Child marriage and its impact on fertility and fertility-related outcomes in South Asian countries

Abstract

Spanish

French

Although South Asia (SA) is reported as a home of child marriage, the association of child marriage with fertility and fertility-related outcomes in this region is poorly explored. The most recent data of the Demographic and Health Survey of six SA countries – Afghanistan, Bangladesh, India, Maldives, Nepal and Pakistan – have been used in this article. The unit of analysis is 584,213 currently married women aged 20–49. The outcomes of interest are fertility and fertility-outcomes. Quantitatively important and reliable estimates were obtained from the statistical analyses. The results are presented by odds ratios with 95% CIs. Findings reveal that, overall, 42.1% of the respondents were married-off before age 18. The prevalence of child marriage was lowest in the Maldives and highest in Bangladesh at 20.5% and 74.4% respectively. The likelihood of early childbirth and repeated childbirth were significantly (p < 0.001) lower and that for high fertility, unintended pregnancy, lifetime pregnancy termination and use of a modern contraceptive method was significantly higher in the child married women compared to their adult married counterparts. Reforms should aim to have more girls remain in schooling for both personal and overall societal development and also to reduce adverse reproductive outcomes caused by child marriage.

Keywords

Child marriage child rights fertility outcomes South Asian countries

Introduction

It is commonly believed that childbearing at adolescent age outside of marriage is rare in South Asia societies; therefore, child marriages and ensuing adolescent motherhood are very common in this region (UNICEF, 2019). South Asia is considered as a home of child marriage (UNICEF, 2014). Child marriage – marriage before age 18 – is considered a human rights violation, which affects millions of women around the world. Ten countries in South Asia and Sub-Saharan Africa possess the highest rate of child marriage; Niger has the highest prevalence of child marriage and Bangladesh has the highest rate of marriage involving girls under age 15. Globally, more than 40% of women aged 20–24 in 2012 had been married before the age of 18, of which half were from South Asia and India alone accounted for one-third (UNICEF, 2014).

Although child marriage, particularly of girls, has long been a global concern due to its various consequences socioeconomically and for maternal and child health, globally over 700 million women each year entered into the formal union before age 18 (Lendhardt, 2016). The literature on child marriages in Bangladesh added that child marriages significantly impede girls’ opportunity to receive education and or to be empowered in a patriarchal society (Mim, 2017). In a study on 36 countries from Sub-Saharan Africa and South West Asia it was found that child marriage driven by socioeconomic and cultural factors is associated with decreased schooling and a higher likelihood of dropping out from secondary level schooling (Delprato et al., 2015). Girls getting married at a young age tend to suffer from psychological disadvantage, such as lack of self-esteem and depression (Ahmed et al., 2013), and sexual abuse, with increasing risks of sexually transmitted diseases and HIV (Clark, 2004; Clark et al., 2006).

Child marriage has wide-ranging adverse effects on girls beyond education. A landmark review of the literature suggests that girls’ child marriage is significantly associated with increased likelihood of early motherhood (Raj et al., 2009), lifetime fertility (Onagoruwa and Wodon, 2018), unintended pregnancy (Godha et al., 2013; Huda et al., 2018), abortion (Godha et al., 2013) and pregnancy termination (Yaya et al., 2019). Research findings suggest that child married women are at higher risk of stillbirth, miscarriage and stunted and underweight children than their adult married counterparts (Godha et al., 2013). Unfortunately, there is a high correlation between child marriage and the reduced likelihood of prenatal and postnatal care by skilled medical care providers. In addition, child marriage notably increases the risk of delivery assistance by unskilled medical providers and also delivery at home (Nasrullah et al., 2013; Sekine and Carter, 2019).

The practice of child marriage prevails in every region of the world, and occurs across cultures, religions and ethnicities. Patriarchy, coercion, social customs and cultural norms are important and major social determinants of child marriage in India (Seth et al., 2018), and in South Asia. Child marriage, often referred to as early marriage, is considered forced marriage and violation of human rights because children are not informed in advance about their marriage partners and they are forced into marriages, willingly accepting marriage as their allotted fate (Bhanji and Punjani, 2014).

