Table 8.

Morbidity, care-seeking and management among infants in the neonatal period at baseline (December 2016 to February 2017) and endline surveys (December 2018 to February 2019) in intervention and comparison areas

	Baseline household survey		Endline household survey		Difference-in-differences*	OR (95% CI)	Adjusted OR (95% CI)†
	Intervention	Comparison	Intervention	Comparison
	% (95% CI) (n)	% (95% CI) (n)	% (95% CI) (n)	% (95% CI) (n)	%
Among all
Baby ill in the first month of life	14 (11 to 19) (n=331)	5 (3 to 9) (n=331)	9 (7 to 13) (n=482)	8 (6 to 11) (n=397)	n/a‡	n/a‡	n/a‡
Among ill
Prevalence of possible neonatal sepsis§	83 (70 to 91) (n=47)	78 (48 to 93) (n=18)	73 (59 to 84) (n=45)	84 (65 to 94) (n=32)	n/a¶	n/a¶	n/a¶
Care-seeking for ill baby**	74 (59 to 86) (n=47)	44 (23 to 69) (n=18)	51†† (34 to 68) (n=43)	68‡‡ (49 to 82) (n=31)	−47	0.04 (0.00 to 0.49)	0.04 (0.00 to 0.60)
Among possible sepsis
Any antibiotic	51 (34 to 68) (n=39)	36 (17 to 60) (n=14)	57 (38 to 73) (n=30)	68 (43 to 85) (n=25)	−26	0.17 (0.01 to 2.06)	0.19 (0.02 to 2.42)

*Difference-in-differences: the difference in the proportion between intervention and comparison areas at endline subtracted from the difference in proportion between intervention and comparison at baseline.

†Adjusted for age of infant.

‡The difference-in-differences, OR and adjusted OR were not calculated for baby ill in the first month of life as this was not an outcome assumed to be influenced by the intervention.

§Possible neonatal sepsis: fever, unable to suckle/feed difficult/fast breathing, severe chest in-drawing, convulsions or lethargy.

¶The difference-in-difference, OR and adjusted OR were not calculated for prevalence of possible neonatal sepsis as this was not an outcome assumed to be influenced by the intervention.

**Care-seeking outside the home.

††Missing data for two individuals.

‡‡Missing data for one individual.