Effect of preconception care on under-five children’s nutritional status: a follow-up study of a randomised controlled trial in a North Maharashtra district, India

Abstract

Background

Child undernutrition has remained a major public health problem. Preconception care is evident to improve pregnancy outcomes. However, the role of preconception care services in long-term child health outcomes is unknown.

Objective

To compare the nutritional status of preschool children born to women who received preconception care with children born to women who did not receive preconception care.

Design

This was an analytical cross-sectional study conducted 2 years after the end of the preconception care services provided.

Settings

Rural area in Nashik district, including two tribal and two non-tribal blocks. We conducted the study in collaboration with the government’s public health system.

Participants

All women desiring pregnancy were enrolled in July–August 2018 to receive preconception care. Women from two blocks received preconception care, while those from the other two blocks did not. Those women and their children were the participants.

Primary and secondary outcomes for the present study

Anthropometric nutritional assessment of the children and estimation of the Composite Index of Anthropometric Failure (CIAF). The association between risk factors and prevalence of CIAF was confirmed by multivariable modified Poisson regression models with robust variance.

Results

We contacted 4141 mother–offspring pairs. The proportion of the combination of stunting plus underweight was highest (35.18%), followed by triple failure (19.10%). Among the children born to mothers who received preconception care, 88.10% had some anthropometric failure, and among children born to mothers in the comparison group, 85.80% had some anthropometric failure (adjusted risk ratio (ARR) 0.99 (95% CI 0.97 to 1.02)). The incidence of some anthropometric failure among children of women residing in tribal areas was 94.38%, and among children of women residing in non-tribal areas, it was 82.54% (ARR 1.11 (95% CI 1.08 to 1.15)). Mothers of second or higher parity had a higher risk of having a child with some anthropometric failure (ARR 1.07 (95% CI 1.01 to 1.11)). Low birth weight babies had a higher risk (ARR 1.04 (95% CI 1.00 to 1.08)); boys had a higher risk (ARR 1.04 (95% CI 1.01 to 1.07)) and risk increased with age.

Conclusions

Numerous factors, particularly residence in a tribal area, affect child nutritional status and mere preconception care for women may not prevent undernutrition among preschool children.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• It was a comparison of children whose mothers received or did not receive preconception care.

• The study used the Composite Index of Anthropometric Failure as an assessment tool.

• The women were from rural areas, both tribal and non-tribal.

• This fairly large study was conducted in collaboration with the government public health system.

• There was a time gap between the provision of preconception care and the nutritional assessment of children.

Introduction

Undernutrition in preschool children has remained a significant global health concern for decades. Child undernutrition not only contributes a substantial economic burden to an individual and family, but also to the healthcare system. Although obesity among children is an emerging problem in some areas, undernutrition remains more important. The global estimate of stunting is 148.1 million (22.3%) and of wasting is 45 million (6.8%) among preschool children in 2022.¹ Undernutrition contributes to about 45% of under-five mortality.² Poor nutritional status during preschool may affect cognitive development and subsequently overall growth and development.^{3 4}

WHO has advocated universal implementation of preconception care for the prevention of maternal and child adverse events.⁵ However, in most countries, preconception care is not universal. Pregnancy outcomes and maternal outcomes after providing preconception care are evident within a short period; hence, studies revolve around these two facets,⁶,⁸ but effects on child nutrition are rarely assessed. Preconception care is crucial in ensuring optimal birth outcomes and child nutritional status by addressing maternal risk factors such as maternal nutritional status and micronutrient deficiencies in low- and middle-income countries.⁹

The authors provided preconception care from April 2018 to October 2021 and observed effects on pregnancy outcomes, Body Mass Index (BMI) and haemoglobin of the women.^{10 11} The original project did not have a follow-up component for their children. Stunting, wasting and underweight among preschool children are classical indicators of assessment of undernutrition. Their proportion in Maharashtra, including Nashik district, has hovered around 25%–50% for decades.¹² In 2000, Peter Svedberg, an economist, argued that children may have simultaneous combinations of the three types of undernutrition and proposed the Composite Index of Anthropometric Failure (CIAF). CIAF is suitable for identifying the most vulnerable children. There are many studies assessing conventional indicators, including National Health Surveys, mostly from developing countries. Some studies have assessed under-five children using the CIAF approach.¹³,¹⁶

