ContributorsKH and AR led conceptualisation and drafting of the paper. AD led the study on nurses in Uttar Pradesh, ND the study on accredited social health activists in Uttar Pradesh, HW and JR the study on community health workers and community health worker policy in Sierra Leone, LM, JK, and AR the study on gender parity in the global physician workforce, and KH, YA, and NS the study on selfhelp groups in India. FS and RF-M led development of the case on the nurse from eSwatini. VP, RH, and EBa did the systematic literature review on health systems models. JGS and AR led the systematic review on gender transformative clinical interventions. KH, LM, JK, FS, RF-M, AD, YA, JY, EBl, NB, JGS, and AR did the critical reviews of the literature on gender inequalities and gender norms affecting health and helped draft pieces of those reviews, with consideration of diverse geographic contexts. All authors offered critical inputs and reviews of this work, contributed intellectual and substantive revisions to the writing, and provided final approval of the submitted version.

a b s t r a c tBackground: Despite the health system effort s, health disparities exist across sub-populations in India. We assessed the effects of health behaviour change interventions through women's self-help groups (SHGs) on maternal and newborn health (MNH) behaviours and socio-economic inequalities. Methods: We did a quasi-experimental study of a large-scale SHG program in Uttar Pradesh, India, where 120 geographic blocks received, and 83 blocks did not receive health intervention. Data comes from two cross-sectional surveys with 4,615 recently delivered women in 2015, and 4,250 women in 2017. The intervention included MNH discussions in SHG meetings and community outreach activities. The outcomes included antenatal, natal and postnatal care, contraceptive use, cord care, skin-to-skin care, and breastfeeding practices. Effects were assessed using multilevel mixed-effects regression adjusted differencein-differences (DID) analysis adjusting for geographic clustering and potential covariates, for all, mostmarginalised and least-marginalised women. Concentration indices examined the socio-economic inequality in health practices over time. Findings: The net improvements (5-11 percentage points [pp]) in correct MNH practices were significant in the intervention areas. The improvements over time were higher among the most-marginalised than least-marginalised for antenatal check-ups (DID: 20pp, p < 0 • 001 versus DID: 6pp, p = 0 • 093), consumption of iron folic acid tablets for 100 days (DID: 7pp, p = 0 • 036 versus DID: -1pp, p = 0 • 671), current use of contraception (DID: 12pp, p = 0 • 046 versus DID: 10pp, p = 0 • 021), cord care (DID: 12pp, p = 0 • 051 versus DID: 7pp, p = 0 • 210), and timely initiation of breastfeeding (DID: 29pp, p = 0 • 001 versus DID: 1pp, p = 0 • 933). Lorenz curves and concentration indices indicated reduction in rich-poor gap in health practices over time in the intervention areas. Interpretation: Disparities in MNH behaviours declined with the effort s by SHGs through behaviour change communication intervention.

The Sustainable Development Goals offer the global health community a strategic opportunity to promote human rights, advance gender equality, and achieve health for all. The inability of the health sector to accelerate progress on a range of health outcomes brings into sharp focus the significant impact of gender inequalities and restrictive gender norms on health risks and behaviours. In this paper we draw on evidence from the Series on Gender Equality, Norms and Health to dispel three myths on gender and health and describe persistent barriers to progress. We propose an agenda for action to reduce gender inequality and shift gender norms for improved health outcomes, calling on leaders in national governments, global health institutions, civil society organisations, academia, and the corporate sector to 1) focus on health outcomes and engage actors across sectors to achieve them; 2) reform the workplace and workforce to be more gender equitable; 3) fill gaps in data and eliminate gender bias in research; 4) fund civil society actors and social movements; and 5) strengthen accountability mechanisms. Paper 5 Lancet Series on Gender Equality, Norms and Health Key Messages of the Series • Gender norms and inequalities affect health outcomes for girls and women, boys and men, and gender minorities.