In the South Asian countries, men enjoy control over household resources and are the sole decision-maker in the family, and women are considered men’s subordinate. According to Subramanian, ‘early marriage is a part of the wider practice of female seclusion through their subordination to men and deprivation of equal access to social and material resources’ (Subramanian, 2008: 161). In Muslim culture, ‘remaining virgin and keeping virgin’ is another factor compelling women to get married before they reach adulthood. Sexual contact and hence procreation outside of marriage is prohibited in Islam. Moreover, in Muslim culture, the elders are respected in the family; hence, parents who believe that marriage will protect their daughters from illicit sexual contact can defend early marriage for their young daughters. Therefore, elderly family members put their children under social or emotional pressure and encourage them to marry as part of their family values and traditions (Bhanji and Punjani, 2014; Kamal et al., 2015). Parental education, women’s education, occupation, place of residence, husband’s education, and religion are also identified as important determinants of child marriage in Bangladesh as well as in other South Asian countries (Hotchkiss et al., 2016; Kamal et al., 2015; Mpilambo et al., 2017).

Despite the pervasiveness of child marriage in South Asian countries, the association between child marriage and adverse social, reproductive and health outcomes has yet to be explored adequately. The scarcity of literature on these issues may hinder the implementation of effective policies and programmes among policy makers in South Asian countries, and urges the need for research and study on the women who are most as risk. The aim of this study is to explore the prevalence of child marriage and its association with reproductive outcomes among currently married women aged 20–49 using the most recent Demographic and Health Surveys (DHS) data sets from six South Asian countries.

Data and methods

Data source

In this study, we used pooled data from the most recent DHS conducted during 2014 to 2018 across six South Asian countries: Afghanistan (2015), Bangladesh (2014), India (2015–16), Maldives (2016–17), Nepal (2016) and Pakistan (2017–18). We did not include the data set for Sri Lanka as it was not up to date. The DHS are nationally representative surveys, which collect a wide variety of self-reported data mainly focusing on various indicators of fertility, fertility preference, contraception, child mortality, reproductive health, maternal and child health, etc. DHS follow standard techniques and protocols prescribed by the World Health Organization (WHO). DHS data sets are available for research purposes upon request (http://dhsprogram.com/data/available-datasets.cfm). We restricted our analyses to currently married women aged 20–49 years. We considered the age range based on the length of childbearing. However, the women younger than 20 years old were not included in the analysis because a large proportion of girls remained unmarried in this group. Therefore, we believe that women aged 20–49 can provide a more accurate picture of the outcomes of interest. The pooled sample size stood at 27,632 for Afghanistan, 15,840 for Bangladesh, 510,758 for India, 6,156 for Maldives, 9,487 for Nepal and 14,340 for Pakistan. Data were merged into a file and finally the total sample size stood at 584,213 currently married women.

Outcomes of interest

The outcomes of interest of this study are:

Early fertility: This is a binary indicator and defined as whether the women have had their first childbirth within 12 months of their marriage.

Rapid repeat of childbirth: This is also a binary outcome, defined as whether the last childbirth occurred within the last 24 months of the previous childbirth.

Unintended pregnancy: This is measured with a binary indicator of whether the respondent has had one or more child in the last five years where the pregnancy was wanted or not wanted at all. The ‘unintended’ pregnancy was coded as 1, and ‘wanted’ pregnancy was coded as 0.

Pregnancy termination: This is a binary indicator of whether a woman reported ever having had a pregnancy termination. Pregnancy termination may result from miscarriage, abortion, or stillbirth.

Use of modern contraceptive method: This has been defined as whether the respondents were using any modern contraceptive method at the time of the survey. The response was recorded either by ‘yes’ or ‘no’. The definition of ‘modern contraceptive methods’ is provided in each report of the respective study countries.

Lifetime fertility: Lifetime fertility is defined by the total number children ever born to women during their lifetime. Respondents were classified as having ‘high fertility’ if they had four or more childbirths. If a woman had given four or more live births during her lifetime, she was coded as ‘1’ and ‘0’ otherwise.

Explanatory variables and covariates

The main explanatory variable of this study is women’s age at first marriage. Following the definition provided by the United Nations (UN), the women who got married before age 18 have been termed as ‘child married’ and those who were married at age 18 or later have been termed as ‘adult married’.

A common set of covariates are used in this study to control the results of the multivariate analyses. The variables used as covariates are: current age of the women (20–24, 25–29, 30–34, 35–39, 40–44 and 45–49), place of residence (urban and rural), women’s education (no education, primary, secondary and higher), sex of household head (male, female), current working status of women (working, not working), access to media (no, yes), religion (Hindu, Muslim, Buddhist, Christian and others) and wealth index (poorest, poorer, middle, richer and richest). Several demographic questions are given in the survey of DHS such as materials used for house roofs, walls, and floors; type of drinking water; accessibility of household durable goods; and some questions related to economic status. Based on these indicators, household wealth and economic status has been measured by using principal component analysis and has been termed as ‘wealth index’. The details of the construction of the household ‘wealth index’ are provided in the DHS reports of each country. This was done to capture the picture of the overall situation in South Asia.