Very few small studies have assessed the effects of comprehensive or components of preconception care on conventional nutritional indicators of the offspring. The age of assessment has considerable variability. Despite the WHO’s recommendation, universal comprehensive preconception care in the public sector in many countries is not implemented. Only a few countries, like the USA, gather information from the Pregnancy Risk Assessment Monitoring System and the Behavioural Risk Factor Surveillance System and provide services. There is a paucity of large-scale studies. To the best of the authors’ knowledge, no study has been undertaken on the effect of comprehensive preconception care on CIAF among their preschool offspring. There is a possibility of the benefits accrued from preconception care continuing to their offspring. The women from the original project, in addition to preconception care, received standard antenatal and postnatal care. The research question was: to what extent can preconception care modify the prevalence of CIAF in preschool children of women who received it, compared with those who did not? The results might provide additional convincing points for policymakers to implement preconception care universally.

Objectives of the study

1. To compare the nutritional status by measuring anthropometric failure among preschool children born to mothers who have received preconception care with children whose mothers did not receive preconception care.

2. To compare the nutritional status of preschool children between tribal areas and non-tribal areas.

3. To determine any association between undernutrition and maternal/child factors.

Methods

Study design

This was a follow-up study of a cluster randomised trial assessing nutritional status as a secondary outcome.

Setting

The study was conducted in Nashik District, India. Out of 15 blocks in the district, nine are notified tribal. A block in India is a subdistrict administrative unit covering about 100 000 population and responsible for rural development. One tribal and one non-tribal area were randomly selected for preconception care implementation. In the next stage, we chose one tribal block adjacent to the selected tribal block for comparison. Similarly, we chose one non-tribal block adjacent to the selected non-tribal blocks for comparison.^{10 11 17}

Participants

All women residing in four blocks and who desired pregnancy within 1 year were enrolled during July–August 2018. The women residing in Peith and Sinnar blocks received preconception care. The women residing in Trimbakeshwar and Niphad blocks did not receive preconception care. All children born to these mothers were the study population. We included children below 60 months. Children who were alive and present at the time of the survey were assessed for nutritional status.

Variables

We collected information about age, parity, etc, of the women (table 1) and sex, date of birth, birth weight, birth order, height and weight of each child (table 2). The study used a 5-year age-group classification for women (from below 20 to 30 and above), three groups of parity (nullipara, primipara and second or more), and two groups of BMI (underweight and others), which are standard categories. Similarly, birth weight was grouped into two (low birth weight and others), and anthropometric measurements were grouped into the standard types and grades. The study identified children having mild, moderate and severe stunting, wasting and underweight and their combinations.

Table 1. Characteristics of the mothers of the surveyed children at enrolment (n=4141).

Maternal characteristics	Intervention Number (%)	No-intervention Number (%)
Age (in years) (n=4094)
<20	402 (20.22)	414 (19.66)
20–24	1141 (57.39)	1187 (56.36)
25–29	388 (19.52)	444 (21.08)
≥30	57 (2.87)	61 (2.90)
Education (n=3919)
No formal education	155 (8.15)	169 (8.38)
Formal education	1747 (91.85)	1848 (91.62)
Employment (3870)
Employed	561 (29.46)	795 (40.44)
Unemployed	1343 (70.54)	1171 (59.56)
Residence (4141)
Tribal	937 (46.85)	592 (27.65)
Non-tribal	1063 (53.15)	1549 (72.35)
Type of family (3685) (n=3685)
Nuclear	189 (10.49)	213 (11.31)
Joint or extended	1612 (89.51)	1671 (88.69)
Consanguinity (n=3771)	339 (18.2)	479 (25.1)
Parity (n=4028)
0	1007 (51.88)	1073 (51.51)
<2	651 (33.54)	773 (37.04)
≥2	283 (14.58)	241 (11.55)
Body Mass Index (n=3795)
<18.5	781 (41.72)	753 (35.17)
≥18.5	1091 (58.28)	1170 (60.84)

Table 2. Child characteristics at the survey time (n=4141).

