Proximate Determinants of Fertility Decline

Abstract

Existing literatures showed the role of proximate determinants (e.g postpartum or use of contraception) affect fertility. The issue of reducing fertility remains of great importance to many countries. Sudan is like African countries that have been slow transition of fertility. This study aims to identify intermediate determinants, and to verify their impact on fertility in Northern Sudan. The study also examines the impact of each proximate factor (Sexual exposure, contraception, postpartum infecundability and abortion) on fertility. The revised Bongaarts model of fertility was applied using the data for the most recent Sudan Demographic and Health Survey (DHS). The Total Fertility Rate (TFR) has decreased by 23% during the period in 1973-1999.The original model analysis showed that nuptiality had an important role in lowering of fertility during the specified period. Postpartum infecundability had the greatest impact on fertility reduction by 0.62, then sexual activity by 0.66, contraceptive use by 0.93. The revised model gave a fit somewhat similar than do earlier models.

Introduction

Current stalls in fertility decline have been observed in a few countries in sub-Saharan Africa, and so far no plausible common reason has been identified in the literature1. The situation of Arab regions has been more bizarre than others, the results of the study showed that, fertility rate in most of the countries of sub-Saharan Africa depends on the progress in education, especially for women of primary school age during the 1980s. In contrast, fertility has been less prevalent in countries where the development of education has not reached advanced levels1, as a situation of Arab countries, including Sudan, whereas the case of Sudan has been even more puzzling because women in the 1980s registered unusually high numbers of children. Looking at the region during the period 1980-1985, most women in the Arab countries had an average of 5.3 to 7.0 children, while in Sudan women registered an average of 4.6 children2. Researchers have assumed several hypotheses to explain the phenomenon of high fertility in the Arab region, including highlighting the specificity of Arab culture, and the role of Islamic religion (for example)3, which has led to a demographic transition in the Arab region. Other researchers4 found that, the low levels of education and the low status of women could explain the high fertility rate. Fertility plays a very important role in demographic studies, because it directly affects age and gender related factors. In addition, information on population size and distribution is vital to development, strategic and economic planning. Fertility patterns in most African and Asian countries were very similar in the 1960s, where Total Fertility Rate (TFR) estimated at 6.6 children per woman, which slightly higher than the average from low-income countries. In 1989, fertility levels in developing countries had varied markedly, these countries have experienced reduced fertility by more than 50%, (TFR) registered 6.2 children per woman in 1965 to 2.7 children per woman in 1989, and also the fertility rates in the United States began to decline in the 1960s and stabilised on a decline of 40% from 6.3 to 4.4 births per woman5. Sudan was one of the countries that registered a clear decline of fertility, data showed a decrease in the total fertility rate from six births per women in 1979 Sudan Fertility Survey (SFS)6, to about five children in the 1990 Sudan Demographic and Health Survey (SDHS)7.

The population of Sudan reached 10.3 million in the first national census carried out in 1956 and the population growth rate was 2.2% per annum. The second census was conducted in 1973 and the total population was 14.8 million with a 2.2% per annum growth rate. At the 1983 census, Sudan’s population reached 19.09 million. The estimated intercensal rate of growth, 1973-1983 was 2.8 % per annum. The fourth population census was carried in 1993 and reported a total population of 24.24 million. The estimated intercensal growth rate 1983-1993 was 2.4% per year. The four national censuses reported a young age structure with about 45% of the population below the age of 15 and 3% of age 60 and over. Table 1 shows a good indicator of prevalence of high fertility. Regarding the dependency ratio, the four censuses reported respectively105, 102, 98.8 and 102. The first census showed a crude birth rate of 51.7. The second population census reported a crude birth rate of 48.8 for the whole country, a mean of 4.8 children born to women (15-49 years), and a nationwide fertility rate of 7.3 births per woman. The second and third population censuses reported a 6.7 and 6.8 total fertility rates (TFRs) respectively. The fourth census showed the mean number of children ever born to all women was 6.5and a TFR of 5 children per woman. When the Demographic Health Survey (DHS)7 data was compiled between 1990 and 1995 and analysed, it was found that, most of the countries in Sub-Saharan Africa had a TFR of more than 6 children per woman8, despite these high fertility rates decrease and this decrease was documented in the number of countries in Africa, including Sudan. Sudan is one of those countries that have undergone fertility decline. Recently, some evidence showed that the total fertility rate decreased from 6 children per woman in (SFS79) (Sudan Fertility Survey) to about 5 children per woman in (Sudan Demographic Health Survey) (SDHS90). For long periods of time, women in Sudan did not use contraception among the population as a Survey of Maternal and Child Health (SMCH, 1993), Through this survey, the proportion of women who did not use contraception has been registered by 9.9%, while the percentage of users during that period was about 5%. If the levels of these ratios are compared with (SFS, 1979), there is a decrease in all the levels. Women, who did not use any method of contraception, were about 12.3%, while on the other hand, women, who use contraceptive methods, were about 6.4%.

