FERTILITY
The first comprehensive national data obtained on fertility and mortality were collected in the 1979 Ghana Fertility Survey (GFS). The 1988 Ghana Demographic and Health Survey (GDHS) was the second such survey. One of the objectives of the 1993 GDHS is to update knowledge on fertility levels and trends.
Two types of fertility data were collected in the 1993 GDHS. First, each woman was asked questions about the number o f sons and daughters living with her, the number living elsewhere, and the number that had died. Second, a complete live birth history was collected from each respondent covering sex, date of birth and survival status of each child or, if dead, age at death.
In this chapter, current fertility levels, trends and differentials; cumulative fertility; birth intervals; age at first birth; and adolescent fertility are considered.
3.1
Current Fertility Levels and Trends
Table 3.1 presents the current age-specific fertility rates and a series o f summary measures. A 5-year reference period has been adopted instead of a 3-year period due to substantial displacement of births from the third year prior to the survey to the fourth, which has the effect of biasing downward the 3-year rate (see Appendix B). For all women, the age-specific fertility rate starts from 119 live births per thousand for the age group 15-19, reaches a peak of 244 at age 25-29, and falls to 29 per thousand for those age 45-49. The same pattem seems to emerge for urban and rural wom- en except that fertility reaches a maximum earlier in rural areas, and is sustained at a level well above 200 per thou- sand into the mid-30's (see Figure 3. I).
The total fertility rate for women age 15-49 for the five years preceding the survey is 5.5 children per woman. If studied by place of residence, rural women have about six- ty percent more children than their urban counterparts. The fertility rate for rural women is 6.4 children per woman while their urban counterparts have 4 children per woman. Differentials in the general fertility rate (GFR) have been registered as well. For the country the GFR is 188 per thou- sand women while by place o f residence marked variations emerge; 138 per thousand and 218 per thousand for urban
Table 3.1 Current fertility
Age-specific and cumulative fertility rates and the crude birth rate for the five years preceding the survey, by urban-rural residence, Ghana 1993
Residence
Age group Urban Rural Total
15-19 83 145 119 20-24 166 273 231 25-29 193 272 244 30-34 165 243 215 35-39 118 186 163 40-44 53 120 99 45-49 20 33 29 TFR 15-49 3.99 6.36 5.50 TFR 15-44 3.89 6.20 5.35 GFR 138 218 188 CBR 32.9 40.2 38.0
Note: Rates are for the period 1-60 months preceding the survey. Rates for age group 45-49 may be slightly biased due to truncation.
TFR: Total fertility rate for ages 15-49, expressed per woman
GFR: General fertility rate (births divided by number of women 15-44), expressed per 1,000 women
CBR: Crude birth rate, expressed per 1,000 population
and rural women, respectively. The crude birth rate (CBR) for the country is 38 per thousand, which indicates some decline in fertility when compared with the rate of 44 per thousand observed by indirect techniques from the 1984 Census.
Figure 3.1
Age-Specific Fertility Rates
By Urban-Rural Residence
Births per 1,000 Women3OO ( )1 200 50 U i i i i i 15-1 g 2 0 - 2 4 2 5 - 2 9 3 0 - 3 4 3 5 - 3 9 4 0 ~ ' 4 Age Group 4 5 - 4 9 GDHS l g g 3
3.2
Fertility Differentials
Table 3.2 shows the total fertility rate (TFR) of women age 15-49 years for the five years preceding the survey and the children ever bom (CEB) for women age 40-49 years by selected background characteristics. As regards the regional "IPR values, Greater Accra has the lowest fertility rate of 3.6 children per woman followed by Eastern and Volta with 5.1 and 5.4 children per woman, respectively. The region with the highest TFR is Northern Region with 7.4 children per woman. Total fertility rates for other regions are close to 6 children per
w o m a n .
The data also show that TFR decreases as the level of education o f the woman increases. Women with no education have 6.7 children while those with secondary/higher education average only 2.9 children. For those with primary education, the TFR is 6.1 children, whereas those with middle/JSS education have about 4.7 children per woman (see Figure 3.2).
