Table 3.

Summary regression results for age trends for exclusive breastfeeding

Stratum name	Stratum characteristics	Intercept (95% confidence interval)	Slope (95% confidence interval)	Standard error of the residual
AfrD	Africa: high adult, high child mortality	-0.98 (-1.33, -0.63)	-0.21 (-0.29, -0.14)	1.66
AfrE	Africa: very high adult, high child mortality	0.68 (0.29, 1.07)	-0.38 (-0.47, -0.29)	1.79
AmrB	Americas: low child, low adult mortality	0.23 (-0.20, 0.66)	-0.39 (-0.50, -0.29)	1.14
AmrD	Americas: high adult, high child mortality	0.91 (0.47, 1.34)	-0.39 (-0.49, -0.28)	1.10
EmrB	Eastern Mediterranean: low child, low adult mortality	-0.84 (-1.75, 0.06)	-0.47 (-0.80, -0.14)	0.57
EmrD	Eastern Mediterranean: high adult, high child mortality	0.26 (-0.25, 0.76)	-0.29 (-0.39, -0.18)	0.90
SearB	South-east Asia: low child, low adult mortality	1.01 (0.55, 1.48)	-0.47 (-0.58, -0.35)	0.92
SearD	South-east Asia: high adult, high child mortality	1.03 (0.83, 1.23)	-0.38 (-0.42, -0.33)	0.63
WprB	Western Pacific: low child, low adult mortality	0.58 (-0.15, 1.31)	-0.45 (-0.68, -0.22)	1.49

Regressions use minimum chi-square estimation with dependent variable logarithm of prevalence odds (that is, Log(p/1 - p), where p is prevalence of exclusive breastfeeding), and independent variable oldest month of age by two-month age groups; regressions for category-A countries stratified by geographical and epidemiologic criteria as noted in the table; exclusive breastfeeding estimated for five age groups 0–1 to 8–9 months of age. Data sources listed in the Annex (Additional file 1).