Methods

Data

The main sources of fertility analysis in Sudan are the Sudan Census (SC) in 1973, Sudan Fertility Survey (SFS) in 1979, the Sudan Demographic and Health Survey (SDHS) in 1990, the Sudan Census (CS) in 1993, the Safe Motherhood Survey (SMS) in 1999 and Sudan Multiple Indicator Cluster Survey (MICS) in 2014. The study is effective to use analytical procedures for proximate determinants, which affects fertility with the original version by Bongaarts [1978] versus the revised models by Stover [1998] and Bongaarts [2015], with the proposed presentation on the model.

Methodology

The Bongaarts model framework is based on intermediate fertility variables and can be defined as biological and behavioural factors through economic, social and cultural factors, as well as environmental variables which affect fertility. These variables are called intermediate fertility variables. This term was first mentioned by (Davis and Blake)9. These are the relationships summed up by Bongaarts in the late 1970s, of the model of the relationship between four intermediate variables of fertility:

TFR= Cm* Cc* Ca* Ci*TF

TFR = total fertility rate, equal to the number of births a woman would have at the end of the reproductive years.

Cm = index of proportion married.

The index is calculated as a weighted average of the age groups of the percentages of Married women m(a), where (a)= age.

This indicator is calculated according to the following equation:

Cm = m(a)g(a)/g(a) (1)

Equation (1) can also be written as Cm= TFR / TM, so that

TFR= Cm*TM (2)

TM = total marital fertility rate, equal to the number of births a woman would have at the end of the reproductive years if she were to bear children at prevailing age-specific marital fertility rates and to remain married during the entire reproductive period (based on the fertility of married women aged 15-45);

TM= Cc*TNM (3)

Where,

CcIndex of noncontraception;

TMTotal Marital fertility rate;

Proximate Determinants of Fertility Decline

Cc=1-1.08 u e (4)

Where

uAverage proportion of married women currently using contraception (average of age-specific use rates);

e Average contraceptive effectiveness (average of use-effectiveness levels by age and method);

1.08 = Sterility correction coefficient;

We also find that Bongaarts focused its discussion on the effect of contraception on marital fertility, to link the indicator of non-contraception to the total fertility rate; equation (3) is substituted in equation (2), to produce the following equation:

TFR= Cm*Cc*TNM (5)

Bongaarts explained that, effective Induced abortion measurement tools are somewhat lacking. The most detailed studies of this topic have been made by Robert Potter10. Where he showed in his work the following:

An induced abortion always averts less than one birth.

The number of births averted per induced abortion is largely independent of the age of the woman.

The number of births averted per induced abortion is strongly influenced by the practice of contraception following the induced abortion. In the absence of contraception, an induced abortion averts about 0.4 births, while about 0.8 births are averted when moderately effective contraception is practiced. In general from the concept has been circulated to infants, the births averted per induced abortion, (b), may be estimated with the equation:

b=0.4(1+u) (6)

Bongaarts identified (A) as an overall measure of the incidence of induced abortion is provided by the total abortion rate

A=b*TA= 0.41+u*TA) (7)

Where

A The average number of births that have been averted per woman by the end of the reproductive years;

Bongaarts also noted that induced abortion among the population is defined as a ratio between the total fertility rate of, TFR, and the total fertility estimated without induced abortion TFR+A.

