Caste-Based Inequality in Fruit and Vegetable Consumption in India.

OBJECTIVE: Fruit and vegetable (F&V) consumption is of central importance to many diet-related health outcomes. In India, caste is a major basis of socioeconomic inequality. Recent analysis shows that more disadvantaged "lower" castes consume less F&V than the rest. This article explores whether this consumption gap arises due to differential distribution of drivers of consumption such as income and education across castes, or whether behavioral differences or discrimination may be at play. DESIGN: The Oaxaca-Blinder regression decomposition is applied to explain the gap in F&V consumption between "upper" castes and "lower" castes, using data from the 68th (2011-2012) round of the National Sample Survey Organization household survey. RESULTS: Differences in the distribution of F&V drivers account for all of the 50 grams/person/day consumption gap between upper and lower castes. In particular, much of the gap is explained by income differential across castes. CONCLUSIONS: In the long run, India's positive discrimination policies in education and employment that seek to equalize income across castes are also likely to help close the F&V consumption gap, leading to health benefits. In the medium run, interventions acting to boost lower caste income, such as cash transfers targeting lower castes, may be effective.

Guidelines for Modeling and Reporting Health Effects of Climate Change Mitigation Actions.

BACKGROUND: Modeling suggests that climate change mitigation actions can have substantial human health benefits that accrue quickly and locally. Documenting the benefits can help drive more ambitious and health-protective climate change mitigation actions; however, documenting the adverse health effects can help to avoid them. Estimating the health effects of mitigation (HEM) actions can help policy makers prioritize investments based not only on mitigation potential but also on expected health benefits. To date, however, the wide range of incompatible approaches taken to developing and reporting HEM estimates has limited their comparability and usefulness to policymakers. OBJECTIVE: The objective of this effort was to generate guidance for modeling studies on scoping, estimating, and reporting population health effects from climate change mitigation actions. METHODS: An expert panel of HEM researchers was recruited to participate in developing guidance for conducting HEM studies. The primary literature and a synthesis of HEM studies were provided to the panel. Panel members then participated in a modified Delphi exercise to identify areas of consensus regarding HEM estimation. Finally, the panel met to review and discuss consensus findings, resolve remaining differences, and generate guidance regarding conducting HEM studies. RESULTS: The panel generated a checklist of recommendations regarding stakeholder engagement: HEM modeling, including model structure, scope and scale, demographics, time horizons, counterfactuals, health response functions, and metrics; parameterization and reporting; approaches to uncertainty and sensitivity analysis; accounting for policy uptake; and discounting. DISCUSSION: This checklist provides guidance for conducting and reporting HEM estimates to make them more comparable and useful for policymakers. Harmonization of HEM estimates has the potential to lead to advances in and improved synthesis of policy-relevant research that can inform evidence-based decision making and practice. https://doi.org/10.1289/EHP6745.

What underlies inadequate and unequal fruit and vegetable consumption in India? An exploratory analysis.

Adequate consumption of fruit and vegetables is key to improved diet-related health in India. We analyse fruit and vegetable consumption in the Indian population using National Sample Survey data. A series of regressions is estimated to characterise the distribution of household fruit and vegetable consumption and explore key socio-economic and food system drivers of consumption. Household income and price are important correlates, but consumption is also higher where households are headed by females, are rural, or involve agricultural livelihoods. Caste is an important source of inequality, particularly amongst those with low consumption, with Scheduled Tribes consuming less F&V than others. We also find preliminary evidence that formal agricultural market infrastructure is positively associated with fruit and vegetable consumption in India.

Automated versus physician assignment of cause of death for verbal autopsies: randomized trial of 9374 deaths in 117 villages in India.

