Consanguineous Marriages in India: Prevalence and Determinants

Abstract

Consanguineous marriage is the legal union of male and female of a common ancestor related by blood. The most common prevailing form of consanguineous marriages is between first cousins. Middle East Asian countries and southern states in India show high prevalence. A comparative analysis between the two rounds of National Family Health Survey 1 and 4 have shown a declining trend for the practice of consanguineous marriages. The highest percentage of consanguineous marriages is seen among the first cousins from both father’s and mother’s side, most commonly practice in the southern states of Andhra Pradesh, Telangana, Tamil Nadu, and Karnataka except Kerala. Importantly, the practice of consanguineous marriage is higher among the Muslims of North India and Hindus of Southern India, among the Other Backward Classes and the less educated population of the middle and richer wealth index. A significant relationship can be noticed among consanguineous marriages and socio-economic variables.

Keywords

Blood relatives consanguineous marriage National Family Health Survey determinants India

Introduction

Consanguineous marriage is defined as the ‘legal union of male and female of a common ancestor or between a man and woman related by blood’ (Al-Salem & Raishdeh, 1993; Bittles and Hussain, 2000; Centerwall, 1965; Centerwall & Centerwall, 1966; Jurdi & Saxena, 2003; Tamim et al., 2003; Yunis et al., 2008). Marriage within the same clan, tribe, caste or biraderi boundaries largely remains the rule in more traditional societies and is strongly favoured in many migrant communities now resident in western countries (Bittles, 2005). Consanguineous marriage is a practice where individuals favour to marry within their own group. The word ‘consanguinity’ comes from two Latin words con meaning shared and sanguis meaning blood. The most common prevailing form of consanguineous marriage is between first cousins (Al-Salem & Raishdeh, 1993). Consanguinity is characterised by early-age marriage and early-age childbearing in females (Bittles et al., 1991; Hussain & Bittles, 2000).

Several communities of North Africa, Middle East, West Asia and Southern India follow the tradition of consanguineous marriages, where intra-familial unions collectively account for 20% to 50% of all marriages, with first cousin’s union accounting for almost one-third of all marriages (Tadmouri et al., 2009). The prevalence of consanguinity and rates of first cousin marriage vary widely within and between populations and communities, depending on ethnicity, religion, culture and geography. Consanguineous marriages are also practiced among emigrant communities from highly consanguineous countries and regions, such as Pakistan, Turkey, North Africa and Lebanon, now resident in Europe, North America and Australia (Schulpen et al., 2006). Generally, the highest rates of consanguineous marriage have consistently been associated with low socioeconomic status, illiteracy and rural residence (Bittles et al., 1991; Bittles, 1994). In some populations, a high prevalence of consanguineous marriages has also been reported among land-owning families (Rami & Chandrasekhar, 1979) and in traditional ruling groups and the highest socioeconomic strata (Al-Thakeb, 1985). Social, religious, cultural, political and economic factors still play roles in favouring consanguineous marriages among the new generations just as strongly as they did among the older generations, particularly in rural areas.

The first National Family Health Survey (NFHS-1; 1992–1993) shows the prevalence of consanguineous marriages in India at 14%, whereas the second Indian Human Development Survey (IHDS-II, 2011–2012) shows its prevalence at 8%. In order to understand the prevalence of consanguineous marriages in India, it is desirable to understand three basic cultural aspects that are the configuration of the linguistic regions, the institution of caste and the family organisation. The institution of caste is one of the important factors which leads to consanguinity (Chakravarti, 1968). Typically, the highest rates of inbreeding are seen in the Hindu population, especially among tribal groups and Scheduled Castes (Ghosh & Majumder, 1979), but inbreeding is also favoured by the Muslim and Christian communities suggesting social compliance with the prevailing Dravidian marriage customs (Bai et al., 1981; Rao, 1983). Interestingly, caste Hindus and tribal groups who originated in North India but now are resident in the southern states maintain their strenuous avoidance of inbreeding (Pingle, 1983; Rao & Murthy, 1984) in keeping with the strict Aryan outbreeding tradition (Centerwall et al., 1969). The major reason advanced by South Indian groups for the popularity of consanguineous marriages is perceived benefit in terms of social stability, resulting from marriage to a close relative, allied to the maintenance of family property. In general, consanguinity is negatively correlated with socio-economic status (Dronamraju, 1964; Khan, 1963) although exceptions have been reported in major land-owning families in Andhra Pradesh. Hindus favour cross-cousin unions, whereas parallel cousin marriages are more common form in the Muslim community. Uncle–niece marriages (mother’s brother with sister’s daughter) are arranged in many parts of Andhra Pradesh, Karnataka and Tamil Nadu, but their prevalence is low in Kerala (Centerwall et al., 1969; Dumont, 1953; Good, 1980; Reid, 1973).

Recently, India has been experiencing large-scale socio-economic transformation. The existing research on consanguineous marriages in India has become quite outdated now which cannot be generalised anymore (Babu et al., 1994; Beck, 1972; Centerwall, 1965; Centerwall & Centerwall, 1966; Devi et al., 1981; George et al., 1992; John & Jayabal, 1971; Khan, 1963; Kulkarni & Kurian, 1990; Rao & Inbaraj, 1977, 1979, 1980; Sureender et al., 1998). Hence, there is a need to update the knowledge and information on consanguineous marriages: their levels, differentials and determinants, by providing a comprehensive assessment of national-level data considering the variation in the levels of consanguineous marriages by all the states and socio-economic groups across India.

