Interrelationships between early antenatal care, health facility delivery and early postnatal care among women in Uganda: a structural equation analysis

Figures & data

Table 1. Operationalization and measurement of variables used in the study

Code	Variable	Description	Coding if any	Data type
Y1	Early Antenatal care	Timing of the first antenatal visit	1. First ANC visit within first trimester. 2. First ANC visit after first trimester	Binary
Y2	Health facility delivery	Supervised delivery at the health facility or institutional delivery	1. Delivered from a health facility 2. Did not deliver from a health facility	Binary
Y3	Early Postnatal care	Timing of first postnatal checkup	1. First PNC within 48 hours after birth 2. First PNC after 48 hours	Binary
X1	Age of the women	Age of the women at the time of the survey	1. 15–18 2. 19–35 3. 36–49	ordinal
X2	Parity	Number of children the woman has already had (CEB)	1. 1–3 2. 4–6 3. 7+	ordinal
X3	Highest maternal education level	Mother’s highest level of education	1. some primary 2. completed primary seven 3. some secondary 4. completed secondary six	ordinal
X4	Wealth	Income level of the household	1. Poor 2. middle 3. Rich	ordinal
X5	Marital status	Marital status of the woman	1. Unmarried 2. Married	nominal
X6	Pregnancy wanted	If the mother wanted the last pregnancy	1. Yes 2. No	nominal
X7	Exposure to mass media	Women who listen to radio, read newspapers or watch television	1. Exposure 2. Non exposure	nominal
X8	Pregnancy complications	If the pregnancy had complications or not	1. Yes 2. No	nominal
X9	Community factor	Availability of community health worker	1. Yes 2. No	nominal
X10	Distance to the health facility	If the distance from home to the health facility is a big problem or not.	1. Big problem 2. not big problem	nominal
X11	Direct costs/fees	If cost paid while accessing a service is a problem or not	1. big problem 2. not	nominal

Table 2. Weighted percentage distribution of women by background characteristics

Characteristic	Category	Number (N = 10,152)	Percent (%)
Age	15–19	2347	23.1
	20–34	5154	50.8
	35–49	2651	26.1
Highest Maternal Education level	Some primary	4406	43.4
	Completed primary seven	1827	18
	Some secondary	3198	31.5
	Completed secondary six	721	7.1
Marital status	Married	3189	31.4
	Unmarried	6963	68.6
Wealth	Poor	4128	40.7
	Middle	1912	18.8
	Rich	4112	40.5
Parity	1–3 4–6 7+	2231 3664 4257	22 36.1 41.9
Distance to health facility	Big problem	3957	38.9
	Not big problem	6195	60.1
Cost of service	Big problem	4763	47
	Not big problem	5389	53
Community health worker	Readily available	7258	71.5
	Not readily available	2894	28.5
Media exposure	Exposed	8110	80
	Not exposed	2042	20
Pregnancy wanted	Yes	6185	60.9
	No	3967	39.1
Pregnancy complications	Yes	660	6.5
	No	9492	93.5

Table 3. Weighted percentage distribution of women by endogenous variables

Factor	Category	Number	Percent (%)
Early Antenatal care n = 10,152	Within first trimester	2897	28.5
	After first trimester	7255	71.5
Health facility delivery n = 10,152	Delivered from the health facility	7512	74
	Did not deliver from the health facility	2640	26
Early postnatal care n = 5,901	Within 48 hours after childbirth	3186	54
	After 48 hours after childbirth	2715	46

Table 4. Logit and GSEM models predicting the Odds Ratios of women utilizing early ANC