Characteristics	Preconception care	No preconception care
	Number (%)	Number (%)
Age in months (n=4141)
0–11	209 (10.45)	192 (8.97)
12–23	312 (15.6)	313 (14.62)
24–35	468 (23.4)	463 (21.63)
36–47	815 (40.75)	880 (41.1)
48–59	196 (9.8)	293 (13.69)
Sex
Boys	1059 (52.95)	1173 (54.79)
Girls	941 (47.05)	968 (45.21)
Birth order (n=4021)
1	1170 (60.25)	1143 (54.98)
≥2	772 (39.75)	936 (45.02)
Birth weight in kg (n=3666)
Low birth weight (<2500 g)	188 (10.88)	200 (10.32)
Normal (≥2500 g)	1540 (89.12)	1738 (89.68)
Exclusive breastfeeding	1991 (99.55)	2110 (98.6)
Complete immunisation	1995 (99.75)	1530 (71.5)
Hospitalisation	112 (5.6)	106 (5)

Data sources/measurement

The authors trained Accredited Social Health Activists (ASHAs) to collect data on selected variables and measure children’s height and weight. ASHAs from January 2023 to May 2023 carried out the data collection and nutritional assessment of all living children born to then-enrolled women and corrections till 30 September 2023. The research team, Primary Health Center (PH) and block personnel supervised the data collection and compilation.

We prepared a format in the local language. The format was validated and then pretested. The information about sociodemographic characteristics, birth information, severe medical condition, immunisation status and exclusive breastfeeding, etc, was collected from mothers. ASHAs weighed the children using portable hanging machines and measured the height in standing positions without shoes and using either tape or against marked walls. The weight was recorded in kilograms and height in centimetres with one decimal point. The data manager scrutinised hard copies of the data; queries/missing information (about 10%) were communicated, and updates were obtained till September 2023.

Nutritional status was assessed using the CIAF based on the combination of stunting, height-for-age Z-score, wasting, weight-for-height Z-score and underweight, weight-for-age Z-score indices. CIAF was categorised as no failure, wasting only, stunting only, underweight only, wasting+stunting, stunting+underweight, underweight+wasting and wasting+stunting+underweight. The WHO’s definitions and grading were used to categorise the indices.¹⁸

Sample size and sampling

The authors did not calculate the sample size, as it was a follow-up study. The required sample size may be 858 children per arm, assuming a CIAF failure rate of 52.18%¹⁹ and a 6% reduction in CIAF failure (one-tailed and superiority design). Even if the double design effect is considered, the number of children was certainly adequate in both arms.

Statistical methods

The data obtained from hard copies were entered into a Microsoft Excel sheet. We used SPSS V.29 for analysis. The prevalence of CIAF was estimated and compared among two groups (received and not preconception care, and residence in tribal or non-tribal areas) by applying the χ² test. Further analyses were conducted using STATA V.14.1. Crude (univariable) risk ratios (RRs) with 95% CIs were first estimated. We assessed potential multicollinearity among predictors included in the multivariable model. Although certain variables, such as maternal age, parity and birth order, were conceptually related, these variables were entered as categorical variables based on clinically meaningful standard groupings. Empirical assessment showed no evidence of problematic multicollinearity, with all variance inflation factors below conventional concern thresholds. Therefore, all selected covariates were retained in the adjusted model based on their epidemiological relevance, and the multivariate model was considered stable and interpretable. We calculated adjusted RRs using multilevel mixed-effects Poisson regression and generalised estimating equations with cluster-robust SE. All statistical tests were two-sided, and a p <0.05 was considered statistically significant. For the calculation of the adjusted prevalence ratio, the maternal variables used are given in table 3, and the child variables are given in table 4.

Table 3. Maternal factors associated with child anthropometric failure (n=4141).