Data and Method

The data for this study has been used from the NFHS-1 and NFHS-4 conducted during 1992–1993 and 2015–2016 respectively. Details of the survey design, instrument and findings are available in the national reports (IIPS, 1995; IIPS & ICF, 2017). The National Family Health Survey is a large-scale, multi-round survey conducted in a representative sample of households throughout India. It includes information on the respondent’s background, reproduction, marriage and cohabitation, contraception and others. In the first round of NFHS-1 (1992–1993), interviews were conducted with a nationally representative sample of 88,562 households and 89,777 ever-married women aged 13–49 years across India, while NFHS-4 (2015–2016) covered a representative sample of 601,509 households and 699,686 women aged 15–49 years. To make a comparative analysis, ever-married women aged 15–49 years have been considered in the study. Therefore, the analytical sample in this study was restricted to 89,426 and 540,670 women who had ever been married aged 15–49 in NFHS-1 and NFHS-4 respectively. The questions taken to determine consanguineous marriage are that whether the respondent was related to her current husband in any way before marriage and what type of relationship it was. The first question was to be answered in yes/no and the second question had the categories of first cousin on father’s side, first cousin on mother’s side, second cousin, uncle, other blood relative, brother-in-law and other non-blood relative. The women who report that their husband was related to them by blood before their marriage are considered as consanguineous marriage.

To find out how the consanguineous marriage varies across the background characteristics, a set of background characteristics has been considered. These are current age of women, caste, religion, educational level, wealth index, place of residence and region. The age group of women has been divided into three categories, that is, 15–24, 25–34 and 35–49. Caste has been divided into four categories, namely Scheduled Castes (SCs), Scheduled Tribes (STs), Other Backward Classes (OBCs) and others. Religion has been grouped into Hindu, Muslim, Christian, Sikh and others. Educational level of women has been divided into four categories, that is, no education, primary, secondary and higher. The wealth index has been categorised as poorest, poor, middle, richer and richest. The place of residence has been categorised as rural and urban. India has been divided into six geographical regions, that is, north, south, east, west, central and north-east. The northern region includes the states Chandigarh, Delhi, Haryana, Himachal Pradesh, Jammu and Kashmir, Punjab, Rajasthan and Uttaranchal. Central India includes three states, namely Chhattisgarh, Madhya Pradesh and Uttar Pradesh. Similarly, the eastern region comprises of four states (Bihar, Jharkhand, Odisha and West Bengal). When it comes to the north-eastern region, it includes eight states, that is, Arunachal Pradesh, Assam, Manipur, Meghalaya, Mizoram, Nagaland, Sikkim and Tripura. The western region covers Dadra and Nagar Haveli, Daman and Diu, Goa, Gujarat and Maharashtra. The southern states, on the other hand, include five states and three union territories (Andaman and Nicobar Islands, Andhra Pradesh, Karnataka, Kerala, Lakshadweep, Puducherry, Tamil Nadu and Telangana). Bivariate analysis is used to understand the differentials in consanguineous marriages by the socio-economic, demographic and regional contexts. Multivariate analysis is used to understand the significant determinants of consanguineous marriage in India. Binary logistic regression is an appropriate technique to analyse the relationship between a set of predictor variables and a dependent variable which is dichotomous. The predictor variables may be either categorical or in ratio scale. As our dependent variable is categorical and consists of two outcomes, consanguineous and non-consanguineous marriages, we have used binary logistic regression. The model allows the effects of the independent variables to differ for each outcome and handles the non-independence of the categories of the dependent variable by simultaneous estimation of the models for both the outcomes. The analysis is carried out using Stata 14.

Results

Prevalence of Consanguineous Marriage in India

Figure 1 shows the prevalence of consanguineous marriages in India for the two datasets of NFHS-1 and NFHS-4. The percentage distribution of consanguineous marriages (marriage to the blood relatives) in NFHS-1 was 14.4% which shows a decline by 2.5% in NFHS-4 to 11.9%. The prevalence of marriage to other non-blood relatives which was 1.9% in 1992–1993 has shown a slight increase to 2.4% in 2015–2016. Whereas the percentage of marriage to non-relatives, which was 83.7% in 1992–1993, has shown an increase to 85.7% in 2015–2016, which may be due to a decline in the percentage of consanguineous marriages.

Figure 1.

Source: Computed from NFHS-1 and NFHS-4 data files.

Table 1.

Relationship to Current (Last) Husband Before Their Marriage, NFHS-1 and NFHS-4.

Relationship	NFHS-1	NFHS-4
First cousin on father’s side	5.5	4.3
First cousin on mother’s side	4.6	4.3
Second cousin	0.9	0.7
Uncle	0.9	0.6
Other blood relative	2.4	2.1
Consanguineous	14.4	11.9
Brother in-law	0.4	0.7
Other non-blood relative	1.5	1.7
Not related	83.7	85.7
N	89426	540670

Source: Computed from NFHS I and IV data files.