Statistical analysis

Descriptive and multivariate analyses were used in this study. Descriptive statistics include univariate statistics on the dependent and independent variables. The bivariate analysis, namely chi-square tests, was conducted to examine the associations between the outcomes of interest and explanatory variable age at marriage. We constructed a series of binary logistic regression models to examine the influence of child marriage on the selected outcomes of interest controlling for the covariates as mentioned above. The results of the logistic regression analyses are presented by odds ratios (ORs) with 95% confidence intervals (CI). The level of significance was set at 5%. Prior to running the multivariate analyses, multicollinearity among the predictor and covariates was checked and found to be nonexistent as the highest variance inflation factor (VIF) was estimated to be 1.96. All the analyses were performed using unweighted data as stratification and weighting factors differ from country to country. All of the statistical analyses were performed by IBM SPSS v21 (IBM SPSS Statistics, IBM Corporation, Somers, NY).

Ethical considerations

This study used de-identified information provided by the respondents. We used publicly available data obtained from MEASURE DHS. MEASURE DHS followed all ethical procedures needed for data collection from the respondents. All DHS have been approved by the ICF International and Institutional Review Board to ensure that the protocols comply with the World Health Organization and the US Department of Health and Human Services for the protection of human subjects. Therefore, this study does not require more ethical approval from any other institution.

Results

Table 1 captures the background characteristics of the sample of currently married women of age 20–49 years in South Asia. The proportion of women aged 20–24 was 15.8% and those aged 25–29 was more than 20%. Slightly fewer than 71% of the women lived in rural areas. The proportion of women with no formal education was 38.1% and more than 38% of the women had secondary level education and only 8.8% had higher education. It is important to note that 87% of the women were from male headed households. Approximately 25% of the women fall into the working category and more than 76% had access to media. The majority (74%) of the women were Hindu by religion and about 14.7% were Muslims. Slightly fewer than 20% of the women were from the poorest and 18.5% were from the richest households.

Table 1.

Descriptive statistics of the respondents by their background characteristics.

Characteristics	Sample		Type of marriage (%)		Chi-square value
Characteristics	N	%	Adult marriage	Child marriage	Chi-square value
Country					9211^a
Afghanistan	27,632	4.7	53.4	46.6
Bangladesh	15,840	2.7	25.8	74.2
India	510,758	87.4	59.0	41.0
Maldives	6,156	1.1	79.5	20.5
Nepal	9,487	1.6	45.6	54.4
Pakistan	14,340	2.5	64.4	35.6
Current age					1845^a
20–24	92,528	15.8	60.7	39.3
25–29	118,980	20.4	61.9	38.1
30–34	107,045	18.3	57.2	42.8
35–39	100,669	17.2	56.0	44.0
40–44	84,812	14.5	53.7	46.3
45–49	80,179	13.7	56.4	43.6
Place of residence					3746^a
Urban	171,435	29.3	64.3	35.7
Rural	412,778	70.7	55.3	44.7
Women’s level of education					48254^a
No education	222,608	38.1	44.9	55.1
Primary	87,631	15.0	47.8	52.2
Secondary	222,338	38.1	66.8	33.2
Higher	51,636	8.8	91.4	8.6
Sex of household head					323^a
Male	508,104	87.0	57.5	42.5
Female	76,109	13.0	61.5	38.5
Working status of women					519^a
Not working	121,098	74.6	57.7	42.3
Working	41,290	25.4	51.1	48.9
Access to media					9064^a
No	139,517	23.9	46.7	53.3
Yes	444,533	76.1	61.5	38.5
Religion					6652^a
Hindu	396,566	74.0	56.8	43.2
Muslim	79,039	14.7	51.5	48.5
Buddhist	6,848	1.3	67.7	32.3
Christian	35,466	6.6	72.5	27.5
Others	18,165	3.4	75.2	24.8
Wealth index					18513^a
Poorest	115,049	19.7	47.1	52.9
Poorer	125,178	21.4	51.4	48.6
Middle	121,743	20.8	56.7	43.3
Richer	114,415	19.6	62.6	37.4
Richest	107,828	18.5	73.3	26.7
Total	584,213	100.0	57.9	42.1

Note: The total number of women for different categories of the variables may not be always 584,213 due to various missing cases. ^aLevel of significance p < 0.001.

