined with a 6-m visual acuity chart. The glaucomatous eyes were used for analysis of IOP, as were eyes with better than 0.01 (decimal) visual acuity. All visual acuity data were transformed to the LogMAR (MAR¼minimal angle resolution) form for analysis.” Unit of analysis (individual or eye): eye or individual (1 eye per person was used for the analysis) Other issues with outcome assessment (e.g., quality control for outcomes if any): N/A.
Notes	Study dates: N/A. Funding source(s): National Science Counsel of Taiwan (NSC 97-2320-B-039-022-MY3 and NSC 98-2815-C-039-097-B), Department of Health, Executive Yuan, Taiwan (DOH98-TD-F-113-098011), and China Medical University (CMU98-CT-10), Taichung, Taiwan, Publication language: English.
Risk of bias
Bias	Authors’ judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Method used to generate random sequence was not described in the article. After contacting the author, we received the following information: “We used computer to generate 40 (our original expected patient collection numbers) simple randomized allocation sequences. We then assigned patients to ’control’ or ’trial’ group according to their clinic visit order. So we knew the next patient would be assigned to which group, but we didn’t know who would be the next one (patient’s visit order was arranged by the clinic assistant, who didn’t know the randomized allocation sequences). The patient him-/herself would not know which group he/she would be assigned.”
Allocation concealment (selection bias)	Low risk	Method used to conceal allocation was not described in the article. We considered that allocation was adequately concealed after contacting the authors because “patient’s visit order was arranged by the clinic assistant, who didn’t know the randomized allocation sequences.”
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Masking of participants and personnel was not described in the article. After contacting the author, we received the information that “patients were blinded of their treatment group whereas the outcome assessors were not blinded.”
Blinding of outcome assessment (detection bias) All outcomes	High risk	See above.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No loss to follow-up was reported.
Selective reporting (reporting bias)	Unclear risk	Original protocol was not available.
Other bias	Low risk	No other bias was identified.