Table 1 shows the relationship of women to current husband prior to marriage with six categories, namely, first cousin on both father’s and mother’s side, second cousin, uncle, other blood relative, brother-in-law, other non-blood relative and not related. The relationship to the current husband prior to marriage is seen to be highest among the first cousin from both father’s and mother’s sides. The percentage of women married to their first cousin from their father’s side was 5.5% as per NFHS-1, which later came down to 4.3% according to NFHS-4. Women married to their first cousin from their mother’s side have also declined from 4.6% in NFHS-1 to 4.3% in NFHS-4. Marriage to other non-blood relative and brother-in-law shows a slight increase from NFHS-1 to NFHS-4, that is, 1.5% to 1.7% and 0.4% to 0.7%, respectively, whereas women getting married to their second cousin (0.9%–0.7%), uncle (0.9%–0.6%) or other blood relative (2.4%–2.1%) has shown a declining trend from NFHS-1 to NFHS-4. The consanguineous marriages are calculated by combining the categories of first cousin from both father’s and mother’s side, second cousin, uncle and other blood relative, which has shown a decline from NFHS-1 to NFHS-4 (14.4%–11.9%).

Figure 2.

Percentage of Consanguineous marriage by States, India, NFHS I.

Figure 3.

Percentage of Consanguineous marriage by States, India, NFHS IV.

Figures 2 and 3 show the percentage of consanguineous marriages by states from the two datasets of NFHS-1 and NFHS-4. According to NFHS-1, the states with less than or equal to 5% of consanguineous marriages are Himachal Pradesh, Punjab, Haryana, Rajasthan, undivided Madhya Pradesh and the north-eastern states except for Arunachal Pradesh. Followed by states with 5.1% to 10% prevalence are Jammu and Kashmir, undivided Uttar Pradesh and Bihar, West Bengal, Arunachal Pradesh, Odisha, Gujarat and Kerala. The states with the highest prevalence of consanguineous marriages are Maharashtra, undivided Andhra Pradesh, Goa, Karnataka and Tamil Nadu. Figure 3 shows that changes in the prevalence that have occurred over the years from the first round of NFHS to the fourth. The states with the highest prevalence percentage remain almost the same with Goa shifting to the lower category followed by the addition of Jammu and Kashmir. The states with the lowest prevalence of consanguineous marriages were added with the states of Jharkhand, Arunachal Pradesh, Uttarakhand, Sikkim and Kerala in the less than or equal to 5% category, with the exclusion of Madhya Pradesh and Chhattisgarh.

Table 2.

Percentage of Consanguineous Marriage by States, India, NFHS-1 and NFHS-4.

State	NFHS-1	NFHS-4
North	2.5	4.8
Chandigarh		1.2
Delhi	6.1	6.9
Haryana	1.8	4.3
Himachal Pradesh	1.5	1.5
Jammu and Kashmir	9.8	18.9
Punjab	1.6	2.2
Rajasthan	2.0	3.6
Uttarakhand	$	4.9
Central	7.5	8.9
Chhattisgarh	@	5.4
Madhya Pradesh	4.9	7.4 (8.1)
Uttar Pradesh	8.8	9.5 (9.8)
East	6.4	5.9
Bihar	6.2	5.1 (5.4)
Jharkhand	#	4.1
Odisha	7.0	8.1
West Bengal	6.3	5.9
Northeast	2.6	1.8
Arunachal Pradesh	8.5	3.8
Assam	2.1	1.8
Manipur	4.6	2.9
Meghalaya	3.6	2.6
Mizoram	1.3	1.9
Nagaland	3.0	2.9
Sikkim		0.9
Tripura	3.0	0.4
West	18.6	12.8
Dadra and Nagar Haveli		0.7
Daman and Diu		2.9
Goa	14.5	8.8
Gujarat	5.7	8.0
Maharashtra	25.4	15.2
South	34.6	25.7
Andaman and Nicobar Islands		13.5
Andhra Pradesh	34.9	27.0 (28.6)
Karnataka	35.6	25.5
Kerala	9.4	4.8
Lakshadweep		25.8
Puducherry		22.1
Tamil Nadu	46.4	32.1
Telangana	+	24.8
India	14.4	11.9

Source: Computed from NFHS I and IV data files.

Notes: Includes ever-married women aged 15–49; $ = included in Uttar Pradesh; @ = included in Madhya Pradesh; # = included in Bihar; + = included in Andhra Pradesh; data included in () = as per the current geographical territory.

