Table 3.

Values/Distributions Used for Parameters in the Economic Model and Study Designs That Would Provide Further Information on Them

Parameter	Description	Value/Distribution	Study Design to Provide Further Information
$d_2^M$	Effect of albumin on mortality	Posterior distribution generated by Bayesian evidence synthesis model	RCT including albumin and control arm
$d_3^M$	Effect of saline on mortality	Posterior distribution generated by Bayesian evidence synthesis model	RCT including saline and control arm
$d_4^M$	Effect of gelofusine on mortality	Posterior distribution generated by Bayesian evidence synthesis model	RCT including gelofusine and control arm
dS	Change to NS risk in “saved” patients	Posterior distribution generated by Bayesian evidence synthesis model	RCT including fluid resuscitation arm(s) and control arm
$C^F$	Fluid cost (per patient)	$1 (saline), $35 (albumin), $12.50 (gelofusine)
$C^H$	Additional in-patient costs associated with survival	$60 (based on 5 d at $12/d)
$\alpha$	Log-odds of death without fluid resuscitation	Normal with implied median probability of death 25%, 95% CI 15%–40%	Cohort study with short-term follow-up
$p^B$	Probability of NS without fluid resuscitation	Beta (1,9) (mean 10%, 95% CI 0.6%–28.5%)	Cohort study with 28-d follow-up
$p^L$	Probability that NS will still be present at 6 mo conditional on NS observed at 28 d	Beta (1,1)	Cohort study with 6-mo follow-up
$C^S$	Long-term discounted costs of NS	$20 000
${\text{q}}^M$	QALY loss per fatality	Normal with CHAR1 = 20 and s = 5	Cohort study on survivors without NS: long-term follow-up
${\text{q}}^{\text{S}}$	QALY loss per case of NS	Truncated normal with CHAR1 = 5, s = 3.16, lower limit = 0	Cohort study on those with NS: long-term follow-up

Note: CI = confidence interval; NS = neurological sequelae; QALY = quality-adjusted life-year; RCT = randomized controlled trial.