The Under-five Mortality Rate (U5MR) represents the number of children who die by the age of five yr per 1,000 live births1. It is the decisive metric in assessing a nation’s healthcare and developmental progress because U5MR provides valuable insights into the health and survival of its youngest population. During the year 2020, ⁓5 million children under the age of five yr lost their lives in the world, primarily due to preventable and treatable causes of preterm birth complications, birth asphyxia/trauma, pneumonia, diarrhoea, and malaria, all of which can be prevented or treated through the provision of accessible and affordable health and sanitation interventions2. The prevailing infant and child mortality rates serve as the reflection of the social and economic advancement of a country. At the global level, the total number of deaths in children under-five yr of age decreased significantly from 12.6 million in 1990 to 5.0 million in 20203. Moreover, the global U5MR demonstrated a remarkable decline of 61 per cent, decreasing from 93 deaths per 1000 live births in 1990 to 37 deaths per 1000 live births in 20203. This improvement depicts the social and economic advancement of countries worldwide.

India holds a pivotal position in global endeavours to end preventable deaths among newborns and children under the age of five. The National Family Health Survey (NFHS)-14 reported that, in India, one in nine children die before reaching the age of five (109.3/1000 live births). This situation changed to 41.9 deaths per 1000 live births in NFHS-5 2019-20215.

Global and Indian studies report associations of different socioeconomic and demographic factors with infant and child mortality6-8. While maternal factors associated with infant and child mortality include the mother’s age at the time of childbirth, education, work status, place of delivery, full antenatal care and safe delivery9, child-related factors consist of sex, birth order, birth interval and birth weight10, and household factors include place of residence, drinking water facility, toilet facility, wealth, religion and ethnicity11.

Even after the massive improvement in the socioeconomic and demographic advancement of the country, the U5MR among Scheduled Tribes (ST) in India remains high compared with the general population5. India is home to 705 ST groups with a collective count of 1045,45,716 individuals, constituting 8.6 per cent of the Indian population12. According to the NFHS-5 survey in 2019-2021, the U5MR was 50 deaths per 1000 live births among the ST population compared to the national average of 41.95. This significant disparity in U5MR highlights the backwardness faced by the ST population concerning child health. In this scenario, it is crucial to identify the socioeconomic and demographic factors that put the tribal population in India at an increased risk of under-five mortality (UFM) compared with the general population. Moreover, it is crucial to prioritize proper attention and care for the children belonging to the ST category to address these disparities effectively. The present study examines the socioeconomic and demographic factors associated with UFM among ST in India by using the NFHS-5 dataset.

Material & Methods

This study was conducted at the department of Social Work, Central University of Tamil Nadu, Thiruvarur, Tamil Nadu from July 2023 to September 2023.

Study design and data source

This study used the NFHS-5 data available in the public domain (https://www.dhsprogram.com) and based on information gathered from 6,36,699 households, 724,155 women in the age groups of 15 to 49 yr, and 101,839 men within 15-54 yr. The NFHS-5 data were collected all over India through two phases: the first phase from June 17, 2019 to January 30, 2020 and the second phase from January 2 to April 30, 202113. The survey was carried out by the International Institute for Population Science (IIPS) under the guidance of the Ministry of Health and Family Welfare (MoHFW), Government of India. NFHS-5 fact sheet of India provides information on key health indicators, including infant and child mortality rates. The study extracted data for under-five children from the ‘Children’s Recode’ dataset’. ‘Children’s Data - Children’s Recode (KR)’ has one record for every child of interviewed women born in the five yr (0-59 months) preceding the survey.

Study sample

The current study is based on data from the Children’s Recode dataset encompassing responses collected from 28 Indian States and eight Union Territories from a total of 2,23,920 responses. The study specifically focuses on 47,118 women between the ages of 1 to 49 yr who met the inclusion criteria of the study. Data sorted based on the NFHS question ‘belong to a scheduled caste, scheduled tribe, and other backward class’. If the answer included ‘Scheduled Tribe’ in the response, the responding women contributed to the samples of this study. The final sample size was 47,118 women between the ages 15 to 49 yr.

