Abstract
Background
The study of birth interval is important for maternal and child health. The long birth interval is favorable for maternal, child health, and nutritional outcomes. The present study is an attempt to explore the relationship between birth intervals and poor nutritional condition of children under five years of age in India.
Methods
The unit of analysis is children under five years of age in India. The data come from the fourth round of Indian National Family Health Survey, 2015–2016. Bivariate and logistic regression model were used to explore the relationship between birth intervals and the poor nutritional status of children.
Results
The logistic regression shows a 28% increase in stunting for those children born with a birth interval of less than 24 months. Also, there is a 26% increase in underweight for children of birth interval less than 24 months. It is evident that low birth weight, poor facilities during pregnancy are statistically associated with poor nutritional status of children.
Conclusion
Therefore, the present study attempts to determine to what extent the length of preceding birth interval influences the child undernutrition and the result revealed that short birth intervals are associated with an increased risk of child stunting and underweight even after controlling the biological, social and behavioral predictors. The study suggests that interventions that aim to increase birth intervals, including family planning and reproductive health services, may be important in improving nutritional status in children.
Keywords
1. Introduction
The study of birth interval is important for maternal and child health.
1
The long birth interval is favorable for maternal, child health, and nutritional outcomes. Birth interval is the length of time between a child's birth and a previous and/or subsequent sibling's birth. A short subsequent birth interval can place the child at risk for several reasons. The short birth interval can lead to preterm birth and low birth weight as the mother may not have recovered her nutritional status. Because of short birth interval mother's nutrient reserves become depleted, which leads to the increased risk of intrauterine growth retardation, that adversely affect infant nutrient stores at birth and nutrient delivery via breast.2
, 3
, 4
, 5
, 6
Due to short birth interval, caring for a new infant also reduces the amount of time that the mother can devote for the older child. The subsequent pregnancy may alter care practices that affect the current child's health.7
Infants should be exclusively breastfed for six months and subsequently breastfeeding be continued alongside the gradual introduction of nutritiously diverse and safe solids at an appropriate frequency.
8
,9
Several studies have found that undernutrition among children is affected by breastfeeding and complementary feeding practices.10
, 11
, 12
Besides, improved water, sanitation and hygiene practices protect against stunting,13
, 14
, 15
indoor air pollution from solid fuel use has been suggested as a risk factor.16
Childhood undernutrition is also affected by maternal characteristics, such as age,17
,18
nutritional status.19
Besides, household characteristics i.e., economic status,17
,19
caste,20
maternal20
,21
and paternal education,21
occupation22
and household decision-making roles23
are major underlying determinants of childhood undernutrition. Healthcare utilization during pregnancy, birth, the postnatal period and continuing into childhood determines a health system's ability to prevent, diagnose and treat chronic undernutrition.17
,24
,25
Birth spacing influences different outcome measures for the mother, newborn and child. The prevalence of stunting and underweight decreases as birth interval increases.
26
,27
Previous birth interval of at least 36 months was associated with a 10–50% reduction in childhood stunting.- Rutstein S.O.
United States. Agency for International Development. Further Evidence of the Effects of Preceding Birth Intervals on Neonatal, Infant, and Under-five-year Mortality and Nutritional Status in Developing Countries: Evidence from the Demographic and Health Surveys.
Macro International,
Calverton, MD2008: 78
28
Birth intervals of less than 12 months and 12–23 months were associated with higher risks for stunting as compared to 24–35 months.29
Because of the socio-cultural and spatial variations, the prevalence of childhood undernutrition varies widely both between and within countries.18
, 30
Mothers who adequately space their pregnancies are able to provide their children with the necessary nutrition for growth development and a strong immune system, thereby reducing the likelihood of childhood undernutrition. Adequate spacing between births allows women to recover and be healthy for their next pregnancy.2. Methods
2.1 Data
The four rounds of India's National Family Health Survey (NFHS) carried during 1992-93, 1998-99, 2004-05 and 2015-16, provide national representative data on child health and nutrition. The present study was based on the latest round of NFHS-4 (2015-16).
31
The survey collected information on socio-economic and hygienic conditions of households, full birth history of eligible women on a retrospective basis, child's survival status and birth intervals. The sampling design adopted is a multi-stage stratified cluster sampling. A total of 699686 eligible women in the reproductive ages 15–49 years completed the interview. As the outcomes are related to the anthropometric measures of a child, the whole data for the present analyses use child as the unit of observation, rather than the mother itself. The NFHS-4 provided related information on 259627 children born in the last five years preceding the survey.As the main objective of the present study is to explore the relationship between preceding birth interval and the outcomes of interest – stunting and underweight, the first births born to eligible women are excluded from the analytical sample due to lack of preceding birth intervals for these indexed children. Also, to eliminate the confounding effect induced by sharing characteristics of multiple births, the analytical sample is restricted to only single births. With these restrictions, the anthropometric measures were available for a total analytical sample of 159862 index children of birth order two or higher. The NFHS-4 provides the normalized z-scores for height-for-age and weight-for-age. The two outcome variables, stunting and underweight are calculated from the normalized scores as per the definition provided by the National Center for Health Statistics (NCHS)/World Health Organization.
