Wong 2012.

Study characteristics
Methods	Study objective: to compare quality of life (QOL) aspects in venous leg ulcer patients over 55 years of age, of short‐stretch compression (SSB), 4‐layer compression bandaging (4LB) and usual care (UC) (moist wound‐healing dressing, no compression) Trial design (e.g. parallel group) including research sites: parallel, randomised controlled trial, multi‐site Follow‐up period: 12 and 24 weeks Number of arms: 3 Study start date and end date: patients recruited from May 2007 to November 2008 Care setting: community (9 general outpatient clinics in the New Territories East Cluster, Kowloon East Cluster and the Kowloon Central Cluster)
Participants	Study population: leg ulcers caused by chronic venous insufficiency Eligibilitycriteria: inclusion of either men or women aged 55 years or older having confirmed venous leg ulcers (with partial or full‐thickness skin loss) by clinical and vascular assessment (i.e. Doppler); without necrotic tissue; ability to understand and communicate in Cantonese; ABPI ≥ 0.8 indicating suitability for compression bandaging. Exclusion of people with ulcers of < 5 cm2 or > 118 cm2, with an ulcer duration of less than 4 weeks or longer than 1 year; those with 2 or multiple leg ulcers, either on 1 or both legs; those with an ABPI < 0.8 and those with concurrent administration of drugs that may affect ulcer healing, such as corticosteroids or chemotherapeutics Sex (M:F): 206:115 overall Age (years): mean 71.7 (SD 8.5) overall Duration of leg ulcers: mean 27.4 (SD 43.7) months overall Baseline leg ulcer area: mean 8.2 (SD 11.0) cm2 overall Group difference: no significant differences between groups in variables explored (e.g. age; sex; ulcer duration; ulcer size) Total number of participants: 321 patients, each with 1 ulcer Unit of analysis (including number of ulcers per person): participants, each with 1 ulcer Unit of randomisation (e.g. leg ulcer, limb, or participant): patients
Interventions	Intervention characteristics Four‐layer bandage Details of interventions (including compression devices used, and duration of interventions applied): 4‐layer compression bandaging (Profore; Smith & Nephew PLC, Hull, UK). Typically applied at weekly intervals for 24 weeks. The sub‐bandage resting pressure between 40 to 50 mmHg Descriptions of any co‐interventions or standard care: used along with the usual topical ulcer care, including a moist wound‐healing dressing Number of participants randomised: 107 participants Number of participants analysed: 87 participants for health‐related quality of life and pain score outcome; 107 participants for ulcer‐healing outcomes Short‐stretch bandage Details of interventions (including compression devices used, and duration of interventions applied): short‐stretch compression bandaging (Rosidal sys; Lohmann & Rauscher GmbH & Co KG, Rengsdorf, Germany). Typically applied at weekly intervals for 24 weeks. The sub‐bandage resting pressure between 40 to 50 mmHg Descriptions of any co‐interventions or standard care: used along with the usual topical ulcer care, including a moist wound‐healing dressing. Number of participants randomised: 107 participants Number of participants analysed: 95 for health‐related quality of life and pain score outcome; 107 for ulcer‐healing outcomes Usual care Details of interventions (including compression devices used, and duration of interventions applied): moist wound‐healing dressing without compression. Applied alginates, films, foams or absorbent dressings, as well as a bio‐cellulose dressing (Suprasorb A or A+Ag, Suprasorb F, Solvaline, Vliwazell, Suprasorb P, Suprasorb X or X+PHMB; Lohmann & Rauscher GmbH & Co KG, Rengsdorf, Germany) Descriptions of any co‐interventions or standard care: not described Number of participants randomised: 107 participants Number of participants analysed: 94 health‐related quality of life and pain score outcome; 107 for ulcer‐healing outcomes
Outcomes	Time‐to‐complete wound healing Outcome type: time‐to‐event Time points: not relevant Reporting: partially reported Measurement method (e.g. scale, self‐reporting): complete ulcer healing defined according to the criteria of the Wound Healing Society as 100 per cent re‐epithelialisation of the wound surface with the absence of exudates (Wong 2008b). Definition: mean time to ulcer healing Missing data and reasons (including the number of people who died): all included in analysis Proportion of wounds completely healed during follow‐up Outcome type: binary Time points: week 12 and 24 Reporting: partially reported Measurement method (e.g. scale, self‐reporting): used a digital planimetry tool (VeV VERGe Videometer) to assess reduction in ulcer area Definition: ulcer healing within 6 months Missing data and reasons (including the number of people who died): all included in analysis. Adverse events Outcome type: binary Time points: not described Measurement method (e.g. scale, self‐reporting): probably not measured using a prespecified method Definition: total numbers of participants with adverse events Notes: data on dropouts due to reasons other than adverse events were not extracted for this review. Data extracted for this review were based on Table 5 of Wong 2012 rather than its Figure 1 that had different numbers of dropouts due to adverse events. Participant health‐related quality of life/health status Outcome type: continuous Time points: week 12 and 24 Reporting: partially reported Measurement method (e.g. scale, self‐reporting): Chinese (Hong Kong) Short Form‐12‐item Health Survey (SF‐12) as a generic QOL tool (higher score = better