Antenatal care and its determinants
Table 1 presents a profile and a comparison of antenatal care visits in Egypt in the two surveys and their main determinants. It shows a significant improvement in the use of antenatal care services: while only 41.7% of women had had at least one antenatal care visit in 1995, this proportion had increased to 90.2% by 2014. A similar improvement is observed for URAC: while just 30.4% of women had received URAC in 1995, that number had increased to 89.2% of women in 2014.
This improvement in use of antenatal care was accompanied by significant changes in the demographic attributes of the women surveyed. Regarding age at birth, the proportions of women who gave birth at both young and old ages decreased significantly between the two surveys; while 75.4% of women gave birth between the age of 20 and 35 in 1995, this proportion increased to 82% in 2014. The majority of the declines occurred among older women (35 and over).
On the socioeconomic front, in 2014 the level of educational attainment achieved by women and their husbands showed significant gains compared to the levels in 1995. In 2014, almost 73% of women had secondary or higher education, compared to 34% in 1995. Similarly, while only 44.3% of the women’s husbands had secondary or higher education in 1995, this proportion had increased to 73.2% in 2014.
Using the lack of a modern toilet facility as a proxy for household standard of living, Table 1 shows significant improvement in these standards of living, with the proportion of households with a modern toilet increasing from 21.6% in 1995 to 52.3% in 2014.
Table 1. Percentage distribution and concentration indexes for URAC and its determinants (1995 and 2014)
Variables
|
1995
|
2014
|
|
|
%
|
CI
|
%
|
CI
|
Antenatal care
|
|
|
|
|
Any antenatal care***
|
41.7
|
0.294
|
90.20
|
0.028
|
Regular antenatal care***
|
30.4
|
0.417
|
82.90
|
0.052
|
Age at birth ***
|
|
|
|
|
<20 years
|
10.00
|
-0.146
|
7.40
|
-0.119
|
20-
|
27.40
|
-0.018
|
30.50
|
0.004
|
25-
|
29.10
|
0.067
|
32.40
|
0.030
|
30-
|
18.90
|
0.026
|
19.10
|
0.010
|
35-
|
10.70
|
-0.016
|
8.60
|
-0.027
|
40+
|
3.90
|
-0.079
|
2.00
|
-0.080
|
Respondent's educational attainment***
|
|
|
|
No education
|
44.30
|
-0.352
|
18.30
|
-0.393
|
Primary
|
22.00
|
-0.019
|
8.90
|
-0.205
|
Secondary
|
28.10
|
0.406
|
57.30
|
0.036
|
Higher
|
5.70
|
0.812
|
15.60
|
0.446
|
Respondent's husband’s educational attainment***
|
|
|
No education
|
27.80
|
-0.407
|
13.20
|
-0.286
|
Primary
|
28.00
|
-0.130
|
13.60
|
-0.133
|
Secondary
|
33.40
|
0.235
|
56.60
|
-0.009
|
Higher
|
10.90
|
0.654
|
16.60
|
0.365
|
Region ***
|
|
|
|
|
Urban governorate
|
18.00
|
0.475
|
10.50
|
0.712
|
Urban Lower Egypt
|
10.30
|
0.430
|
9.30
|
0.616
|
Rural Lower Egypt
|
30.00
|
-0.124
|
39.40
|
-0.154
|
Urban Upper Egypt
|
11.30
|
0.307
|
11.10
|
0.479
|
Rural Upper Egypt
|
29.40
|
-0.437
|
28.80
|
-0.438
|
Frontier governorate
|
1.00
|
0.142
|
1.00
|
0.123
|
Toilet type (modern)***
|
21.60
|
0.732
|
52.30
|
0.350
|
Experience of terminated pregnancy ***
|
26.70
|
-0.022
|
20.70
|
-0.066
|
Birth attributes
|
|
|
|
|
Birth interval>24 months)***
|
78.80
|
0.027
|
84.50
|
0.011
|
Single birth**
|
98.40
|
0.001
|
97.80
|
0.001
|
Birth order ***
|
3.90
|
-0.115
|
2.70
|
-0.074
|
Previous birth experience
|
|
|
|
|
Delivery at medical unit***
|
11.00
|
0.250
|
28.60
|
0.012
|
Delivered by C-section***
|
1.90
|
0.328
|
15.80
|
0.085
|
Live birth
|
75.00
|
-0.008
|
75.80
|
-0.028
|
*** Significant < 0.001 ** Significant <0.01
On the health needs front, except for the death of the previous child and having had multiple births, Table 1 shows significant improvement on all indicators. Experience of terminated pregnancies among the sample decreased significantly, from 26.7% in 1995 to 20.7% in 2014. The prevalence of appropriate preceding birth interval (more than 24 months) increased from 78.8% in 1995 to 84.5% in 2014. Birth order also showed a significant decrease between 1995 and 2014, from 3.9 to 2.7. Delivery in medical centers also increased from 11% in 1995 to 28.6% in 2014. By contrast, the prevalence of cesarean section increased from 1.9% in 1995 to 15.8% in 2014.
