Deciphering disparities in childhood stunting in an underdeveloped state of India: an investigation applying the unconditional quantile regression method.

BACKGROUND: Unacceptably high rate of childhood stunting for decades remained a puzzle in the eastern Indian state of Bihar. Despite various programmatic interventions, nearly half of the under-five children (numerically about 10 million) are still stunted in this resource-constrained state. DATA AND METHODS: Using four successive rounds of National Family Health Survey (NFHS) data spread over more than two decades and by employing unconditional quantile regressions and counterfactual decomposition (QR-CD), the present study aims to assess effects of various endowments as well as returns to those endowments in disparities in childhood stunting over the period. RESULTS: The results show that although the child's height-for-age Z-scores (HAZ) disparity largely accounted for differing levels of endowments during the earlier decades, in the later periods, inadequate access to the benefits from various development programmes was also found responsible for HAZ disparities. Moreover, effects of endowments and their returns varied across quantiles. We argue that apart from equalizing endowments, ensuring adequate access to different nutrition-centric programmes is essential to lessen the burden of childhood stunting. CONCLUSION: The state must focus on intersectoral convergence of different schemes in the form of state nutrition mission, and, strengthen nutrition-centric policy processes and their political underpinnings to harness better dividend.

Distinct mortality patterns at 0-2 days versus the remaining neonatal period: results from population-based assessment in the Indian state of Bihar.

BACKGROUND: The objectives of this study were to understand the differences in mortality rate, risk factors for mortality, and cause of death distribution in three neonatal age sub-groups (0-2, 3-7, and 8-27 days) and assess the change in mortality rate with previous assessments to inform programmatic decision-making in the Indian state of Bihar, a large state with a high burden of newborn deaths. METHODS: Detailed interviews were conducted in a representative sample of 23,602 live births between January and December 2016 (96.2% participation) in Bihar state. We estimated the neonatal mortality rate (NMR) for the three age sub-groups and explored the association of these deaths with a variety of risk factors using a hierarchical logistic regression model approach. Verbal autopsies were conducted using the PHMRC questionnaire and the cause of death assigned using the SmartVA automated algorithm. Change in NMR from 2011 to 2016 was estimated by comparing it with a previous assessment. RESULTS: The NMR 0-2-day, 3-7-day, and 8-27-day mortality estimates in 2016 were 24.7 (95% CI 21.8-28.0), 13.2 (11.1 to 15.7), 5.8 (4.4 to 7.5), and 5.8 (4.5 to 7.5) per 1000 live births, respectively. A statistically significant reduction of 23.3% (95% CI 9.2% to 37.3) was seen in NMR from 2011 to 2016, driven by a reduction of 35.3% (95% CI 18.4% to 52.2) in 0-2-day mortality. In the final regression model, the highest odds for mortality in 0-2 days were related to the gestation period of ≤ 8 months (OR 16.5, 95% CI 11.9-22.9) followed by obstetric complications, no antiseptic cord care, and delivery at a private health facility or home. The 3-7- and 8-27-day mortality was driven by illness in the neonatal period (OR 10.33, 95% CI 6.31-16.90, and OR 4.88, 95% CI 3.13-7.61, respectively) and pregnancy with multiple foetuses (OR 5.15, 95% CI 2.39-11.10, and OR 11.77, 95% CI 6.43-21.53, respectively). Birth asphyxia (61.1%) and preterm delivery (22.1%) accounted for most of 0-2-day deaths; pneumonia (34.5%), preterm delivery (33.7%), and meningitis/sepsis (20.1%) accounted for the majority of 3-7-day deaths; meningitis/sepsis (30.6%), pneumonia (29.1%), and preterm delivery (26.2%) were the leading causes of death at 8-27 days. CONCLUSIONS: To our knowledge, this is the first study to report a detailed neonatal epidemiology by age sub-groups for a major Indian state, which has highlighted the distinctly different mortality rate, risk factors, and causes of death at 0-2 days versus the rest of the neonatal period. Monitoring mortality at 0-2 and 3-7 days separately in the traditional early neonatal period of 0-7 days would enable more effective programming to reduce neonatal mortality.