As a measure o f change, the TFRs observed in the survey are compared with the average number of children ever bom to women age 40-49, at the end of their childbearing period. The data in Table 3.2 show
Table 3.2 Fertility by background characteristics Total fertility rate for the five years preceding the survey and mean number of children ever born to women age 40-49, by selected background characteristics, Ghana 1993
Mean number of children Total ever born Background fertility to women characteristic rate I age 40-49 Residence Urban 3.99 5.35 Rural 6.36 6.55 Region Western 5.54 6.12 Central 5.57 6.79 Greater Accra 3.56 4.53 Volta 5.41 5.71 Eastern 5.10 6.19 Ashanti 5.60 6.43 Brong-Ahafo 5.46 6.84 Northern 7.39 7.44 Upper West 6.02 6.11 Upper East 6.44 5.88 Education No education 6.67 6.66 Primary 6.10 6.34 Middle/JSS 4.71 5.66 Secondary/Higher 2.90 3.67
Total Fertility Rate by
Selected Background Characteristics
RESIDENCE Urban Rural REGION Western Central Greater Accra Volta Eastern Ashanti Brong-Ahafo Northern Upper West Upper East EDUCATION No education Primary Middle/JSS Secondary/Higher 14 I 6.4 5.5 5,6 3.6 5.4 5.1 5.6 5.5 6 6.4 ~ ~ \ X X \ \ \ \ \ \ \ N 1 6.7 ~ ~ \ ~ X ~ l 4.7 ~ ~ , ~ , , ~ , ' ~ 2.9 ) 2 4 6
Total Fertility Rate
7.4
GDHS 1993
northernmost regions, that is Northern, Upper West and Upper East. However, there appears to have been little change in fertility levels in rural areas and among women with no education.
3.3
Fertility Trends
The 1960Post Enumeration Survey (Gil et al., 1971), the 1971 Supplementary Enquiry (CBS, 1971), the Ghana Fertility Survey, 1979-1980 (CBS, 1983) and the Demographic and Health Survey o f 1988 (IRD and GSS, 1989) have suggested a TFR of more than 6 children per woman. For a comparison o f current fertility rates derived from the 1988 GDHS with those from the 1993 GDHS, see Figure 3.3. The TFR calculated for the 1988 GDHS was 6.4 children per woman, compared to 5.5 derived from the present survey.
Age-specific fertility rates for successive five-year periods preceding the survey are presented in Table 3.3. The trend in fertility up to age 34 is clearly seen. Rates for periods further into the past are, of course, truncated for women at older ages. Evidence from the table suggests that there have been declines in fertility over the last 15 years preceding the survey. For the age group 20-24, the decline in fertility during the two five-year periods prior to the survey is 13 percent. For women age 30-34 years a significant decline o f 21 percent is observed over a decade. For women 35-39, the decline is 34 percent. Thus, it is clear that significant declines in fertility have occurred after age 30.
In Table 3.4 the age-specific fertility rates of ever-married women by duration o f marriage are displayed for successive five-year periods preceding the survey. The data show declines in fertility at all but the shortest marital duration. The decline over a decade for women married 15-19 years is 32 percent, and for those married 20-24 years, 43 percent.
Figure 3.3
Age-Specific Fertility Rates,
GDHS 1988 and GDHS 1993
45-49 Births per 1,000 Women
3O0 0 , o , ° • " ° " ~ " ° o . o o 200 150 **° 0 i i i p i 15-19 20-24 25-29 30-34 35-39 40-44 Age Group "(F'GDHS 1988 4E'GDHS 1993 LBB
I
Table 3.3 Fertility trendsAge-specific fertility rates for five-year periods preceding
the survey, Ghana 1993
Years preceding survey Maternal age at birth 0-4 5-9 10-14 15-19 15-19 119 113 128 118 20-24 231 248 264 243 25-29 244 266 274 263 30-34 215 236 273 272 35-39 163 197 248 40-44 99 139 4 5 4 9 29
Note: Age-specific fertility rates are per 1,000 women.