Improvement in the prevalence of URAC was accompanied by significant improvement/reduction in the levels of wealth-based inequality; while in 1995 CI for URAC was 0.417, it had declined to 0.052 in 2014, a reduction of 87.5%.
Table 1 also shows the concentration indexes for the main determinants of URAC. In 1995, wealth-related inequalities were large for almost all the determinants. On one hand, women aged below 25 years and those 35 years or older while giving birth, those with low levels of education, low levels of husband’s education, and rural residents of both Lower Egypt and Upper Egypt, were highly concentrated among relatively poor respondents. The factor of a large number of children was also highly concentrated among poor respondents. On the other hand, proper birth intervals, delivery in medical units and cesarean section were less prevalent among relatively poor respondents.
In 2014, similar patterns of the wealth-related inequalities in the determinants of URAC can be observed, with the majority of the inequalities declining in magnitude, reflecting improvements in their wealth-related differentials. For example, while CI for the respondents’ secondary education was 0.406 in 1995, in 2014, CI had decreased to 0.036. Similarly, wealth-related differentials in birth order declined from CI=-0.115 in 1995 to CI=-0.066 in 2014. The concentration of the poor who live in rural Upper Egypt showed a stability between the two years (CI=-0.438). By contrast, an increase in inequality was observed in the experience of terminated pregnancies, and for respondents living in urban Lower and Upper Egypt.
Determinants of use of regular antenatal care
Table 2 presents the odds ratios for the logit coefficients for URAC in the years 1995 and 2014. For 1995, older age at birth and high levels of education in both the respondent and her husband were significantly more likely to correlate with URAC. Regional patterns for URAC exhibited lower odds of use among all regions, in particular the rural regions. For example, living in rural Upper Egypt decreased the odds of URAC by 71%.
Proper preceding birth intervals were associated with significantly higher odds of URAC, while a higher birth order was significantly related to lower odds of regular use of antenatal care (OR=0.89). Delivery in a medical unit for the previous birth was also associated with 2.16 higher odds of regular use of antenatal care. A previous live birth decreased the odds of URAC by 45%.
In 2014, the odds of URAC showed a similar pattern as in 1995, but the effects of the different determinants on URAC were attenuated. For age at birth, only those giving birth aged 30-35, or over 40, exhibited higher odds of use compared to other age groups. Education for both the respondents and their husbands continued to be positively related to increased odds of URAC. Residents of regions other than urban governorates and urban Lower Egypt showed lower odds of URAC. High standards of living were also associated with 26% higher odds of antenatal care. Delivery of a previous birth by cesarean section showed 84% increases in URAC.