Deferred and referred deliveries contribute to stillbirths in the Indian state of Bihar: results from a population-based survey of all births.

BACKGROUND: The India Newborn Action Plan (INAP) aims for < 10 stillbirths per 1000 births by 2030. A population-based understanding of risk factors for stillbirths compared with live births that could assist with reduction of stillbirths is not readily available for the Indian population. METHODS: Detailed interviews were conducted in a representative sample of all births between January and December 2016 from 182,486 households (96.2% participation) in 1657 clusters in the Indian state of Bihar. A stillbirth was defined as foetal death with gestation period of ≥ 7 months wherein the foetus did not show any sign of life. The association of stillbirth was investigated with a variety of risk factors among all births using a hierarchical logistic regression model approach. RESULTS: A total of 23,940 births including 338 stillbirths were identified giving the state stillbirth rate (SBR) of 15.4 (95% CI 13.2-17.9) per 1000 births, with no difference in SBR by sex. Antepartum and intrapartum SBR was 5.6 (95% CI 4.3-7.2) and 4.5 (95% CI 3.3-6.1) per 1000 births, respectively. Detailed interview was available for 20,152 (84.2% participation) births including 275 stillbirths (81.4% participation). In the final regression model, significantly higher odds of stillbirth were documented for deliveries with gestation period of ≤ 8 months (OR 11.36, 95% CI 8.13-15.88), for first born (OR 5.79, 95% CI 4.06-8.26), deferred deliveries wherein a woman was sent back home and asked to come later for delivery by a health provider (OR 5.51, 95% CI 2.81-10.78), and in those with forceful push/pull during the delivery by the health provider (OR 4.85, 95% CI 3.39-6.95). The other significant risk factors were maternal age ≥ 30 years (OR 3.20, 95% CI 1.52-6.74), pregnancies with multiple foetuses (OR 2.82, 95% CI 1.49-5.33), breech presentation of the baby (OR 2.70, 95% CI 1.75-4.18), and births in private facilities (OR 1.75, 95% CI 1.19-2.56) and home (OR 2.60, 95% CI 1.87-3.62). Varied risk factors were associated with antepartum and intrapartum stillbirths. Birth weight was available only for 40 (14.5%) stillborns. Among the facility deliveries, the women who were referred from one facility to another for delivery had significantly high odds of stillbirth (OR 3.32, 95% CI 2.03-5.43). CONCLUSIONS: We found an increased risk of stillbirths in deferred and referred deliveries in addition to demographic and clinical risk factors for antepartum and intrapartum stillbirths, highlighting aspects of health care that need attention in addition to improving skills of health providers to reduce stillbirths. The INAP could utilise these findings to further strengthen its approach to meet the stillbirth reduction target by 2030.

Assessment of quality of antenatal care services in public sector facilities in India.

OBJECTIVES: We undertook assessment of quality of antenatal care (ANC) services in public sector facilities in the Indian state of Bihar state delivered under the national ANC programme (Pradhan Mantri Surakshit Matritva Abhiyan, PMSMA). SETTING: Three community health centres and one subdistrict hospital each in two randomly selected districts of Bihar. PARTICIPANTS: Pregnant women who sought ANC services under PMSMA irrespective of the pregnancy trimester. PRIMARY AND SECONDARY MEASURES: Quality ANC services were considered if a woman received all of these services in that visit-weight, blood pressure and abdomen check, urine and blood sample taken, and were given iron and folic acid and calcium tablets. The process of ANC service provision was documented. RESULTS: Eight hundred and fourteen (94.5% participation) women participated. Coverage of quality ANC services was 30.4% (95% CI 27.3% to 33.7%) irrespective of pregnancy trimester, and was similar in both districts and ranged 3%-83.1% across the facilities. Quality ANC service coverage was significantly lower for women in the first trimester of pregnancy (6.8%, 95% CI 3.3% to 13.6%) as compared with those in the second (34.4%, 95% CI 29.9% to 39.1%) and third (32.9%, 95% CI 27.9% to 38.3%) trimester of pregnancy. Individually, the coverage of weight and blood pressure check-up, receipt of iron folic acid (IFA) and calcium tablets, and blood sample collection was >85%. The coverage of urine sample collection was 46.3% (95% CI 42.9% to 49.7%) and of abdomen check-up was 62% (95% CI 58.6% to 65.3%). Poor information sharing post check-up was done with the pregnant women. Varied implementation of ANC service provision was seen in the facilities as compared with the PMSMA guidelines, in particular with laboratory diagnostics and doctor consultation. Task shifting from doctors to ANMs was observed in all facilities. CONCLUSIONS: Grossly inadequate quality ANC services under the PMSMA needs urgent attention to improve maternal and neonatal health outcomes.