Ca= TFR/TFR+A (8)

Where

CaThe proportion by which fertility is reduced as the consequence of the practice of induced abortion;

According to the equation between TFR andTNM, modifying equation (5) tobecome now:

TFR= Cm*Cc*Ca*TNM (9)

Bongaarts's vision in Lactational Infecundability Ci is that breastfeeding has a significant effect in increasing the period of birth, and reducing natural fertility, by Comparing average length of period between pregnancy and the subsequent pregnancy in the case of breastfeeding or not, so the period of birth (pregnancy and other pregnancy) has been divided into four components11: An infecundable interval immediately following a birth. In the absence of lactation, this segment averages about 1.5 months, while prolonged lactation results in infecundable periods of up to two years. The duration of this birth-interval segment is usually measured from birth to the first postpartum menses, because the return of menses closely coincides with the return of ovulation. Waiting time to conception, this starts at the first ovulation following birth and ends with a conception. Although few measurements are available, existing observations indicate that population averages for this interval range from a low of about 5 months to high values that only rarely exceed 10 months, with typical values around 7.5 months12. Time added by spontaneous intrauterine mortality. In cases where a conception does not end in a live birth, the duration of a shortened pregnancy and another waiting time to conception are added to the birth interval. On average the time added by intrauterine mortality equals about 2 months per birth interval. A nine-month gestation period ending in a live birth.

In the case of non- lactation, Bongaarts also stated that, the standard average period of birth can be estimated to equal 20 months as follows: 1.5 + 7.5 + 2 + 9 - 20 months, In the case of lactation it equals the average total duration of the infecundable period plus 18.5 months (7.5 + 2 + 9). The average period of birth (by lactation and without and with lactation) is called lactational Infecundability:

Ci= 20/18.5+i (10)

Where

CiIndex of lactational infecundability;

i Average duration (in months) of infecundability from birth to the first postpartum ovulation (menses)

Bongaarts also found that, the relationship between lactation and the natural marital fertility rate could be represented by the following equation :

TNM= Ci* TF (11)

Where

TF TotalFecundity rate equal to the total natural marital fertility rate in the absence of lactation

Given the equations (10) and (11) and in the absence of lactation, (i=1.5) month and thus the (Ci=1). This results in equal total fertility rate TF with the total natural marital fertility rate TNM, i.e. that TNM=TF. If the average interval of non- Reproductive(i=18.9), it gives the value of index of lactational infecundability equal to Ci=0.53 and is also taken into account that

TNM=0.53TF.

All of these variables {m(a), ex(a), u(a), o(a), e(a), r(a), i(a), f(a), fm (a)}14can be estimated from DHS surveys using procedures clarified by (Bongaarts & Potter)13 except two of which are r(a) and ab(a).

Marriage IndexCm :

Since both)Bongaarts & Stover(15 take into account the issue of extramarital sexual activity, the revised models are the same as the original model because marriage is the only relationship that binds couples together, and leads to having children, cases of having children outside the marital relationship rarely Occurs in Sudan and even it does exist it is not officially registered.

Contraception Index Cc :

Since the fertility Adjustment Factor r(a) was developed to modify the contraception prevalence among married women, to calculate the higher fertility prevalence of low-fertility women, and to estimate this factor we applied a revised version of the method developed by (Bongaarts and Kirmeyer)16, the method introduces a new variable fnca which is defined as the fertility rate of sexually exposed women, and can be observed in the absence of abortion and postpartum infertility as follows:

fnca=f(a)/Cm*aCi*aCa*a

As stipulated in the table of revised equations for the model of approximate determinants by age cohorts:

Cc*a=1-r*a(u*a-o(a))e*(a)]

Then

fnca=ff*a[1-r*a(u*a-o(a))e*(a)]

Postpartum Infecundability Ci :

This indicator has not been revised and no observations have been made, but the nature breast-feeding index equations should be taken into consideration in the model of the specific age groups. The revised indicators estimate data obtained from various sources for the period (1973-1999), and attention was also given to summarising the differences between the three models (the original model, the revised models).