Table 2. Odds ratio for use of antenatal care in 1995 and 2014
Variables
|
1995
|
2014
|
|
Age at birth
|
|
|
|
|
20-
|
1.21
|
|
1.05
|
|
25-
|
1.73
|
***
|
1.24
|
|
30-
|
1.93
|
***
|
1.43
|
*
|
35-
|
1.98
|
***
|
1.34
|
|
40+
|
2.19
|
**
|
1.84
|
*
|
Respondent's educational attainment
|
|
|
|
|
Primary
|
1.38
|
**
|
1.11
|
|
Secondary
|
2.54
|
***
|
1.51
|
***
|
Higher
|
4.68
|
***
|
2.61
|
***
|
Respondent's husband’s educational attainment
|
|
|
|
Primary
|
1.40
|
**
|
1.39
|
**
|
Secondary
|
1.44
|
**
|
1.51
|
***
|
Higher
|
1.92
|
***
|
1.95
|
***
|
Region
|
|
|
|
|
Urban Lower Egypt
|
0.89
|
|
0.99
|
|
Rural Lower Egypt
|
0.43
|
***
|
0.82
|
|
Urban Upper Egypt
|
0.74
|
*
|
0.59
|
***
|
Rural Upper Egypt
|
0.29
|
***
|
0.48
|
***
|
Frontier governorate
|
0.43
|
***
|
0.52
|
***
|
Toilet type (modern)
|
2.47
|
***
|
1.20
|
*
|
Experience of terminated pregnancy
|
1.73
|
***
|
1.26
|
**
|
Birth attributes
|
|
|
|
|
Birth interval>24 months)
|
1.46
|
***
|
1.46
|
***
|
Single birth
|
0.78
|
|
0.58
|
*
|
Birth order
|
0.89
|
***
|
0.83
|
***
|
Previous birth experience
|
|
|
|
|
Delivery at medical unit
|
2.16
|
***
|
0.93
|
|
Delivered by C-section
|
0.98
|
|
1.84
|
***
|
Live birth
|
0.55
|
***
|
0.45
|
***
|
Pseudo R2
|
0.26
|
|
0.11
|
|
Log likelihood
|
-3317.39
|
|
-4451.88
|
|
*** Significant < 0.001 ** Significant <0.01 *Significant at <0.05
Decomposition of use of regular antenatal care inequality
Table 3 presents the results of the decomposition of the inequalities in URAC into its determinants. Figures 1A and 1B show the different shares of the inequality in the determinants of receiving regular antenatal care in 1995 and 2014, respectively. Figure 1A reveals that, in 1995, six determinants of URAC accounted for 90% of its inequality. In 1995, these determinants (arranged by their relative share) are: living in rural Upper Egypt (25.3%), household standard of living indicator (20.8%), women’s secondary or higher education (17.4% and 12.4%, respectively), husbands with higher education (7.1%), and birth order of the child (6.9%).
Table 3. Decomposition of the wealth-related inequalities in URAC
|
Variables
|
2014
|
1995
|
dy/dx
|
CI
|
Mean
|
Elasticity
|
Share
|
% Share
|
dy/dx
|
CI
|
Mean
|
Elasticity
|
Share
|
% Share
|
Age at birth
|
|
|
|
|
|
|
|
|
|
|
|
|
20-
|
0.007
|
0.004
|
0.305
|
0.003
|
0.0000
|
0.0
|
0.025
|
-0.018
|
0.274
|
0.023
|
-0.0004
|
-0.1
|
25-
|
0.029
|
0.030
|
0.324
|
0.011
|
0.0003
|
0.8
|
0.078
|
0.067
|
0.291
|
0.074
|
0.0050
|
1.5
|
30-
|
0.046
|
0.010
|
0.192
|
0.011
|
0.0001
|
0.3
|
0.094
|
0.026
|
0.189
|
0.058
|
0.0015
|
0.5
|
35-
|
0.038
|
-0.027
|
0.086
|
0.004
|
-0.0001
|
-0.3
|
0.098
|
-0.016
|
0.107
|
0.034
|
-0.0005
|
-0.2
|
40+
|
0.073
|
-0.080
|
0.020
|
0.002
|
-0.0001
|
-0.3
|
0.114
|
-0.079
|
0.039
|
0.015
|
-0.0012
|
-0.4
|
Respondent's educational attainment
|
|
|
|
|
|
|
Primary
|
0.015
|
-0.205
|
0.089
|
0.002
|
-0.0003
|
-0.8
|
0.048