Implications of the availability and distribution of birth weight on addressing neonatal mortality: population-based assessment from Bihar state of India.

OBJECTIVE: A large proportion of neonatal deaths in India are attributable to low birth weight (LBW). We report population-based distribution and determinants of birth weight in Bihar state, and on the perceptions about birth weight among carers. DESIGN: A cross-sectional household survey in a state representative sample of 6007 live births born in 2018-2019. Mothers provided detailed interviews on sociodemographic characteristics and birth weight, and their perceptions on LBW (birth weight <2500 g). We report on birth weight availability, LBW prevalence, neonatal mortality rate (NMR) by birth weight and perceptions of mothers on LBW implications. SETTING: Bihar state, India. PARTICIPANTS: Women with live birth between October 2018 and September 2019. RESULTS: A total of 5021 (83.5%) live births participated, and 3939 (78.4%) were weighed at birth. LBW prevalence among those with available birth weight was 18.4% (95% CI 17.1 to 19.7). Majority (87.5%) of the live births born at home were not weighed at birth. LBW prevalence decreased and birth weight ≥2500 g increased significantly with increasing wealth index quartile. NMR was significantly higher in live births weighing <1500 g (11.3%; 95% CI 5.1 to 23.1) and 1500-1999 g (8.0%; 95% CI 4.6 to 13.6) than those weighing ≥2500 g (1.3%, 95% CI 0.9 to 1.7). Assuming proportional correspondence of LBW and NMR in live births with and without birth weight, the estimated LBW among those without birth weight was 35.5% (95% CI 33.0 to 38.0) and among all live births irrespective of birth weight availability was 23.0% (95% CI 21.9 to 24.2). 70% of mothers considered LBW to be a sign of sickness, 59.5% perceived it as a risk of developing other illnesses and 8.6% as having an increased probability of death. CONCLUSIONS: Missing birth weight is substantially compromising the planning of interventions to address LBW at the population-level. Variations of LBW by place of delivery and sociodemographic indicators, and the perceptions of carers about LBW can facilitate appropriate actions to address LBW and the associated neonatal mortality.

Ensem-HAR: An Ensemble Deep Learning Model for Smartphone Sensor-Based Human Activity Recognition for Measurement of Elderly Health Monitoring.

Biomedical images contain a huge number of sensor measurements that can provide disease characteristics. Computer-assisted analysis of such parameters aids in the early detection of disease, and as a result aids medical professionals in quickly selecting appropriate medications. Human Activity Recognition, abbreviated as 'HAR', is the prediction of common human measurements, which consist of movements such as walking, running, drinking, cooking, etc. It is extremely advantageous for services in the sphere of medical care, such as fitness trackers, senior care, and archiving patient information for future use. The two types of data that can be fed to the HAR system as input are, first, video sequences or images of human activities, and second, time-series data of physical movements during different activities recorded through sensors such as accelerometers, gyroscopes, etc., that are present in smart gadgets. In this paper, we have decided to work with time-series kind of data as the input. Here, we propose an ensemble of four deep learning-based classification models, namely, 'CNN-net', 'CNNLSTM-net', 'ConvLSTM-net', and 'StackedLSTM-net', which is termed as 'Ensem-HAR'. Each of the classification models used in the ensemble is based on a typical 1D Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) network; however, they differ in terms of their architectural variations. Prediction through the proposed Ensem-HAR is carried out by stacking predictions from each of the four mentioned classification models, then training a Blender or Meta-learner on the stacked prediction, which provides the final prediction on test data. Our proposed model was evaluated over three benchmark datasets, WISDM, PAMAP2, and UCI-HAR; the proposed Ensem-HAR model for biomedical measurement achieved 98.70%, 97.45%, and 95.05% accuracy, respectively, on the mentioned datasets. The results from the experiments reveal that the suggested model performs better than the other multiple generated measurements to which it was compared.