Table 1 shows that the incidence of child marriage among women by current age was highest in the age group 40–44 (46.3%) and lowest, at 38.1%, in the age group 25–29. Child marriage was significantly more prevalent in rural than urban areas. Women’s level of education was significantly inversely associated with child marriage; the rate of child marriage was highest, at 55.1%, among the women with no formal education and lowest, at 8.6%, among the women who had a higher level of education. Male headed households had a higher incidence of child marriage than the female headed households (42.5% vs 38.5%). The prevalence of child marriage differed by 6.6% for working status of women, the incidence being higher in the working group (48.9%) than in the non-working group of women (42.3%). The distribution of child marriage according to access to media was found to be significantly lower among women who had access to media (38.5%) than among women who had no access to media (53.3%). The incidence of child marriage was more prevalent among the Muslim women than for their counterparts in the other religious groups. The wealth index showed a negative association with child marriage. Child marriage reported in the poorest households constituted 52.9% of the sample of women, whereas this proportion was lowest, at 26.7%, in the households in the richest wealth quintiles. Overall, the prevalence of child marriage in the study countries was 42.1%; indicating that 42.1% women get married before they reach the age of 18.

Although child marriage in some South Asian countries has decreased over time, it is still high compared to other developing countries. The variables that we considered in our study showed that education can weaken the institution of marriage and it undermines men’s authority. This outcome becomes a threat to men’s authority and educating young girls delays marriage. Therefore, poor families tend to reject educating their daughters. As a result, young girls who have not been educated lead a life dependent upon their spouse. They cannot work in any job and continue their life as housewives. This scenario is more common in rural areas.

Young girls who cannot achieve their independence economically tend to have more children than those who can work and become economically independent. Our findings are supported by the sociological view which states that child marriage leads to other social problems that require regionally specific solutions such as poverty, education and health. Moreover, women who married at a young age are denied the right to education. Another important outcome is that child marriages are much more common in poor families than in wealthy families. Stronger protective laws are needed to end child marriage, and addressing poverty can play a critical role in reducing child marriage.

Figure 1 displays the prevalence of child marriage by country for women aged 20–49. The prevalence of child marriage in the women aged 20–49 was highest in Bangladesh (74.4%), followed by Nepal (54.4%) and Afghanistan (46.6%), India (41%), Pakistan (35.6%) and the Maldives (20.5%). Figure 2 displays the prevalence of child marriage by country for women aged 20–24. When these two figures are compared, it is seen that the percentage of child marriage in Bangladesh, Nepal, India and the Maldives decreased over time. There is a sharp decrease in the percentage of child marriage in the Maldives from 20.5% to 3.4%. In contrast, the percentage of child marriage has increased in Afghanistan and Pakistan. Out of these six countries, only the Maldives has made significant progress in reducing child marriage. Although the percentage of child marriage decreases in some countries, the current percentage still remains very high. More reforms, such as empowering girls, educating girls, encouraging supportive laws, etc., are needed to reduce child marriage in these countries. To eliminate the discrimination of women, most of the South Asian countries have passed laws specifying a minimum marriage age of 18 years, consistent with the definition of childhood articulated in the Convention on the Rights of the Child. However, these laws are not fully implemented in these countries. As the results of our study show, child marriage can adversely affect women along several dimensions. Some of these variables are religion, household head, education and work status. When child marriages are evaluated by religion, it is clear that such marriages are less frequent in Christian societies than in other religions. In Muslim culture, it is found that child marriages are higher than for other religions. The main reason why child marriages are more common in the Muslim community is the difference in family structure. Parents marry their daughters at a young age to protect them from illegal sexual contact because sexual contact and hence childbirth outside of marriage is prohibited in Islam.

Figure 1.

Prevalence of child marriage by country for women aged 20–49 (%).

Figure 2.

Prevalence of child marriage by country for women aged 20–24 (%).

In addition, most of the marriages are arranged or orchestrated by the parents. Therefore, these young girls have less choice at which age they will marry or with whom they will marry. In addition, as the educational level of the society decreases, the rate of child marriages increases. Therefore, increasing the educational levels of these societies is important in reducing child marriages. Furthermore, we found a strong correlation between a woman’s age at marriage and the level of education she receives. Early marriage negatively affects women in terms of early termination of education. Most of the women who had child marriages have no education or have only primary education. Very few women have received higher education. In contrast, a woman who married as an adult received more education compared to those who were child married.