|
-0.019
|
0.220
|
0.035
|
-0.0007
|
-0.2
|
Secondary
|
0.057
|
0.036
|
0.573
|
0.040
|
0.0014
|
3.4
|
0.153
|
0.406
|
0.281
|
0.142
|
0.0575
|
17.4
|
Higher
|
0.115
|
0.446
|
0.156
|
0.022
|
0.0096
|
22.9
|
0.271
|
0.812
|
0.057
|
0.051
|
0.0410
|
12.4
|
Respondent's husband’s educational attainment
|
|
|
|
|
|
|
Primary
|
0.047
|
-0.133
|
0.136
|
0.008
|
-0.0010
|
-2.4
|
0.050
|
-0.130
|
0.279
|
0.046
|
-0.0059
|
-1.8
|
Secondary
|
0.057
|
-0.009
|
0.566
|
0.039
|
-0.0003
|
-0.8
|
0.054
|
0.235
|
0.334
|
0.059
|
0.0139
|
4.2
|
Higher
|
0.087
|
0.365
|
0.166
|
0.017
|
0.0064
|
15.2
|
0.099
|
0.654
|
0.109
|
0.036
|
0.0233
|
7.1
|
Region
|
|
|
|
|
|
|
|
|
|
|
|
|
Urban Lower Egypt
|
-0.001
|
0.616
|
0.093
|
0.000
|
-0.0001
|
-0.2
|
-0.023
|
0.430
|
0.103
|
-0.008
|
-0.0033
|
-1.0
|
Rural Lower Egypt
|
-0.021
|
-0.154
|
0.394
|
-0.010
|
0.0016
|
3.7
|
-0.142
|
-0.124
|
0.300
|
-0.140
|
0.0174
|
5.3
|
Urban Upper Egypt
|
-0.064
|
0.479
|
0.111
|
-0.008
|
-0.0041
|
-9.7
|
-0.054
|
0.307
|
0.113
|
-0.020
|
-0.0062
|
-1.9
|
Rural Upper Egypt
|
-0.093
|
-0.438
|
0.287
|
-0.032
|
0.0141
|
33.5
|
-0.198
|
-0.437
|
0.294
|
-0.191
|
0.0836
|
25.3
|
Frontier governorate
|
-0.081
|
0.123
|
0.009
|
-0.001
|
-0.0001
|
-0.3
|
-0.142
|
0.142
|
0.010
|
-0.005
|
-0.0007
|
-0.2
|
Toilet type (modern)
|
0.023
|
0.350
|
0.523
|
0.015
|
0.0051
|
12.1
|
0.132
|
0.732
|
0.216
|
0.094
|
0.0686
|
20.8
|
Experience of terminated
|
|
|
|
|
|
|
|
|
|
|
|
pregnancy
|
0.030
|
-0.066
|
0.207
|
0.007
|
-0.0005
|
-1.2
|
0.080
|
-0.022
|
0.267
|
0.070
|
-0.0016
|
-0.5
|
Birth attributes
|
|
|
|
|
|
|
|
|
|
|
|
|
Birth interval>24 months)
|
0.048
|
0.011
|
0.845
|
0.049
|
0.0005
|
1.2
|
0.055
|
0.027
|
0.788
|
0.142
|
0.0039
|
1.2
|
Single birth
|
-0.069
|
0.001
|
0.978
|
-0.081
|
-0.0001
|
-0.2
|
-0.036
|
0.001
|
0.984
|
-0.117
|
-0.0001
|
0.0
|
Birth order
|
-0.024
|
-0.074
|
2.675
|
-0.079
|
0.0058
|
13.9
|
-0.016
|
-0.115
|
3.714
|
-0.199
|
0.0228
|
6.9
|
Previous birth experience
|
|
|
|
|
|
|
|
|
|
|
|
Delivery at medical unit
|
-0.009
|
0.012
|
0.286
|
-0.003
|
0.0000
|
-0.1
|
0.113
|
0.250
|
0.110
|
0.041
|
0.0102
|
3.1
|
Delivered by C-section
|
0.077
|
0.085
|
0.158
|
0.015
|
0.0013
|
3.0
|
-0.004
|
0.328
|
0.019
|
0.000
|
-0.0001
|
0.0
|
Live birth
|
-0.103
|
-0.028
|
0.758
|
-0.094
|
0.0026
|
6.2
|
-0.087
|
-0.008
|
0.750
|
-0.215
|
0.0017
|
0.5
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Figure 1B shows that, in 2014, five the previous six determinants continued to play a major role in explaining the inequalities in URAC and account for 97.6% of the inequalities. The highest share was attributed to inequalities in living in rural Upper Egypt (33.5%), followed by inequalities in the respondent’s higher education (22.9%) and husband’s higher education (15.2%). Inequalities in birth order accounted for 13.9% and household standard of living retreated, accounting for 12.1% of the inequality in antenatal care.
