Publications

Background

Early identification of high-risk pregnancies could reduce stillbirths, yet remains a challenge in low- and middle-income countries (LMICs). This study aims to estimate the associations between easily observable risk factors and stillbirths, and construct a risk score which could be adopted in LMICs to identify pregnancies with high risk of stillbirths.

Methods

Using the most recent Demographic and Health Surveys from 50 low- and middle-income countries (LMICs) with available data between January 1, 2010 and December 31, 2021, we analysed a total of 22 factors associated with stillbirths in a series of single-adjusted and mutually adjusted logistic regression models. Upon identification of the risk factors with the strongest associations, we constructed a risk score on the basis of the magnitude of the β coefficient to examine the cumulative effects of risk factors on stillbirths. To assess whether the associations between risk scores and stillbirths were moderated by protective factors, we added an interaction term between the identified protective factor and risk scores to the regression model. We also conducted two sets of subgroup analyses for previous history of pregnancy and maternal age at pregnancy and four sets of supplementary analyses to test the robustness of the results.

Findings

Among the 795,642 women identified for analysis with at least one pregnancy within the five years before the survey, the most recent pregnancy of 8968 (1.13%) ended as stillbirths. Using a mutually adjusted regression model, we found that the top factors showing the strongest associations with stillbirths were short maternal height (odds ratio [OR]: 1.99, 95% confidence interval [CI]: 1.48–2.67, P < 0.001), interpregnancy interval less than six months (OR: 1.84, 95% CI: 1.42–2.38, P < 0.001), previous stillbirth history (OR: 1.55, 95% CI: 1.07–2.26, P < 0.020), low maternal education (OR: 1.50, 95% CI: 1.01–2.24, P = 0.045), and lowest household wealth (OR: 1.32, 95% CI: 1.08–1.61, P = 0.008). A female household head was a protective factor with an OR of 0.71 (95% CI: 0.55–0.90, P = 0.005). Single-adjusted models, subgroup analyses, and sensitivity analyses showed generally consistent results. We also found that the odds of stillbirths increased with a larger risk score with a P trend <0.001. Compared with women without any risk factors, women with a risk score of 5 or more were 4.11 (95% CI: 2.83–5.97, P < 0.001) times more likely to have their pregnancies ending up as stillbirths. However, these associations were weakened if the head of household was female.

Interpretation

Our study suggested that short maternal height, low socioeconomic status, previous stillbirth history, low maternal education, and very short interpregnancy interval had the strongest associations with stillbirths. The construction of risk scores using easily observable risk factors could be an effective way to identify high-risk pregnancies in resource-poor settings.

The minimum dietary diversity (MDD) indicator as defined by the WHO is commonly used to assess micronutrient deficiency in young children. However, individual food item-specific consumption patterns may be overlooked when focusing solely on this indicator. We provide a comprehensive view on food item and food group consumption patterns of children aged 6–23 months old using DHS data from 59 low- and middle-income countries. Consumption levels of food items ranged from 79.0 % for breastfeeding to 5.9 % for organ meats, showing particularly low levels for protein rich food items. There were significant differences in food item consumption levels for different countries as well as household correlates’ relevance such as a household’s wealth decile and the child’s age group, hinting towards potential underlying mechanisms such as regional availability, household’s available resources and awareness of food group’s importance from early age. The results suggest that the analysis of MDD should be complemented with information on individual food item consumption to identify priorities for policy makers aiming to fight undernutrition across the globe..

Question  
What share of deaths among children younger than 5 years occur before 2 years of age in low- and middle-income countries?

Findings  
In this cross-sectional study of 2 827 515 children younger than 5 years from 77 countries, a large majority (81.5%) of deaths occurred before 2 years of age in all countries, among boys and girls, and in the households with the highest and lowest living standards.

Meaning  
These findings suggest that coverage of potentially life-saving interventions should be ensured for children younger than 2 years of age in low- and middle-income countries.

Dietary diversity is an important indicator of child malnutrition. However, little is known about the geographic variation of diet indicators across India, particularly within districts and across states. As such, the purpose of this paper was to elucidate the small area variations in diet indicators between clusters within districts of India. Overall, we found that clusters were the largest source of variation for children not eating grains, roots, and tubers, legumes and nuts, dairy, vitamin A-rich vegetables and fruits, and other vegetables and fruits. We also found positive correlations between the district percent and cluster standard deviations of children not breastfeeding or eating grains, roots, and tubers, but negative correlations between the district percent and cluster standard deviation for the remaining seven outcomes. These findings underscore the importance of targeting clusters to improve child dietary diversity.