BACKGROUND: Verbal autopsies with physician assignment of cause of death (COD) are commonly used in settings where medical certification of deaths is uncommon. It remains unanswered if automated algorithms can replace physician assignment. METHODS: We randomized verbal autopsy interviews for deaths in 117 villages in rural India to either physician or automated COD assignment. Twenty-four trained lay (non-medical) surveyors applied the allocated method using a laptop-based electronic system. Two of 25 physicians were allocated randomly to independently code the deaths in the physician assignment arm. Six algorithms (Naïve Bayes Classifier (NBC), King-Lu, InSilicoVA, InSilicoVA-NT, InterVA-4, and SmartVA) coded each death in the automated arm. The primary outcome was concordance with the COD distribution in the standard physician-assigned arm. Four thousand six hundred fifty-one (4651) deaths were allocated to physician (standard), and 4723 to automated arms. RESULTS: The two arms were nearly identical in demographics and key symptom patterns. The average concordances of automated algorithms with the standard were 62%, 56%, and 59% for adult, child, and neonatal deaths, respectively. Automated algorithms showed inconsistent results, even for causes that are relatively easy to identify such as road traffic injuries. Automated algorithms underestimated the number of cancer and suicide deaths in adults and overestimated other injuries in adults and children. Across all ages, average weighted concordance with the standard was 62% (range 79-45%) with the best to worst ranking automated algorithms being InterVA-4, InSilicoVA-NT, InSilicoVA, SmartVA, NBC, and King-Lu. Individual-level sensitivity for causes of adult deaths in the automated arm was low between the algorithms but high between two independent physicians in the physician arm. CONCLUSIONS: While desirable, automated algorithms require further development and rigorous evaluation. Lay reporting of deaths paired with physician COD assignment of verbal autopsies, despite some limitations, remains a practicable method to document the patterns of mortality reliably for unattended deaths. TRIAL REGISTRATION: ClinicalTrials.gov , NCT02810366. Submitted on 11 April 2016.

Future diets in India: A systematic review of food consumption projection studies.

Against a backdrop of a rapidly changing food system and a growing population, characterisation of likely future diets in India can help to inform agriculture and health policies. We systematically searched six published literature databases and grey literature repositories up to January 2018 for studies projecting the consumption of foods in India to time points beyond 2018. The 11 identified studies reported on nine foods up to 2050: the available evidence suggests projected increases in per capita consumption of vegetables, fruit and dairy products, and little projected change in cereal (rice and wheat) and pulse consumption. Meat consumption is projected to remain low. Understanding and mitigating the impacts of projected dietary changes in India is important to protect public health and the environment.

Greenhouse gas emissions and water footprints of typical dietary patterns in India.

Agriculture is a major contributor to India's environmental footprint, particularly through greenhouse gas (GHG) emissions from livestock and fresh water used for irrigation. These impacts are likely to increase in future as agriculture attempts to keep pace with India's growing population and changing dietary preferences. Within India there is considerable dietary variation, and this study therefore aimed to quantify the GHG emissions and water usage associated with distinct dietary patterns. Five distinct diets were identified from the Indian Migration Study - a large adult population sample in India - using finite mixture modelling. These were defined as: Rice & low diversity, Rice & fruit, Wheat & pulses, Wheat, rice & oils, Rice & meat. The GHG emissions of each dietary pattern were quantified based on a Life Cycle Assessment (LCA) approach, and water use was quantified using Water Footprint (WF) data. Mixed-effects regression models quantified differences in the environmental impacts of the dietary patterns. There was substantial variability between diets: the rice-based patterns had higher associated GHG emissions and green WFs, but the wheat-based patterns had higher blue WFs. Regression modelling showed that the Rice & meat pattern had the highest environmental impacts overall, with 0.77 (95% CI 0.64-0.89) kg CO2e/capita/day (31%) higher emissions, 536 (95% CI 449-623) L/capita/day (24%) higher green WF and 109 (95% CI 85.9-133) L/capita/day (19%) higher blue WF than the reference Rice & low diversity pattern. Diets in India are likely to become more diverse with rising incomes, moving away from patterns such as the Rice & low diversity diet. Patterns such as the Rice & meat diet may become more common, and the environmental consequences of such changes could be substantial given the size of India's population. As global environmental stress increases, agricultural and nutrition policies must recognise the environmental impacts of potential future dietary changes.

Projected health effects of realistic dietary changes to address freshwater constraints in India: a modelling study.