32
As per the standards of WHO, a child is classified as stunted or underweight if his/her z-score is two or more standard deviations below the mean. Using this standard criterion the prevalence of stunting and underweight in the entire population of under-five children was 38.4% and 34.5%, respectively. The analytical sub-sample in the present study gives 57103 cases (40.9%) of stunting and 50985 cases (36.6%) of underweight. Overall, including wasting, the analytical sample indicates that 79754 children (49.9%) suffer from some form of undernutrition.The earlier three rounds of NFHS (1992-93, 1998-99, and 2004-05) reported median birth intervals of 32, 33 and 31 months respectively. The present analytical sample from the NFHS-4 shows 27% of the index children were born following short birth intervals of less than 24 months, 32% after intervals of 24–35 months, 28% after 36–59 months, and 13% after intervals of 60 months or more (Table 1).
Table 1Percentage distributions of predictors for stunting and underweight.
| Variables | Frequency(n = 159,862) | Percentage |
| Birth intervals | ||
| 0–23 months | 42,787 | 26.76 |
| 24–35 moths | 50,867 | 31.82 |
| 36–59 months | 44,640 | 27.92 |
| 60 months and more | 21,568 | 13.49 |
| Wealth index | ||
| Poorest | 49,361 | 30.88 |
| Poorer | 39,611 | 24.78 |
| Middle | 30,542 | 19.11 |
| Richer | 23,385 | 14.63 |
| Richest | 16,963 | 10.61 |
| Place of residence | ||
| Urban | 34,877 | 21.82 |
| Rural | 124,985 | 78.18 |
| Mother's educational level | ||
| No education | 62,093 | 38.84 |
| Primary | 25,840 | 16.16 |
| Secondary | 62,583 | 39.15 |
| Higher | 9346 | 5.85 |
| Birth order | ||
| Order 2 | 78,309 | 48.99 |
| Order 3 | 40,615 | 25.41 |
| Order 4 and above | 40,938 | 25.61 |
| Age of child | ||
| 0–11 months | 30,765 | 19.24 |
| 12–23 months | 31,801 | 19.89 |
| 24–35 months | 31,506 | 19.71 |
| 36–47 months | 33,259 | 20.8 |
| 48–59 months | 32,531 | 20.35 |
| Survival status | ||
| No previous child died | 149,024 | 93.41 |
| Previous child died | 10,520 | 6.59 |
| Intention to pregnancy | ||
| Wanted | 149,615 | 93.67 |
| Unwanted | 10,111 | 6.33 |
| Prenatal Care | ||
| Standard | 81,167 | 63.79 |
| Below standard | 19,538 | 15.36 |
| No care | 26,535 | 20.85 |
| Birth weight (kg.) | ||
| Low | 40,722 | 25.47 |
| Normal | 69,526 | 43.49 |
| Missing | 49,614 | 31.04 |
| Place of delivery | ||
| Delivery at health facility | 110,774 | 69.35 |
| Delivery at home | 48,952 | 30.65 |
2.2 Statistical analysis
The association between outcome variables (stunting and underweight) and a set of predictors was examined by two-way bivariate analyses using chi-squared tests. Next multiple logistic regression model was used to explore the relationship between birth intervals and outcome variables stunting and underweight controlling for several characteristics of child and mother. Some of the children in the analytical sample of 159862 observations are from the same household and same mother, so they share some of the household and maternal characteristics. This may inflate the standard errors of the estimated odds ratio from the fitted logistic regression, therefore we controlled for this clustering effect by using the “robust cluster” in the regression model to obtain unbiased standard errors. Analyses were performed using STATA version 13.
2.3 Predictors
The prominent risk factors for determining the adverse nutritional outcomes during infancy and childhood include a child's prenatal and post-natal practices, household's socio-economic condition, breastfeeding practices and size of household. The present study also considers similar categories of risk factors used in other studies.
32
The present study classifies the risk factors as household resources, household structure, reproductive history and outcomes, and the social environment of the household.32
A complete description and treatment of outcomes and predictors used in the present analysis are shown in Table 1.3. Results
3.1 Stunting and underweight by birth interval
Table 2 presents the percent distribution of stunted and underweight within the categories of birth intervals by regions and union territories of India. The results show that across the regions the percentage of stunting and underweight among the children born after an interval of less than 24 months is higher than the percentage of stunting and underweight among the children born after an interval of greater than 59 months. At the all India level, the percentages of stunting and underweight of children born after an interval of less than 24 months are 46 and 41 respectively. The bivariate analysis shows a significant association between stunting and preceding birth interval for most of the states (p < 0.001). Also, a similar significant association was observed between underweight and preceding birth interval across the regions. But there is no significant association between the outcomes (stunting and underweight) and preceding birth interval in the Union Territories.
