De Conno 1987.

Methods	Design: Randomised cross‐over trial Year: Not reported Country: Italy
Participants	Patients: 120 patients were randomised and 29 did not complete treatment (10/29 patients suspended both treatments, 16/29 patients suspended pentazocine and 3/29 suspended buprenorphine): No patient characteristics are reported. Inclusion criteria: Patients with cancer pain of moderate to severe intensity aged > 18 years who have given informed consent. Exclusion criteria: Severe renal or hepatic impairment, severe respiratory failure, chronic treatment with high doses of agonist analgesics, increase in intracranial pressure, pregnancy.
Interventions	Buprenorphine: Drug: Buprenorphine SL tablets 0.2 mg. Dose/dosing: 1 to 2 tablets of 0.2 mg administered every 6 to 8 hours depending on need. Formulation: SL. Route of administration: SL. Length of treatment: 7 days. Titration schedule: No information reported. Rescue medication: No information provided. Other medication: No information provided. Pentazocine: Drug: Tablets 50 mg. Dose/dosing: 1 to 2 tablets of 50 mg administered every 6 to 8 hours depending on need. Formulation: Oral. Route of administration: Oral. Length of treatment: 7 days. Titration schedule: No information reported. Rescue medication: No information provided. Other medication: No information provided. For cross‐over trials: Cross‐over schedule: There was no wash‐out period either at the beginning of the study or between treatments. The first 7‐day stage was followed immediately by the second 7‐day stage.
Outcomes	Pain intensity: Assessed by patients and their relatives daily on a 5‐point categorical scale: 1 = slight pain, 2.5 = moderate/troublesome pain, 5 = severe/exhausting pain, 7.5= terrible pain, 10 = excruciating/killing pain. Number of hours slept, number of hours pain free, and number of hours assessed as any of the pain categories above. The total number of hours must add up to 24 hours. Number of hours spent standing, sitting or lying. Side effects: Nausea, vomiting, drowsiness, agitation vertigo, tremors, dry mouth, sweating, itching , allergy and headache (all recorded daily). 16 patients of the whole sample who completed both treatments data on the quality of life were also collected through: 1) Karnofsky performance status; 2) hours of work activities; 3) hours of evasion; 4) hours of inactivity. These data are not reported as it is unclear how these patients were selected and what their characteristics are.
Notes	Study free of commercial funding? Not clear, the information reported for the author affiliations is ambiguous and may include the manufacturer of buprenorphine. Groups comparable at baseline? Unclear, no information reported. ITT analyses undertaken? It does not appear that the analyses were conducted as ITT. This study is published in two papers, the earlier paper is in English and includes a smaller sample than the later paper, which is published in Italian and lay‐translated by MSH. The data from this newer paper which reports the larger sample have been included.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information reported.
Allocation concealment (selection bias)	Unclear risk	No information reported.
Blinding of participants and personnel (performance bias) Pain	Unclear risk	No information reported.
Blinding of participants and personnel (performance bias) Adverse events	Unclear risk	No information reported.
Blinding of outcome assessment (detection bias) Pain	Unclear risk	No information reported.
Blinding of outcome assessment (detection bias) Adverse events	Unclear risk	No information reported.
Incomplete outcome data (attrition bias) Pain	High risk	Data only available for 91/120 randomised patients.
Incomplete outcome data (attrition bias) Adverse events	High risk	See cell above.
Selective reporting (reporting bias)	Unclear risk	The outcomes are not well‐reported.
Other bias	Unclear risk	It is unclear whether the study is at high risk of other biases.
Were the patients adequately titrated?	High risk	The patients were not titrated at the beginning of the actual study.
For cross‐over trials: Are data available for both time periods?	Low risk	Yes, for a number of participants.