Table 2 presents the prevalence of consanguineous marriages by state/union territories and relationship to current husband prior to marriage for the two datasets of NFHS-1 and NFHS-4. Combining the six categories of relationship to current husband the category of consanguineous marriages has been formed. The states have been divided into six regions of north, central, east, northeast, west and south. The highest prevalence of consanguineous marriages can be seen in the southern region, 34.6% for NFHS-1 and 25.7% for NFHS-4. The states with the highest prevalence of consanguineous marriages according to NFHS-1 are Tamil Nadu (46.4%), Karnataka (35.6%), Andhra Pradesh (34.9%) and Maharashtra (25.4%). And the states with the lowest prevalence of consanguineous marriages according to NFHS-1 are Mizoram (1.3%), Himachal Pradesh (1.5%), Punjab (1.6%) and Haryana (1.8%). It is observable that the higher prevalence states are mostly from the southern region and the lower are from the northern region. A prevalence of more than 5% can also be observed in most of the states of the central and eastern regions. Uttar Pradesh marks the highest percentage of consanguineous marriages among all the central states. The overall percentage for consanguineous marriages in India is 14.4% according to NFHS-1 which shows a decline to 11.9% in NFHS-4, with the highest prevalence being in the southern region (25.7%). Tamil Nadu maintains its position at the top with 32.1% followed by Andhra Pradesh (27%), Karnataka (25.5%) and Telangana (24.8%), except for Kerala which remains at a low prevalence of 4.8%. The north-eastern region marks the lowest percentage of consanguineous marriages at 1.8% including Tripura (0.4%), Sikkim (0.9%), Arunachal Pradesh (1.2%) and Mizoram (1.9%). The union territory with the lowest percentage of consanguineous marriages is Dadra and Nagar Haveli (0.7%) followed by Chandigarh (1.2%). Except for the northern and central regions, all the other regions have shown a decline in the percentage of consanguineous marriages from NFHS-1 to NFHS-4. Almost all the states have shown increasing percentages except the states of Bihar, Arunachal Pradesh, Assam, Manipur, Meghalaya, Nagaland, Tripura, Goa, Maharashtra, Andhra Pradesh, Kerala and Tamil Nadu, which have shown declining percentages for consanguineous marriages from NFHS-1 to NFHS-4.

District-Wise Variations in the Prevalence of Consanguineous Marriage

As the NFHS-4 sample was designed to allow district-level estimates, we have calculated the prevalence of consanguineous marriage at the district level for all the 640 districts of India (complete table not shown here). The state-level estimates confirm the district-level variation in consanguineous marriage. The proportions of consanguineous marriages are higher (above 30%) in most of the districts of Southern India. The low prevalence (less than 1%) districts in consanguineous marriage are mostly distributed in northern parts of the country, confirming the north-south divide in the practice of consanguineous marriage in India (Figure 4). Table 3 shows the top and bottom 20 districts in terms of proportion of consanguineous marriage. Among the top 20 districts (above 34%), 19 are from the three southern states of Tamil Nadu, Karnataka and Andhra Pradesh and an additional one is from Jammu and Kashmir, whereas the bottom 20 districts (0.8% or less) are mainly spread in the northern regions of India indicating the less prevalence of consanguineous marriage. Besides, the prevalence of consanguineous marriage is found to be 0% in 25 districts of India namely Lahul and Spiti (Himachal Pradesh), Kinnaur (Himachal Pradesh), North District (Sikkim), West District (Sikkim), South District (Sikkim), Tawang (Arunachal Pradesh), Upper Siang (Arunachal Pradesh), Changlang (Arunachal Pradesh), Dibang Valley (Arunachal Pradesh), Lohit (Arunachal Pradesh), Zunheboto (Nagaland), Longleng (Nagaland), Mamit (Mizoram), Kolasib (Mizoram), Champhai (Mizoram), Serchhip (Mizoram), Lawngtlai (Mizoram), Saiha (Mizoram), Dhalai (Tripura), West Khasi Hills (Meghalaya), Ribhoi (Meghalaya), Jaintia Hills (Meghalaya), Lakhimpur (Assam), Dima Hasao (Assam) and Nicobars (Andaman and Nicobar Islands).

Table 3.

Top 20 and Bottom 20 Districts in the Prevalence of Consanguineous Marriage in India, NFHS-4.

Top 20 Districts			Bottom 20 Districts
District	State	Percentage	District	State	Percentage
Ramanathapuram	Tamil Nadu	42.5	Jalor	Rajasthan	0.8
Theni	Tamil Nadu	41.0	Solan	Himachal Pradesh	0.7
Madurai	Tamil Nadu	40.5	Amritsar	Punjab	0.7
Dharmapuri	Tamil Nadu	40.4	North Tripura	Tripura	0.7
Yadgir	Karnataka	39.1	Dadra & Nagar Haveli	Dadra & Nagar Haveli	0.7
Salem	Tamil Nadu	39.0	Bilaspur	Chhattisgarh	0.6
Gadag	Karnataka	38.0	Panipat	Haryana	0.6
Bijapur	Karnataka	37.9	Rohtak	Haryana	0.6
Ramban	Jammu and Kashmir	37.6	Lalitpur	Uttar Pradesh	0.6
Tiruvannamalai	Tamil Nadu	37.5	West Tripura	Tripura	0.6
Pudukkottai	Tamil Nadu	37.2	Karbi Anglong	Assam	0.6
Sivaganga	Tamil Nadu	36.9	Hoshiarpur	Punjab	0.5
Srikakulam	Andhra Pradesh	36.8	Sahibzada Ajit Singh Nagar	Punjab	0.5
Tiruchirappalli	Tamil Nadu	36.2	Rupnagar	Punjab	0.4
Koppal	Karnataka	36.0	Golaghat	Assam	0.4
Chitradurga	Karnataka	35.6	Sonipat	Haryana	0.3
Perambalur	Tamil Nadu	35.6	East Khasi Hills	Meghalaya	0.3
Viluppuram	Tamil Nadu	35.4	Nagaon	Assam	0.3
Krishnagiri	Tamil Nadu	35.3	South Tripura	Tripura	0.2
Virudhunagar	Tamil Nadu	34.4	Dibrugarh	Assam	0.2

Source: Computed from NFHS-4 data file.