Technical assistance: a practical account of the challenges in design and implementation.

Technical assistance is provided to country governments as part of international development programmes to support policymaking or strengthen state capability. This article presents the conceptual evolution of 'technical assistance' linked to capacity development, starting with programmes aiming exclusively to enhance individual capacity in the 1950s to 1970s and progressing to complex systems approaches in the past ten years. It also presents some of the frequent challenges in designing and implementing technical assistance, drawing from the existing literature and the authors' experience in international development. The article summarises the latest thinking about delivering more effective development, including the adaptive management practices and the initiatives to strengthen evidence about what works. Finally, we complement this article with a follow-up open letter reflecting on the current policy options and opportunities for change.

Towards reimagined technical assistance: the current policy options and opportunities for change.

Technical assistance has been at the heart of development assistance provided to country governments by donor agencies over the past several decades. The current debates on reimagining technical assistance focus on the existing challenges of the different types of technical assistance and the (re)construction of an ideal model for delivering this type of support, with little discussion about the dilemmas involved in making day-to-day decisions and trade-offs in implementation. This article presents technical assistance as a policy option for governments and details the existing models of delivering technical assistance, their limitations, and the required enabling conditions. The models presented focus on the type of role for the technical advisers- as doers (performing government functions), partners (working with the government to perform a specific role) and facilitators (enabling and facilitating change programmes to address wicked problems). Finally, the paper provides a practical account of the implications of the programme design and suggests potential opportunities for change particularly in the context of COVID-19 pandemic. It complements an open letter on the practical account of the current challenges in the design and implementation of technical assistance programmes.

Trends in reproductive, maternal, newborn and child health and nutrition indicators during five years of piloting and scaling-up of Ananya interventions in Bihar, India.

BACKGROUND: The Ananya program in Bihar implemented household and community-level interventions to improve reproductive, maternal, newborn and child health and nutrition (RMNCHN) in two phases: a first phase of intensive ancillary support to governmental implementation and innovation testing by non-government organisation (NGO) partners in eight focus districts (2012-2014), followed by a second phase of state-wide government-led implementation with techno-managerial assistance from NGOs (2014 onwards). This paper examines trends in RMNCHN indicators in the program's implementation districts from 2012-2017. METHODS: Eight consecutive rounds of cross-sectional Community-based Household Surveys conducted by CARE India in 2012-2017 provided comparable data on a large number of indicators of frontline worker (FLW) performance, mothers' behaviours, and facility-based care and outreach service delivery across the continuum of maternal and child care. Logistic regression, considering the complex survey design and sample weights generated by that design, was used to estimate trends using survey rounds 2-5 for the first phase in the eight focus districts and rounds 6-9 for the second phase in all 38 districts statewide, as well as the overall change from round 2-9 in focus districts. To aid in contextualising the results, indicators were also compared amongst the formerly focus and the non-focus districts at the beginning of the second phase. RESULTS: In the first phase, the levels of 34 out of 52 indicators increased significantly in the focus districts, including almost all indicators of FLW performance in antenatal and postnatal care, along with mother's birth preparedness, some breastfeeding practices, and immunisations. Between the two phases, 33 of 52 indicators declined significantly. In the second phase, the formerly focus districts experienced a rise in the levels of 14 of 50 indicators and a decline in the levels of 14 other indicators. There was a rise in the levels of 22 out of 50 indicators in the non-focus districts in the second phase, with a decline in the levels of 13 other indicators. CONCLUSIONS: Improvements in indicators were conditional on implementation support to program activities at a level of intensity that was higher than what could be achieved at scale so far. Successes during the pilot phase of intensive support suggests that RMNCHN can be improved statewide in Bihar with sufficient investments in systems performance improvements. STUDY REGISTRATION: ClinicalTrials.gov number NCT02726230.

