Table 4.

Regression models to be fitted

Model	Exposure	Outcome*	Adjustment(s)	Interpretation
1	AAFI	Life expectancy	Nil	Primary evaluation of austerity hypothesis.
2	Real per capita government expenditure	Life expectancy	Nil	Sensitivity analysis 1 using alternative austerity measure.
3	Public social spending	Life expectancy	Nil	Sensitivity analysis 2 using alternative austerity measure.
4	Cyclically adjusted primary balance	Life expectancy	Nil	Sensitivity analysis 3 using alternative austerity measure.
5–8	As per models 1–4	Mortality rates	Nil	Evaluation of austerity hypotheses across primary and alternative measures using mortality rate outcome.
9–12	As per models 1–4	Underemployment	Nil	Impact of austerity on underemployment.
13–16	As per models 1–4	Mean household income	Nil	Impact of austerity on mean household income.
17–20	As per models 1–4	Life expectancy	GDP per capita	Impact of austerity after accounting for recession, but noting the potential for austerity to cause recession.
21–24	As per models 1–4	Life expectancy	Underemployment	Estimate of the mediating role of underemployment.
25–28	As per models 1–4	Life expectancy	Mean household income	Estimate of the mediating role of household incomes.
29–32	As per models 1–4	Life expectancy	Nil	Sensitivity analyses changing lag time to 0 years.
33–36	As per models 1–4	Life expectancy	Nil	Sensitivity analyses changing lag time to 5 years.
37–40	As per models 1–4	Life expectancy	Nil	Sensitivity analyses limiting the impacts to 2 years after the austerity measure returns to baseline.

*Life expectancy will be calculated for the total population and separately for men and women. The mortality rates will be age standardised for the total population, separately for men and women, and for separate age strata.

AAFI, Alesina-Ardagna Fiscal Index; GDP, gross domestic product.