BACKGROUND: The availability of freshwater for irrigation in the Indian agricultural sector is expected to decline over the coming decades. This might have implications for food production in India, with subsequent effects on diets and health. We identify realistic and healthy dietary changes that could enhance the resilience of the Indian food system to future decreases in water availability. METHODS: In this modelling study, we optimised typical dietary patterns in an Indian population sample to meet projected decreases in the availability of water per person for irrigation (blue water footprint) due to population growth (to 2025 and 2050). The optimised diets met nutritional guidelines and minimised deviation from existing patterns. Resulting changes in life-years lost due to coronary heart disease, stroke, diabetes, and cancers were modelled using life tables, and changes in greenhouse gas emissions associated with the production of diets were estimated. The primary outcomes of the model were changes in life-years per 100 000 total population over 40 years (to 2050). FINDINGS: The optimised diets had up to 30% lower blue water footprints and generally contained lower amounts of wheat, dairy, and poultry, and increased amounts of legumes. In the 2050 scenario, adoption of these diets would on average result in 6800 life-years gained per 100 000 total population (95% CI 1600-13 100) over 40 years. The dietary changes were accompanied by reductions in greenhouse gas emissions. The magnitude of the health and environmental effects varied between dietary patterns. INTERPRETATION: Modest changes in diets could help to address projected reductions in the availability of freshwater for irrigation in India. These dietary changes could also simultaneously reduce diet-related greenhouse gas emissions and improve diet-related health outcomes. FUNDING: Wellcome Trust.

Comparison of food consumption in Indian adults between national and sub-national dietary data sources.

Accurate data on dietary intake are important for public health, nutrition and agricultural policy. The National Sample Survey is widely used by policymakers in India to estimate nutritional outcomes in the country, but has not been compared with other dietary data sources. To assess relative differences across available Indian dietary data sources, we compare intake of food groups across six national and sub-national surveys between 2004 and 2012, representing various dietary intake estimation methodologies, including Household Consumption Expenditure Surveys (HCES), FFQ, food balance sheets (FBS), and 24-h recall (24HR) surveys. We matched data for relevant years, regions and economic groups, for ages 16-59. One set of national HCES and the 24HR showed a decline in food intake in India between 2004-2005 and 2011-2012, whereas another HCES and FBS showed an increase. Differences in intake were smallest between the two HCES (1 % relative difference). Relative to these, FFQ and FBS had higher intake (13 and 35 %), and the 24HR lower intake (-9 %). Cereal consumption had high agreement across comparisons (average 5 % difference), whereas fruit and nuts, eggs, meat and fish and sugar had the least (120, 119, 56 and 50 % average differences, respectively). Spearman's coefficients showed high correlation of ranked food group intake across surveys. The underlying methods of the compared data highlight possible sources of under- or over-estimation, and influence their relevance for addressing various research questions and programmatic needs.

Dietary patterns and non-communicable disease risk in Indian adults: secondary analysis of Indian Migration Study data.

OBJECTIVE: Undernutrition and non-communicable disease (NCD) are important public health issues in India, yet their relationship with dietary patterns is poorly understood. The current study identified distinct dietary patterns and their association with micronutrient undernutrition (Ca, Fe, Zn) and NCD risk factors (underweight, obesity, waist:hip ratio, hypertension, total:HDL cholesterol, diabetes). DESIGN: Data were from the cross-sectional Indian Migration Study, including semi-quantitative FFQ. Distinct dietary patterns were identified using finite mixture modelling; associations with NCD risk factors were assessed using mixed-effects logistic regression models. SETTING: India. SUBJECTS: Migrant factory workers, their rural-dwelling siblings and urban non-migrants. Participants (7067 adults) resided mainly in Karnataka, Andhra Pradesh, Maharashtra and Uttar Pradesh. RESULTS: Five distinct, regionally distributed, dietary patterns were identified, with rice-based patterns in the south and wheat-based patterns in the north-west. A rice-based pattern characterised by low energy consumption and dietary diversity ('Rice & low diversity') was consumed predominantly by adults with little formal education in rural settings, while a rice-based pattern with high fruit consumption ('Rice & fruit') was consumed by more educated adults in urban settings. Dietary patterns met WHO macronutrient recommendations, but some had low micronutrient contents. Dietary pattern membership was associated with several NCD risk factors. CONCLUSIONS: Five distinct dietary patterns were identified, supporting sub-national assessments of the implications of dietary patterns for various health, food system or environment outcomes.