Impact of the Ananya program on reproductive, maternal, newborn and child health and nutrition in Bihar, India: early results from a quasi-experimental study.

BACKGROUND: The Government of Bihar (GoB) in India, the Bill and Melinda Gates Foundation and several non-governmental organisations launched the Ananya program aimed to support the GoB to improve reproductive, maternal, newborn and child health and nutrition (RMNCHN) statewide. Here we summarise changes in indicators attained during the initial two-year pilot phase (2012-2013) of implementation in eight focus districts of approximately 28 million population, aimed to inform subsequent scale-up. METHODS: The quasi-experimental impact evaluation included statewide household surveys at two time points during the pilot phase: January-April 2012 ("baseline") including an initial cohort of beneficiaries and January-April 2014 ("midline") with a new cohort. The two arms were: 1) eight intervention districts, and 2) a comparison arm comprised of the remaining 30 districts in Bihar where Ananya interventions were not implemented. We analysed changes in indicators across the RMNCHN continuum of care from baseline to midline in intervention and comparison districts using a difference-in-difference analysis. RESULTS: Indicators in the two arms were similar at baseline. Overall, 40% of indicators (20 of 51) changed significantly from baseline to midline in the comparison districts unrelated to Ananya; two-thirds (n = 13) of secular indicator changes were in a direction expected to promote health. Statistically significant impact attributable to the Ananya program was found for 10% (five of 51) of RMNCHN indicators. Positive impacts were most prominent for mother's behaviours in contraceptive utilisation. CONCLUSIONS: The Ananya program had limited impact in improving health-related outcomes during the first two-year period covered by this evaluation. The program's theories of change and action were not powered to observe statistically significant differences in RMNCHN indicators within two years, but rather aimed to help inform program improvements and scale-up. Evaluation of large-scale programs such as Ananya using theory-informed, equity-sensitive (including gender), mixed-methods approaches can help elucidate causality and better explain pathways through which supply- and demand-side interventions contribute to changes in behaviour among the actors involved in the production of population-level health outcomes. Evidence from Bihar indicates that deep structural constraints in health system organisation and delivery of interventions pose substantial limitations on behaviour change among health care providers and beneficiaries. STUDY REGISTRATION: ClinicalTrials.gov number NCT02726230.

Health layering of self-help groups: impacts on reproductive, maternal, newborn and child health and nutrition in Bihar, India.

BACKGROUND: Self-help group (SHG) interventions have been widely studied in low and middle income countries. However, there is little data on specific impacts of health layering, or adding health education modules upon existing SHGs which were formed primarily for economic empowerment. We examined three SHG interventions from 2012-2017 in Bihar, India to test the hypothesis that health-layering of SHGs would lead to improved health-related behaviours of women in SHGs. METHODS: A model for health layering of SHGs - Parivartan - was developed by the non-governmental organisation (NGO), Project Concern International, in 64 blocks of eight districts. Layering included health modules, community events and review mechanisms. The health layering model was adapted for use with government-led SHGs, called JEEViKA+HL, in 37 other blocks of Bihar. Scale-up of government-led SHGs without health layering (JEEViKA) occurred contemporaneously in 433 other blocks, providing a natural comparison group. Using Community-based Household Surveys (CHS, rounds 6-9) by CARE India, 62 reproductive, maternal, newborn and child health and nutrition (RMNCHN) and sanitation indicators were examined for SHGs with health layering (Pavivartan SHGs and JEEViKA+HL SHGs) compared to those without. We calculated mean, standard deviation and odds ratios of indicators using surveymeans and survey logistic regression. RESULTS: In 2014, 64% of indicators were significantly higher in Parivartan members compared to non-members residing in the same blocks. During scale up, from 2015-17, half (50%) of indicators had significantly higher odds in health layered SHG members (Parivartan or JEEViKA+HL) in 101 blocks compared to SHG members without health layering (JEEViKA) in 433 blocks. CONCLUSIONS: Health layering of SHGs was demonstrated by an NGO-led model (Parivartan), adapted and scaled up by a government model (JEEViKA+HL), and associated with significant improvements in health compared to non-health-layered SHGs (JEEViKA). These results strengthen the evidence base for further layering of health onto the SHG platform for scale-level health change. STUDY REGISTRATION: ClinicalTrials.gov number NCT02726230.