Table 2Percent of stunted and underweight by birth intervals and region.
| Nutritional status/Birth Intervals | Stunted | Underweight | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0–23 | 24–35 | 36–59 | ≥60 | Total | p* | 0–23 | 24–35 | 36–59 | ≥60 | Total | p* | |
| North | ||||||||||||
| Haryana | 40.3 | 37.4 | 34.6 | 30.6 | 37.0 | 0.002 | 34.3 | 34.8 | 29.5 | 24.9 | 32.4 | <0.001 |
| Himachal Pradesh | 31.6 | 32.5 | 27.2 | 15.7 | 28.3 | <0.001 | 26.9 | 24.1 | 21.9 | 16.7 | 23.1 | 0.004 |
| Jammu & Kashmir | 35.8 | 33.1 | 27.3 | 23.8 | 30.3 | <0.001 | 23.4 | 19.8 | 15.4 | 14.4 | 18.3 | <0.001 |
| Delhi | 39.3 | 37.5 | 32.8 | 27.7 | 34.2 | 0.131 | 45.0 | 27.3 | 28.4 | 22.0 | 30.0 | <0.001 |
| Punjab | 32.0 | 31.0 | 21.3 | 25.1 | 27.6 | <0.001 | 25.7 | 23.4 | 21.4 | 22.1 | 23.2 | 0.296 |
| Rajasthan | 44.9 | 43.2 | 38.0 | 30.6 | 41.0 | <0.001 | 42.0 | 41.1 | 37.0 | 31.9 | 39.3 | <0.001 |
| Uttarakhand | 41.2 | 37.9 | 31.4 | 28.5 | 35.6 | <0.001 | 33.2 | 30.8 | 25.2 | 21.7 | 28.6 | <0.001 |
| Central | ||||||||||||
| Chhattisgarh | 41.5 | 37.7 | 39.2 | 33.8 | 38.5 | 0.010 | 41.9 | 40.6 | 39.4 | 35.8 | 39.9 | 0.07 |
| Madhya Pradesh | 49.2 | 45.1 | 38.1 | 34.8 | 43.7 | <0.001 | 51.6 | 46.4 | 40.7 | 35.9 | 45.6 | <0.001 |
| Uttar Pradesh | 52.4 | 49.6 | 45.6 | 39.8 | 48.3 | <0.001 | 44.2 | 41.3 | 38.2 | 34.1 | 40.5 | <0.001 |
| East | ||||||||||||
| Bihar | 52.7 | 50.5 | 47.3 | 44.1 | 49.9 | <0.001 | 48.9 | 46.3 | 43.4 | 39.5 | 45.9 | <0.001 |
| Jharkhand | 50.4 | 47.6 | 45.4 | 38.9 | 46.4 | <0.001 | 53.7 | 50.6 | 47.4 | 42.8 | 49.2 | <0.001 |
| Odisha | 43.7 | 41.7 | 36.2 | 32.2 | 37.8 | <0.001 | 42.4 | 41.6 | 38.4 | 33.7 | 38.8 | <0.001 |
| West Bengal | 47.6 | 39.9 | 36.7 | 30.2 | 37.7 | <0.001 | 43.2 | 40.5 | 33.3 | 32.7 | 36.7 | <0.001 |
| North-East | ||||||||||||
| Arunachal Pradesh | 40.8 | 32.7 | 26.9 | 27.3 | 31.4 | <0.001 | 21.8 | 19.4 | 16.9 | 18.8 | 19.0 | 0.189 |
| Assam | 43.9 | 42.8 | 37.0 | 33.4 | 38.5 | <0.001 | 34.4 | 33.9 | 30.7 | 24.4 | 30.4 | <0.001 |
| Manipur | 39.6 | 34.9 | 29.6 | 26.6 | 32.4 | <0.001 | 17.5 | 14.8 | 12.4 | 15.0 | 14.5 | 0.039 |
| Meghalaya | 50.8 | 46.7 | 39.9 | 43.8 | 45.5 | <0.001 | 33.5 | 30.7 | 28.9 | 29.3 | 30.7 | 0.258 |
| Mizoram | 36.7 | 33.2 | 30.9 | 26.9 | 32.2 | 0.002 | 17.9 | 15.4 | 12.5 | 10.9 | 14.3 | 0.002 |
| Nagaland | 34.2 | 32.7 | 28.0 | 18.8 | 30.4 | <0.001 | 18.4 | 19.5 | 16.8 | 12.8 | 17.7 | 0.053 |
| Sikkim | 40.4 | 47.7 | 27.8 | 27.9 | 32.7 | 0.010 | 17.5 | 18.5 | 19.4 | 12.0 | 15.7 | 0.317 |
| Tripura | 33.8 | 42.7 | 30.8 | 20.2 | 29.8 | 0.001 | 28.6 | 32.0 | 25.4 | 25.0 | 27.0 | 0.57 |
| West | ||||||||||||
| Goa | 17.5 | 39.6 | 14.3 | 20.4 | 22.7 | 0.013 | 12.5 | 39.6 | 25.0 | 25.9 | 26.3 | 0.039 |
| Gujarat | 49.0 | 45.5 | 38.0 | 32.9 | 42.4 | <0.001 | 48.8 | 47.5 | 38.6 | 36.5 | 43.7 | <0.001 |
| Maharashtra | 43.6 | 38.3 | 35.2 | 25.5 | 37.3 | <0.001 | 43.4 | 40.9 | 37.2 | 28.2 | 39.0 | <0.001 |
