Beilin 2008.

Methods	RCT Approved by the ethics committee and informed consent obtained Setting: Israel Funding: charity
Participants	82 ASA 1 to 3 patients, aged 20 to 72 years, undergoing general, gynaecological, urological or orthopaedic surgery Exclusion criteria were not fluent in Hebrew, serious hearing or visual impairment precluding neuropsychological testing, absence of consent, history of head trauma, neurological disease, alcoholism, drug abuse and consumption of psychotropic drugs or antidepressants
Interventions	Treatment group: patient‐controlled lumbar (L2‐L4; catheter advanced 3 to 4 cm cephalad) epidural analgesia with 3 mL of 2% lidocaine as a test followed by 12 mL of 0.5% bupivacaine and 50 to 100 mcg of fentanyl 15 minutes before surgical incision, and bupivacaine 0.1% and fentanyl 2 mcg/mL for 24 hours (n = 30) Control group: IV patient‐controlled analgesia with morphine for 24 hours (n = 30) General anaesthesia for all participants
Outcomes	Pain: "Pain intensity was significantly greater in patients of the IV group throughout the observation period, both at rest and during coughing"
Notes	Data not extractable for abdominal surgery only. Study authors contacted 23 February 2016, but did not reply
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"randomly assigned", "by a schedule based on a random‐numbers table"
Allocation concealment (selection bias)	Unclear risk	"Preoperative evaluation was performed by an anaesthesiologist, 3 to 7 days before surgery", "Additional information, including age, weight, demographic status, and education, were documented for patients who consented to participate in the study", "randomized immediately prior to surgery", "by a schedule based on a random‐numbers table"
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Not mentioned
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not mentioned
Incomplete outcome data (attrition bias) All outcomes	High risk	High rate of drop‐outs: "A total of 22 patients dropped from the study because of failure to complete the second (postoperative) testing session for various reasons: A total of 6 patients did not feel well enough to cooperate; 5 were disconnected from the patient‐controlled pump early due to nausea and vomiting; and 11 had other reasons (fever, lack of optical glasses, noisy setting, refusal to participate in second testing, etc.)"
Selective reporting (reporting bias)	Low risk	All results reported
Other bias	Low risk	Groups well balanced: "The groups were similar in demographic characteristics, including body weight, age, male‐to‐female ratio, years of education, type and duration of surgery"