Impact of mHealth interventions for reproductive, maternal, newborn and child health and nutrition at scale: BBC Media Action and the Ananya program in Bihar, India.

BACKGROUND: Mobile health (mHealth) tools have potential for improving the reach and quality of health information and services through community health workers in low- and middle-income countries. This study evaluates the impact of an mHealth tool implemented at scale as part of the statewide reproductive,maternal, newborn and child health and nutrition (RMNCHN) program in Bihar, India. METHODS: Three survey-based data sets were analysed to compare the health-related knowledge, attitudes and behaviours amongst childbearing women exposed to the Mobile Kunji and Dr. Anita mHealth tools during their visits with frontline workers compared with those who were unexposed. RESULTS: An evaluation by Mathematica (2014) revealed that exposure to Mobile Kunji and Dr. Anita recordings were associated with significantly higher odds of consuming iron-folic acid tablets (odds ratio (OR) = 2.3, 95% confidence interval (CI) = 1.8-3.1) as well as taking a set of three measures for delivery preparedness (OR = 2.8, 95% CI = 1.9-4.2) and appropriate infant complementary feeding (OR = 1.9, 95% CI = 1.0-3.5). CARE India's Community-based Household Surveys (2012-2017) demonstrated significant improvements in early breastfeeding (OR = 1.64, 95% CI = 1.5-1.78) and exclusive breastfeeding (OR = 1.46, 95% CI = 1.33-1.62) in addition to birth preparedness practices. BBC Media Action's Usage & Engagement Survey (2014) demonstrated a positive association between exposure to Mobile Kunji and Dr. Anita and exclusive breastfeeding (58% exposed vs 43% unexposed, P < 0.01) as well as maternal respondents' trust in their frontline worker. CONCLUSIONS: Significant improvements in RMNCHN-related knowledge and behaviours were observed for Bihari women who were exposed to Mobile Kunji and Dr. Anita. This analysis is unique in its rigorous evaluation across multiple data sets of mHealth interventions implemented at scale. These results can help inform global understanding of how best to use mHealth tools, for whom, and in what contexts. STUDY REGISTRATION: ClinicalTrials.gov number NCT02726230.

Health impact of self-help groups scaled-up statewide in Bihar, India.

BACKGROUND: The objective of this study was to assess the impact of self-help groups (SHGs) and subsequent scale-up on reproductive, maternal, newborn, child health, and nutrition (RMNCHN) and sanitation outcomes among marginalised women in Bihar, India from 2014-2017. METHODS: We examined RMNCHN and sanitation behaviors in women who were members of any SHGs compared to non-members, without differentiating between types of SHGs. We analysed annual surveys across 38 districts of Bihar covering 62 690 women who had a live birth in the past 12 months. All analyses utilised data from Community-based Household Surveys (CHS) rounds 6-9 collected in 2014-2017 by CARE India as part of the Bihar Technical Support Program funded by the Bill & Melinda Gates Foundation. We examined 66 RMNCHN and sanitation indicators using survey logistic regression; the comparison group in all cases was age-comparable women from the geographic contexts of the SHG members but who did not belong to SHGs. We also examined links between discussion topics in SHGs and changes in relevant behaviours, and stratification of effects by parity and mother's age. RESULTS: SHG members had higher odds compared to non-SHG members for 60% of antenatal care indicators, 22% of delivery indicators, 70% of postnatal care indicators, 50% of nutrition indicators, 100% of family planning and sanitation indicators and no immunisation indicators measured. According to delivery platform, most FLW performance indicators (80%) had increased odds, followed by maternal behaviours (57%) and facility care and outreach service delivery (22%) compared to non-SHG members. Self-report of discussions within SHGs on specific topics was associated with increased related maternal behaviours. Younger SHG members (<25 years) had attenuated health indicators compared to older group members (≥25 years), and women with more children had more positive indicators compared to women with fewer children. CONCLUSIONS: SHG membership was associated with improved RMNCHN and sanitation indicators at scale in Bihar, India. Further work is needed to understand the specific impacts of health layering upon SHGs. Working through SHGs is a promising vehicle for improving primary health care. STUDY REGISTRATION: ClinicalTrials.gov number NCT02726230.