Characteristic	Category	unadjusted OR (95% CI)	Adjusted OR (95% CI)
Age	15–19 (r)	0.89 (0.81–0.94)*
	20–34	1.31 (1.12–1.42)*	0.92 (0.79–0.97)**
	35–49	0.97 (0.88–1.05)	1.18 (1.10–1.35)**
Highest Maternal Education level	Some primary (r)	1.32 (0.92–1.40)
	Completed primary seven	1.69 (1.45–1.84)*	1.68 (1.56–1.84)**
	Some secondary	0.70 (0.64–0.96)*	0.90 (0.80–1.10)
	Completed secondary six	1.13 (0.93–1.05)	1.17 (0.94–1.35)
Marital status	Unmarried (r)	0.85 (0.75–0.94)*
	Married	1.24 (1.19–1.43)*	1.12 (0.94–1.26)**
Wealth	Poor (r)	0.90 (0.87–1.1)	-
	Middle	0.86 (0.71–1.02)	-
	Rich	0.88 (0.78–1.05)	-
Distance to health facility	Big problem (r)	0.87 (0.89–0.95)*
	Not big problem	1.51 (1.23–1.65)*	1.35 (1.23–1.73)**
Cost of service	Big problem (r)	0.61 (0.47–0.82)*
	Not big problem	1.13 (1.10–1.32)*	1.85 (1.31–2.12)**
Community health worker	Not readily available (r)	1.05 (1.03–1.19)*
	Readily available	1.13 (1.08–1.35)*	1.06 (1.04–1.17)**
Media exposure	Not exposed (r)	0.9 (0.75–1.13)	-
	Exposed	1.08 (0.98–1.31)	-
Pregnancy wanted	No (r)	0.82 (0.76–0.98)*
	Yes	1.17 (1.05–1.38)*	1.15 (1.07–1.36)**
Pregnancy complications	No (r)	0.65 (0.45–0.87)*
	Yes	1.95 (1.73–2.45)*	2.04 (1.85–2.26)**
Parity	1–3(r)	0.76 (0.65–0.86)*	0.71 (0.62–0.80)**
	4–6	0.86 (0.78–1.1)	0.81 (0.72–1.00)
	7+	1.2 (0.92–1.40)	1.03 (0.96–1.23)

(OR): odds ratios; CI: confidence intervals; * indicates variables with p < 0.1, ** indicates significant effects with p < 0.05, and (r) indicates the reference category and (-) indicates variables not considered at multivariable level

Table 5. Logit and GSEM model predicting the Odds Ratios of women utilizing a health facility

Characteristic	Category	unadjusted OR (95% CI)	Adjusted OR (95% CI)
Early ANC	Went after the first trimester	1.10 (1.05–1.17)*
	Went within first trimester	1.08 (1.03–1.15)*	1.04 (1.01–1.14)**
Age	15–19 (r)	1.23 (1.10–1.40)*
	20–34	0.87 (0.78–0.94)*	0.83 (0.79–0.97)**
	35–49	1.57 (1.32–2.10)*	1.52 (1.32–2.05)**
Highest Maternal Education level	Some primary (r)	1.52 (1.31–1.76)*
	Completed primary seven	2.47 (2.29–2.68)*	1.47 (1.29–1.68)**
	Some secondary	2.20 (1.87–2.58)*	2.20 (1.87–2.58)**
	Completed secondary six	3.45 (2.64–4.45)*	3.58 (2.68–4.44)**
Marital status	Unmarried (r)	1.17 (0.95–1.32)
	Married	1.12 (0.92–1.26)	-
Wealth	Poor (r)	1.01 (0.89–1.50)
	Middle	1.12 (0.99–1.26)	1.21 (0.89–1.34)
	Rich	1.67 (1.50–1.84)*	1.15 (0.78–1.29)
Distance to health facility	Big problem (r)	0.81 (0.75–1.01)
	Not big problem	1.23 (1.00–1.34)	-
Cost of service	Big problem (r)	0.96 (0.81–1.07)
	Not big problem	0.87 (0.88–1.12)	-
Community health worker	Not readily available (r)	1.08 (1.04–1.24)*
	Readily available	2.01 (1.94–2.56)*	2.05 (1.87–2.47)**
Media exposure	Not exposed (r)	0.90 (0.81–0.97)*
	Exposed	1.06 (0.97–1.31)	1.26 (1.14–1.40)**
Pregnancy wanted	No (r)	0.85 (0.79–1.12)
	Yes	0.95 (0.89–1.08)	-
pregnancy complications	No (r)	0.82 (0.72–1.06)
	Yes	1.27 (0.97–1.47)	-
Parity	1–3(r) 4-6 7+	1.10 (0.93–1.24) 0.89 (0.80–105) 0.94 (0.882–0.97)*	0.95 (0.82–1.07) 0.71 (0.67–1.02) 0.92 (0.86–0.95)**

