Page 2017.

Methods	RCT comparing simvastatin vs placebo on duration of delirium coma Study took place in a single ICU (mixed medical and surgical) in the UK
Participants	Participants included 142 (N = 71 simvastatin, mean age 61.9 ± 15.3 years, 45/71 (63%) male; N = 71 placebo, mean age 62.1 ± 17.3 years, 37/71 (52%) male) critically ill patients requiring mechanical ventilation within 72 hours of ICU admission Study enrolment between February 2013 and January 2015
Interventions	Participants received 80 mg simvastatin or placebo within 72 hours of admission to the ICU, irrespective of the presence of coma or delirium. Study drug was given daily, orally or by feeding tube. Treatment was discontinued at ICU discharge, after a maximum of 28 days, at death, with creatine kinase concentrations > 10 times the upper limit of normal, with alanine transaminase concentrations > 8 times the upper limit of normal, with development of a clinical condition requiring immediate treatment with statins, upon discontinuation of active medical treatment, with request for discontinuation by patient or legal representative, or upon request for discontinuation by attending clinician or contraindication to enteral drug administration. Patients were kept on fentanyl and propofol infusions, titrated to a RASS of 0 to ‐1, unless deeper sedation was required Delirium status was assessed via the CAM‐ICU twice during each 12‐hour shift, with a minimum of 4 hours between 2 consecutive assessments. Delirium was defined if RASS was ‐2 to +4 and CAM‐ICU was positive
Outcomes	Primary Number of delirium‐free and coma‐free days, defined as the number of days in the first 14 days after randomization during which the patient was alive without delirium and not in coma Secondary Delirium‐free and coma‐free days to day 28 Ventilator‐free days to day 28 Mortality at 6 months Length of ICU and hospital stay Safety with regard to elevated creatine kinase and alanine transaminase concentrations Serious adverse events
Notes	Funding The National Institute for Health Research provided funding Sponsored by The study was sponsored by West Hertfordshire Hospitals National Health Service Trust Co‐ordinated by The study was co‐ordinated by the Northern Ireland Clinical Trials Unit Registration The trial was registered with the International Standard Randomised Controlled Trial Registry under number ISRCTN89079989 Study authors were contacted for clarification, and requested information was provided Erratum in Corrections (Lancet Respir Med 2018)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The study statistician generated the randomization schedule in advance using nQuery Advisor version 4.0; randomization was done by variable block sizes of 2, 4, 6, and 8, without stratification
Allocation concealment (selection bias)	Low risk	No details were provided pertaining to the individual responsible for randomizing participants. Patient drug packs were prepared by Victoria Pharmaceuticals (Belfast, Northern Ireland) according to the pre‐arranged randomization schedule and were distributed to the hospital pharmacy, which stored the packs in a secure area and dispensed them to the ICU as required. Each pack was numbered with a unique patient trial identifier that had been allocated to each participant at the time of random assignment to a group
Blinding of participants and personnel (performance bias) All outcomes	Low risk	This was a double‐blind study. Simvastatin or placebo tablets were packaged in white opaque high‐density polyethylene plastic containers sealed with a tamper‐evident seal. Placebo and simvastatin tablets were indistinguishable when crushed and dispersed in water for enteral administration
Blinding of outcome assessment (detection bias) All outcomes	Low risk	All ICU clinical and research staff, legal representatives, and participants were masked to study drug. The data monitoring and safety committee reviewed blinded data reports. Statisticians were not masked to allocation
Incomplete outcome data (attrition bias) All outcomes	Low risk	Data were analysed according to an intention‐to‐treat principle. All randomized participants were included in the primary analysis
Selective reporting (reporting bias)	Low risk	All outcomes reported in methods were included in results. Data presented for all participants were included in the analysis
Other bias	Low risk	The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report