Figure 4.

Percentage of Consanguineous marriage by Districts, India, NFHS IV.

Differentials in Consanguineous Marriage in India

The differentials in the prevalence of consanguineous marriages in India across selected demographic and socio-economic variables for NFHS-1 and NFHS-4 are shown in Table 4. The prevalence varied largely with age group, caste, religion, educational level, wealth index, place of residence and region. According to NFHS-1, the highest prevalence of consanguineous marriages can be seen in the 15–24 age group (15.5%), which further declines for the 25–34 age group (13.4%) and finally rises for the 35+ age group (14.4%). In 2015–2016 (NFHS-4), the highest prevalence of consanguineous marriages could be seen among the 15–24 age group (14.6%), which declined for age groups 25–34 (12.4%) and 35+ (12%). In case of caste, the highest prevalence of consanguineous marriages can be observed for the SCs (15.1% for NFHS-1), whereas the highest prevalence is found among OBCs (13.5%) in NFHS-4. The STs show a rise from 9.6% (NFHS-1) to 10.4% (NFHS-4). The practice of consanguineous marriages was a common practice of Hindus in NFHS-1 (26.9%) followed by Sikhs (13.9%), others (12.9%), Muslims (12.7%) and the lowest prevalence among the Christians (2.3%). Whereas according to NFHS-4, the Muslims have topped the list in the practice of consanguineous marriages (17.4%), followed by Christians (13.4%), others (12.5%), Hindu (11.2%) and Sikhs (2.9%).

Table 4.

Percentage of Ever-Married Women age 15–49 Who Had Consanguineous Marriage by Background Characteristics, India, NFHS-1 and NFHS-4.

Background Characteristics	NFHS-1		NFHS-4
Background Characteristics	Percentage	N	Percentage	N
Age group
15–24	15.5	27,064	13.7	100,347
25–34	13.4	32,101	11.8	199,654
35+	14.4	30,261	11.4	240,669
Caste
Scheduled Caste	15.1	10,815	12.0	110,106
Scheduled Tribe	9.6	7,884	10.4	49,962
OBC			13.2	235,251
Others	14.8	70,727	10.5	145,351
Religion
Hindu	26.9	10,735	11.2	440,149
Muslim	12.7	2,136	17.4	70,896
Christian	2.3	1,679	13.4	12,355
Sikh	13.9	1,566	2.9	8,625
Others	12.9	73,310	12.5	8,647
Educational level
No education	14.6	55,243	11.1	182,909
Primary	16.5	14,578	12.4	78,070
Secondary	13.2	16,593	12.9	226,529
Higher	6.2	3,011	10.0	53,162
Wealth Index
Poorest	11.5	17,040	7.7	98,813
Poorer	13.6	17,857	11.5	107,160
Middle	16.8	18,125	14.5	111,541
Richer	16.6	18,209	14.3	113,532
Richest	13.1	18,195	11.2	109,626
Place of residence
Urban	15.6	23,413	13.4	181,707
Rural	13.9	66,012	11.2	358,964
Region
North	2.5	10,618	4.8	71,444
South	34.6	20,784	25.7	127,653
East	6.4	19,773	5.9	123,713
West	18.6	12,928	12.8	78,784
Central	7.5	21,950	8.9	120,479
North East	2.6	3,373	1.8	18,597
Total	14.4	89,426	11.9	540,670

Source: Computed from NFHS-1 and NFHS-4 data files.

Note: N is weighted.

It can well be observed that the prevalence of consanguineous marriages by the level of educational attainment is a hill-shaped curve for both datasets. The highest prevalence can be seen for the population who have completed primary level of education (16.5%) for NFHS-1, followed by those who never went to school (14.6%), secondary (13.2%) and higher (6.2%), whereas the highest prevalence of consanguineous marriages according to NFHS-4 can be seen for those who have completed secondary level of education (12.9%), followed by primary (12.4%), no education (11.1%) and higher (10%). For both the rounds, the pattern in the prevalence of consanguineous marriages by wealth index remains the same; the population belonging to the middle category shows the highest prevalence (16.5% and 14.5% respectively for NFHS-1 and NFHS-4), which is followed by the richer (16.6% and 14.3%), poorer (13.6% and 11.5%), richest (13.1% and 11.2%) and the lowest prevalence of consanguineous marriages could be observed for the poorest category (11.5% and 7.7%). The prevalence of consanguineous marriage is higher in urban areas (15.6% and 13.4%) than the rural areas (13.9% and 11.2%). Consanguineous marriage is a usual practice in southern region of India with 34.6% in NFHS-1 and 25.7% in NFHS-4, followed by western region (18.6% and 12.8%), central region (7.5% and 8.9%), eastern region (6.4% and 5.9%), northern region (2.5% and 4.8%) and north-eastern region (2.6% and 1.8%). The northern and central regions have shown an increase in the prevalence of consanguineous marriages whereas all the other regions have shown a decline in their prevalence.