Maternal characteristics		CIAF failure	No failure	RR (95% CI)	ARR (95% CI)
Age (in years) (n=4094)	<20	725	91	0.96 (0.88 to 1.05)	1.11 (1.04 to 1.19)
	20–24	2031	297	1.05 (0.99 to 1.11)	1.11 (1.02 to 1.20)
	25–29	705	127	1.03 (1.01 to 1.05)	1.07 (0.99 to 1.15)
	≥30	96	22	1	1
Education (n=3919)	No formal education	295	29	0.95 (0.90 to 1.00)	1.01 (0.96 to 1.07)
	Formal education	3099	496	1	1
Occupation (n=3870)	Unemployed	2223	291	0.95 (0.91 to 0.99)	1.01 (0.99 to 1.03)
	Employed	1134	222	1	1
Tribal area	Yes	1443	86	1.14 (1.14 to 1.15)	1.12 (1.10 to 1.14)
	No	2156	456	1	1
Type of family (n=3685)	Joint or extended	2853	430	1.01 (0.98 to 1.05)	1.00 (0.97 to 1.04)
	Nuclear	345	57	1	1
Consanguinity (n=3771)	Yes	718	100	1.02 (0.99 to 1.05)	1.03 (1.01 to 1.05)
	No	2545	408	1	1
Parity (n=4028)	1	1210	214	0.98 (0.94 to 1.01)	0.99 (0.97 to 1.02)
	≥2	476	48	1.04 (1.00 to 1.09)	1.06 (1.01 to 1.11)
	0	1812	268	1	1
Body Mass Index (n=3795)	<18.5	1358	176	0.97 (0.95 to 0.99)	1.02 (1.00 to 1.04)
	≥18.5	1935	326	1	1
Mother receiving preconception care	No	1837	304	0.97 (0.85 to 1.12)	1.00 (0.99 to 1.01)
	Yes	1762	238	1	1

1 represents the reference category.

ARR, adjusted risk ratio; CIAF, Composite Index of Anthropometric Failure; RR, risk ratio.

Table 4. Child factors associated with child anthropometric failure (n=4141).

Child characteristics		CIAF failure	No failure	RR (95% CI)	ARR (95% CI)
Age in months	12–23	511	114	1.09 (1.06 to 1.13)	1.10 (1.04 to 1.16)
	24–35	818	113	1.17 (1.16 to 1.18)	1.18 (1.13 to 1.24)
	36–47	1522	173	1.20 (1.14 to 1.26)	1.22 (1.12 to 1.33)
	48–59	448	41	1.22 (1.17 to 1.28)	1.24 (1.16 to 1.34)
	0–11	300	101	1	1
Sex	Boys	1972	260	1.04 (1.00 to 1.07)	1.05 (1.00 to 1.10)
	Girls	1627	282	1	1
Tribal area	Yes	1443	86	1.14 (1.14 to 1.15)	1.12 (1.10 to 1.14)
	No	2156	456	1	1
Birth order	≥2	1501	207	1.01 (0.99 to 1.04)	1.04 (1.01 to 1.08)
	1	1995	318	1	1
Birth weight (g)	<2500	348	40	1.05 (1.01 to 1.08)	1.05 (1.02 to 1.09)
	≥2500	2805	473	1	1
Exclusive breastfeeding	Not given	34	6	0.98 (0.86 to 1.11)	0.99 (0.86 to 1.15)
	Given	3565	536	1	1
Immunisation status	Partial	577	39	1.09 (1.02 to 1.16)	0.98 (0.95 to 1.00)
	Complete	3022	503	1	1
Severe medical conditions/hospitalisation	Yes	192	26	1.01 (0.96 to 1.07)	1.00 (0.93 to 1.07)
	No	3407	516	1	1
Mother receiving preconception care	No	1837	304	0.97 (0.85 to 1.12)	1.00 (0.99 to 1.01)
	Yes	1762	238	1	1

1 represents the reference category.

ARR, adjusted risk ratio; CIAF, Composite Index of Anthropometric Failure; RR, risk ratio.

Patient and public involvement

Patients and/or the public were involved in the design, conduct, reporting or dissemination plans of this research.

Results

The present study enrolled 4141 (74.83%) mother–offspring pairs from four study blocks. Figure 1 gives the area-wise details. The mean age of mothers at enrolment was 22.45 years (SD=3.23). The majority of the women had formal education (91.73%), were not formally employed (64.96%), were residing in non-tribal areas (63.08%), and lived in joint or extended families (89.09%). One-fifth of women reported consanguineous marriage (table 1). Mean height, weight and BMI of women were 151.21 cm (SD 14.22), 44.11 kg (SD 6.89) and 19.49 (SD 3.33), respectively.

Figure 1. Block-wise details of contacted and traced women and their children surveyed from Nashik district, Maharashtra, 2023.

Boys constituted 53.90%. The majority of the children were firstborn (57.52%), received exclusive breastfeeding (99.03%) and were completely immunised (85.12%) (table 2). The mean age of children was 33.15 months (SD 13.49), and the mean birth weight, height and weight of the children were 2.75 kg (SD 0.38), 85.45 cm (SD 12.97) and 11.22 kg (SD 2.54), respectively.