status); Charing Cross Venous Ulcer Questionnaire (Chinese) as a disease‐specific QOL measurement instrument (lower score = better status); both collected at weeks 0, 12 and 24 Definition (including ulcer stage): health‐related quality of life. Missing data and reasons (including the number of people who died): 20 of 107 in 4LB; 12 of 107 in SSB; and 13 of 107 in UC missed (all withdraws due to adverse events; specific adverse events given for 17 in 4LB; 12 in SSB and 16 in UC). Notes: outcome assessor unspecified. The study authors also measured Frenchay Activity Index (for functional status) but reported this and quality of life separately. The review authors did not extract the Frenchay Activity Index data for this review. Cost effectiveness Not reported; cost data presented but not extracted for this review Mean pain score Outcome type: continuous Time points: week 12 and 24 Reporting: partially reported Measurement method (e.g. scale, self‐reporting): self‐rated using Brief Pain Inventory instrument with a 10‐point VAS scale with grades from 0 to 10; 0 = least painful. Collected at weeks 0, 12 and 24 (lower score = less pain) Definition (including ulcer stage): ulcer‐related pain severity Missing data and reasons (including the number of people who died): not reported Notes: outcome assessor unspecified. Pain interference with various aspects of QOL measured but not extracted for this review. Outcomes that are not considered in this review but reported in trials: The average ulcer area Pain interference with various aspects of QOL Interface pressure Frenchay Activitiy Index
Identification	Publication type/ status (e.g. conference abstract): 2 full papers (Wong 2012; So 2014), retraction record, conference abstract Trial protocol: not described Source of funding: funded by the Health, Welfare and Food Bureau of Hong Kong (HHSRF #404060481) and a scientific grant of Lohmann & Rauscher GmbH & Co KG, Rengsdorf, Germany. Country of origin: Hong Kong Comments: Wong 2012 published in J Vass Surg 2012;55:1376‐85 was retracted by the Journal’s Editor‐in‐Chief because the study has been previously published in J Eur Acad Dermatol Venereol 2012;26:102‐10. Data extracted for this review were based on the paper from J Eur Acad Dermatol Venereol 2012;26:102‐10. Two authors (the leading and correspondence authors) participate in various scientific projects with medical device companies, such as Smith & Nephew and Lohmann & Rauscher. One author is an employee of Lohmann & Rauscher, the company that provided all the study products. Contact information: A. Andriessen. E‐mail: anneke.a@tiscali.nl
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: “Eligible patients were randomly assigned to either one of the experimental groups (compression treatment) or the control group using a computer generated list after pre‐test measurements were taken” Comment: low risk of bias because of the use of a proper method of generating random sequence
Allocation concealment (selection bias)	Unclear risk	Quote: “After confirming eligibility of a patient and obtaining informed consent, the clinical investigator digitally received the information on the allocation of the patient to one of the treatment groups” [based on retracted paper]. Comment: unclear risk of bias because information provided is insufficient for this domain’s judgement
Blinding of participants and personnel (performance bias) All outcomes	High risk	Outcome group: all outcomes Quote: "participants and interviewers (data collectors) were not informed of the treatment allocation of the study participants for the duration of the study" (Wong 2008b) Quote: "the interveners who performed wound dressing and/or bandaging were not blinded to the treatment groups. Blinding the interveners is not achievable ..." (Wong 2008b) Comment: high risk of bias because in our judgement it is clearly not possible to blind either participants or personnel and a related study (Wong 2008b) suggest that there is non‐blinding of participants
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Outcome group: all outcomes Quote: "This was a double‐blind study, in which the interviewers who were responsible for the pre‐test and post test data collections were not given any information regarding the group to which the participant was assigned" Comment: low risk of bias
Incomplete outcome data (attrition bias) All outcomes	Low risk	Outcome group: ulcer healing and time to healing Quote: “All 321 patients were included in the survival analysis on ulcer healing” Comment: low risk of bias because of the use of ITT analysis Outcome group: health of quality and pain Quote: “All withdrawn cases were regarded as unsuccessful in terms of treatment and all variables, including size and pain score” Quote: “Forty‐five patients (14%) were withdrawn before the second data collection at week 24.” Comment: unclear risk of bias because completed case data used for qualify of life and pain outcomes and the rate of dropouts is not high
Selective reporting (reporting bias)	High risk	Comment: high risk of bias because costs data presented in results ‐ not mentioned in methods ‐ of the retracted paper of Wong 2012 but not presented in the published paper. Additionally, patient’s flow and outcome data presented in Wong (JEADV 2012, 26, 102–110) and the retracted paper (Wong JVS, 2012, 55,1376‐1385) do not match
Other bias	Low risk	Comment: the study appears to be free of other sources of bias

4LB: four‐layer bandage; ABPI: ankle:brachial pressure index; IQR: interquartile range; ITT: intention‐to‐treat; QOL: quality of life; SEM: standard error of the mean.