Cox 1990.

Methods	Location of trial: US. Number of centres: multicentre (number unclear). Funding: National Cancer Institute, National Institute for Health (grants 21661, 32115, 12258, 13457, 20235, 21439, 29565, 12262). Trial ID: RTOG 83‐13.
Participants	Inclusion criteria: squamous cell carcinoma of the upper aerodigestive tract, Stages III & IV considered inoperable, with no prior resection or radiotherapy. Adequate bone marrow and renal function. T1‐T4 included. Exclusion criteria: history of previous malignant tumour. Prior chemotherapy within 6 weeks of randomisation. Recruitment period: April 1983 to February 1986 (Scheme A). OC: 47/237 (20%). OP: 104/237 (44%). OC+OP: 151/237 (64%). Number randomised: 260. Number analysed: 237.
Interventions	Different doses of hyperfractionated radiotherapy Gr A (n = 63): 67.2 Gy. Gr B (n = 58): 72.0 Gy. Gr C (n = 116): 76.8 Gy. All fractions were 1.2 Gy given twice daily 5 days per week. Interval between fractions was permitted to be 4‐8 hours. Radiotherapy was administered with photons of 1.25 MV or greater with minimum source axis distance of 80 cm.
Outcomes	Primary: locoregional control. Secondary: overall survival, toxicity, late effects. Duration of follow‐up: minimum 2 years.
Notes	Trial also randomised patients to 81.6 Gy or 72.0 Gy between February 1986 to November 1987 (Scheme B). However, the trial focuses on data from Scheme A. HR data taken from Kaplan‐Meier graphs (no numbers at risk).
Risk of bias
Bias	Authors' judgement	Support for judgement
Adequate sequence generation?	Unclear risk	"Randomized to 1 of 3 total doses". No details of method of sequence generation given.
Allocation concealment?	Unclear risk	Insufficient information.
Blinding ‐ Outcome Assessors	High risk	Not mentioned.
Incomplete outcome data addressed?	Unclear risk	9% of patients randomised are excluded from analysis, but reasons and group allocation not described.
Free of selective reporting?	Unclear risk	Multiple publications addressing different outcomes and exposures.
Free of other bias?	Unclear risk	Possible contamination due to some patients having had prior chemotherapy.