Outcome and explanatory variables

The outcome variable of the study was UFM among ST in India. The explanatory variables were broadly classified into maternal, child, sociodemographic, and household factors. Mother’s current age (≤24, 25-29, 30-34, and ≥35 yr), level of education (no education, primary, secondary, and higher), age at first birth (≤24, 25-29, 30-34, and ≥35 yr), place of delivery (home and institution), delivery by caesarean section (yes/no) and complete antenatal care (yes/no) were taken into account under maternal factors. The sex of the child (male or female), birth order (1-3, ≥4), and incidence of diarrhoea (yes/no) were included under child factors. Type of state [Empowered Action Group (EAG)/Non-EAG], place of residence (urban/rural), source of drinking water (improved/unimproved), and toilet facility (improved/unimproved), religion (Hindu, Christian, Muslim and others), number of household members (1-5, ≥6), and wealth status (poor, middle and rich), were included under the sociodemographic and household factors affecting U5MR. The explanatory variables to be explored were decided upon based on prior literature search9,10.

Statistical analysis

Frequency and percentage distribution table was prepared for different maternal, child, socio demographic and household factors examined in this study. Logistic regression was initially performed to estimate the independent variables’ unadjusted odds ratio (UOR). All those variables found to be statistically significant (P<0.05) were considered for the multiple logistic regression to estimate the adjusted odds ratio (AOR) in association with U5MR among the ST population in India. The significance level was defined at P<0.05. All statistical analyses were performed using the IBM Statistical Package for the Social Sciences (SPSS) software, version 26 (IBM Corp., Armonk, N.Y., USA)

Results Cross-tabulation of maternal, child, sociodemographic and household factors and UFM

The cross-tabulation of maternal, child, sociodemographic, and household factors and UFM among ST population in India is shown in Table I. In 47,118 maternal responses, under-five mortality reported was 1,810. In a total of 47,118 respondents, 13,276 belonged to the age group of <24 yr, and 36.43 per cent of the respondents were in the age group of 25-29 yr, (17,166); 3.5 per cent of this experienced UFM. The education status of the respondents varied between mothers with no formal education, primary, secondary and higher education. The incidence of UFM was 661 among women who had no formal education, and it was 64 among higher educated women. Out of 11,793 home deliveries, 588 children, constituting five per cent of the total, faced under-five mortality. Similarly, among 4,859 women who did not receive antenatal care during pregnancy, 237, accounting for 4.9 per cent, lost their children before reaching the age of five. Most ST respondents (42,082/47118; 89%) were rural dwellers, while 5,036 were residents in urban settings. Four per cent of rural dwellers and 2.7 per cent among urban dwellers faced the problem of under-five death. The majority of the respondents belonged to the Hindu religion (25599/47118; 54%), while 16386, 1322, and 3811 belonged to Christianity (35%), Muslim (3%), and other religions (8%), respectively. The incidence of under-five mortality was high among Hindus (1171). In terms of wealth index, UFM was high among the poor wealth index family (1507), and the number was less among the middle and rich wealth index households (189 and 114, respectively). While considering the child factors of under-five mortality, it was identified that the death was higher among male children than the female. About 1024 male children died under the age of five, and the number was 78 among female children. Furthermore, 4.3 per cent of 8859 children born in a birth order of >4 experienced under-five death.

Table I. Cross-tabulation of maternal, child, sociodemographic, and household factors and UFM among the Scheduled Tribe (ST) population in India