| South | ||||||||||||
| Andhra Pradesh | 36.4 | 34.3 | 32.5 | 33.8 | 34.5 | 0.719 | 33.4 | 33.3 | 32.4 | 31.0 | 32.9 | 0.952 |
| Karnataka | 44.2 | 43.3 | 37.9 | 33.3 | 41.0 | <0.001 | 40.3 | 40.9 | 37.1 | 29.0 | 38.4 | <0.001 |
| Kerala | 18.0 | 22.8 | 21.7 | 19.2 | 20.6 | 0.603 | 19.5 | 16.9 | 15.9 | 17.1 | 16.9 | 0.817 |
| Tamil Nadu | 35.5 | 29.1 | 25.0 | 23.6 | 28.9 | <0.001 | 31.8 | 27.1 | 22.7 | 20.6 | 26.2 | <0.001 |
| Telangana | 38.4 | 30.3 | 28.1 | 36.0 | 33.2 | 0.022 | 33.4 | 29.7 | 27.3 | 41.2 | 31.8 | 0.028 |
| UTs | ||||||||||||
| A & N Islands | 24.7 | 26.0 | 23.4 | 28.4 | 25.6 | 0.910 | 19.2 | 19.2 | 31.2 | 10.8 | 20.2 | 0.02 |
| Chandigarh | 41.7 | 23.1 | 31.0 | 30.0 | 31.3 | 0.566 | 16.7 | 34.6 | 24.1 | 30.0 | 26.3 | 0.514 |
| D & N Haveli | 48.3 | 52.1 | 41.2 | 20.0 | 42.5 | 0.060 | 44.8 | 62.5 | 35.3 | 28.0 | 44.4 | 0.013 |
| Daman & Diu | 22.2 | 38.0 | 29.7 | 16.7 | 28.3 | 0.153 | 25.9 | 28.0 | 31.3 | 22.2 | 27.7 | 0.804 |
| Lakshadweep | 33.3 | 20.8 | 17.4 | 23.8 | 22.3 | 0.614 | 20.0 | 25.0 | 23.9 | 14.3 | 19.6 | 0.548 |
| Puducherry | 31.9 | 27.5 | 22.6 | 25.0 | 26.6 | 0.407 | 23.9 | 26.7 | 18.8 | 19.2 | 22.1 | 0.391 |
| India | 46.1 | 43.0 | 37.9 | 32.2 | 40.9 | <0.001 | 41.0 | 38.5 | 33.9 | 28.7 | 36.5 | <0.001 |
Note: A & N = Andaman and Nicobar; D & N = Dadar and Nagar; *chi-square test for significance difference between stunting/underweight and birth intervals for each state.
3.2 Predictors of stunting
The main relationship that we want to explore in the present study is between the birth interval and the nutritional outcomes - stunting and underweight. It is evident that as the birth interval decreases the rate of stunting increases. Children born after an interval of less than 24 months experience 46% of stunting (Table 3). Among children born after 60 months or more, 32% are stunted. We retained the multiple logistic regression controlled for several backgrounds against the model without controlling the backgrounds (BIC: 138520.7 vs 139748.8). The interpretations are for the model with controlled variables. The multiple logistic regression model once again confirms the increase in the rate of stunting with a decrease in birth intervals after controlling for other characteristics in the model. Children born after less than 24 months (OR = 1.28, 95%CI: 1.24, 1.33) were significantly more likely to be stunted than those born after 36–59 months. Similarly, the odds of being stunted for children born after 24–35 months (OR = 1.14, 95% CI:1.10, 1.18) were significantly higher than those born after 36–59 months. Increase in birth interval shows lower chances of stunting where children born after 60 months or more (OR = 0.89,95% CI:0.85, 0.93) were significantly less likely to be stunted compared to those born after 36–59 months.
Table 3Rate of stunting by predictors and likelihood of stunting for under-five children.
