Table 3.

Summary of financial indicators’ rate of change of country groups

Indicator (US$)	Year range	Estimated intercept coefficients		Estimated slope coefficients and p values					Trend comparison p values
		LIC+ α00	LIC− α00+α01	LIC+ α10	LIC− α10+α11	χ2	KR	PB	χ2	KR	PB
Government health spending per capita*	2000–2016	6.06 (1.81)	9.14 (2.22)	0.30 (0.05)	−0.16 (0.06)	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001
DAH per birth on newborn and child health vaccines*	2000–2017	5.32 (3.20)	1.90 (3.92)	1.75 (0.19)	1.59 (0.24)	0.50	0.51	0.50	0.28	0.31	0.30
Total spending per birth on routine immunisation vaccines†	2006–2017	14.17 (2.97)	6.30 (3.79)	1.19 (0.32)	1.34 (0.42)	0.72	0.72	0.72	0.057	0.063	0.068
Government spending per birth on routine immunisation vaccines†		2.32 (0.55)	1.46 (0.68)	0.22 (0.04)	0.10 (0.05)	0.0090	0.0093	0.0087	0.0021	0.0038	0.0033

Each financial indicator was fitted by a linear mixed-effects regression model. The table shows the intercept and slope coefficients of LIC+ (α₀₀ and α₁₀) and LIC− (α₀₀+α₀₁ and α₁₀+α₁₁) with their SE in parentheses. χ², KR and PB represent the p values of an asymptotic χ² test, a Kenward-Roger approximation for F tests for reduction of mean structure and a parametric bootstrap method (10 000 simulations), respectively.²⁵ There are two types of p values presented: the first corresponds to the significance of the slope coefficient α₁₁, that is, if the yearly rate of change of LIC+ is significantly different from the yearly rate of change of LIC−. The second type of p values compare the overall trends of LIC+ and LIC−, that is, it does not refer to the significance of a specific coefficient but to the significance of intercept and slope combined. P values below 0.05 are highlighted in bold.

*Constant 2018 US$.

†Constant 2010 US$.

DAH, development assistance for health; LIC, low-income countries.