Determinants of Consanguineous Marriage in India

To find out the determining factor on the prevalence of consanguineous marriages in India, logistic regression analyses have been carried out to make a comparative study in both datasets. The logistic regression analyses have been carried out with a 95% confidence interval. It can be observed from Table 5 that controlling the effect of other variables, the odds of consanguineous marriage decreases with the increase in age of women. The odds of having consanguineous marriage among women aged 25–34 were 15% and 17% less for women aged more than 35 years than the women aged 15–24 years in NFHS-1. The odds of consanguineous marriage were 21% and 30% less likely among women aged 25–34 and more than 35 years respectively, than the women aged 15–24 years in NFHS-4. Controlling the effect of other variables, the odds of consanguineous marriage is 2.9 times higher among Muslims than Hindus in NFHS-4. The odds of religion in NFHS-4 have shown a rise as compared to NFHS-1 data. With respect to educational attainment, the odds for women who have attained primary, secondary and higher educational level are 25%, 39% and 63%, respectively, less likely to go for consanguineous marriage than the illiterate women in NFHS-1, whereas in NFHS-4, women with secondary and higher educational level are 7% and 63% less likely to go for consanguineous marriage than the illiterate women.

Table 5.

Odds Ratio Showing the Effect of Background Variables on the Consanguineous Marriage: Results from Logistic Regression Analyses.

Background Variables	NFHS-1			NFHS-4
	Exp(B)	95% CI		Exp(B)	95% CI
	Exp(B)	Lower	Upper	Exp(B)	Lower	Upper
Age group
15–24®
25–34	0.851***	0.805	0.899	0.790***	0.769	0.811
35+	0.830***	0.784	0.878	0.697***	0.678	0.717
Caste
Scheduled Caste®
Scheduled Tribe	0.892**	0.804	0.989	1.018	0.981	1.055
OBC				0.880***	0.856	0.905
Others	0.787***	0.733	0.845	0.944***	0.914	0.974
Religion
Hindu®
Muslim	0.204***	0.179	0.232	2.947***	2.874	3.023
Christian	0.308***	0.24	0.395	1.030	0.974	1.090
Sikh	0.516***	0.435	0.613	0.595***	0.531	0.666
Others	0.333***	0.313	0.353	1.174***	1.085	1.272
Educational level
No education®
Primary	0.751***	0.705	0.800	0.997	0.967	1.028
Secondary	0.609***	0.567	0.654	0.926***	0.902	0.951
Higher	0.372***	0.314	0.440	0.738***	0.705	0.772
Wealth index
Poorest®
Poorer	0.929*	0.859	1.005	1.159***	1.121	1.198
Middle	0.889***	0.823	0.96	1.156***	1.117	1.196
Richer	0.852***	0.785	0.925	1.068***	1.028	1.11
Richest	0.843***	0.763	0.933	0.983	0.94	1.027
Place of residence
Urban®
Rural	0.886***	0.835	0.941	1.026**	1.002	1.051
Region
North®
Central	13.506***	12.362	14.755	5.776***	5.589	5.969
East	1.610***	1.446	1.794	0.965*	0.928	1.004
North East	5.511***	5.000	6.074	2.181***	2.095	2.27
West	1.834***	1.659	2.027	1.491***	1.441	1.541
South	0.969	0.835	1.124	0.352***	0.332	0.373
Constant	0.190***			0.069***

Source: Computed from NFHS I and IV data files.

Note: ® = Reference category; ***, **, * = <1%, 5% and 10% levels of significance.

Controlling the effect of other variables, the wealth index had a negative relation with the consanguineous marriage in NFHS-1, that is, as the wealth index increases the prevalence of consanguineous marriage decreases which was statistically significant. But the likelihood to go for consanguineous marriage is about 1.1 times higher among women with poorer, middle and richer categories than the women with the poorest category in NFHS-4. Place of residence is a significant determinant of consanguineous marriages. Taking urban as the reference category the odds for the women residing in the rural areas are 11% less likely to go for consanguineous marriage than their urban counterparts in NFHS-1 while it is 1.026 times higher among rural women in NFHS-4. With respect to the region, the odds for the NFHS-1 remain comparatively higher than NFHS-4. The central region shows higher odds with 13.5 and 5.8 times more than the women residing in the northern region. The odds of the eastern region have declined from 1.6 times in NFHS-1 to 0.96 times in NFHS-4. Similarly, the north-eastern region has also declined from 5.5 times to 2.2 times. The western region, too, shows a declining trend from NFHS-1 to NFHS-4 which was 1.834 times to 1.491 times during NFHS-1 to NFHS-4.