Indexes are calculates as follows:

First: Marriage Index(Cm): Marriage index of (SC1973):

Number of women, births and marriages in Sudan according to 1973 Census

FaASFR= Scheduling of age group fertility rates at age (a)

ma= The proportions of the currently married, as themashould include the question of consent but exposures and others give a weight of 0.5.

Cm= 6.820894/9.58972=0.711271

Second: Contraception Index(Cc):

The effectiveness estimates for contraception are difficult to obtain because of the scarcity of availability. Effective average use (e) is estimated as a weighted average of the method of classification use of effectiveness levelse(m), with weights equal to the proportion of women who use a given method

Cc=0.8756

Third: Induced Abortion Index(Ca):

In this indicator we find that the value of Ca equals

Ca= TFRTFR+A

A=b*TA= 0.41+u*TA

Ca= TFRTFR+0.41+u*TA

Since the documentation of the induced abortion statistics is not available in the Sudan because its prevention (illegal), we assumed the absence of induced abortion and therefore the value of Ca=1.0 .

Fourth: Postpartum Infecundability Index (Ci):

In this indicator we find that the value of Ci equals

Ci= 2018.5+i

Therefore the estimate is direct, but the value i(average of duration to Postpartum Infecundability) is not available, but we can get an approximate value of the duration of lactation (B), from the following equation:

i=1.735 exp⁡(0.1396*B-0.001872*B2)

Postpartum Infecundability index of (SMS1999):

According to the Federal Ministry of Health in the Sudan, the average duration of breastfeeding is 20 months.

B=20 , i=1.753exp⁡(0.1396*(20)-0.001872*(20)2i=13.5362

Ci= 2018.5+13.5=0.625

Indexes and total fertility rates Estimated &observed in Sudan according to data sources (1973 – 1999)

Results

Estimates of proximate determinants and revised indicators were obtained from the data collected from DHS surveys during the period 1973-1999 in Sudan. Differences between three models examined (i.e., the original version and the revisions by JS and JB) were studied. Figures (1) plot the (John Stover revision) (JS) and (John Bongaarts revision)(JB) indicators versus the original models, and from the Figures we note that the indicators are positively correlated, the strongest correlations observed in Cm, Ci, and then Cc. We note that, the issue of sexual intercourse is not permitted in Sudan, so the revised model indicators have been similar to those of the original form. Sudan is also a country, where sex is forbidden outside of marriage, Table (1) illustrates that, the three models gave a very similar variation to the Cm ratio was 0.66. We also find that, the three models gave a very similar variation to the Cm index, and the differences between JS and JB models equal (0.66 versus 0.66), and slight differences simple in both Ccand Ci, where the difference between JS and JB models in the index Cc value (0.94 versus 0.96), in the Ci index is worth (0.62 versus 0.61). This slight variation of the factors of religion and culture that exist in the Sudan can be attributed, if the models are applied in several countries, Differences in factors of religion and culture will greatly influence the factors involved., and therefore we will find very clear and significant differences in all the indicators of the fundamental difference in the causes of these differences. Figure (2) represents three indicator charts by TFR and as expected for Cm indicators and Cc decreases at different time intervals (data source surveys) and their higher-to-lower fertility transmission, the Ci indicator was fairly stable with higher-to-lower fertility movement as well.

