Table 3. Key parameters and possible estimates and default assumptions.

Key parameters		Possible estimates and default assumptions
Health opportunity costs of health care expenditure in each period (t)	kht	Estimates reported in Ochalek et al. (2018) could provide initial default estimates of cost per DALY averted Projections of these estimates based on estimates of health expenditure and consumption growth are possible These initial estimates can be refined and updated as other country specific estimates emerge
Consumption opportunity costs of health care expenditure in each period (t)	kct	A conservative default assumption of 1 (1 dollar spent on health care delivers 1 dollar in net production) A conservative default assumption that the real value of the net production effects will grow at the same rate as consumption
Consumption value of health in each period (t)	υht	Country specific estimates of Vht, see Robinson et al. (2019) and Robinson and Hammitt (2018) Growth in Vht can be based on growth in consumption (gc) and the income elasticity of demand for health A conservative scenario using an income elasticity of demand for health of 1 (Vht will grow at gc) Scenario using an income elasticity of demand of 1.5
Opportunity costs for other sectors (x) in each period (t) Social rate of time preference for consumption	υxt/kxt	Default assumption that vht/kht = υxt/kxt when considering impacts on other public sectors
	rc	A normative assumption of zero pure time preference for social choices is not unreasonable (δ = 0) Two scenarios based on alternative assumptions of inequality aversion can be used A conservative scenario based on η = 1 (rc = gc) Alternative scenario based on η = 2 (rc = 2gc) Other scenarios can be based on evidence of why η is likely to differ in specific contexts or different judgements about gc
Catastrophic risk		If catastrophic risk is included it should be based an estimate of the probability of truly catastrophic events where recovery would not be possible (≤0.1%)
Macroeconomic risk		Use of rc without adjustment for macroeconomic risk may be a reasonable for projects with shorter time horizons (⩽30 years) For longer time horizons or where macroeconomic risk is greater, declining rates may be required but should be based on uncertainty in consumption growth
Project specific risk		Where possible project specific risks should be included in how the time streams of consumption equivalent effects of the project are estimated A qualitative indication of whether projects are likely to be strongly pro or counter cyclical should be provided Further research is required on how the interaction of project specific and macroeconomic risk might be best quantified for the types of project relevant to LMICs