In India, districts serve as central policy unit for program development, administration and implementation. The one-size-fits-all approach based on average prevalence estimates at the district level fails to capture the substantial small area variation. In addition to district average, heterogeneity within districts should be considered in policy design. The objective of this study was to quantify the extent of small area variation in child stunting, underweight and wasting across 36 states/Union Territories (UTs), 640 districts (and 543 PCs), and villages/blocks in India. We utilized the 4th round of Indian National Family Health Survey (NFHS-4) conducted in 2015–2016. The study population included 225,002 children aged 0–59 months whose height and weight information were available. Stunting was defined as height-for-age z-score below 2 SD from the World Health Organization child growth reference standards. Similarly, underweight and wasting were each defined as weight-for-age < -2 SD and weight-for-height < -2 SD from the age- and sex-specific medians. We adopted a four-level logistic regression model to partition the total variation in stunting, underweight and wasting. We computed precision-weighted prevalence of child anthropometric failures across districts and PCs as well as within-district/PC variation using standard deviation (SD) measures. For stunting, 56.4% (var: 0.237; SE: 0.008) of the total variation was attributed to villages/blocks, followed by 25.8% (var: 0.109; SE: 0.030) to states/UTs, and 17.7% (Var: 0.074; SE: 0.006) to districts. For underweight and wasting, villages/blocks accounted for 38.4% (var: 0.224; SE: 0.007) and 50% (var: 0.285; SE: 0.009), respectively, of the total contextual variance in India. Similar findings were shown in multilevel models incorporating PC as a geographical unit instead of districts. We found high positive correlations between mean prevalence and SD for stunting (r = 0.780, p < 0.001), underweight (r = 0.860, p < 0.001), and wasting (r = 0.857, p < 0.001) across all districts in India. A similar pattern of correlation was found for PCs. Within-district and within-PC variation are the primary source of variation for child malnutrition in India. Our results suggest the importance of considering heterogeneity within districts and PCs when planning and administering child nutrition policies.

Objective

National averages obscure geographic variation in program performance. We determined Parliamentary Constituency (PC)-wise estimates of TB notification to guide political engagement.

Methods

We extracted district-level TB notification data from the 2018 annual TB report. We derived PC-level estimates by building a ‘cross-walk’ between districts and PCs using boundary shapefiles. We described the spatial distribution of the PC-wise estimates of Total Notification Rate and percentage of Private Sector Notification.

Results

The median PC-wise Total Notification Rate was 126.24/100 000 (IQR: 94.86/100 000, 162.22/100 000). The median PC-wise Percentage Private Sector Notification was 18.03% (IQR: 9.56%, 26.84%). Only 16 (2.94%) PCs met the target of 50% private sector notification. Most of high notification rates in PCs were driven by high notification in public sector. There was geographic – both interstate and within state inter-PC – variation in the estimates of these indicators. The study identified some geographic patterns of notification – high positive outlier PCs with adjoining PCs in lower deciles of notification rates, intra-state differences in PC performance, and similarities in notification rates of adjoining PCs in different states.

Conclusion

In addition to regional inequality, the study identified geospatial patterns that can aid in the formulation of suitable interventions. These include decongestion of overburdened facilities by strengthening poorly performing units. The PCs with a high percentage Private Sector Notification can act as role models for neighbouring PCs to improve private sector engagement. MPs can play a crucial role in mobilising additional resources, creating awareness, and establishing inter-PC and inter-state collaboration to improve TB program performance.

Prior research has identified a number of risk factors ranging from inadequate household sanitation to maternal characteristics as important determinants of child malnutrition and health in India. What is less known is the extent to which these individual-level risk factors are geographically distributed. Assessing the geographic distribution, especially at multiple levels, matters as it can inform where, and at what level, interventions should be targeted. The three levels of significance in the Indian context are villages, districts, and states. Thus, the purpose of this paper was to (a) examine what proportion of the variation in 21 risk factors is attributable to villages, districts, and states in India and (b) elucidate the specific states where these risk factors are clustered within India. Using the fourth National Family Health Survey dataset, from 2015 to 2016, we found that the proportion of variation attributable to villages ranged from 14% to 63%, 10% to 29% for districts and 17% to 62% for states. Furthermore, we found that Bihar, Jharkhand, Madhya Pradesh, and Uttar Pradesh were in the highest risk quintile for more than 10 of the risk factors included in our study. This is an indication of geographic clustering of risk factors. The risk factors that are clustered in states such as Bihar, Jharkhand, Madhya Pradesh and Uttar Pradesh underscore the need for policies and interventions that address a broader set of child malnutrition determinants beyond those that are nutrition specific.