3.4
Children Ever Born and Living
Table 3.4 Fertility by marital duration
Fertility rates for ever-married women by duration since first marriage in years, for five-year periods preceding the survey, Ghana 1993
Marriage Years preceding survey duration at birth 0-4 5-9 10-14 15-19 0 4 303 300 329 308 5-9 264 277 286 276 10-14 216 255 278 280 15-19 176 209 260 310 20-24 125 154 219 25-29 49 96
Note: Duration-specific fertility rates are per 1,000 women.
The distribution of number of children ever born and those living by age of women is presented in Table 3.5. The table shows that the mean number of children ever born among 'all women is 2.9. The number of children ever born is a function of age, rising from a low of 0.2 for women age 15-19, to a high of 6.6 for women in the age group 45-49. This suggests that in the past by the time a woman finished childbearing she
P e r c e n t distribution o f all w o m e n a n d c u r r e n t l y m a r r i e d w o m e n by n u m b e r o f c h i l d r e n ever b o r n ( C E B ) a n d m e a n n u m b e r ever b o r n a n d l i v i n g , a c c o r d i n g to five-year a g e g r o u p s , G h a n a 1993
Number of children ever born (CEB) Number Mean no. Mean no.
Age of of of living
group 0 1 2 3 4 5 6 7 8 9 10+ Total women CEB children
A L L W O M E N 15-19 81.4 16.6 1.6 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 I00.0 803 0.21 0.20 20-24 31.0 34.4 25.5 7.1 1.6 0.5 0.0 0.0 0.0 0.0 0.0 100.0 829 1.15 1.02 25-29 10.1 19.9 27.9 21.9 13.3 5.2 1.5 0.2 0.0 0.0 0.0 100.0 845 2.31 2.02 30-34 2.4 6.2 12.7 20.2 24.5 20.1 6.9 4.8 1.7 0.5 0.0 100.0 743 3.84 3.33 35-39 4.0 4.0 7.4 14.6 19.3 17.6 14.3 10.8 5.5 1.4 1.2 1(30.0 581 4.58 3.91 40-44 2.6 3.5 5.9 8.5 9.6 11.1 15.3 17.2 11.8 8.2 6.4 100.0 425 5.82 4.92 45-49 2.7 1.5 3.3 7.7 7.4 9.8 11.3 15.5 13.4 12.5 14.9 100.0 336 6.64 5.35 Total 23.2 14.8 13.9 11.9 10.6 8.3 5.5 5.0 3.1 2.0 1.8 100.0 4562 2.91 2.49 CURRENTLY MARRIED W O M E N 15-19 36.0 55.9 6.2 1.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 161 0.74 0.70 20-24 13.3 39.1 34.5 10.2 2.2 0 2 0.0 0.0 0.0 0.0 0.0 100.0 550 1.51 1.34 25-29 5.3 18.1 28.9 24.0 15.6 6.0 1.9 0.3 0.0 0.0 0.0 100.0 700 2.53 2.21 30-34 1.2 4.9 12.0 19.7 25.6 21.5 7.3 5.3 1.8 0.6 0.0 100.0 659 3.99 3.45 35-39 3.2 2.8 6.4 15.5 17.9 17.7 15.3 12.3 6.0 1.4 1.4 1(30.0 497 4.76 4.05 40-44 2.5 2.2 5.3 8.1 9.5 12.0 14.8 18.2 11.5 8.7 7.0 100.0 357 5.96 5.06 45-49 1.8 1.1 2.5 7.5 6.4 10.0 11.4 17.5 14.6 12.5 14.6 100.0 280 6.80 5.47 Total 6.4 15.3 16.8 15.1 13.5 10.8 6.9 6.6 3.9 2.4 2.3 100.0 3204 3.67 3.13
number of living children also increases with age: the value for all women is 2.5 and the number of children ever bom is 2.9. The 0.4 difference indicates the effects of mortality. A lower proportion of women age 45- 49 had 10 or more children in 1993 than in 1988 (15 percent compared to 22 percent)
With respect to currently married women, the mean number of children ever bom is 3.7. This value is slightly higher by almost one child relative to that for all women. Also the number of currently married women who have no child is lower by 17 percentage points than for all women. Even though the mean number of children ever born follows the same pattem as for all women, the value is higher for currently married women, especially in the first three age groups, i.e., ages 15-29.