Changes in the determinants’ shares between 1995 and 2014 reveal the role played by their concentration indexes and elasticities. For example, the analysis shows that living in rural Upper Egypt maintained its place as providing the largest contribution to inequality in URAC in both years. In 1995, this large share was mainly the product of a high concentration of poor women in this region (CI=-0.437) and its large elasticity in URAC =-0.191), resulting in a 25.3% share. Between 1995 and 2014, this share increased to 33.5%. This increase was the product of the lower value CI for URAC in 2014 compared to the level in 1995. At the same time, living in rural Upper Egypt maintained its high concentration of poor women (CI=-0.438), but experienced some improvement in its elasticity ( =-0.093). For household standard of living (toilet type), the 20.8% share in 1995 was the product of its high concentration among the rich (CI=0.732) and relatively moderate elasticity ( =0.093). By contrast, in 2014, the concentration index declined by almost 50% (CI=0.350) and elasticity also decreased, reaching ( =0.015), producing a decline in its share to 12.1%. The respondent’s higher education showed an increase in its share between 1995 and 2014 from 12.4% to 22.9%, respectively, despite the improvement in its concentration index (from 0.812 to 0.446) and elasticity (from 0.051 to 0.022).
Decomposition of the change in the wealth-related inequalities of use of regular antenatal care between 1995 and 2014
Understanding the main source for the decline in wealth-related inequalities in URAC is an important exercise to assess the contribution of the different policies over the period under consideration. Table 4 shows the changes in the determinants of URAC that contributed to the decline in its inequality between 1995 and 2014. These changes are further divided between its two components, namely changes in their elasticity and changes in their concentration index ( . The assessment of a determinant’s contribution to the change in the inequalities in regular use of antenatal care is carried out in terms of the overall sign and the magnitude of this contribution. A large absolute value indicates a large contribution. As the difference in the inequality of URAC is assessed in terms of inequality in 1995 minus inequality in 2014, a positive sign for any determinant indicates that this determinant has contributed to the decline in inequality, while a negative sign indicates that this determinant has counteracted the decline in the inequality.
Table 4. Decomposition of changes in the wealth-related inequalities in URAC
|
Differences in
|
Share
|
% Share
|
Variables
|
Elasticity
|
CI
|
Elasticity
|
CI
|
Total
|
Elasticity
|
CI
|
Total
|
|
Age at birth
|
|
|
|
|
|
|
|
|
|
20-
|
0.020
|
-0.022
|
0.000
|
-0.001
|
0.000
|
0.0
|
-0.2
|
-0.2
|
|
25-
|
0.063
|
0.038
|
0.002
|
0.003
|
0.005
|
0.6
|
1.0
|
1.6
|
|
30-
|
0.048
|
0.016
|
0.000
|
0.001
|
0.001
|
0.2
|
0.3
|
0.5
|
|
35-
|
0.030
|
0.011
|
-0.001
|
0.000
|
0.000
|
-0.3
|
0.1
|
-0.2
|
|
40+
|
0.013
|
0.001
|
-0.001
|
0.000
|
-0.001
|
-0.4
|
0.0
|
-0.4
|
|
Respondent's educational attainment
|
|
|
|
|
|
|
|
Primary
|
0.033
|
0.185
|
-0.007
|
0.006
|
0.000
|
-2.4
|
2.3
|
-0.1
|
|
Secondary
|
0.102
|
0.370
|
0.004
|
0.052
|
0.056
|
1.3
|
18.2
|
19.5
|
|
Higher
|
0.029
|
0.366
|
0.013
|
0.019
|