Child undernutrition remains high in India with far-reaching consequences for child health and development. Anthropometry reflects undernutrition. We examined the state-level trends in underweight, stunting, and wasting prevalence and inequality by living standards using four rounds of the National Family Health Surveys in 26 states in India, conducted in 1992–1993, 1998–1999, 2005–2006, and 2015–2016. The average annual reduction (AAR) for underweight ranged from 0.04 percentage points (pp) (95% CI − 0.12, 0.20) in Haryana to 1.05 pp (95% CI 0.88, 1.22) in West Bengal for underweight; 0.35 pp (95% CI 0.11, 0.59) in Manipur to 1.47 (95% CI 1.19, 1.75) in Himachal Pradesh for stunting; and − 0.65 pp (95% CI − 0.77, − 0.52) in Haryana to 0.36 pp (95% CI 0.22, 0.51) in Bihar & Jharkhand for wasting. We find that change in the pp difference between children with the poorest and richest household living standards varied by states: statistically significant decline (increase) was observed in 5 (3) states for underweight, 5 (4) states for stunting, and 2 (1) states for wasting. Prevalence of poor anthropometric outcomes as well as disparities by states and living standards remain a problem in India.

Importance  Geographic targeting of public health interventions is needed in resource-constrained developing countries.

Objective  To develop methods for estimating health and development indicators across micropolicy units, using assembly constituencies (ACs) in India as an example.

Design, Setting, and Participants  This cross-sectional study included children younger than 5 years who participated in the fourth National Family and Health Survey (NFHS-4), conducted between January 2015 and December 2016. Participants lived in 36 states and union territories and 640 districts in India. Children who had valid weight and height measures were selected for stunting, underweight, and wasting analysis, and children between age 6 and 59 months with valid blood hemoglobin concentration levels were included in the anemia analysis sample. The analysis was performed between February 1 and August 15, 2020.

Exposures  A total of 3940 ACs were identified from the geographic location of primary sampling units in which the children’s households were surveyed in NFHS-4.

Main Outcomes and Measures  Stunting, underweight, and wasting were defined according to the World Health Organization Child Growth Standards. Anemia was defined as blood hemoglobin concentration less than 11.0 g/dL.

Results  The main analytic sample included 222 172 children (mean [SD] age, 30.03 [17.01] months; 114 902 [51.72%] boys) from 3940 ACs in the stunting, underweight, and wasting analysis and 215 593 children (mean [SD] age, 32.63 [15.47] months; 112 259 [52.07%] boys) from 3941 ACs in the anemia analysis. The burden of child undernutrition varied substantially across ACs: from 18.02% to 60.94% for stunting, with a median (IQR) of 35.56% (29.82%-42.42%); from 10.40% to 63.24% for underweight, with a median (IQR) of 32.82% (25.50%-40.96%); from 5.56% to 39.91% for wasting, with a median (IQR) of 19.91% (15.70%-24.27%); and from 18.63% to 83.05% for anemia, with a median (IQR) of 55.74% (48.41%-63.01%). The degree of inequality within states varied across states; those with high stunting, underweight, and wasting prevalence tended to have high levels of inequality. For example, Uttar Pradesh, Jharkhand, and Karnataka had high mean AC-level prevalence of child stunting (Uttar Pradesh, 45.29%; Jharkhand, 43.76%; Karnataka, 39.77%) and also large SDs (Uttar Pradesh, 6.90; Jharkhand, 6.02; Karnataka, 6.72). The Moran I indices ranged from 0.25 to 0.80, indicating varying levels of spatial autocorrelation in child undernutrition across the states in India. No substantial difference in AC-level child undernutrition prevalence was found after adjusting for possible random displacement of geographic location data.

Conclusions and Relevance  In this cross-sectional study, substantial inequality in child undernutrition was found across ACs in India, suggesting the importance of considering local electoral units in designing targeted interventions. The methods presented in this paper can be further applied to measuring health and development indicators in small electoral units for enhanced geographic precision of public health data in developing countries.

Background

Child malnutrition remains a major public health issue in India. Along with myriad upstream and social determinants of these adverse outcomes, recent studies have highlighted regional differences in mean child malnutrition rates. This research helps policy makers look between urban and rural communities and states to take a population-level approach to addressing the root causes of child malnutrition. However, one gap in this between-population approach has been the omission of households as a unit of analysis. Households could represent important sources of variation in child malnutrition within communities, districts, and states.

Methods

Using the fourth round of India’s National Family Health Survey from 2015 to 2016, we analyzed four and five-level multilevel models to estimate the proportion of variation in child malnutrition attributable to states, districts, communities, households, and children.

Results

Overall, we found that of the four levels that children were nested in (households, communities, districts, and states), the greatest proportion of variation in child height-for-age Z score, weight-for-age Z score, weight-for-height Z score, hemoglobin, birthweight, stunting, underweight, wasting, anemia, and low birthweight was attributable to households. Furthermore, we found that when the household level is omitted from models, the variance estimates for communities and children are overestimated.

Conclusions

These findings highlight the importance of households as an important source of clustering and variation in child malnutrition outcomes. As such, policies and interventions should address household-level social determinants, such as asset and social deprivations, in order to prevent poor child growth outcomes among the most vulnerable households in India.