Among the children born to mothers who received preconception care, 88.10% had some anthropometric failure, and among children born to mothers in the comparison group, 85.80% had some anthropometric failure (χ² with Yates correction=4.60; p=0.03). The presence of some anthropometric failure among children of women residing in tribal areas was 94.38%; and among children of women residing in non-tribal areas, the proportion was 82.54% (χ² with Yates=117.68; p<0.001). The details are given in table 5. Overall, only 13.09% of children did not manifest any form of anthropometric failure, and 19.10% of children had all three classical forms of undernourishment (figure 2).

Table 5. Undernutrition in different forms among under-five children in Nashik district, India.

		No failure	Type of failure
			Wasting only	Stunting only	Underweight only	Stunting+underweight	Wasting+underweight	All three
Intervention	Yes	238	143	289	12	730	202	386
	No	304	139	326	10	727	230	405
	Total	542	282	615	22	1457	432	791
		χ2 =4.60; p=0.03
Tribal	Yes	86	78	178	6	695	136	350
	No	456	204	437	16	762	296	441
	Total	542	282	615	22	1457	432	791
		χ2 =117.68; p ≤0.0001

Figure 2. Composite Index of anthropometric failure among children in Nashik district, India, 2023.

Some maternal factors and their association with child anthropometric failure are given in table 3. We dichotomised no failure and some failure. Residing in a tribal area and parity two and above were significantly associated factors.

Similarly, when dichotomised, the highest adjusted prevalence ratio was observed among children in the 48–59 months age group (APR 1.26 (95% CI 1.17 to 1.35)). Anthropometric failure was almost 1.04 times higher among boys, 1.15 times higher among tribal children, 1.03 times higher among children with birth order 2 and above and 1.04 times higher among low birth weight babies. Factors like exclusive breastfeeding, complete immunisation and severe medical conditions were not associated with child anthropometric failure (table 4).

Discussion

For the implementation of preconception care, one tribal and one non-tribal block were selected, and adjacent blocks were selected to match sociogeographical variables. We observed desired changes in pregnancy outcomes, haemoglobin level and BMI of the women after preconception care (as per WHO guidelines).^{10 11} As a follow-up, after about 2 years, we conducted the present study to assess the effect of preconception care on the nutritional status of the children born to those women. We chose the CIAF classification as it helps to identify the most vulnerable children (having triple failure) who need immediate and intense attention. The prevalence of undernutrition using CIAF is likely to be higher as it considers single or combinations of any failure.

Our study reported that over four-fifths of children had anthropometric failure in one or more CIAF categories, higher than most of the Indian studies (24.4%–75.3%), including the pooled estimate from 52 Indian studies^{1416 20},²⁴; NFHS-5 data (50%–60%)¹³²⁵,²⁸ and many studies from African and Southeast Asian countries.²⁹,³⁸ The differences in the estimates may be due to regional differences and study settings.^{39 40} But the ranking of types of failures is similar to national findings, with stunting+underweight the highest and only underweight the lowest. ³⁸It is documented that ‘wasting and stunting’ is a theoretical possibility and may not be observed.^{28 38} We did not find any child with wasting+stunting in this study and in an earlier study,¹⁴ or in studies from National Family Health Surveys (NFHS) data of various rounds.^{13 25 26 28 38} Even if one finds such children, the proportion is usually low.⁴¹,⁴³

A few studies showed that maternal nutrition influences birth outcomes, growth and development of the newborn.⁴⁴,⁴⁶ Similar to the present study findings, a prospective cohort study did not reveal any association between preconceptional maternal weight and nutritional status of the child at birth and 5 years of age.⁴⁷ Even one randomised clinical trial did not find a difference in stunting among children whose mothers received preconception care or did not.⁴⁸ The preconception care services may not benefit the early childhood life of the offspring due to issues like unsafe water, lack of sanitation, food security and inadequate health infrastructure;⁴⁹ the families might have suffered from other challenges like poverty, unemployment and food insecurity due to the COVID-19 pandemic.⁵⁰ The probable reason for non-observance of the effect of preconception care may be that the benefits accrued after preconception care are short-lived. The second reason might be the differential distribution of other factors that determine undernutrition among children, including the services of the Integrated Child Development Scheme (ICDS), the availability of food, weaning practices and exclusive breastfeeding.⁵¹ Residence in tribal areas is a dominant determinant. The CIAF proportion has always remained higher among tribal children compared with the upper caste during the last three rounds of NFHS.⁵² In the present study, the proportion of women from the tribal area was higher in the intervention group.

