Alam 1999.

Methods	Randomized controlled trial. Duration:1 year 10 months, from July 1995 to May 1997.
Participants	Number of participants: 300 randomized (168 men; 131 women). Inclusion criteria: adult men and women aged 15 to 55 years; history of acute watery diarrhoea for < 24 hours before admission; severe dehydration; stool positive for Vibrio cholerae under dark‐field illumination; successful rehydration with intravenous infusion within 6 hours of admission. Exclusion criteria: suspected pregnancy; bloody diarrhoea; systemic infection requiring intravenous antibiotics; inability to rehydrate with intravenous infusion within 6 hours after admission.
Interventions	(1) ORS ≤ 270 (glucose‐based). (2) ORS ≥ 310 (glucose‐based). See Additional Table 2 for the ORS compositions. Co‐interventions: erythromycin (500 mg orally every 6 hours for 3 days). Food: bread, banana (immediately after rehydration), and standard meals (rice, fish, or meat, vegetables and lentils) 3 times daily. Water: given as desired, usually with meals.
Outcomes	(1) Need for unscheduled intravenous infusion.* (2) Duration of diarrhoea (after randomization, hours).* (3) Vomiting during rehydration during initial 24 hours.* (4) Stool weight (g/kg bodyweight) in first 24 hours after admission/randomization.* (5) Total stool weight, g/kg body weight. (6) Urine volume during initial 24 hours (ml/kg bodyweight). (7) Total urine volume (ml/kg body weight). (8) Initial 24 hours ORS intake (ml/kg body weight). (9) Total ORS intake (ml/kg body weight). (10) Initial 24 hours water intake (ml/kg body weight). (11) Total water intake (ml/kg body weight). (12) Biochemical hyponatraemia (< 130 mmol/L) 24 hours after admission.* (13) Biochemical hyponatraemia (< 125 mmol/L) 24 hours after admission.* (14) Biochemical hyponatraemia (< 120 mmol/L) 24 hours after admission.*
Notes	Location: Bangladesh. (Trial started at two sites, in Bangladesh and Indonesia, but it was discontinued at the latter due to inadequate participant supervision. No data from Indonesia included in the analysis.) Date: July 1995 to May 1997.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"The randomization list ... was prepared ... by use of permuted blocks of variable length."
Allocation concealment (selection bias)	Unclear risk	Not described.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	"double‐blind" but method not described.
Incomplete outcome data (attrition bias) All outcomes	Low risk	"Data collected on patients withdrawn from the study were included in the analysis up to the time of withdrawal." Loss to follow‐up: Reduced 13/147 (8.8%) and Higher 16/153 (10.5%).
Selective reporting (reporting bias)	Unclear risk	Protocol not available.
Other bias	Low risk	The study appears to be free of other bias.