| Predictors | Frequency (n = 159,862) | Rate per 100 Children | Odds ratios | ||
|---|---|---|---|---|---|
| Value | p-value* | Value | 95%CI | ||
| Birth intervals | |||||
| 0–23 months | 16,865 | 46.12 | 1.28 | 1.24, 1.33 | |
| 24–35 moths | 19,255 | 43.00 | 1.14 | 1.10, 1.18 | |
| 36–59 months | 14,922 | 37.95 | <0.001 | 1.00 | |
| 60 months or more | 6061 | 32.22 | 0.89 | 0.85, 0.93 | |
| Wealth index | |||||
| Poorest | 21,790 | 51.34 | 1.38 | 1.32, 1.43 | |
| Poorer | 15,285 | 44.17 | 1.20 | 1.15, 1.24 | |
| Middle | 10,082 | 37.53 | <0.001 | 1.00 | |
| Richer | 6390 | 30.90 | 0.82 | 0.79, 0.86 | |
| Richest | 3556 | 23.89 | 0.68 | 0.65, 0.72 | |
| Place of residence | |||||
| Urban | 10,374 | 34.03 | <0.001 | 1.00 | |
| Rural | 46,729 | 42.87 | 0.94 | 0.90, 0.97 | |
| Mother's educational level | |||||
| No education | 26,604 | 49.85 | <0.001 | 1.00 | |
| Primary | 9854 | 43.67 | 0.90 | 0.86, 0.93 | |
| Secondary | 18,835 | 34.06 | 0.76 | 0.74, 0.79 | |
| Higher | 1810 | 21.96 | 0.59 | 0.55, 0.64 | |
| Birth order | |||||
| Order 2 | 25,356 | 36.87 | <0.001 | 1.00 | |
| Order 3 | 14,849 | 41.80 | 1.09 | 1.05, 1.12 | |
| Order 4 and above | 16,898 | 48.05 | 1.23 | 1.19, 1.28 | |
| Age of child | |||||
| 0–11 months | 5731 | 22.25 | <0.001 | 1.00 | |
| 12–23 months | 12,690 | 45.29 | 3.00 | 2.88, 3.12 | |
| 24–35 months | 12,679 | 45.62 | 2.90 | 2.79, 3.03 | |
| 36–47 months | 13,758 | 46.85 | 2.95 | 2.83, 3.08 | |
| 48–59 months | 12,245 | 42.89 | 2.42 | 2.31, 2.53 | |
| Survival status | |||||
| No previous child died | 53,347 | 40.86 | 0.005 | 1.00 | |
| Previous child died | 3668 | 42.40 | 0.91 | 0.86, 0.96 | |
| Intention to pregnancy | |||||
| Wanted | 53,160 | 40.62 | <0.001 | 1.00 | |
| Unwanted | 3943 | 45.83 | 1.04 | 0.99, 1.09 | |
| Prenatal Care | |||||
| Standard | 25,557 | 35.37 | <0.001 | 1.00 | |
| Below standard | 7208 | 41.21 | 1.05 | 1.02, 1.09 | |
| No care | 10,407 | 45.84 | 1.08 | 1.04, 1.12 | |
| Birth Weight (kg.) | |||||
| Normal | 15,868 | 44.52 | <0.001 | 1.00 | |
| Low | 21,388 | 34.43 | 1.44 | 1.40, 1.49 | |
| Missing | 19,847 | 47.58 | 1.21 | 1.16, 1.26 | |
| Place of delivery | |||||
| Delivery at health facility | 37,433 | 38.48 | <0.001 | 1.00 | |
| Delivery at home | 19,670 | 46.62 | 0.98 | 0.94, 1.02 | |
a controlling for caste, sex of index child, religion, region, and mother's age at birth of child; *chi-square test for significance difference between stunting and birth intervals for each state.
Among the household resources, there is a statistically significant relationship between the household wealth index (standard of living index) and stunting. Almost 51% and 24% of children in the poorest and richest quintiles of wealth index are respectively stunted. The risk of stunting decreases as the wealth index quintile increases, where children who are in the poorest quintile (OR = 1.38, 95% CI:1.32, 1.43) were significantly more likely to be stunted compared to those in the middle quintile. Those children who are in better off household i.e. in the richest quintile are 32% (OR = 0.68, 95% CI:0.65, 0.72) less likely to be stunted compared to those who are in the middle quintile. Maternal education is also associated with stunting. The bivariate analysis shows that children whose mothers do not have any education are 50% stunted and this figure goes down to 22% when children are from mothers with higher education. The odds of stunting for children whose mothers have primary (OR = 0.90, 95% CI:0.86, 0.93), secondary (OR = 0.76, 95% CI:0.74, 0.79) and higher (OR = 0.59, 95% CI:0.55, 0.64) education were significantly less likely than those mothers who do not have any education.
Both the factors related to the household structure are statistically associated with the rate of stunting. The odds of stunting ranges between 2.42 and 3.00 for older children compared to infants. The high-birth order is also related to stunting; the bivariate analysis shows that higher-birth order children experience higher rates of stunting. This is again confirmed by the logistic regression model where children of birth orders three (OR = 1.09, 95% CI:1.05, 1.12) and four or more (OR = 1.23, 95% CI:1.19, 1.28) were significantly more likely to be stunted than children of birth-order two.
Most of the factors related to reproductive and outcomes are significantly associated with stunting. The odd of stunting for children of unwanted pregnancy (OR = 1.04, 95% CI: 0.99, 1.09) were significantly higher as compared to children of wanted pregnancy. But the relation is not statistically significant. Children having older sibling's death have lesser odds of stunting compared to their counterparts. Children who have low birth weight (OR = 1.44, 95% CI:1.40, 1.49) were significantly having higher odds of experiencing stunting than children of normal birth weight. Quality cares given to mothers before birth of a child is also very important; children of mothers who got prenatal care of below-standard have a 1.09 time higher odds of being stunted compared to those who got standard prenatal care. The relationship is also statistically significant. After controlling for several confounders, the study reveals a higher percentage of stunting for those children who were born less than 24 months (Table 5, Appendix).