The proportion of presence of any anthropometric failure among tribal children is usually high, and the range may be 70%–75%.^{2328 53},⁵⁵ A child belonging to a tribe may have odds of 1.5 times the anthropometric failure compared with a child in the general category.¹³ There may be a significant difference in the proportion of anthropometric failure between various tribes.^{54 55} The reasons for the high proportion in our study may be the inclusion of about 50% of children from difficult-to-reach tribal areas and, second, sociocultural factors that may lead to limited food diversity.

Similar to other studies, lower age at conception and higher parity were significantly associated.^{26 28 56 57} The majority of the educated enrolled women may nullify the significance of the association, as reported in other studies.^{58 59} Our study findings did not provide sufficient evidence of the association between maternal prepregnancy BMI and child nutritional status, as evident in other studies.^{21 35 59} The inconsistent results may be due to the presence of other underlying risk factors. The studies that have analysed NFHS data and have shown a significant association, which may be due to the very large number of children.^{26 28} In the present study, the BMI was certainly before conception.

Anthropometric failure increases with increasing age and is higher among boys, tribal children, those of higher birth order and low birth weight babies, consistent with systematic reviews.^{58 60} Higher prevalence of anthropometric failure among boys is consistent in almost all studies; moreover, it is observed that the gap between boys and girls is increasing.¹³ As observed in the present study, the children born with low birth weight continue to have a higher prevalence of anthropometric failure.^{26 28} The increased prevalence in older children and among second-birth-order children may be attributable to food insecurity,⁶¹ or short birth interval⁶²; and that among boys may be attributable to biological factors.⁶³ Lower socioeconomic status and poverty were reported in many studies, including India, Nepal, Indonesia and East Africa.^{32 35 64} Low birth weight is usually a significant determinant of undernutrition, as evident from existing literature.²¹

The lack of association between exclusive breastfeeding, immunisation and chronic conditions with the nutritional status of children contradicts several other studies.^{20 21} The proportion of exclusive breastfeeding and complete immunisation in the study population in all four blocks was high, while that of chronic conditions was very low. Hence, the differences in these factors may not be adequate to reveal a measurable association. To improve maternal and child health outcomes, a comprehensive approach focusing on the first 1000 days is essential; we should also include the 100 days before conception. The preconception component, apart from improvement in the health of the women, will take care of low birth weight babies, prematurity, congenital anomalies and early neonatal death. The prenatal and postnatal services of good quality need to be continued. At each interaction, breastfeeding, weaning, etc, should find a place. The national-level surveys and ICDS can identify the most vulnerable children, that is, those with all three forms of undernutrition. We strongly recommend that national-level surveys and ICDS data should disseminate information on anthropometric failure among preschool children. The Angan-wadi Workers will need some basic training. The training may be initiated from tribal areas, and the government should give top priority to treating children with all three forms of undernutrition. It will certainly reduce childhood mortality.

It was a fairly large study conducted in rural areas, including tribal areas. It was conducted with government health personnel. The study assessed the effects of preconception care on the nutritional status of preschool children. Information on antenatal and postnatal services was not collected, but we believe they are not differentially distributed across the two groups of women. The mothers of children less than 6 months of age had not received the last 3 months of preconception care. The study did not consider overweight children. As the number was minuscule, they were included in the normal weight range for children. The number of clusters was only four; hence, the CIs were narrow. Despite these limitations, the study provides insights into the magnitude of undernutrition among children below 5 years of age in rural and tribal areas.

Conclusion

The main study comparing preconception care found differences in pregnancy outcomes and women’s health between those who received preconception care and those who did not. Due to the rarity of studies assessing the effect of preconception care on childhood undernutrition, the present study, like a ‘phase two’, was conducted. The study concludes that mere preconception care for women may not protect their children from undernutrition. The complexity of numerous factors affecting a child’s nutritional status needs to be considered. There should be a continuum of care from preconception to the first 1000 days, including feeding/supplementation counselling. The very high proportion of under-five children having one or more forms of anthropometric failure was primarily due to the rural study setting, and about 50% of the children were from tribal areas.

Data availability statement

Data are available upon reasonable request.