Greenhouse gas emissions from agricultural food production to supply Indian diets: Implications for climate change mitigation.

Agriculture is a major source of greenhouse gas (GHG) emissions globally. The growing global population is putting pressure on agricultural production systems that aim to secure food production while minimising GHG emissions. In this study, the GHG emissions associated with the production of major food commodities in India are calculated using the Cool Farm Tool. GHG emissions, based on farm management for major crops (including cereals like wheat and rice, pulses, potatoes, fruits and vegetables) and livestock-based products (milk, eggs, chicken and mutton meat), are quantified and compared. Livestock and rice production were found to be the main sources of GHG emissions in Indian agriculture with a country average of 5.65 kg CO2eq kg-1 rice, 45.54 kg CO2eq kg-1 mutton meat and 2.4 kg CO2eq kg-1 milk. Production of cereals (except rice), fruits and vegetables in India emits comparatively less GHGs with <1 kg CO2eq kg-1 product. These findings suggest that a shift towards dietary patterns with greater consumption of animal source foods could greatly increase GHG emissions from Indian agriculture. A range of mitigation options are available that could reduce emissions from current levels and may be compatible with increased future food production and consumption demands in India.

The Impacts of Dietary Change on Greenhouse Gas Emissions, Land Use, Water Use, and Health: A Systematic Review.

Food production is a major driver of greenhouse gas (GHG) emissions, water and land use, and dietary risk factors are contributors to non-communicable diseases. Shifts in dietary patterns can therefore potentially provide benefits for both the environment and health. However, there is uncertainty about the magnitude of these impacts, and the dietary changes necessary to achieve them. We systematically review the evidence on changes in GHG emissions, land use, and water use, from shifting current dietary intakes to environmentally sustainable dietary patterns. We find 14 common sustainable dietary patterns across reviewed studies, with reductions as high as 70-80% of GHG emissions and land use, and 50% of water use (with medians of about 20-30% for these indicators across all studies) possible by adopting sustainable dietary patterns. Reductions in environmental footprints were generally proportional to the magnitude of animal-based food restriction. Dietary shifts also yielded modest benefits in all-cause mortality risk. Our review reveals that environmental and health benefits are possible by shifting current Western diets to a variety of more sustainable dietary patterns.

How do sustainable diets fit into the climate agenda?

Food production is a major driver of greenhouse gas (GHG) emission and other environmental footprints, and dietary risk factors are contributors to non-communicable diseases. A growing body of evidence has shown that changes in what and how much we eat can offer benefits for both the environment and health. However, several data gaps and complexities remain in this research area. A better understanding and increased uptake of sustainable diets will require further research, investment, and interdisciplinary collaboration.

Childhood Cancer Mortality in India: Direct Estimates From a Nationally Representative Survey of Childhood Deaths.

PURPOSE: Although most children with cancer live in low- and middle-income countries, measurements of childhood cancer burden in such countries have been restricted to incidence rates from a few subnational cancer registries and mortality rates from vital statistics. We aimed to provide alternative burden estimates by using nationally representative longitudinal survey-derived mortality rates. METHODS: We examined cancer deaths in childhood (1 month to 14 years of age) in the Million Death Study, a cohort of > 27,000 pediatric deaths in India on the basis of enhanced verbal autopsies. All deaths potentially due to childhood cancer were identified. Two pediatric specialists independently categorized deaths as definite, probable, possible, or unlikely cancer related. From definite and probable deaths, we estimated national and regional mortality rates attributable to childhood malignancies. Data on symptoms and health care-seeking behavior were abstracted from closed-ended questions and caregiver narratives. RESULTS: Of 700 included deaths, 189 were classified as definite or possibly cancer related. The κ-statistic between reviewers was 0.75 (95% CI, 0.71 to 0.78). From these deaths, we estimated that in 2010, 13,700 were a result of childhood cancer in India, which led to a mortality rate of 37 (95% CI, 31 to 42) per million population per year, which exceeds many prior estimates of mortality and even some estimates of incidence. Disparities between mortality estimates were widest in northeast India and for brain tumors. A preponderance of male deaths was seen (male:female ratio, 1.6:1). CONCLUSION: The burden of childhood cancer in India is substantially higher than previously suggested. This information will aid advocacy for national strategies aimed at improving outcomes for Indian children with cancer.