Average of three model indicators in Sudan during the period (1973-1999)

Discussion

Study is based on the assessment of the proximate determinants of fertility and the role of the selected socio-economic variables in influencing the proximate determinants and fertility in Northern Sudan. The study is also based on an analytical basis to estimate the impact of approximate determinants on fertility, so the study of quantitative fertility was based on the study using the Bongaarts framework19, the results of the study based on its original model through revisions that will provide Stover29, where the effect of each proximate determinant on fertility (Cm,Cc,CiandCa) was calculated.Determining fertility behaviour accurately plays a key role in the development and implementation of programmes, the results of the study confirmed that the approximate determinants as outlined by Bongaarts had a significant impact on fertility in Sudan. According to the revision of Bongaarts effects of postpartum infecundability was the most influential (Ci=0.61), followed by sexual activity(Cm=0.66) then contraceptive use (Cc=0.96) and abortion (Ca=1.0). The results didn't differ much according to the revision of Stover effects of postpartum infecundability was the most influential (Ci=0.62), followed by sexual activity(Cm=0.66) then contraceptive use (Cc=0.94) and abortion (Ca=1.0). It is noted that women in Sudan rely heavily on postpartum infecundability to reducefertility21. Nuptiality plays an important role in reduction of fertility during the specified period. The index of marriage (Cm) was 0.80 during SFS-79, 0.68 during the SDHS-90, and 0.60 during SMS-99 resulting in overall fertility reduction by 22.5 %.

Conclusion

The original approximate determinant model developed by Bongaarts opens the door wide for multiple uses to determine fertility levels and trends, and revisions on the model have refined in calculating fertility indicators. When applied to Sudan data, the revised models gave somewhat similar results to the original models due to the fact that certain determinants, such as Abortion, which was adopted as (Ca=0.1), were proven to be prohibited in Sudan. The specified age model was also considered the proper basis for the model as a whole, during which the fertility impact was estimated. Study analysis showed that, the approximate determinants of both original and revised models had a significant impact on fertility in Sudan. Therefore, strategies have to be designed to promote the breast feeding practices and the contraceptive use among the reproductive women.

References

Goujon A, Lutz W and Kc S. Education stalls and subsequent stalls in African fertility: A descriptive overview. Demographic Research 2015 ;33():1281-96.

Courbage Y. Economic and political issues of fertility transition in the Arab world: Answers and open questions. Population and Environment 1999; 20(4): 353-80.

Cohen B, and House WJ. Women's urban labour market status in developing countries: how well do they fare in Khartoum, Sudan?. The Journal of Development Studies1993;29(3): 461-483.

Department of Statistics. Sudan Demographic and Health Survey, Principal Report Volume II, Khartoum. 1992.

Mboup G, and Saha, T. Fertility levels, trends, and Differentials: Demographic and Health Surveys. Comparative Study. Calverton, Maryland: Macro International 1998; (28): 77.

Davis K and Blake J. Social structure and fertility: An analytic framework. Economic Development and Cultural Change 1956; 4(4): 211-35.

Potter Jr RG, Wyon JB, Parker M and Gordon JE. A case study of birth interval dynamics. Population Studies 1965; 19(1): 81-96.

Bongaarts J. Modeling the fertility impact of the proximate determinants: Time for a tune-up. Demographic Research 2015; 33(19): 535-60.

Bongaarts J andKirmeyer S. Estimating the impact of contraceptive prevalence on fertility: Aggregate and age-specific versions of a model. Population Council 1980.

Bongaarts J and Westoff CF. The potential role of contraception in reducing abortion. Studies in Family Planning 2000; 31(3): 193-202.

Bongaarts J. The fertility-inhibiting effects of the intermediate fertility variables. Studies in family planning 1982; 179-89.

Department of Statistics. The Fourth Population Census, 1993, Principal Report Volume I & II, Khartoum 1995.

Hobcraft J and Little RJ. Fertility exposure analysis: A new method for assessing the contribution of proximate determinants to fertility differentials. Population Studies 1984; 38(1): 21-45.

Kirk D, Pillet B. Fertility levels, trends, and differentials in sub-Saharan Africa in the 1980s and 1990s. Studies in family planning 1998; 1-22.

Mahfouz MS. Fertility in Northern Sudan (1979–1999): levels, trends and determinants. In Proceedings of the 26th International Population Conference of the IUSSP. PU Office of Population Research 2009.

Stover J. Revising the proximate determinants of fertility framework: What have we learned in the past 20 years?. Studies in family planning 1998; 255-67.