There is a positive relationship between marriage age and schooling. Child marriage is likely to be a significant barrier to women’s education. In addition, the increasing interest of women in education increases the age of marriage. Increasing the educational level of women positively affects their working status. Most of the women who had higher education tended to have a job. This outcome may affect the power, status and autonomy of women within the household and woman can become household head. The variables household head, education and work status are positively correlated with each other and these variables should be carefully considered in order to reduce child marriages in the South Asian countries.

The distribution of fertility and fertility-outcomes cross classified by type of marriage is reported in Table 2. It is found that 17% of the child married women have had their first childbirth within 12 months of their marriage and the rate among the adult married women was 28.9%. Almost 29.2% of the child married women have had repeated childbirths as against 30.1% among those of adult marriage. Of the child married women, 36.5% reported that they had higher fertility; the corresponding figure in the adult married women was only 17.6%. Unintended pregnancy was significantly higher in the women married as a child than those married as adults. Lifetime pregnancy termination was also significantly higher in the child married women compared to their adult married counterparts (18% vs 16%). Current use of a modern contraceptive method significantly differed by type of marriage; the prevalence was higher in the child married women than among those married as adults (49% vs 39%).

Table 2.

Distribution of women by fertility and fertility-outcomes and type of marriage.

Outcomes of interest	Sample		Type of marriage (%)		Chi-square value
Outcomes of interest	N	%	Adult marriage	Child marriage	Chi-square value
Early childbirth					9706^a
No	381,237	76.5	71.1	83.0
Yes	117,243	23.5	28.9	17.0
Repeated childbirth					41^a
No	307,825	70.3	69.9	70.8
Yes	129,988	29.7	30.1	29.2
High fertility					25325^a
No	435,470	74.5	82.4	63.5
Yes	148,743	25.5	17.6	36.5
Wantedness of last child					505^a
Wanted	199,196	90.0	91.1	88.1
Unintended	22,145	10.0	8.9	11.9
Ever had terminated pregnancy					179^a
No	486,287	83.2	83.5	82.2
Yes	97,897	16.8	16.5	17.8
Currently using modern FPM					5169^a
No	334,613	57.3	60.7	51.0
Yes	249,600	42.7	39.3	49.0

Note: The total number of women for different categories of the variables may not be always 584,213 due to various missing cases. FPM = family planning method. ^aLevel of significance p < 0.001.

The results of the multivariate logistic regression analyses of the effect of child marriage on fertility and fertility-related outcomes are presented in Table 3. The estimated OR and associated CI were adjusted for current age, place of residence, women’s level of education, sex of household head, women’s working status, access to media, religion and wealth index. Findings reveal that the women having experience of child marriage were significantly at decreased risk of childbirth in the first year of marriage than their counterparts women who were married as adults (OR = 0.51, 95% CI = 0.49–0.43). The likelihood of repeated childbirths significantly decreased by 11% (OR = 0.89, 95% CI = 0.86–0.92) compared to adult married women. The estimated OR of high fertility was 2.27 times higher (OR = 2.27, 95% CI = 2.10–2.49) in the child married women than their counterparts who were married as adults. Compared to the adult married women, child married women were significantly (p < 0.001) at higher risk of unintended pregnancy (OR = 1.31, 95% CI = 1.22–1.41), lifetime pregnancy termination (OR = 1.15, 95% CI = 1.11–1.19) and current use of any FPM (OR = 1.46, 95% CI = 1.42–1.50).

Table 3.

Results of multivariate logistic regression analyses showing the effect of child marriage on fertility and fertility-outcomes.

Outcomes of interest	Unadjusted OR with 95% CI			Adjusted OR with 95% CI
	OR	95% CI		AOR	95% CI
	OR	Lower	Upper	AOR	Lower	Upper
Early childbirth
No	1.00	—	—	1.00	—	—
Yes	0.50^a	0.50	0.51	0.51^a	0.49	0.53
Repeated childbirth
No	1.00	—	—	1.00	—	—
Yes	0.96^a	0.95	0.97	0.89^a	0.86	0.92
High fertility
No	1.00	—	—	1.00	—	—
Yes	2.68^a	2.65	2.72	2.27^a	2.10	2.49
Wantedness of last child
Wanted	1.00	—	—	1.00	—	—
Unintended	1.38^a	1.35	1.42	1.31^a	1.22	1.41
Ever had terminated pregnancy
No	1.00	—	—	1.00	—	—
Yes	1.09^a	1.10	1.12	1.15^a	1.11	1.19
Currently using any modern FPM
No	1.00	—	—	1.00	—	—
Yes	1.48^a	1.47	1.50	1.46^a	1.42	1.50

Level of significance p < 0.001.