Variables Under-five mortality Total Yes, n(%) No, n(%) Maternal factors Respondent’s current age (yr) <24 564(4.2) 12712(95.8) 13276 25-29 606(3.5) 16560(96.5) 17166 30-34 396(4) 9507(96) 9903 >35 244(3.6) 6529(96.4) 6773 Education of the respondent No education 661(5.1) 12324(94.9) 12985 Primary 308(4) 7401(96) 7709 Secondary 777(3.3) 22485(96.7) 23262 Higher 64(2) 3098(98) 3162 Delivery by caesarean section Yes 158(3.4) 4525(96.6) 4683 No 1652(3.9) 40783(96.1) 42435 Place of delivery Home 588(5) 11205(95) 11793 Institution 1222(3.5) 34103(96.5) 35325 Antenatal care Yes 1573(3.7) 40686(96.3) 42259 No 237(4.9) 4622(95.1) 4859 Age at first birth <24 1520(4) 36573(96) 38093 25-29 224(3.3) 6571(96.7) 6795 30-34 49(2.7) 1785(97.3) 1834 >35 17(4.3) 379(95.7) 396 Child factors Sex of the child Male 1024(4.3) 22976(95.7) 24000 Female 786(3.4) 22332(96.6) 23118 Birth order 1-3 1427(3.7) 36832(96.3) 38259 >4 383(4.3) 8476(95.7) 8859 Had diarrhoea recently Yes 137(4.2) 3141(95.8) 3278 No 1673(3.8) 42167(96.2) 43840 Sociodemographic and household factors Wealth index Poor 1507(4.3) 33258(95.7) 34765 Middle 189(2.7) 6825(97.3) 7014 Rich 114(2.1) 5225(97.9) 5339 Household members 1-5 1171(4.8) 23113(95.2) 24284 >6 639(2.8) 20537(97.2) 22834 Religion Hindu 1171(4.6) 24428(95.4) 25599 Christian 492(3) 15894(97) 16386 Muslim 40(3) 1282(97) 1322 Others 107(2.8) 3704(97.2) 3811 Source of drinking water Improved 1525(3.8) 38215(96.2) 39746 Unimproved 285(3.9) 7093(96.1) 7372 Toilet facility Improved 1112(3.4) 31596(96.6) 32708 Unimproved 698(4.8) 13712(95.2) 14410 State type EAG 917(5.1) 16953(94.9) 17870 Non-EAG 893(3.1) 28355(96.9) 29248 Type of place of residence Rural 1672(4) 40410(96) 42082 Urban 138(2.7) 4898(97.3) 5036 Bivariate analysis-UFM among the ST population in India

The Bivariate analysis of maternal, child, sociodemographic and household factors affecting UFM among the ST population in India is shown in Table II. In terms of the maternal factors, the likelihood of death of children below the age of five yr was low among women in the age group 25-29 yr [crude odds ratio (COR): 0.825; 95% confidence interval (CI): 0.734-0.927], and ≥35 yr (COR: 0.842; 95% CI: 0.723-0.982), compared with women in the age group of ≤24 yr. Compared to lesser age at first birth (≤24 yr), the likelihood of death of under-five children was low among women in the age group 25-29 yr (COR: 0.82; 95% CI: 0.711-0.946) and 30-34 yr (COR: 0.661; 95% CI: 0.495-0.881). Also, the death among under-five children differed significantly with the level of the mother’s education. UFM was less likely when the mothers had primary (COR: 0.776; 95% CI: 0.676-0.891), secondary (COR: 0.644; 95% CI: 0.579-0.716) or higher (COR: 0.385; 95% CI: 0.297-0.499) level of education, compared with children born to mothers who did not have any formal education. The children born at home were 1.46 times more susceptible to UFM (COR: 1.464; 95% CI: 1.324-1.62) than those born at a health facilities/centres. Antenatal care (ANC) during pregnancy was also found to be significantly associated with the death of under-five children. Compared to the mothers who received ANC, UFM was 1.32 (COR: 1.326; 95% CI: 1.153-1.525) times higher in mothers who did not receive any ANC during pregnancy.

Table II. Bivariate analysis of maternal, child, sociodemographic, and household factors affecting UFM among the ST population in India

Variables COR 95% CI P value Maternal factors Current age (yr) ≤24 Ref 25-29 0.825 0.734 - 0.927 0.001** 30-34 0.939 0.823 - 1.07 0.346 ≥35 0.842 0.723 - 0.982 0.028* Level of education No education Ref Primary 0.776 0.676 - 0.891 0.001** Secondary 0.644 0.579 - 0.716 0.001** Higher 0.385 0.297 - 0.499 0.001** Age at first birth (yr) ≤24 Ref 25-29 0.82 0.711 - 0.946 0.006* 30-34 0.661 0.495 - 0.881 0.005* ≥35 1.079 0.662 - 1.759 0.76 Delivery by caesarean section Yes Ref No 1.16 0.983 - 1.37 0.08 Place of delivery Home 1.464 1.324 - 1.62 0.001** Institution Ref Received antenatal care for pregnancy Yes Ref No 1.326 1.153 - 1.525 0.001** Child factors Sex of the child Male Ref Female 0.79 0.718 - 0.868 0.001** Birth order number 1-3 Ref ≥4 1. 166 1.039 - 1.309 0.001** Had diarrhoea recently Yes Ref No 0.91 0.761 - 1.087 0.297 Sociodemographic and household factors State type EAG Ref Non-EAG 0.582 0.53 - 0.64 0.001** Type of place of residence Rural Ref Urban 0.681 0.571 - 0.812 0.001** Source of drinking water Improved Ref Unimproved 1.007 0.885 - 1.146 0.917 Type of toilet facility Improved Ref Unimproved 1.446 1.313 - 1.593 0.001** Religion Hindu Ref Christian 0. 646 0. 58 - 0.719 0.001** Muslim 0.651 0.473 - 0.896 0.009* Others 0.603 0.493 - 0.737 0.001** Number of household members 1-5 Ref ≥6 0.568 0.515 - 0.627 0.001** Wealth index Poor Ref Middle 0.611 0.524 - 0.713 0.001** Rich 0.482 0.397 - 0.584 0.001**