3.5 B i r t h I n t e r v a l s
The timing of births has significant influence on both fertility and mortality. There is evidence that women with closely spaced births have higher fertility than women with longer birth intervals. Similarly, studies in diverse settings consistently show that shorter birth intervals increase the incidence of infant and child mortality.
Table 3.6 shows the percent distribution of births in the five years preceding the survey by length of interval since the previous birth, classified by selected demographic and background characteristics of the women. Ingeneral, themedianbirthintervalis slightly over 3 years. About 1 in 4 births occurs fourormore years after a previous birth, and one-sixth of the births occur within two years of a previous birth.
Table 3.6 Birth intervals
Percent distribution of births in the five years preceding the survey by number of months since previous birth, according to demographic and background characteristics, Ghana 1993
Number of months since previous birth
Charactedstic 7-17 18-23 24-35 36-47 48+
Median
number of Number months since of Total previous birth births Age of mother 15-19 * * * * * * * I5 20-29 4.7 13.7 37.6 24.6 19.4 100.0 34.2 1154 30-39 4.5 9.0 30.3 26.4 29.8 100.0 38.2 1371 40+ 6.2 8.9 28.5 21.8 34.5 100.0 39.0 417 Birth order 2-3 3.6 11.2 33.0 23,8 28.5 100.0 36.7 1317 4-6 5.4 10.9 31.8 26.6 25.3 100.0 36.6 1167 7+ 7.6 9.7 35.5 24.3 22.8 100.0 35.0 473
Sex or prior birth
Male 4.3 11.0 34.0 24.3 26.3 100.0 36.2 1526
Female 5.6 10.6 31.7 25.8 26.3 100.0 36.8 1431
Survlva/or prior birth
Living 3.5 9.6 33.1 26.0 27.9 100.0 37.1 2542 Deed 13.5 18.6 31.8 19.3 16.9 100.0 30.5 415 Residence Urban 3.5 9.4 27.2 21.4 38.5 100.0 41.1 746 Rural 5.4 11.3 34.8 26.2 22.2 100.0 35.6 2211 Region Western 6.2 8.5 36.3 23.9 25.1 100.0 35.6 259 Central 5.7 14.9 32.6 27.0 19.9 100.0 35.2 282 Greater Accra 2.6 12.2 28.7 19.1 37.4 100.0 39.2 230 Volta 2.7 9.7 33.1 19.5 35.0 100.0 38.1 329 Eastern 6.6 8.2 29.8 25.2 30.2 100.0 37.5 305 Ashanti 5.2 10.8 33.0 24.7 26.3 100.0 36.3 518 Brong-Ahafo 4.5 13.0 33.6 22.6 26.4 100.0 35.6 292 Nortbem 5.9 12.5 34.9 29.5 17.3 100.0 35.2 393 Upper West 4.8 8.7 39.7 32.5 14.3 100.0 35.3 126 Upper East 4.5 7.6 29.1 29~6 29.1 100.0 39.4 223 Education No education 5.5 12.2 35.0 25.7 21.6 100.0 35.3 1371 Primary 5.1 11.5 32.3 26.3 24.8 100.0 36.3 532 Middle/JSS 4.3 8.7 31.3 24.3 31.3 1(30.0 38.0 916 Secondary/Higher 2,9 8.7 24.6 18.1 45.7 100.0 44.5 138 Total 4.9 10,8 32.9 25,0 26.3 100.0 36,4 2957
Note: First-order births are excluded. The interval for multiple births is the number of months since the preceding pregnancy that ended in a live birth. An aztefisk signifies percentages based on fewer than 25 births, which have been suppressed.
median birth interval for women age 40 or more is 39 months but is only 34 months for those age 20-29 years. The median duration of birth interval increases with age, an indication of decreasing fecundity as women get older.
Birth intervals in relation to birth order or parity of the children are also examined. Apart from the seventh-order and higher children who have a lower median birth interval, there are no important differences in the median length of birth interval by birth order.
Since the death of a child curtails lactation, women who experience child loss are more likely to have a shorter birth interval than those who do not experience child loss. As indicated in the table, there appears to be a reduction o f 6.6 months in the birth interval for children whose prior sibling has died. With regard to sex, no marked variations are observed in the median birth interval.
