Sykes 1991a.

Methods	RCT, single‐centre, cross‐over group design
Participants	51 people with cancer in UK hospice who had not under gone bowel diversion, were not clinically obstructed and who required a laxative. Participants were receiving either > 80 or < 80 mg of strong opioid a day
Interventions	Drug 1: senna plus lactulose (in equal quantities) liquid Drug 2: equivalent volume of co‐danthramer plus poloxamer Starting doses of laxatives were set by the protocol in relation to opioid dosage and subsequently modulated according to clinical response. No further details on doses Duration of treatment: 1 week twice daily Cross‐over: switched to the alternative for 1 further week
Outcomes	Stool form and frequency, failure (absence of a single stool passed spontaneously during 1 treatment week), use of rescue laxatives, participant's assessment of bowel function, participant preference and adverse events
Notes	No details provided
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No details provided
Allocation concealment (selection bias)	Unclear risk	No details provided
Blinding (performance bias and detection bias) All outcomes	High risk	"Not possible because of physical characteristics of drugs"
Incomplete outcome data (attrition bias) All outcomes	High risk	58/117 participants completed the cross‐over (of the 58, 6 participants were excluded from analysis because of breaches in the protocol and 1 as "data unclear") 0 dropped out because of inefficacy Not reported if intention‐to‐treat analysis
Selective reporting (reporting bias)	Unclear risk	No details provided