Naive Bayes classifiers for verbal autopsies: comparison to physician-based classification for 21,000 child and adult deaths.

BACKGROUND: Verbal autopsies (VA) are increasingly used in low- and middle-income countries where most causes of death (COD) occur at home without medical attention, and home deaths differ substantially from hospital deaths. Hence, there is no plausible "standard" against which VAs for home deaths may be validated. Previous studies have shown contradictory performance of automated methods compared to physician-based classification of CODs. We sought to compare the performance of the classic naive Bayes classifier (NBC) versus existing automated classifiers, using physician-based classification as the reference. METHODS: We compared the performance of NBC, an open-source Tariff Method (OTM), and InterVA-4 on three datasets covering about 21,000 child and adult deaths: the ongoing Million Death Study in India, and health and demographic surveillance sites in Agincourt, South Africa and Matlab, Bangladesh. We applied several training and testing splits of the data to quantify the sensitivity and specificity compared to physician coding for individual CODs and to test the cause-specific mortality fractions at the population level. RESULTS: The NBC achieved comparable sensitivity (median 0.51, range 0.48-0.58) to OTM (median 0.50, range 0.41-0.51), with InterVA-4 having lower sensitivity (median 0.43, range 0.36-0.47) in all three datasets, across all CODs. Consistency of CODs was comparable for NBC and InterVA-4 but lower for OTM. NBC and OTM achieved better performance when using a local rather than a non-local training dataset. At the population level, NBC scored the highest cause-specific mortality fraction accuracy across the datasets (median 0.88, range 0.87-0.93), followed by InterVA-4 (median 0.66, range 0.62-0.73) and OTM (median 0.57, range 0.42-0.58). CONCLUSIONS: NBC outperforms current similar COD classifiers at the population level. Nevertheless, no current automated classifier adequately replicates physician classification for individual CODs. There is a need for further research on automated classifiers using local training and test data in diverse settings prior to recommending any replacement of physician-based classification of verbal autopsies.

Performance criteria for verbal autopsy-based systems to estimate national causes of death: development and application to the Indian Million Death Study.

BACKGROUND: Verbal autopsy (VA) has been proposed to determine the cause of death (COD) distributions in settings where most deaths occur without medical attention or certification. We develop performance criteria for VA-based COD systems and apply these to the Registrar General of India's ongoing, nationally-representative Indian Million Death Study (MDS). METHODS: Performance criteria include a low ill-defined proportion of deaths before old age; reproducibility, including consistency of COD distributions with independent resampling; differences in COD distribution of hospital, home, urban or rural deaths; age-, sex- and time-specific plausibility of specific diseases; stability and repeatability of dual physician coding; and the ability of the mortality classification system to capture a wide range of conditions. RESULTS: The introduction of the MDS in India reduced the proportion of ill-defined deaths before age 70 years from 13% to 4%. The cause-specific mortality fractions (CSMFs) at ages 5 to 69 years for independently resampled deaths and the MDS were very similar across 19 disease categories. By contrast, CSMFs at these ages differed between hospital and home deaths and between urban and rural deaths. Thus, reliance mostly on urban or hospital data can distort national estimates of CODs. Age-, sex- and time-specific patterns for various diseases were plausible. Initial physician agreement on COD occurred about two-thirds of the time. The MDS COD classification system was able to capture more eligible records than alternative classification systems. By these metrics, the Indian MDS performs well for deaths prior to age 70 years. The key implication for low- and middle-income countries where medical certification of death remains uncommon is to implement COD surveys that randomly sample all deaths, use simple but high-quality field work with built-in resampling, and use electronic rather than paper systems to expedite field work and coding. CONCLUSIONS: Simple criteria can evaluate the performance of VA-based COD systems. Despite the misclassification of VA, the MDS demonstrates that national surveys of CODs using VA are an order of magnitude better than the limited COD data previously available.

Comparison of physician-certified verbal autopsy with computer-coded verbal autopsy for cause of death assignment in hospitalized patients in low- and middle-income countries: systematic review.