Evaluation of a large-scale reproductive, maternal, newborn and child health and nutrition program in Bihar, India, through an equity lens.

BACKGROUND: Despite increasing focus on health inequities in low- and middle income countries, significant disparities persist. We analysed impacts of a statewide maternal and child health program among the most compared to the least marginalised women in Bihar, India. METHODS: Utilising survey-weighted logistic regression, we estimated programmatic impact using difference-in-difference estimators from Mathematica data collected at the beginning (2012, n = 10 174) and after two years of program implementation (2014, n = 9611). We also examined changes in disparities over time using eight rounds of Community-based Household Surveys (CHS) (2012-2017, n = 48 349) collected by CARE India. RESULTS: At baseline for the Mathematica data, least marginalised women generally performed desired health-related behaviours more frequently than the most marginalised. After two years, most disparities persisted. Disparities increased for skilled birth attendant identification [+16.2% (most marginalised) vs +32.6% (least marginalized), P < 0.01) and skin-to-skin care (+14.8% vs +20.4%, P < 0.05), and decreased for immediate breastfeeding (+10.4 vs -4.9, P < 0.01). For the CHS data, odds ratios compared the most to the least marginalised women as referent. Results demonstrated that disparities were most significant for indicators reliant on access to care such as delivery in a facility (OR range: 0.15 to 0.48) or by a qualified doctor (OR range: 0.08 to 0.25), and seeking care for complications (OR range: 0.26 to 0.64). CONCLUSIONS: Disparities observed at baseline generally persisted throughout program implementation. The most significant disparities were observed amongst behaviours dependent upon access to care. Changes in disparities largely were due to improvements for the least marginalised women without improvements for the most marginalised. Equity-based assessments of programmatic impacts, including those of universal health approaches, must be undertaken to monitor disparities and to ensure equitable and sustainable benefits for all. STUDY REGISTRATION: ClinicalTrials.gov number NCT02726230.

Improving primary health care delivery in Bihar, India: Learning from piloting and statewide scale-up of Ananya.

In 2010, the Bill and Melinda Gates Foundation (BMGF) partnered with the Government of Bihar (GoB), India to launch the Ananya program to improve reproductive, maternal, newborn and child health and nutrition (RMNCHN) outcomes. The program sought to address supply- and demand-side barriers to the adoption, coverage, quality, equity and health impact of select RMNCHN interventions. Approaches included strengthening frontline worker service delivery; social and behavior change communications; layering of health, nutrition and sanitation into women's self-help groups (SHGs); and quality improvement in maternal and newborn care at primary health care facilities. Ananya program interventions were piloted in approximately 28 million population in eight innovation districts from 2011-2013, and then beginning in 2014, were scaled up by the GoB across the rest of the state's population of 104 million. A Bihar Technical Support Program provided techno-managerial support to governmental Health as well as Integrated Child Development Services, and the JEEViKA Technical Support Program supported health layering and scale-up of the GoB's SHG program. The level of support at the block level during statewide scale-up in 2014 onwards was approximately one-fourth that provided in the pilot phase of Ananya in 2011-2013. This paper - the first manuscript in an 11-manuscript and 2-viewpoint collection on Learning from Ananya: Lessons for primary health care performance improvement - seeks to provide a broad description of Ananya and subsequent statewide adaptation and scale-up, and capture the background and context, key objectives, interventions, delivery approaches and evaluation methods of this expansive program. Subsequent papers in this collection focus on specific intervention delivery platforms. For the analyses in this series, Stanford University held key informant interviews and worked with the technical support and evaluation grantees of the Ananya program, as well as leadership from the India Country Office of the BMGF, to analyse and synthesise data from multiple sources. Capturing lessons from the Ananya pilot program and statewide scale-up will assist program managers and policymakers to more effectively design and implement RMNCHN programs at scale through technical assistance to governments.