Discussion and conclusion

To our knowledge, this is the first study in six South Asian countries which has extensively examined the prevalence of child marriage and its association with fertility and fertility-related outcomes. We examined the impact of child marriage on various indicators of fertility, such as childbirth in the first year of marriage, rapid repeated childbirth, i.e. spacing between pregnancies, and the use of any modern contraceptive method. Low control over fertility outcomes such as high fertility, unintended pregnancy and pregnancy termination was also examined. As found in this study, child marriage is common in South Asia. The findings revealed large differentiation in the prevalence of child marriage across the South Asian countries. Bangladesh is the country in South Asia where child marriage is most prevalent: 74.4% of women get married before they reach age 18. In two countries, Nepal and Afghanistan, about half of women aged 20–49 are married-off as a child. Approximately 41% women in India and 35.6% women in Pakistan experience child marriages. Child marriage frequency was low in the Maldives compared to the other countries, at 20.5%. Overall, more than 42% women in these South Asian countries experience marriage before they reach the age of 18. In an earlier study on four South Asian countries it was reported that the median age at first marriage of women aged 20–24 was lowest in Bangladesh, followed by India, Nepal and Pakistan (Godha et al., 2013). Although we have similar results for Bangladesh, we have observed different results for India, Nepal and Pakistan in our study. In our study, the median age at first marriage of women aged 25–49 was lowest in Bangladesh, followed by Nepal, Afghanistan, India, Pakistan and the Maldives.

The findings of our study suggest that, compared to the adult married women, the child married women were at decreased risk of childbirth in the first year of marriage and repeated childbirth. These findings are consistent with those of the study conducted on four South Asian countries by Godha et al. (2013), and also in another study of 34 countries in Sub-Saharan Africa (Yaya et al., 2019). This may be partly ascribed to the biological and physical immaturity among the child brides compared to their counterparts married as adults. Furthermore, the negative association between early childbirth and repeated pregnancy can possibly be attributed to low fecundity at very young ages of the child married women and low coital frequency, or both (Godha et al., 2013).

A plausible reason for the increased likelihood of higher fertility among the child married women compared to their adult counterparts may be that the child married women have been married for a longer time than the women married as adults. Moreover, most of the child married women compared to those married as adults may have achieved desired family size or wanted to space out more time for the next pregnancy, resulting in their most recent childbirth being an unintended one. In addition, use of a contraceptive method, particularly more effective modern contraception usage, in the initial years of marriage is lower in the child married women compared with later when they have one or more children (Aynul and Amin, 2015). However, our study findings are quite in accord with those conducted elsewhere (Godha et al., 2013; Yaya et al., 2019). Studies on early-marrying societies reveal that child brides enter into sexual life with poor knowledge regarding sexuality and protecting themselves from unintended pregnancy and sexually transmitted infection (STI), and their circumstances often prevent them from asserting their preferences with regard to the timing of pregnancy (Jejeebhoy and Bott, 2005), resulting in a higher incidence of unintended pregnancy among them. Increased use of modern contraceptive methods among the child married women compared to their adult married counterparts could be partially attributed to the attainment of the number and composition of children desired. This finding is also consistent with those of studies conducted earlier (Godha et al., 2013; Yaya et al., 2019).

In conclusion, the results of this study provide important information regarding the association of child marriage with various fertility and reproductive outcomes. Although studies on child marriage and its adverse effects on reproductive outcomes are not rare, to our knowledge, this study is the first which used conjointly the most recent data sets of six South Asian countries to examine the prevalence and impact of child marriage on fertility and fertility-related outcomes. Beyond child marriage, the outcomes of interest we considered for analysis in our study are also significantly associated with various socioeconomic and cultural factors, including place of residence, education, religion, household wealth index, working status of women, access to mass media and poverty. A number of studies have explained how these factors mediate fertility and reproductive outcomes of women in low- and middle-income countries (Beyene, 2019; Kassahun et al., 2019; Lerch, 2019; Yimer and Modiba, 2019).