BACKGROUND: Computer-coded verbal autopsy (CCVA) methods to assign causes of death (CODs) for medically unattended deaths have been proposed as an alternative to physician-certified verbal autopsy (PCVA). We conducted a systematic review of 19 published comparison studies (from 684 evaluated), most of which used hospital-based deaths as the reference standard. We assessed the performance of PCVA and five CCVA methods: Random Forest, Tariff, InterVA, King-Lu, and Simplified Symptom Pattern. METHODS: The reviewed studies assessed methods' performance through various metrics: sensitivity, specificity, and chance-corrected concordance for coding individual deaths, and cause-specific mortality fraction (CSMF) error and CSMF accuracy at the population level. These results were summarized into means, medians, and ranges. RESULTS: The 19 studies ranged from 200 to 50,000 deaths per study (total over 116,000 deaths). Sensitivity of PCVA versus hospital-assigned COD varied widely by cause, but showed consistently high specificity. PCVA and CCVA methods had an overall chance-corrected concordance of about 50% or lower, across all ages and CODs. At the population level, the relative CSMF error between PCVA and hospital-based deaths indicated good performance for most CODs. Random Forest had the best CSMF accuracy performance, followed closely by PCVA and the other CCVA methods, but with lower values for InterVA-3. CONCLUSIONS: There is no single best-performing coding method for verbal autopsies across various studies and metrics. There is little current justification for CCVA to replace PCVA, particularly as physician diagnosis remains the worldwide standard for clinical diagnosis on live patients. Further assessments and large accessible datasets on which to train and test combinations of methods are required, particularly for rural deaths without medical attention.

Performance of four computer-coded verbal autopsy methods for cause of death assignment compared with physician coding on 24,000 deaths in low- and middle-income countries.

BACKGROUND: Physician-coded verbal autopsy (PCVA) is the most widely used method to determine causes of death (CODs) in countries where medical certification of death is uncommon. Computer-coded verbal autopsy (CCVA) methods have been proposed as a faster and cheaper alternative to PCVA, though they have not been widely compared to PCVA or to each other. METHODS: We compared the performance of open-source random forest, open-source tariff method, InterVA-4, and the King-Lu method to PCVA on five datasets comprising over 24,000 verbal autopsies from low- and middle-income countries. Metrics to assess performance were positive predictive value and partial chance-corrected concordance at the individual level, and cause-specific mortality fraction accuracy and cause-specific mortality fraction error at the population level. RESULTS: The positive predictive value for the most probable COD predicted by the four CCVA methods averaged about 43% to 44% across the datasets. The average positive predictive value improved for the top three most probable CODs, with greater improvements for open-source random forest (69%) and open-source tariff method (68%) than for InterVA-4 (62%). The average partial chance-corrected concordance for the most probable COD predicted by the open-source random forest, open-source tariff method and InterVA-4 were 41%, 40% and 41%, respectively, with better results for the top three most probable CODs. Performance generally improved with larger datasets. At the population level, the King-Lu method had the highest average cause-specific mortality fraction accuracy across all five datasets (91%), followed by InterVA-4 (72% across three datasets), open-source random forest (71%) and open-source tariff method (54%). CONCLUSIONS: On an individual level, no single method was able to replicate the physician assignment of COD more than about half the time. At the population level, the King-Lu method was the best method to estimate cause-specific mortality fractions, though it does not assign individual CODs. Future testing should focus on combining different computer-coded verbal autopsy tools, paired with PCVA strengths. This includes using open-source tools applied to larger and varied datasets (especially those including a random sample of deaths drawn from the population), so as to establish the performance for age- and sex-specific CODs.

Cancer mortality in India: a nationally representative survey.