Use of mobile technology by frontline health workers to promote reproductive, maternal, newborn and child health and nutrition: a cluster randomized controlled Trial in Bihar, India.

BACKGROUND: mHealth technology holds promise for improving the effectiveness of frontline health workers (FLWs), who provide most health-related primary care services, especially reproductive, maternal, newborn, child health and nutrition services (RMNCHN), in low-resource - especially hard-to-reach - settings. Data are lacking, however, from rigorous evaluations of mHealth interventions on delivery of health services or on health-related behaviors and outcomes. METHODS: The Information Communication Technology-Continuum of Care Service (ICT-CCS) tool was designed for use by community-based FLWs to increase the coverage, quality and coordination of services they provide in Bihar, India. It consisted of numerous mobile phone-based job aids aimed to improve key RMNCHN-related behaviors and outcomes. ICT-CCS was implemented in Saharsa district, with cluster randomization at the health sub-center level. In total, evaluation surveys were conducted with approximately 1100 FLWs and 3000 beneficiaries who had delivered an infant in the previous year in the catchment areas of intervention and control health sub-centers, about half before implementation (mid-2012) and half two years afterward (mid-2014). Analyses included bivariate and difference-in-difference analyses across study groups. RESULTS: The ICT-CCS intervention was associated with more frequent coordination of AWWs with ASHAs on home visits and greater job confidence among ASHAs. The intervention resulted in an 11 percentage point increase in FLW antenatal home visits during the third trimester (P = 0.04). In the post-implementation period, postnatal home visits during the first week were increased in the intervention (72%) vs the control (60%) group (P < 0.01). The intervention also resulted in 13, 12, and 21 percentage point increases in skin-to-skin care (P < 0.01), breastfeeding immediately after delivery (P < 0.01), and age-appropriate complementary feeding (P < 0.01). FLW supervision and other RMNCHN behaviors were not significantly impacted. CONCLUSIONS: Important improvements in FLW home visits and RMNCHN behaviors were achieved. The ICT-CCS tool shows promise for facilitating FLW effectiveness in improving RMNCHN behaviors.

Effects of team-based goals and non-monetary incentives on front-line health worker performance and maternal health behaviours: a cluster randomised controlled trial in Bihar, India.

INTRODUCTION: We evaluated the impact of a 'Team-Based Goals and Incentives' (TBGI) intervention in Bihar, India, designed to improve front-line (community health) worker (FLW) performance and health-promoting behaviours related to reproductive, maternal, newborn and child health and nutrition. METHODS: This study used a cluster randomised controlled trial design and difference-in-difference analyses of improvements in maternal health-related behaviours related to the intervention's team-based goals (primary), and interactions of FLWs with each other and with maternal beneficiaries (secondary). Evaluation participants included approximately 1300 FLWs and 3600 mothers at baseline (May to June 2012) and after 2.5 years of implementation (November to December 2014) who had delivered an infant in the previous year. RESULTS: The TBGI intervention resulted in significant increases in the frequency of antenatal home visits (15 absolute percentage points (PP), p=0.03) and receipt of iron-folic acid (IFA) tablets (7 PP, p=0.02), but non-significant changes in other health behaviours related to the trial's goals. Improvements were seen in selected attitudes related to coordination and teamwork among FLWs, and in the provision of advice to beneficiaries (ranging from 8 to 14 PP) related to IFA, cord care, breast feeding, complementary feeding and family planning. CONCLUSION: Results suggest that combining an integrated set of team-based coverage goals and targets, small non-cash incentives for teams who meet targets and team building to motivate FLWs resulted in improvements in FLW coordination and teamwork, and in the quality and quantity of FLW-beneficiary interactions. These improvements represent programmatically meaningful steps towards improving health behaviours and outcomes. TRIAL REGISTRATION NUMBER: NCT03406221.