3.3 Predictors of underweight
The results of bivariate and logistic regression analyses are shown in Table 4. Once again it is seen that children born after an interval of less than 24 months experience 41% of being underweight as compared to 28% of those children born after an interval of 60 months or more. The multiple logistic regression model with controlled variables (BIC: 134464.5 vs 137141.3) was selected as the final model for interpretation. The multiple logistics regression model shows the increase in the rate of underweight with a decrease in birth intervals after controlling for other characteristics in the model. Children born after less than 24 months (OR = 1.26, 95% CI: 1.22, 1.31) were significantly more likely to be underweight than those born after 36–59 months. The chances of being underweight for children born after 24–35 months (OR = 1.13, 95% CI:1.09, 1.17) were significantly higher than those born after 36–59 months. Lower chances of underweight were associated with higher birth interval, children born after 60 months or more (OR = 0.93, 95% CI:0.89, 0.97) were significantly less likely to be underweight compared to those born after 36–59 months.
Table 4Rate of underweight by predictors and likelihood of underweight for under-five children.
| Variables | Frequency (n = 159,862) | Rate per 100 Children | Odds ratios | ||
|---|---|---|---|---|---|
| Value | P-value* | Value | 95%CI | ||
| Birth interval | |||||
| 0–23 months | 14,992 | 41.00 | 1.26 | 1.22, 1.31 | |
| 24–35 moths | 17,241 | 38.50 | 1.13 | 1.09, 1.17 | |
| 36–59 months | 13,354 | 33.96 | <0.001 | 1.00 | |
| 60 months or more | 5398 | 28.69 | 0.93 | 0.89, 0.97 | |
| Wealth index | |||||
| Poorest | 20,443 | 48.16 | 1.51 | 1.44, 1.57 | |
| Poorer | 13,309 | 38.46 | 1.21 | 1.16, 1.26 | |
| Middle | 8609 | 32.05 | <0.001 | 1.00 | |
| Richer | 5575 | 26.96 | 0.85 | 0.81, 0.89 | |
| Richest | 3049 | 20.49 | 0.72 | 0.68, 0.76 | |
| Place of residence | |||||
| Urban | 9268 | 30.40 | <0.001 | 1.00 | |
| Rural | 41,717 | 38.28 | 0.87 | 0.84, 0.91 | |
| Mother's educational level | |||||
| No education | 24,280 | 45.49 | <0.001 | 1.00 | |
| Primary | 8708 | 38.59 | 0.89 | 0.86, 0.93 | |
| Secondary | 16,447 | 29.74 | 0.77 | 0.74, 0.80 | |
| Higher | 1550 | 18.81 | 0.61 | 0.57, 0.66 | |
| Birth order | |||||
| Order 2 | 22,932 | 33.34 | <0.001 | 1.00 | |
| Order 3 | 13,294 | 37.42 | 1.06 | 1.02, 1.09 | |
| Order 4 and above | 14,759 | 41.96 | 1.14 | 1.09, 1.19 | |
| Age of child | |||||
| 0–11 months | 6943 | 26.96 | <0.001 | 1.00 | |
| 12–23 months | 10,210 | 36.44 | 1.58 | 1.52, 1.64 | |
| 24–35 months | 10,995 | 39.56 | 1.89 | 1.82, 1.97 | |
| 36–47 months | 11,654 | 39.69 | 1.85 | 1.77, 1.93 | |
| 48–59 months | 11,183 | 39.17 | 1.75 | 1.68, 1.83 | |
| Survival status | |||||
| No previous child died | 47,531 | 36.41 | 0.005 | 1.00 | |
| Previous child died | 3375 | 39.01 | 0.95 | 0.90, 1.01 | |
| Intention to pregnancy | |||||
| Wanted | 47,588 | 36.36 | <0.001 | 1.00 | |
| Unwanted | 3397 | 39.49 | 0.97 | 0.92, 1.02 | |
| Prenatal care | |||||
| Standard | 23,138 | 32.02 | <0.001 | 1.00 | |
| Below standard | 6937 | 39.66 | 1.07 | 1.03, 1.11 | |
| No care | 9360 | 41.23 | 1.01 | 0.97, 1.04 | |
| Birth weight (kg.) | |||||
| Normal | 18,023 | 29.01 | <0.001 | 1.00 | |
| Low | 15,721 | 44.11 | 1.75 | 1.70, 1.81 | |
| Missing | 17,241 | 41.33 | 1.24 | 1.19, 1.29 | |
| Place of delivery | |||||
| Delivery at health facility | 33,646 | 34.59 | <0.001 | 1.00 | |
| Delivery at home | 17,339 | 41.09 | 1.05 | 1.01, 1.09 | |
a : controlling forcaste, sex of index child, religion, region, and mother's age at birth of child; *chi-square test for significance difference between underweight and birth intervals for each state.