0.031
|
4.5
|
6.4
|
10.9
|
|
Respondent's husband’s educational attainment
|
|
|
|
|
|
|
|
Primary
|
0.038
|
0.003
|
-0.005
|
0.000
|
-0.005
|
-1.7
|
0.0
|
-1.7
|
|
Secondary
|
0.020
|
0.243
|
0.000
|
0.014
|
0.014
|
-0.1
|
5.0
|
4.9
|
|
Higher
|
0.018
|
0.288
|
0.007
|
0.010
|
0.017
|
2.3
|
3.6
|
5.9
|
|
Region
|
|
|
|
|
|
|
|
|
|
Urban Lower Egypt
|
-0.007
|
-0.186
|
-0.005
|
0.001
|
-0.003
|
-1.6
|
0.5
|
-1.1
|
|
Rural Lower Egypt
|
-0.130
|
0.030
|
0.020
|
-0.004
|
0.016
|
7.0
|
-1.5
|
5.5
|
|
Urban Upper Egypt
|
-0.012
|
-0.172
|
-0.006
|
0.003
|
-0.002
|
-1.9
|
1.2
|
-0.7
|
|
Rural Upper Egypt
|
-0.159
|
0.001
|
0.070
|
0.000
|
0.070
|
24.2
|
-0.1
|
24.1
|
|
Frontier governorate
|
-0.004
|
0.019
|
0.000
|
0.000
|
-0.001
|
-0.2
|
0.0
|
-0.2
|
|
Toilet type (modern)
|
0.079
|
0.382
|
0.028
|
0.036
|
0.064
|
9.6
|
12.4
|
22.0
|
|
Experience of terminated pregnancy
|
0.063
|
0.044
|
-0.004
|
0.003
|
-0.001
|
-1.4
|
1.1
|
-0.3
|
|
Birth attributes
|
|
|
|
|
|
|
|
|
|
Birth interval>24 months)
|
0.093
|
0.017
|
0.001
|
0.002
|
0.003
|
0.3
|
0.8
|
1.1
|
|
Single birth
|
-0.036
|
0.000
|
0.000
|
0.000
|
0.000
|
0.0
|
0.0
|
0.0
|
|
Birth order
|
-0.120
|
-0.040
|
0.009
|
0.008
|
0.017
|
3.1
|
2.8
|
5.9
|
|
Previous birth experience
|
|
|
|
|
|
|
|
|
|
Delivery at medical unit
|
0.044
|
0.237
|
0.001
|
0.010
|
0.010
|
0.2
|
3.4
|
3.6
|
|
Delivered by C-section
|
-0.015
|
0.243
|
-0.001
|
0.000
|
-0.001
|
-0.4
|
0.0
|
-0.4
|
|
Live birth
|
-0.122
|
0.020
|
0.003
|
-0.004
|
-0.001
|
1.2
|
-1.5
|
-0.3
|
|
Figure 2 shows that socioeconomic determinants were the largest contributors to the decline in wealth-based inequality in URAC between 1995 and 2014. Of the 12 determinants with positive signs, nine were socioeconomic determinants. Living in rural Upper Egypt, the household’s standard of living, and the woman’s attainment of secondary or higher education were the largest contributors to the decline in inequality, with a total share of 76.6% (24.1%, 22.1%, 19.5% and 10.9%, respectively).
These were followed by husbands’ higher or secondary education and living in rural Lower Egypt, which accounted for a total share of 16.3% (5.9%, 4.9% and 5.5%, respectively). By contrast, only three health need determinants showed positive contributions to the decline in URAC, namely birth order (5.9%), delivery in a medical center (3.6%) and appropriate birth interval (>24 months) (1.2%).
Examination of the two components of the determinant’s contribution to the decline in inequality revealed different patterns for different determinants. Table 4 shows that the contributions of living in rural Upper Egypt and living in rural Lower Egypt were mainly due to declines in elasticity between 1995 and 2014. By contrast, the contribution of a woman’s and her husband’s secondary education was mainly due to declines in CI during the same period. For the other positively contributing determinants, their contributions were divided between declines in elasticity ( ) and declines in CI, with the share of the declines in CI commonly larger than that for elasticity. For example, the 22.1% contribution of the household standard of living was divided into 12.4% for declines in CI and 9.7% for declines in elasticity. Similarly, the 10.9% contribution of women’s higher education was divided into 6.4% for declines in CI and 4.5% for declines in elasticity.