The study has several strengths besides a few limitations. Of the limitations, firstly it used a cross-sectional data set and respondents’ self-reported information, which might incur underreporting in age and age at marriage. Since in the study countries birth, death and marriage registrations are not properly followed, there might have been bias in reporting exact age and exact age at first marriage. Secondly, there might have also been underreporting of unintended pregnancy. In the Muslim culture, a pregnancy or childbirth is often considered a gift from God; although the couples have achieved desired family size they hence justify their unintended pregnancy as a wanted one. Thirdly, we were unable to distinguish between miscarriage and induced pregnancy termination, which may blur the exact association that we found between the forms of pregnancy termination and type of marriage. Fourthly, owing to the cross-sectional nature of the data, we examined only associations, and causality cannot be assessed. Fifthly, the results obtained in this study represent the picture of a South Asian region, not any individual country, as we merged the data set and were interested to see the situation as a whole for South Asia. Despite these limitations, the study has several strengths. Firstly, we used international standard data sets which were gathered following standard procedures and ensure reliability and accuracy of the health situation. Secondly, to collect data, the DHS follow multistage sampling procedures in selecting clusters and households that cover entire geographic regions within a country and that capture the real picture of the population. Third, as to our knowledge this is the first time data from six countries have been analysed together, the results would help to inform policy makers at a glance of the situation in the South Asian region.

It is important to note that there is a high correlation between child marriage and fertility. Child marriages lead to higher total fertility: women who have been married earlier tend to have more children than women who married at an older age. Jensen and Thornton (2003) showed this relationship in their study. They found that women who married before age 15 have their first baby, on average, two years after marriage. On the other hand, women who married between age 15 and 20 have their first baby around three years later than women who married before age 15. In addition, women who married between the age of 21 and 25 have their first baby, on average, approximately seven years later than women who married before age 15. Another study including 15 countries found that women married at the age of 13 tend to have approximately 26% more live births than women married over the age of 17, and they also found that stopping child marriage might decrease the national rate of total fertility by 11% in these countries (Wodon et al., 2017). Ending child marriage and young births may lead to different results for different countries. However, stopping these factors could reduce population growth in Niger by up to 0.39% of the total population (Wodon et al., 2017). The findings in our study similarly show that as the age of marriage decreases, fertility increases. Therefore, we may highlight that the main reason for the impact of child marriages on total fertility is the early marriage of women. These findings also showed that marriage is one of the most important factors in population growth.

This study’s results suggest that, with a few exceptions, child marriage is more prevalent across all South Asian countries. Well-known mediating factors of child marriage such as low or no formal education and poverty should be addressed to encourage personal as well as social development among the girls to avert child marriage and hence adverse reproductive and sexual health outcomes. Marriage legislation in these countries should be adhered to properly to prevent child marriage in this region. To curb the social custom of child marriage, there should be awareness made among people, including parents, guardians and girl-children, regarding the adverse reproductive and sexual health outcomes. Child married women should be advised to use long acting modern contraceptive methods to avoid unintended pregnancy and to make enough space between two pregnancies and repeated childbirth. Reforms should aim to keep girls in education, not only for their personal development, but also for overall societal development and to reduce the harmful practices of child marriage. Health programmes and innovative interventions should be developed aimed at discouraging child marriage among young girls, and child married women should be provided the necessary health services as a priority.

Footnotes

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Efehan Ulas

Author biographies

SM Mostafa Kamal is a Full Professor in the Department of Mathematics at the Islamic University, Kushtia, Bangladesh. He is also the Dean of the Faculty of Sciences and Chair of the Department of Environmental Science and Geography. He has published in leading international journals on maternity care, child marriage, fertility, public health issues and so on.

Efehan Ulas is an Assistant Professor of Statistics at the Cankiri Karatekin University and a post-doctoral fellow in Simon Fraser University. His research interests include functional data analysis, Phase I clinical trials and environmental statistics. He has published articles in leading international journals.

References

Ahmed

Kant

Alia

Noushad

(2013) Psychological impact evaluation of early marriages. International Journal of Health Science Research 1(2): 84–86.

Aynul

Amin

(2015) Early marriage as a risk factor for mis timed pregnancy among married adolescents in Bangladesh. Asia Pacific Population Journal 31(1): 7–34.

Beyene

(2019) Prevalence of unintended pregnancy and associated factors among pregnant mothers in Jimma town, southwest Ethiopia: A cross sectional study. Contraception and Reproductive Medicine 4(1): Article 8.

Bhanji

Punjani

(2014) Determinants of child (early) marriages among young girls: A public health issue. Journal of Women’s Health Care 3(3): 1–3.

