Demarre 2012.

Study characteristics
Methods	Study objective: to compare the effectiveness of an alternating low pressure air mattress with a standard single‐stage inflation and deflation cycle of the air cells with an alternating low pressure air mattress with multi‐stage inflation and deflation cycle of the air cells Study design: randomised controlled trial Study grouping: parallel group Duration of follow‐up: 14 days Number of arms: 2 Single centre or multi‐site: multi‐site Study start date and end date: December 2007 to January 2010 Setting: 25 wards in 5 hospitals in Belgium.
Participants	Baseline characteristics Inclusion criteria: patients at risk for pressure ulcer development according to the Braden scale (less than 17), including those with non‐blanchable erythema Exclusion criteria: patients with a pressure ulcer Grade II to IV on admission, the expected admission time < 3 days; aged < 18 years; with a "do not resuscitate code" specifying ending all therapeutic interventions, weight < 30 kg or > 160 kg, and informed consent not obtained Sex (M:F): overall 241 (39.4%): 369 (60.6%); 111:187 in multi‐stage group; 130: 182 in single‐stage group Age (years): overall mean 76.3 (SD 14.0); 76.15 (14.82) in multi‐stage alternating air mattress; 76.50 (13.20) in single‐stage alternating air mattress Baseline skin status: overall median Braden score 14.0 (interquartile range (IQR) 12.0 to 15.0); 14.0 (12 to 15) in multi‐stage; 14.0 (12 to 15) in single‐stage Group difference: no difference Total number of participants: 610 Unit of analysis: individuals Unit of randomisation (per patient): individuals
Interventions	Intervention characteristics Multi‐stage alternating air mattress Description of interventions: alternating air mattress with the multi‐stage inflation and deflation of air cells (Hill‐Rom ClinActiv). Three air cells with a continuous low pressure on head zone. Seven cells with a continuous ultra low pressure on heel zone. Ten alternating low pressure cells on back and sacrum zone. Ten‐ to twelve‐minute cycle times for inflation and deflation and the air cells width 10 centimetres NPUAP S3I classification: powered, alternating pressure (active) air surface Co‐interventions: not described Number of participants randomised: n = 298 Number of participants analysed: n = 298 (intention‐to‐treat (ITT) analysis) Single‐stage alternating air mattress Description of interventions: standard alternating air mattress (Hill‐Rom Alto mattress), an alternating air mattress with a standard single‐stage, steep inflation and deflation of the air cells. All air cells were alternating, the cycle time was 10 minutes and the air cell width was 10 centimetres. NPUAP S3I classification: powered, alternating pressure (active) air surface Co‐interventions: not described Number of participants randomised: n = 312 Number of participants analysed: n = 312 (ITT analysis)
Outcomes	Proportion of participants developing a new pressure ulcer Outcome type: binary Time points: 14 days Reporting: fully reported Measurement method (e.g. scale, self‐reporting): assessed by the ward nurses Definition (including ulcer stage): percentage of patients developing a new pressure ulcer of grade 2 to 4 on any location, graded by EPUAP 1999 classification system (Grade I = non‐blanchable erythema; Grade II = an abrasion or a blister; Grade III = superficial ulcer; Grade IV = a deep ulcer) Dropouts: ITT analysis Notes (e.g. other results reported): 17 of 298 individuals (including 13 Grade II, 4 Grade III, 0 Grade IV) in multi‐stage group; 18 of 312 individuals (including 11 Grade II, 2 Grade III, 5 Grade IV) in single‐stage group. Extra data: 51 with new Grade I in multi‐stage; 38 with new Grade I in single‐stage. Ulcers by sites reported but not extracted. Time to pressure ulcer development Outcome type: Time‐to‐event Time points: not relevant Reporting: fully reported Measurement method (e.g. scale, self‐reporting): see above Definition (including ulcer stage): time to develop a pressure ulcer Grade II ‐ IV Dropouts: ITT analysis Notes: median time 5.0 days (IQR 3.0 to 8.5) in multi‐stage group; 8.0 (3.0 to 8.8) in single‐stage group (Mann‐Whitney U‐test = 113, P = 0.182); Kaplan Meier plot reported (log‐rank Chi2 = 0.013, df = 1, P = 0.911); HR 0.96 (95% CI 0.50 to 1.87) estimated by the review authors using the methods in Tierney 2007. Support‐surface‐associated patient comfort Outcome type: binary Time points: 14 days Reporting: fully reported Measurement method (e.g. scale, self‐reporting): probably self‐reported Definition: patient acceptability assessed directly by the number of participants withdrawing their consent to participate during observation period Drop outs: ITT analysis Notes: presented as exclusion reasons. Eleven of 298 individuals withdrawing due to discomfort in multi‐stage group and 0 exclusion due to consent; 16 of 312 due to discomfort and 1 due to consent in single‐stage. All reported adverse events using allocated support surfaces Reporting: not reported Health‐related quality of life (HRQOL) Reporting: not reported Cost‐effectiveness Reporting: not reported
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Included patients were randomly assigned to the study groups using simple randomisation. The random allocation sequence was based on a computer‐generated list of random numbers" Comment: low risk of bias because of the use of a proper random sequence generation method.
Allocation concealment (selection bias)	Low risk	Quote: "Patients were enrolled by the ward nurses ... assigned to one of the mattresses by contacting the researcher. The ward nurse received a number of the type of allocated mattress" Comment: low risk of bias because it is likely that allocation sequence was concealed.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Outcome group: All outcomes Quote: "Both mattresses were covered with an identical mattress cover ..." Quote: "The study could not be blinded, because of the visible differences of the external control unit of the study mattresses" Comment: high risk of bias because it is unlikely that participants were blinded.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Outcome group: primary outcome Quote: "Daily skin assessment was performed by the ward nurses ... in each patient, in the morning" Quote: "No information was provided to the ward nurses about the differences between the experimental and control study device" Quote: "The lack of a blinded outcome assessment is a first limitation ... the nurses were not informed about the differences in the mattresses in order to minimize the effect of non‐blinding" Comment: unclear risk of bias because efforts were made to reduce bias. Outcome group: comfort outcome Comment: high risk of bias because it is unlikely that patients who reported this outcome were blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Outcome group: all outcomes Comment: low risk of bias because ITT analysis was undertaken.
Selective reporting (reporting bias)	Low risk	Comment: the study protocol is not available but it is clear that the published reports include all expected outcomes, including those that were pre‐specified.
Other bias	Low risk	Comment: the study appears to be free of other sources of bias.