ESGCMCRF 2 1990.

Study characteristics
Methods	Study design: parallel RCT (554 assessed for eligibility; 336 randomised); stratified before randomisation into two groups GFR < 20 mL/min/1.73 m2; run‐in for three months to demonstrate declining GFR Study duration: date of first recruitment not stated. Last follow up September 1991 Study follow‐up period: about 50 months
Participants	Countries: Italy, Germany, Sweden Setting: multicentre (number of sites not reported) Patients with GFR < 20 to 60 mL/min/1.73 m2; run‐in for 3 months Number: very low protein diet group (99); low protein diet group (103) Mean age: 50 years Sex (M:F ratio): 1.4:1 Exclusion criteria: stable GFR during run‐in period
Interventions	Very low protein diet group Prescribed protein intake: 0.3 g/kg/day protein + amino acid/keto acid supplement Median calculated protein intake: 0.35 g/kg/day during 1 year Low protein diet group Prescribed protein intake: 0.6 g/kg/day protein Median calculated protein intake: 0.56 g/kg/day during 1 year Co‐interventions All groups received 35 Kcal/kg/day
Outcomes	ESKD measured by number commencing dialysis
Notes	Funded by grant from Bundesministerium fȕr Forschung und Technologie Primary endpoint of reports was dietary compliance No report of final study results identified Fewer than 50 of 554 assessed for eligibility had diabetic kidney disease Information on sequence generation and allocation concealment kindly provided by Professor Norbert Gretz
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Block randomisation scheme for each centre
Allocation concealment (selection bias)	Low risk	Sealed and numbered envelopes
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not blinded and lack of blinding may influence patient management
Blinding of outcome assessment (detection bias): End or change in GFR End of change in GFR	Unclear risk	No information provided. Outcome not reported in this study
Blinding of outcome assessment (detection bias): Need to start dialysis Need to start dialysis	Unclear risk	Unclear how end point of the onset of ESKD and the need for dialysis was determined
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Not clear whether data on all participants were reported
Selective reporting (reporting bias)	High risk	No report of death or GFR; medians of BMI only available
Other bias	Low risk	Grant from Bundesministerium für Forschung und Technologie, FRG (No. 0704743)