Background

The rate of caesarean delivery has increased markedly both globally and within India. However, there is considerable variation within countries. No previous studies have examined the relative importance of multiple geographic levels in shaping the distribution of caesarean delivery and to what extent they can be explained by individual-level risk factors.

Objectives

To describe geographic variation in caesarean delivery and quantify the contribution of individual-level risk factors to the variation in India.

Methods

We conducted four-level logistic regression analysis to partition total variation in caesarean delivery to three geographic levels (states, districts and communities) and quantify the extent to which variance at each level was explained by a set of 20 sociodemographic, medical and institutional risk factors. Stratified analyses were conducted by the type of delivery facility (public/private).

Results

Overall prevalence of caesarean delivery was 19.3% in India in 2016. Most geographic variation was attributable to states (44%), followed by communities (32%), and lastly districts (24%). Adjustment for all risk factors explained 44%, 52% and 46% of variance for states, districts and communities, respectively. The proportion explained by individual risk factors was larger in public facilities than in private facilities at all three levels. A substantial proportion of between-population variation still existed even after clustering of individual risk factors was comprehensively adjusted for.

Conclusions

Diverse contextual factors driving high or low rate of caesarean delivery at each geographic level should be explored in future studies so that tailored intervention can be implemented to reduce the overall variation in caesarean delivery.

Importance  
High out-of-pocket expenditure (OOPE) on health in India may limit achieving universal health coverage. A clear insight on the components of health expenditure may be necessary to make allocative decisions to reduce OOPE, and such details by sociodemographic group and state have not been studied in India.

Objective  
To analyze the relative contribution of drugs, diagnostic tests, doctor and surgeon fees, and expenditure on other medical services and nonmedical health-related services, such as transport, lodging, and food, by sociodemographic characteristics of patients, geography, and type of illness.

Design, Setting, and Participants  A population-based cross-sectional health consumption survey conducted by the National Sample Survey Organisation in 2018 was analyzed in this cross-sectional study. Respondents who provided complete information on costs of medicine, doctors, diagnostics tests, other medical costs, and nonmedical costs were selected. Data were analyzed from August through September 2021.

Main Outcomes and Measures  
Mean and median share of components (ie, medicine, diagnostic tests, doctor fees, other medical costs, and nonmedical costs) in total health care expenditure and income were calculated. Bivariate survey-weighted mean (with 95% CI) and median (IQR) expenditures were calculated for each component across sociodemographic characteristics. The proportion of total expenditure and income contributed by each cost was calculated for each individual. Mean and median were then used to summarize such proportions at the population level. The association between state net domestic product per capita and component share of each health care service was graphically explored.

Results  
Health expenditure details were analyzed for 43 781 individuals for inpatient costs (27 272 [64.3%] women; 26 830 individuals aged 25-64 years [59.9%]) and 8914 individuals for outpatient costs (4176 [48.2%] women; 4901 individuals aged 25-64 years [54.2%]); most individuals were rural residents (24 106 inpatients [67.0]; 4591 outpatients [63.9%]). Medicines accounted for a mean of 29.1% (95% CI, 28.9%-29.2%) of OOPE among inpatients and 60.3% (95% CI, 59.7%-60.9%) of OOPE among outpatients. Doctor consultation charges were a mean of 15.3% (95% CI, 15.1%-15.4%) of OOPE among inpatients and 12.4% (95% CI, 12.1%-12.6%) of OOPE among outpatients. Diagnostic tests accounted for a mean of 12.3% (95% CI, 12.2%-12.4%) of OOPE for inpatient and 9.2% (95% CI, 8.9%-9.5%) of OOPE for outpatient services. Nonmedical costs accounted for a mean of 23.6% (95% CI, 23.3%-23.8%) of OOPE among inpatients and 14.6% (95% CI, 14.1%-15.1%) of OOPE among outpatients. Mean share of OOPE from doctor consultations and diagnostic test charges increased with socioeconomic status. For example, for the lowest vs highest monthly per capita income quintile among inpatients, doctor consultations accounted for 11.5% (95% CI, 11.1%-11.8%) vs 21.2% (95% CI, 20.8%-21.6%), and diagnostic test charges accounted for 10.9% (95% CI, 10.6%-11.1%) vs 14.3% (95% CI, 14.0%-14.5%). The proportion of mean annual health expenditure from mean annual income was $299 of $1918 (15.6%) for inpatient and $391 of $1788 (21.9%) for outpatient services.

Conclusions and Relevance  
This study found that nonmedical costs were significant, share of total health care OOPE from doctor consultation and diagnostic test charges increased with socioeconomic status, and annual cost as a proportion of annual income was lower for inpatient than outpatient services.