Increasing in the household wealth status is statistically significantly related to lower risk of childhood underweight. Almost 48 and 20 %s of children in the poorest and richest quintiles of wealth index are respectively underweight. Children who are in the poorest quintile (OR = 1.51, 95% CI:1.44, 1.57) were significantly more likely to be underweight compared to those in the middle quintile. Those children who are in the wealthier household, i.e., in the richest quintile were (OR = 0.72, 95% CI:0.68, 0.76) less likely to be underweight compared to those who are in the middle quintile. Maternal education is also associated with underweight. The bivariate analysis shows that 45 and 19 %s of children from mothers who do not have any education and from mothers with higher are respectively underweight. The odds of underweight for children whose mothers have primary (OR = 0.89, 95% CI:0.86, 0.93), secondary (OR = 0.77, 95% CI:0.74, 0.80) and higher (OR = 0.61, 95% CI:0.57, 0.66) education were significantly less likely than those mothers who do not have any education.
The odds of underweight for children age 12–23 months (OR = 1.58, 95% CI:1.52, 1.64) and age 48–59 months (OR = 1.75, 95% CI:1.68, 1.83) were more likely as compared to infants. The bivariate analysis shows that higher birth order is linked with an increased risk of underweight. Multiple logistic regression model confirms that children of birth orders three (OR = 1.06, 95% CI:1.02, 09) and four or more (OR = 1.14, 95% CI:1.09, 1.19) were significantly more likely to be underweight than children of birth-order two.
The odd of underweight for children of unwanted pregnancy (OR = 0.97, 95% CI: 0.92, 1.02) were significantly lesser as compared to children of wanted pregnancy. Children having older siblings’ death have lesser odds of underweight compared to their counterparts. Low birth weight children (OR = 1.75, 95% CI:1.70, 1.81) were significantly having higher odds of underweight than children of normal birth weight. Children were more likely to be underweight if the mothers received prenatal care of below-standard (OR = 1.07, 95% CI:1.03, 1.11) and delivered a child at home (OR = 1.05, 95% CI:1.01, 1.09). Similar to stunting there is a high percentage of underweight for those born less than 24 months after controlling for several confounders (Table 6, Appendix). We did not find much difference in the odds ratios for stunting and underweight between controlled and uncontrolled background characteristics (Figure 1 and Figure 2, Appendix).
4. Discussion
It has been a topic of discussion in the literatures that undernutrition leads to child mortality and morbidity in most developing countries. Therefore, it is important to investigate the biological, social, and behavioral mechanisms by which adequate birth spacing might contribute to child health. Of these birth interval plays an important role in child undernutrition. The finding shows that short birth intervals are associated with an increased risk of child stunting and underweight. A child of birth interval 0–23 months has a higher odds of experiencing stunting and underweight as compared to a child of higher birth interval. Older children experience a higher chance of stunting and underweight as compared to infants. A child age 12–23 has a higher chance of experience stunting whereas a child age 24–35 months has a higher chance of underweight. Low birth weight is another predictor of stunting and underweight. The other significant associations with stunting and underweight were maternal education, household wealth index, prenatal care and place of delivery. The risk of a child experience stunting and underweight decreases as the mother's level of education increases. Children whose mothers were belonging from the poorest wealth quintile have a higher chance of being stunting and underweight. The odds of underweight for children of unwanted pregnancy were significantly lesser as compared to children of wanted pregnancy. Children of those mothers who have received quality prenatal care were less likely to experience stunting and overweight.
Another predictor of stunting and underweight is the low birth weight which has adverse consequences on infant and child health. In corroboration with earlier studies, our study once again confirms a statistically significant association between low birthweight and poor nutritional status during infancy and early childhood. This finding is an indication to plan for intervention during pregnancy/prior to pregnancy to prevent low birthweight infants. However, the relationships between low birthweight, short birth intervals and poor childhood nutrition are complex and hence further research is immediate to better understand the relationships.
Our results indicate that short proceeding birth intervals are associated with diminished height by early childhood. Our results suggest that interventions that aim to increase birth intervals, including family planning and reproductive health services, may still be important in improving stunting in children (particularly at early ages) as well as positively contributing to child health more generally. Encouraging women to space births through family planning services and educational awareness could contribute to reducing childhood undernutrition, improve maternal health, and provide healthy childhood development. Birth intervals can be lengthened through various approaches, but are principally increased through the use of family planning methods, extended exclusive breast-feeding, spontaneous or induced abortions. Longer spacing between two births allows for the optimum use of the parent time inputs and resources for each child, which in turn improves child health.
Role of funding source
None of the authors received any funding for the article.
Ethics approval and consent to participate
The Ethics Review Board at the International Institute for Population Sciences, Mumbai, India granted Measure DHS/ICF International ethical approvals before the surveys were conducted, with written informed consent obtained from participants during the surveys. The questionnaires used for the survey were reviewed and approved by ICF International Institutional Review Board (IRB) to ensure they met the United States Department of Health and Human Services regulations for the protection of human participants, as well as the host country's IRB, to ensure compliance with national laws. Approval was sought from Measure DHS and permission was granted for this use.
Contributors
HSC and HS designed the study. HSC analysed the data. HSC wrote the first version. All authors revised it critically and interpreted the data and they have also seen and approved the final version.
Declaration of competing interest
The authors declare that they have no conflict of interest.