Discussion and Conclusion

This study focuses upon the practice of consanguineous marriages, differentials and determinants in India for the two rounds of NFHS, that is, 1 and 4. For this purpose, eight categories of relationship to current or last husband have been considered, where first cousin on both father’s and mother’s side, second cousin, uncle and other blood relative have been combined to form consanguineous marriages and the other three categories are brother-in-law, other non-blood relative and not related to the respondent. The prevalence of consanguineous marriages seems to follow a declining trend from NFHS-1 to NFHS-4, that is, 14.4% to 11.9%. The most preferred relationship for such union is among first cousins from both father’s and mother’s side which show the highest prevalence among all the other categories. The findings accord with the previous studies (Beck, 1972) that the highest prevalence of consanguineous marriages can be seen in the southern states of Tamil Nadu, Andhra Pradesh, Telangana and Karnataka. The state-wise picture remains almost the same for both the rounds with only some slight changes with time for the states of Kerala, Arunachal Pradesh, Uttarakhand, Goa and Jharkhand that have shown a decline in their prevalence, whereas Madhya Pradesh, Chhattisgarh and Jammu and Kashmir have shown an increase in their prevalence. The distribution of consanguinity in India is lop-sided towards the southern states with some exceptions like that of Jammu and Kashmir due to its religious composition (Chakravarti, 1968) and Maharashtra, where consanguinity is a frequent practice. As far as the district-wise picture is concerned, according to the NFHS-4 dataset, the high prevalence districts are mostly from the Deccan plateau as well as the northern part of Uttar Pradesh and the low prevalence districts are mainly from the north-eastern and north-western parts of the country. Consanguinity is characterised by early-age marriage and early-age childbearing in females (Bittles, 1991), a declining trend can very well be noticed between the two rounds. The practice of consanguinity when seen from the caste lens portrays an interesting picture where the SCs show a declining trend while the OBCs remain the highest among all the categories. The STs have shown an increase of around 1% between the two time periods. It was observed that the Hindus had a great practice of marrying their blood relatives according to the first round of our data, which has certainly declined in NFHS-4, whereas the Muslims marrying their close cousins has risen in India. The level of educational attainment showed an inverse relationship for the first round with the rising level of educational attainment the prevalence of consanguineous marriages declines but interesting this trend is not very clearly visible for the fourth round of NFHS. The findings of this study are similar to the previous studies that show the highest rates of consanguineous marriage are consistently associated with lower levels of illiteracy and rural residence (Bittles et al., 1991; Bittles, 1994). The socio-economic status of the prevalence of consanguineous marriages follows the same pattern of a symmetrical curve for both rounds which certainly shows some decline in the proportions over time. The logistic regression analyses show a clear decline in the odds from the first till the fourth round indicating the decline in this practice, which may probably be due to increased awareness about genetic disorders and dissolving caste rigidities, reduced endogamy practices in the society

The findings demonstrate that consanguineous marriages are unevenly distributed throughout the country. Besides first cousin marriages with four forms, the practice of uncle–niece marriage is also prevalent in several parts of South India which is completely non-existent in the north. Analysis suggests that consanguinity has a strong association with early age marriage of the women who are mostly Muslims and belong to the OBCs. A very strong association can also be observed with the women of lower levels of educational attainment as well as those who belong to the lower socio-economic status in the rural areas. It is an age-old practice that has been carried on for years together as endogamy (marriage within the same community). It has mainly been practiced to retain the ancestral property within the same community, and the high prevalence of consanguineous marriages in the southern states can also be attributed to the comparatively better position of women than that of the women of north Indian states.

Footnotes

Declaration of Conflicting Interests

Funding

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

References

Al-Salem

, & Raishdeh

(1993). Consanguinity in north Jordan: Prevalence and pattern. Journal of Biosocial Science , 25(4), 553–556.

Al-Thakeb

(1985). The Arab family and modernity: Evidence from Kuwait. Current Anthropology , 26(5), 575–580.

Babu

B. V.

, Kusuma

Y. S.

, & Naidu

J. M.

(1994). Genetic load among four Andhra caste populations. Social Biology , 41(1–2), 127–129.

Bai Asha

P. V.

, Jacob John

, & Subramaniam

V. R.

(1981). Reproductive wastage and develop mental disorders in relation to consanguinity in south India. Tropical and Geographical Medicine , 33, 275–280.

Beck

B. E. F.

(1972). Peasant Society in Konku . University of British Columbia Press.

Bittles

A. H.

, Mason

W. M.

, Greene

, & Rao

N. A.

(1991). Reproductive behavior and health in consanguineous marriages. Science , 252(5007), 789–794.

Bittles

A. H.

(1991). Consanguinity: A major variable in the study of North African reproductive behavior, morbidity and mortality? In Proceedings of the Demographic and Health Surveys World Conference (Vol. 1, pp. 321–341). Institute for Research Development.

Bittles

A. H.

(1994). The role and significance of consanguinity as a demographic variable. Population and Development Review , 20(3), 561–583.

Bittles

A. H.

(2005). Endogamy, consanguinity and community disease profile. Community Genetics , 8(1), 17–20.

10.

Bittles

A. H.

, & Hussain

(2000). An analysis of consanguineous marriage in the Muslim population of India at regional and state level. Annals of Human Biology , 27(2), 163–171.

11.

Centerwall

W. R.

, Savarinathan

, Mohan

L. R.

, Booshanan

, & Zachariah

(1969). Inbreeding patterns in south India. Social Biology , 16, 81–91.

12.

Centerwall

W. R.

(1965). A preliminary study of consanguinity and congenital anomalies in South India. Paediatrica Indonesiana , 5, 906–914.

13.

Centerwall

W. R.

, & Centerwall

S. A.

(1966). Consanguinity and congenital anomalies in South India: A pilot study. Indian Journal of Medical Research , 54, 1160–1167.