There is a marked difference in birth intervals for women in urban and rural areas. Urban women wait 5.5 months longer than their rural counterparts ~ f o r e having another child.
From the table, it becomes clear that birth intervals vary by region. The regions of Volta, Greater Accra, Eastern, and Upper East have the longest birth interval, ranging between 38 to 39 months. The median length of the birth interval for the others shows no marked differences and averages about 35 to 36 months.
It had been conjectured that women with secondary or higher education have shorter birth intervals due to shorter periods of breastfeeding if they are not using contraception. On the contrary, the G D H S 1993 data demonstrate that women with secondary or higher education have the longest median birth interval, 45 months, followed by those with middle/JSS of 38 months.
3.6
Age at First Birth
Research has suggested that women who have their first birth early tend to have a greater number of children than those who delay their first birth, Table 3.7 gives the percentage distribution of age at first birth by current age of the mother. For the six cohorts with experience up to age 20, there is almost no difference in the percentage who had a birth before age 20. Among women currently age 20-24, 49 percent had given birth by age 20; among those currently age 40-44, 50 percent had given birth before age 20. Furthermore, the median age at first birth shows no tendency toward either a younger or older age at first birth.
Table 3.7 Age at first birth
Percent distribution of women 15-49 by age at first birth, according to current age, Ghana 1993
Current age
Women Median
with Age at first birth Number age at
no of first
births <15 15-17 18-19 20-21 22-24 25+ Total women birth
15-19 81.4 0.7 NA NA NA NA NA 100.0 803 a 20-24 31.0 2.3 22.2 24.0 NA NA NA 100.0 829 a 25-29 10.1 3.0 21.9 22.1 21.1 15.6 NA 100.0 845 20.3 30-34 2.4 4.0 24.1 22.3 21.8 16.3 9.0 100.0 743 20.0 35-39 4.0 3.3 20.3 21.5 19.3 16.7 15.0 100.0 581 20.5 40-44 2.6 5.6 22.8 21.6 17.4 17.9 12.0 100.0 425 20.0 4 5 4 9 2.7 2.4 22.6 22.3 18.2 16.1 15.8 1190.0 336 20.2 NA = Not applicable
Table 3.8 shows the median age at first birth among women age 25-49 years by selected background characteristics. According to the table, the median age at first birth for the women interviewed is 20 years. When considering place of residence, age at first birth for women in urban areas is about a year higher than that observed for their rural counterparts.
Regional variations in the median ages at first birth appear not to be significant, apart from Greater Accra, which has the highest median age at first birth, 21.6 years. The median age at first birth by level of education shows that women with secondary/higher education are more likely to delay childbearing than their counterparts with primary and middle/JSS education. It is also observed that highly educated women on the average have had their first birth at age 24, whereas their counterparts with less education first start giving birth at around 20 years.
Table 3,8 Median age at first birth
Median age at first birth among women age 25-49 years, by current age and selected background characteristics, Ghana 1993 Current age Background Ages chaxacteristic 25-29 30 34 35-39 40-44 45-49 25-49 Residence Urban 21.3 20.6 21.2 20.6 20.4 20,9 Rural 19.8 19.7 20.1 19.7 20.2 19.9 Region Western 19.5 19.0 (19.5) (18.5) * 19.1 Central 19.4 19.1 20.3 (20.0) (20.2) 19.7 Greater Accra 22.5 21.4 21.2 21.2 (20.4) 21,6 Volta 19.6 20.3 21.4 18.5 21.3 20.0 Eastern 20.3 20.4 20.5 (20.8) (19,4) 20.4 Ashanti 20.3 20,0 19.5 18.9 (19.5) 19.9 Brong-Ahafo 20.6 19,2 19.6 (19.01 (18.81 19,6 Northern 20.3 21.1 21.0 (21.41 (20.6) 20,8 Upper West (20.7) (21.11 * * * 21.0 Upper East 19.8 (18.4) (20.9) (20.9) (22.5) 20,1 Education No education 19.6 19.8 20.6 19.7 20.3 20.0 Primary 19.5 18.7 19.7 19.4 (19.6) 19.3 Middle/JSS 20.3 20.0 19.8 20.0 19.2 20.0 Secondary/Higher a 23.2 23.1 (24.1) * 23.8 Total 20.3 20.0 20.5 20.(I 20.2 20.2
Note: The medians for cohorts 15-19 and 20-24 could not be determined because half the women have not yet had a birth.
aMedians were not calculated for these cohorts because less than 5C1 percent of women in the age group x t o x + 4 have had a birth by age x. Numbers in parentheses axe based on 25-49 women; art asterisk indicates that the figure is based on fewer than 25 women and has been suppressed.