Appendix A. Supplementary data
The following are the Supplementary data to this article:
- Multimedia component 1
- Multimedia component 2
References
- Report of a WHO Technical Consultation on Birth Spacing.World Health Organization, Geneva, Switzerland2005 (Technical Report)
- Positive effect of large birth intervals on early childhood hemoglobin levels in africa is limited to girls: cross-sectional DHS study.PloS One. 2015; 10: 1-14
- Nutrition for Health and Development: A Global Agenda for Combating Malnutrition.WHO, Geneva2000
- WHO Anthro for Personal Computers Manual: Software for Assessing Growth of the World's Children.WHO, Geneva2009
- Factors causing malnutrition among under five children in Bangladesh.Pakistan J Nutr. 2006; 5: 558-562
- Birth Spacing Literature: Maternal and Child Nutrition Out Comes. Programme in International Nutrition.University of California, Davis2004
- The effect of birth interval on malnutrition in Bangladeshi infants and young children.J Biosoc Sci. 2000; 32: 289-300
- Infant and Young Child Feeding: A Tool for Assessing National Practices, Policies and Programmes. vol. 140. World Health Organization, Geneva2003
- Determinants of inappropriate complementary feeding practices in young children in India: secondary analysis of National Family Health Survey 2005– 2006.Matern Child Nutr. 2012; 8: 28-44
- Mainstreaming nutrition into maternal and child health programmes: scaling up of exclusive breastfeeding.Matern Child Nutr. 2008; 4: 5-23
- Impact of maternal education about complementary feeding and provision of complementary foods on child growth in developing countries.BMC Publ Health. 2011; : S25
- Community-based supplementary feeding for promoting the growth of children under five years of age in low and middle income countries.Cochrane Database Syst Rev. 2012; 6: CD005039
- Narrowing socioeconomic inequality in child stunting: the Brazilian experience, 1974–2007.Bull World Health Organ. 2010; 88: 305-311
- Some risk factors associated with overweight, stunting and wasting among children under 5 years old.J Pediatr (Rio J). 2008; 84: 251-257
- The effect of water and sanitation on child health: evidence from the demographic and health surveys 1986–2007.Int J Epidemiol. 2011; 40: 1196-1204
- Does biofuel smoke contribute to anaemia and stunting in early childhood?.Int J Epidemiol. 2007; 36: 117-129
- Reducing child malnutrition in Nigeria: combined effects of income growth and provision of information about mothers' access to health care services.Soc Sci Med. 2010; 71: 1973-1980
- Maternal smoking, biofuel smoke exposure and child height-for-age in seven developing countries.Int J Epidemiol. 2009; 38: 1342-1350
- Socioeconomic inequality in malnutrition in developing countries.Bull World Health Organ. 2008; 86: 282-291
- Socioeconomic inequalities in childhood undernutrition in India: analyzing trends between 1992 and 2005.PloS One. 2010; 5: 14-30
- Effect of parental formal education on risk of child stunting in Indonesia and Bangladesh: a cross-sectional study.Lancet. 2008; 371: 322-328
- Heavy agricultural workloads and low crop diversity are strong barriers to improving child feeding practices in the Bolivian Andes.Soc Sci Med. 2012; 75: 1673-1684
- Does maternal autonomy influence feeding practices and infant growth in rural India?.Soc Sci Med. 2011; 73: 447-455
- Handbook Integrated Management of Childhood Illness. vol. 163. World Health Organization, Geneva2005
- The effect of a vaccination program on child anthropometry: evidence from India's Universal Immunization Program.J Public Health. 2012; 34: 489-497
- Effects of preceding birth intervals on neonatal, infant and under-five years mortality and nutritional status in developing countries: evidence from the demographic and health surveys.Int J Gynaecol Obstet. 2005; 89 (Suppl 1:S7–24)
- United States. Agency for International Development. Further Evidence of the Effects of Preceding Birth Intervals on Neonatal, Infant, and Under-five-year Mortality and Nutritional Status in Developing Countries: Evidence from the Demographic and Health Surveys.Macro International, Calverton, MD2008: 78
- Does birth spacing affect maternal or child nutritional status? A systematic literature review.Matern Child Nutr. 2007; 3: 151-173
- Socio-economic differences in health, nutrition and population within developing countries: an overview.Niger J Clin Pract. 2007; 10: 272-282
- ORC Macro.National Family Health Survey India, Mumbai: IIPS2018 (2015–16 NFHS-4)
- Use and interpretation of anthropometric indicators of nutritional status.Bull World Health Organ. 1986; 64: 929-941
- Reconsidering childhood undernutrition: can birth spacing make a difference? An analysis of the 2002-2003 EI Salvador National Family Health Survey.Matern Child Nutr. 2009; 5: 49-63
Article info
Publication history
Published online: April 27, 2020
Accepted:
April 19,
2020
Received in revised form:
April 3,
2020
Received:
December 23,
2019
Identification
Copyright
© 2020 INDIACLEN. Published by Elsevier, a division of RELX India, Pvt. Ltd. All rights reserved.