14.

Chakravarti

M. R.

(1968). Consanguinity in India. Zeitschrift für Morphologie und Anthropologie , Bd. 60(2), 170–183.

15.

Devi

R. R.

, Appaji Rao

A. N.

, & Bittles

A. H.

(1981). Consanguinity, fecundity and post-natal mortality in Karnataka, south India. Annals of Human Biology , 8(5), 469–472.

16.

Dronamraju

K. R.

(1964). Mating systems of the Andhra Pradesh people. Cold Spring Harbor Symposia on Quantitative Biology , 29, 8I–84.

17.

Dumont

(1953). The Dravidian kinship terminology as an explanation of marriage. Man 53, 34–39.

18.

George

, Vedamony

, Idikulla

, & Rao

P. S.

(1992). The effect of consanguinity on pregnancy-induced hypertension. Australian and New Zealand Journal of Obstetrics and Gynecology , 32, 231–232.

19.

Ghosh

A. K.

, & Majumder

P. P.

(1979). Genetic load in an isolated tribal population of south India. Human Genetics , 51(2), 203–208.

20.

Good

(1980). Elder sister’s daughter marriage in South Asia. Journal of Anthropological Research , 36, 474–500.

21.

Hussain

, & Bittles

A. H.

(2000). Sociodemographic correlates of consanguineous marriage in the Muslim population of India. Journal of Biosocial Science , 32(4), 433–442.

22.

International Institute for Population Science. (1995). National Family Health Survey (MCH and Family Planning), India, 1992–93 . IIPS.

23.

International Institute for Population Sciences (IIPS) & ICF. (2017). National Family Health Survey (NFHS-4), 2015–16: India . IIPS.

24.

John

J. T.

, & Jayabal

(1971). Foetal and child loss in relation to consanguinity in Southern India. Indian Journal of Medical Research , 59, 1050–1053.

25.

Jurdi

, & Saxena

P.C.

(2003). The prevalence and correlates of consanguineous marriages in Yemen: Similarities and contrasts with other Arab countries. Journal of Biosocial Science , 35(1), 1-13.

26.

Khan

(1963). A study of Andhra marriages: Consanguinity, caste, illiteracy, and bridal age. Acta Genetica et Statistica Medica , 19, 21–29.

27.

Kulkarni

M. L.

, & Kurian

(1990). Consanguinity and its effect on fetal growth and development: A south Indian study. Journal of Medical Genetics , 27(6), 348–352.

28.

Pingle

(1983). Comparative analysis of mating systems and marriage distance patterns in five tribal groups of Andhra Pradesh, India. Social Biology , 30(1), 67–74.

29.

Reddy

Rami

V. Reddy B. K.

(1979). Consanguinity effects on fertility and mortality among the Reddis of Chittoor District (A.P.) South India. Indian Journal of Heredity , 11, 77–88.

30.

Rao

P. S. S.

(1983). Religion and the intensity of inbreeding in Tamil Nadu, South India. Social Biology , 30, 413–422.

31.

Rao

P. S. S.

, & Inbaraj

S. G.

(1977). Inbreeding effects on human reproduction in Tamil Nadu of South India. Annals of Human Genetics , 41(1), 87–98.

32.

Rao

P. S. S.

, & Inbaraj

S. G.

(1979). Trends in human reproductive wastage in relation to long-term practice of inbreeding. Annals of Human Genetics , 42, 401–413.

33.

Rao

P. S. S.

, & Inbaraj

S. G.

(1980). Inbreeding effects on fetal growth and development. Journal of Medical Genetics , 17, 27–33.

34.

Rao

V. V.

, & Murty

J. S.

(1984). Selection intensities and inbreeding among some caste groups of Andhra Pradesh, India. Social Biology , 31(1–2), 114–119.

35.

Reid

R. M.

(1973). Social structure and inbreeding in a south Indian caste. In Morton

N. E.

(Ed.), Genetic structure of populations (pp. 92–99). University of Hawaii Press.

36.

Schulpen

T. W.

, van Wieringen

J. C.

, van Brummen

P. J.

, van Riel

J. M.

, Beemer

F. A.

, Westers

, & Huber

(2006). Infant mortality, ethnicity, and genetically determined disorders in The Netherlands. European Journal of Public Health , 16, 291–294.

37.

Sureender

, Prabakaran

, & Khan

A. G.

(1998). Mate selection and its impact on female marriage age, pregnancy wastages, and first child survival in Tamil Nadu, India. Social Biology , 45(3–4), 289–301.

38.

Tadmouri

G. O.

, Nair

, Obeid

, Al Ali

M. T.

, Al Khaja

, & Hamamy

H. A.

(2009). Consanguinity and reproductive health among Arabs. Reproductive Health , 6(17), 17.

39.

Tamim

, Khogali

, Beydoun

, Melki

, Yunis

, & Network.

National Collaborative Perinatal Neonatal

(2003). Consanguinity and apnea of prematurity. American Journal of Epidemiology , 158(10), 942–946.

40.

Yunis

, El Rafei

, & Mumtaz

(2008). Consanguinity: Perinatal outcomes and prevention—a view from the Middle East. Neo Reviews , 9(2), e59–e65.