Child factors, including sex and birth order, were found statistically significantly associated with UFM. Compared to male under-five children, female children had better chances of survival (COR: 0.79; 95% CI: 0.718-0.868) among the ST population in India. Furthermore, UFM was higher among children born in a birth order of ≥4 (COR: 1.166; 95% CI: 1.039-1.309) than in a 1-3 birth order.

Among the sociodemographic and household factors, type of state, place of residence, availability of toilet facilities, religion, number of household members and wealth index were significantly associated with UFM among the ST population in India (Table II).

Multivariate analysis-UFM among the ST population in India

Among the maternal factors, mothers who had secondary [adjusted odds ratio (AOR): 0.872; 95% CI: 0.774-0.982), or higher (AOR: 0.666; 95% CI: 0.501-0.885) levels of education had low odds of UFM of their children when compared to those respondents who did not have any formal education. The odds of UFM was 1.4 times higher among children born at home (AOR: 1.42; 95% CI: 1.268-1.59) than the children born at health facilities/centres. Similarly, in the children of mothers who did not receive any ANC during their pregnancy, the odds of UFM was 1.228 times higher (AOR: 1.228; 95% CI: 1.058-1.426) than in mothers who had received ANC (Table III). Among the child factors, the birth order of ≥4 (AOR: 1.422; 95% CI: 1.246-1.624) was observed to have increased odds of UFM than those children born between the birth order of 1-3. Also, compared to male children, female children had lower odds of under-five death (AOR: 0.795; 95% CI: 0.722-0.874). Among the sociodemographic and household factors, the children from the non-EAG of India were less likely to experience UFM (AOR: 0.716; 95% CI: 0.635-0.807) compared to children from the EAG States. Similarly, compared with children from a smaller family size (1-5 family members), the odds of UFM was low (AOR: 0.477; 95% CI: 0.428-0.531) among children who belonged to a family with more than six members. Furthermore, compared to those children in the poor wealth index, those in the middle and rich wealth quintile had reduced odds (AOR: 0.827; 95% CI: 0.701-0.976 and AOR: 0.742; 95% CI: 0.592-0.93, respectively) of UFM (Table III).

Table III. Multivariate analysis of maternal, child, socio-demographic, and household factors affecting UFM among the ST population in India

Variables AOR 95% CI P value Maternal factors Level of education No education Ref Primary 0.916 0.794 - 1.056 0.227 Secondary 0.872 0.774 - 0.982 0.024* Higher 0.666 0.501 - 0.885 0.005* Age at first birth (yr) ≤24 Ref 25-29 0.938 0.811 - 1.086 0.394 30-34 0.812 0.605 - 1.088 0.163 ≥35 1.246 0.76 - 2.042 0.384 Received antenatal care for pregnancy Yes Ref No 1.228 1.058 - 1.426 0.007* Place of delivery Institution Ref Home 1.42 1.268 - 1.59 0.001** Child factors Sex of child Male Ref Female 0.795 0.722 - 0.874 0.001** Birth order number 1-3 Ref ≥4 1.422 1. 246 - 1.624 0.001** Sociodemographic and household factors State type EAG Ref Non-EAG 0.716 0.635 - 0.807 0.001** Type of place of residence Rural Ref Urban 0.986 0.812 - 1.197 0.886 Type of toilet facility Improved Ref Unimproved 1.022 0.916 - 1.139 0.701 Religion Hindu Ref Christian 0.699 0.608 - 0.805 0.001** Muslim 0.814 0.586 - 1.129 0.217 Others 0.636 0.516 - 0.783 0.001** Household members 1-5 Ref ≥6 0.477 0.428 - 0.531 0.001** Wealth index Poor Ref Middle 0.827 0.701 - 0.976 0.025* Rich 0.742 0.592 - 0.93 0.01* Discussion