3.7
Adolescent Fertility
22 percent of teenage girls have started childbearing. Of this number, about 19 percent are already mothers, and 3 percent are pregnant with their first child. Of those who are 19 years old, 45 percent are mothers or pregnant for the first time.
Marked differentials exist in the background characteristics of teenage girls becoming mothers. In rural areas teenagers are more likely to become mothers earlier than their urban counterparts. Thus, in the rural areas, as many as 26 percent of teenage girls have begun childbearing, but only 16 percent in the urban areas.
Table 3.9 Teenage pregnancy and motherhood
Percentage of teenagers 15-19 who are mothers or pregnant with their first child, by selected background characteristics, Ghana 1993
Percentage who are: Percentage who have
Pregnant begun Number
Background Never with first child- of
characteristic pregnant Mothers child bearing teenagers Age 15 98.5 1.5 0.0 1.5 133 16 93.3 4.3 2.5 6.7 163 17 87.0 11.7 1.2 13.0 162 18 63.4 31.4 5.2 36.6 194 19 55.0 39.7 5.3 45.0 151 Residence Urban 83,6 14'.2 2.2 16.4 366 Rural 74.1 22.2 3.7 25.9 437 Region Western 73.5 21.7 4.8 26.5 83 Centa'al 66.7 28.4 4.9 33.3 81 Greater Accra 84.3 12.4 3.3 15,7 121 Volta 88.9 8.6 2.5 11.1 81 Eastern 77.8 22,2 0.0 22.2 108 Ashanti 77.5 19.2 3.3 22.5 120 Brong-Ahafo 74.7 22.0 3.3 25.3 91 Northern 78.8 18.2 3.0 21.2 66 Upper West * * * * 24 Upper East 82.1 17.9 0.0 17.9 28 Education No education 66.7 29.2 4.2 33.3 144 Primary 69.8 25.4 4.8 30.2 126 Middle/JSS 81.5 15.8 2.7 18.5 437 Secondary/Higher 93.7 6.3 0.0 6.3 96 Total 78.4 18.6 3.0 21.6 803
Note: An asterisk signifies that the figure is based on fewer than 25 cases, and has been suppressed.
The level of education also has a significant effect on teenage pregnancy. Of those with no education, 33 percent have started childbearing; the corresponding figures for those with primary and middle/JSS are 30 and 19 percent, respectively. Only 6 percent of teenagers with secondary or higher education have started childbearing. Regional variations are also observed in relation to teenage mothers. The highest proportion of teenage childbearing is in the Central Region, 33 percent, followed by Westem and Brong-Ahafo regions. The region with the lowest proportion of teenage childbearing is the Volta Region (11 percent).
Table 3.10 presents the distribution o f teenagers 15-19 years old by the number of children ever bom to them. The mean number of births is 0.2 for all, but rises to 0.5 for girls 19 years old. The most startling f'mding is that 5 percent o f the 19-year-olds have already given birth to two or more children.
Table 3.10 Children born to teenagers
Percent dis~ibution of teenagers 15-19 by number of children ever born (CEB), Ghana 1993
Age 0 1 2+
Number of Mean
children ever born number Number
of of
Total CEB teenagers
15 98.5 1.5 0.0 100.0 0.02 133 16 95.7 3.7 0.6 100.0 0.05 163 17 88.3 11.1 0.6 100.0 0.12 162 18 68.6 28.4 3.1 100.0 0.35 194 19 60.3 34.4 5.3 100.0 0.47 151 Total 81.4 16.6 2.0 100.0 0.21 803