BACKGROUND: The age-specific mortality rates and total deaths from specific cancers have not been documented for the various regions and subpopulations of India. We therefore assessed the cause of death in 2001-03 in homes in small areas that were chosen to be representative of all the parts of India. METHODS: At least 130 trained physicians independently assigned causes to 122,429 deaths, which occurred in 1·1 million homes in 6671 small areas that were randomly selected to be representative of all of India, based on a structured non-medical surveyor's field report. FINDINGS: 7137 of 122,429 study deaths were due to cancer, corresponding to 556,400 national cancer deaths in India in 2010. 395,400 (71%) cancer deaths occurred in people aged 30-69 years (200,100 men and 195,300 women). At 30-69 years, the three most common fatal cancers were oral (including lip and pharynx, 45,800 [22·9%]), stomach (25,200 [12·6%]), and lung (including trachea and larynx, 22,900 [11·4%]) in men, and cervical (33,400 [17·1%]), stomach (27,500 [14·1%]), and breast (19,900 [10·2%]) in women. Tobacco-related cancers represented 42·0% (84,000) of male and 18·3% (35,700) of female cancer deaths and there were twice as many deaths from oral cancers as lung cancers. Age-standardised cancer mortality rates per 100,000 were similar in rural (men 95·6 [99% CI 89·6-101·7] and women 96·6 [90·7-102·6]) and urban areas (men 102·4 [92·7-112·1] and women 91·2 [81·9-100·5]), but varied greatly between the states, and were two times higher in the least educated than in the most educated adults (men, illiterate 106·6 [97·4-115·7] vs most educated 45·7 [37·8-53·6]; women, illiterate 106·7 [99·9-113·6] vs most educated 43·4 [30·7-56·1]). Cervical cancer was far less common in Muslim than in Hindu women (study deaths 24, age-standardised mortality ratio 0·68 [0·64-0·71] vs 340, 1·06 [1·05-1·08]). INTERPRETATION: Prevention of tobacco-related and cervical cancers and earlier detection of treatable cancers would reduce cancer deaths in India, particularly in the rural areas that are underserved by cancer services. The substantial variation in cancer rates in India suggests other risk factors or causative agents that remain to be discovered. FUNDING: Bill & Melinda Gates Foundation and US National Institutes of Health.

Trends in selective abortions of girls in India: analysis of nationally representative birth histories from 1990 to 2005 and census data from 1991 to 2011.

BACKGROUND: India's 2011 census revealed a growing imbalance between the numbers of girls and boys aged 0-6 years, which we postulate is due to increased prenatal sex determination with subsequent selective abortion of female fetuses. We aimed to establish the trends in sex ratio by birth order from 1990 to 2005 with three nationally representative surveys and to quantify the totals of selective abortions of girls with census cohort data. METHODS: We assessed sex ratios by birth order in 0·25 million births in three rounds of the nationally representative National Family Health Survey covering the period from 1990 to 2005. We estimated totals of selective abortion of girls by assessing the birth cohorts of children aged 0-6 years in the 1991, 2001, and 2011 censuses. Our main statistic was the conditional sex ratio of second-order births after a firstborn girl and we used 3-year rolling weighted averages to test for trends, with differences between trends compared by linear regression. FINDINGS: The conditional sex ratio for second-order births when the firstborn was a girl fell from 906 per 1000 boys (99% CI 798-1013) in 1990 to 836 (733-939) in 2005; an annual decline of 0·52% (p for trend=0·002). Declines were much greater in mothers with 10 or more years of education than in mothers with no education, and in wealthier households compared with poorer households. By contrast, we did not detect any significant declines in the sex ratio for second-order births if the firstborn was a boy, or for firstborns. Between the 2001 and 2011 censuses, more than twice the number of Indian districts (local administrative areas) showed declines in the child sex ratio as districts with no change or increases. After adjusting for excess mortality rates in girls, our estimates of number of selective abortions of girls rose from 0-2·0 million in the 1980s, to 1·2-4·1 million in the 1990s, and to 3·1-6·0 million in the 2000s. Each 1% decline in child sex ratio at ages 0-6 years implied 1·2-3·6 million more selective abortions of girls. Selective abortions of girls totalled about 4·2-12·1 million from 1980-2010, with a greater rate of increase in the 1990s than in the 2000s. INTERPRETATION: Selective abortion of girls, especially for pregnancies after a firstborn girl, has increased substantially in India. Most of India's population now live in states where selective abortion of girls is common. FUNDING: US National Institutes of Health, Canadian Institute of Health Research, International Development Research Centre, and Li Ka Shing Knowledge Institute.