The present study identified that the likelihood of under-five death among children was significantly low in educated mothers compared with mothers who did not have any formal education. Similar findings were reported from other Indian studies where UFM was associated with the level of the mother’s education10,11. A study conducted in rural northern Ghana also identified a similar phenomenon where mothers’ education emerged as a strong predictor of UFM14. Noticeably, educated mothers are more likely to have a permanent source of income, good health knowledge and the capacity to make decisions on children’s health15. It has also been recorded that through regular health checkups, timely vaccinations, proper hygiene and ensuring a nutritious diet for her children, an educated mother contributes to reducing mortality and morbidity rates10.

ANC utilization was another maternal factor observed to be significantly associated with UFM among the ST population in India. Mothers who received full ANC during their pregnancy demonstrated a reduced likelihood of experiencing UFM among their children compared to mothers who did not receive any ANC as other researchers in India found9. Furthermore, a study that used Ethiopian Demographic and Health Survey (EDHS) data in 2016 revealed a decreased odds of (AOR: 0.27) UFM among women who had ANC visits during pregnancy16. This investigation further underlined that the mothers receiving ANC had a higher chance of receiving post-natal care for their children, which was considered crucial for under-five child survival16. Notably, many investigators have urged that UFM can be reduced by maintaining the stipulated frequency of ANC17 under standard of care.

The present study identified that the child’s birth order was a significant factor influencing UFM among the child factors. Other studies conducted among the STs in rural India similarly revealed that the risk of UFM was 42 per cent higher among four or more birth-order children than the first birth-order child11. A study conducted in the Southern Nations, Nationalities and Peoples’ Region (SNNPR) of Ethiopia with high U5MR regions identified that children with a birth order of second or above were associated with increased odds of under-five death compared to children with a birth order of first18. It was noted that when birth order increased, there was a decrease in the care given by the mother because of having more children18.

Similar to our findings, an analysis of cross-sectional data from Nigeria Demographic and Health Surveys (NDHS) in the years 2003, 2008 and 2013 identified that, compared to the female gender, the chances for UFM were 1.24 times higher with male gender19. This sex difference associated with UFM indicated greater capacity of survival among females resulting from a fundamental genetic advantage15,18.

The influence of wealth quintile on UFM, as observed by us, was recorded in different studies across India10,11. Children of mothers from lower-income households, in these investigations, had a higher risk of UFM than children of mothers from families with higher wealth index20,21. On the other hand, it was also identified that compared to children from a smaller family size, the children from a larger family size were less likely to die before the age of five yr. This was consistent with the findings of other previous studies22,23, which could be due to the possibility that larger households being capable of providing an expanded support network for childcare.

EAG States are the classification made by the Ministry of Health and Family Welfare, Government of India in 2001 including the States of Bihar, Chhattisgarh, Jharkhand, Madhya Pradesh, Orissa, Rajasthan, Uttar Pradesh and Uttarakhand for achieving the national health goals by monitoring and facilitating the Millennium Development Goal24. These States experience a higher infant mortality rate than the country’s national average25. Numerous studies have also documented increased rate of UFM among the EAG States in India24,26 and have identified associated challenge27. Furthermore, the prevalence of under nutrition among children in the form of stunting, wasting, and underweight have been recorded in more than 30 per cent in most EAG States28, which might have contributed to increased infant and child mortality rates29.

Our findings should be taken into account in the light of study limitations. Some important predictors of U5M, including the mother’s occupation and the child’s birth weight, could not be included in the study due to the lack of information in the NFHS-5 dataset. Also, due to the retrospective nature of the data, recall bias could have affected the mothers’ responses.

In conclusion, this study among ST in India using the NFHS-5 dataset showed an association between maternal, child, socio demographic and household factors affecting UFM in India. Based on these findings, we suggest that grass-roots-level efforts are necessary to reduce UFM, with a need to target individual-level intervention focusing on mothers. Promoting female education, addressing the vast wealth differentials and providing family planning services would be important steps in improving under-five survival among Scheduled Tribes in India.