(OR): odds ratios; CI: confidence intervals; * indicates variables with p < 0.1, ** indicates significant effects with p < 0.05, and (r) indicates the reference category and (-) indicates variables not considered at multivariable level

Table 6. Logit and GSEM model predicting the Odds Ratios of women utilizing early PNC

Characteristic	Category	unadjusted OR (95% CI)	Adjusted OR (95% CI)
Health facility delivery	Did not deliver from health facility Delivered from health facility	0.87 (0.85–1.08) 1.50 (1.41–1.87)*	0.98 (0.87–1.13)
Early ANC	Went after first trimester	0.80 (0.67–0.92)*
	Went within the first trimester	1.12 (1.07–1.23)*	1.10 (1.05–1.26)*
Age	15–19 (r)	1.89 (1.76–1.98)*
	20–34	0.97 (0.88–0.14)	0.91 (0.84–1.21)
	35–49	1.27 (1.17–1.40)	1.14 (0.94–1.35)
Highest Maternal Education level	Some primary (r)	0.89 (0.73–1.23)
	Completed primary seven	0.80 (0.68–1.18)	1.04 (0.92–1.27)
	Some secondary	1.13 (0.92–1.38)*	1.19(1.04–1.38)*
	Completed secondary six	1.42 (1.00–1.82)	1.11 (0.94–1.35)
Marital status	Unmarried (r)	1.23 (0.98–1.51)	-
	Married	1.02 (0.87–1.16)	-
Wealth	Poor (r)	0.69 (0.59–0.81)*
	Middle	0.81 (0.72–1.16)	0.74 (0.68–1.34)
	Rich	0.93 (0.98–1.34)	1.24 (1.13–1.34)*
Distance to health facility	Big problem (r)	0.91 (0.72–0.99)*
	Not big problem	1.34 (1.17–1.67)*	1.90 (1.53–2.35)*
Cost of service	Big problem (r)	0.87 (0.76–0.96)*
	Not big problem	0.98 (0.88–1.12)	0.92 (0.85–1.16)
Community health worker	Not readily available (r)	0.85 (0.75–0.94)*
	Readily available	2.06 (1.88–2.45)*	1.15 (1.03–1.25)*
Media exposure	Not exposed (r)	0.87 (0.78–1.42)
	Exposed	1.51 (1.24–1.85)*	0.83(0.72–1.21)
Pregnancy wanted	No (r)	0.89 (0.80–1.10)
	Yes	1.03(0.93–1.28)	-
pregnancy complications	No (r)	1.14 (0.91–1.33)
	Yes	1.18 (1.07–1.67)*	1.17 (1.05–1.31)*
Parity	1–3 4-6 7+	0.76 (0.61–1.01) 0.92 (0.84–1.2) 0.82 (0.73–0.93)*	0.70 (0.61–1.1) 0.86 (0.78–1.09) 0.73 (0.67–0.81)**

(OR): odds ratios; CI: confidence intervals; * indicates variables with p < 0.1, ** indicates significant effects with p < 0.05, and (r) indicates the reference category and (-) indicates variables not considered at multivariable level

Table 7. Indirect effects on health facility delivery and EPNC through early ANC

Characteristic	Category	health facility delivery	Early PNC
Age	15–19
	20–34		(1.01*1.1) = 1.01*
	35–49
Highest Maternal Education level	Some primary
	Completed primary seven		(1.69*1.1) = 1.85*
	Some secondary
	Completed secondary six
Marital status	Unmarried
	Married	(1.04*1.1) = 1.16*
Distance to health facility	Big problem
	Not big problem	(1.04*1.35) = 1.40*
Cost of service	Big problem
	Not big problem		(1.85*1.1) = 2.04*
pregnancy complications	No (r)
	Yes	(1.04*2.04) = 2.12*

* indicates significant effects with p < 0.05

Figure 1. Causal loop showing predictors of early ANC, health facility delivery and EPNC

Figure A1. Shows the interrelationships between the three endogenous variables in the model

Data availability statement

The datasets generated and/or analyzed during this study are not publicly available due to a requirement of approval from ICF Macro International U.S.A to use the data. The authors were authorized to use ‘Survey’ data from the Demographic and Health Surveys (DHS) Program accessed on: http://www.dhsprogram.com/data/dataset_admin/login_main.cfm.