Clark

(2004) Early marriage and HIV risks in Sub-Saharan Africa. Studies in Family Planning 35(3): 149–160.

Clark

Bruce

Dude

(2006) Protecting young women from HIV/AIDS: The case against child and adolescent marriage. International Family Planning Perspective 32(2): 79–88.

Delprato

Akyeampong

Sabates

Hernandez-Fernandez

(2015) On the impact of early marriage on schooling outcomes in Sub-Saharan Africa and South West Asia. International Journal of Educational Development 44: 42–55.

Godha

Hotchkiss

Gage

(2013) Association between child marriage and reproductive health outcomes service utilization: A multi-country study from South Asia. Journal of Adolescents Health 52(5): 552–558.

Hotchkiss

Godha

Gage

Cappa

(2016) Risk factors associated with the practice of child marriage among Roma girls in Serbia. BMC International Health and Human Rights 16(1): Article 6.

10.

Huda

Ahmed

Mahmood

, et al. (2018) Unintended pregnancy among married adolescents in urban slums of Bangladesh: A situation analysis. Journal of Health Research 32(4): 298–306.

11.

Jejeebhoy

Bott

(2005) Non-consensual sexual experiences of young people in developing countries: An overview. In: Jejeebhoy

Shah

Thapa

(eds) Sex without Consent: Young People in Developing Countries. London and New York: Zed Books, pp. 3–45.

12.

Jensen

Thornton

(2003) Early female marriage in the developing world. Gender & Development 11(2): 9–19.

13.

Kamal

Hassan

Alam

Ying

(2015) Child marriage in Bangladesh: Trends and determinants. Journal of Biosocial Science 47(1): 120–139.

14.

Kassahun

Zeleke

Dessie

, et al. (2019) Factors associated with unintended pregnancy among women attending antenatal care in Maichew Town, Northern Ethiopia, 2017. BMC Research Notes 12(1): Article 381.

15.

Lendhardt

(2016) Every Last Girl. London: Save the Children.

16.

Lerch

(2019) Fertility decline in urban and rural areas of developing countries. Population and Development Review 45(2): 301–320.

17.

Mim

(2017) Effects of child marriage on girls’ education and empowerment. Journal of Education and Learning 11(1): 9–16.

18.

Mpilambo

Appunni

Kanavo O

Stiegler

(2017) Determinants of early marriage among young women in Democratic Republic of Congo. Journal of Social Sciences 52(1–3): 82–91.

19.

Nasrullah

Zakar

Kramer

(2013) Effect of child marriage on use of maternal health care services in Pakistan. Obstetrics & Gynecology 122(3): 517–524.

20.

Onagoruwa

Wodon

(2018) Measuring the impact of child marriage on total fertility: A study for fifteen countries. Journal of Biosocial Science 50(5): 626–639.

21.

Raj

Saggurti

Balaiah

Silverman

(2009) Prevalence of child marriage and its effect on fertility and fertility-control outcomes of young women in India: A cross-sectional, observational study. The Lancet 373(9678): 1883–1889.

22.

Sekine

Carter

(2019) The effect of child marriage on the utilization of maternal health care in Nepal: A cross-sectional analysis of Demographic and Health Survey 2016. PloS One 14(9): e0222643.

23.

Seth

Bose

Qaiyum

, et al. (2018) Social determinants of child marriage in rural India. Ochsner Journal 18(4): 390–394.

24.

Subramanian

(2008) Determinants of the age at marriage of rural women in India. Family and Consumer Sciences Research Journal 37(2): 160–166.

25.

UNICEF (United Nations Children’s Fund) (2014) Ending Child Marriage: Progress and Prospects. New York: UNICEF, Data and Analytics Section.

26.

UNICEF (United Nations Children’s Fund) (2019) Child Marriage, Adolescent Pregnancy and School Dropout in South Asia. Kathmandu: UNICEF Regional Office for South Asia.

27.

Wodon

Male

Nayihouba

, et al. (2017) Economic Impacts of Child Marriage: Global Synthesis Report. Washington, DC: The World Bank and International Center for Research on Women.

28.

Yaya

Odusina

Bishwajit

(2019) Prevalence of child marriage and its impact on fertility outcomes in 34 sub-Saharan African countries. BMC International Health and Human Rights 19(1): Article 33.

29.

Yimer

Modiba

(2019) Modern contraceptive methods knowledge and practice among blind and deaf women in Ethiopia: A cross-sectional survey. BMC Women’s Health 19(1): Article 151.