Table 4.

Significant determinants for fatal foetal outcomes from multivariable regression analysis (adjusted for age, socioeconomic status and residency of the mother)

Explanatory variablea	aORb	95% CI
Residency (rural)	2.87	1.31, 6.29
Socioeconomic status (poorest)	1.79	1.02, 3.14
Earlier miscarriage/stillbirth	23.19	14.71, 36.55
Delivery (by caesarean section)	13.03	4.24, 40.08
Delivery (instrumental)	5.05	1.50, 16.96
Gestational age (< 37 weeks)	4.45	2.82, 7.01
Delivery cost (chargeable)	0.39	0.25, 0.74
Absence of two doses of tetanus vaccination	2.59	1.56, 4.30
Absence of iron/folic acid supplementation	3.15	1.71, 5.80
Malaria (reported or confirmed)	1.94	1.21, 3.11

Multivariable logistic regression models using the penalized maximum likelihood estimation (Firth method) [27], to account for rare events, and a stepwise backward elimination approach were utilised to identify explanatory variables, which most significantly influence the foetal outcome. Initial models included (i) sociodemographic (i.e. age, socioeconomic status, and residency of the mother); (ii) birth circumstances and antenatal care; and (iii) pregnancy-related morbidity and concomitant health condition variables

Remaining explanatories were included at a significance level of p < 0.2

^aReference categories for explanatory variables: socioeconomic status, least poor; residency, urban; earlier miscarriage, none; delivery, normal; gestational age, ≥37 weeks; delivery cost, free of charge; two doses of tetanus vaccination, received; iron/folic acid supplementation, received; malaria (reported or confirmed), none

^bAdjusted odds ratios