BackgroundThis study evaluates an eight-session behavior change health intervention with women’s self-help groups (SHGs) aimed to promote healthy maternal and newborn practices among the more socially and economically marginalized groups.MethodsUsing a pre-post quasi-experimental design, a total of 545 SHGs were divided into two groups: a control group, which received the usual microcredit intervention; and an intervention group, which received additional participatory training around maternal, neonatal, and child health issues. Women members of SHGs who had a live birth in the 12 months preceding the survey were surveyed on demographics, practices around maternal, neonatal and child health (MNCH), and collectivization. Outcome effects were assessed using difference-in-difference (DID) methods.ResultsWomen from the SHGs with health intervention, relative to controls over time (time 1 to time 2), were more likely to: use contraceptive methods (DID: 9 percentage points [pp], p<0.001), have institutional delivery (DID: 9pp, p<0.05), practice skin-to-skin care (DID: 17pp, p<0.05), delay bathing for 3 or more days (DID: 19pp, p<0.001), initiate timely breastfeeding (DID: 21pp, p<0.001), exclusively breastfeed the child (DID: 27pp, p<0.001), and provide age-appropriate immunization (DID: 9pp, p<0.001). Additionally, women from the SHGs with health intervention when compared to the control group over time were more likely to report: collective efficacy (DID: 17pp, p<0.001), support through accompanying SHG members for antenatal care (DID: 8pp, p<0.05), receive a visit from SHG member within 2 days post-delivery (DID: 32pp, p<0.001), and receive reproductive, maternal, neonatal and child health information from an SHG member (DID: 45pp, p<0.001).ConclusionFindings demonstrate that structured participatory communication on MNCH with women’s groups improve positive health practices. In addition, SHGs can reach a substantial proportion of women while providing an avenue for pregnant women and young mothers to be assisted by others in learning and practicing healthy behaviors, thus building social cohesion on health.

Objectives This study assesses associations between mistreatment by a provider during childbirth and maternal complications in Uttar Pradesh, India. Methods Cross-sectional survey data were collected from women (N = 2639) who had delivered at 68 public health facilities in Uttar Pradesh, participating in a quality of care study. Participants were recruited from April to July 2015 and surveyed on demographics, mistreatment during childbirth (measure developed for this study, Cronbach's alpha = 0.70), and maternal health complications. Regression models assessed associations between mistreatment during childbirth and maternal complications, at delivery and postpartum, adjusting for demographics and pregnancy complications. Results Participants were aged 17-48 years, and 30.3% were scheduled caste/scheduled tribe. One in five (20.9%) reported mistreatment by their provider during childbirth, including discrimination and abuse; complications during delivery (e.g., obstructed labor) and postpartum (e.g., excessive bleeding) were reported by 45.8 and 41.5% of women, respectively. Health providers at delivery included staff nurses (81.8%), midwives (14.0%), and physicians (2.2%); Chi square analyses indicate that women were significantly more likely to report mistreatment when their provider was a nurse rather than a physician or midwife. Women reporting mistreatment by a provider during childbirth had higher odds of complications at delivery (AOR = 1.32; 95% CI 1.05-1.67) and postpartum (AOR = 2.12; 95% CI 1.67-2.68). Conclusions for Practice Mistreatment of women by their provider during childbirth is a pervasive health and human rights violation, and is associated with increased risk for maternal health complications in Uttar Pradesh. Efforts to improve quality of maternal care should include greater training and monitoring of providers to ensure respectful treatment of patients.

No abstract

BackgroundUttar Pradesh (UP) accounts for the largest number of neonatal deaths in India. This study explores potential socio-economic inequities in household-level contacts by community health workers (CHWs) and whether the effects of such household-level contacts on receipt of health services differ across populations in this state.MethodsA multistage sampling design identified live births in the last 12 months across the 25 highest-risk districts of UP (N = 4912). Regression models described the relations between household demographics (caste, religion, wealth, literacy) and CHW contact, and interactions of demographics and CHW contact in predicting health service utilization (> = 4 antenatal care (ANC) visits, facility delivery, modern contraceptive use).ResultsNo differences were found in likelihood of CHW contact based on caste, religion, wealth or literacy. Associations of CHW contact with receipt of ANC and facility delivery were significantly affected by religion, wealth and literacy. CHW contact increased the odds of 4 or more ANC visits only among non-Muslim women, increased the odds of both four or more ANC visits and facility delivery only among lower wealth women, increased the odds of facility delivery to a greater degree among illiterate vs. literate women.ConclusionCHW visits play a vital role in promoting utilization of critical maternal health services in UP. However, significant social inequities exist in associations of CHW visits with such service utilization. Research to clarify these inequities, as well as training for CHWs to address potential biases in the qualities or quantity of their visits based on household socio-economic characteristics is recommended.

BackgroundBihar, India has higher rates of intimate partner violence (IPV) and maternal and infant mortality relative to India as a whole. This study assesses whether IPV is associated with poor reproductive and maternal health outcomes, as well as whether poverty exacerbates any observed associations, among women who gave birth in the preceding 23 months in Bihar, India.MethodsA cross-sectional analysis of data from a representative household sample of mothers of children 0–23 months old in Bihar, India (N = 13,803) was conducted. Associations between lifetime IPV (physical and/or sexual violence) and poor reproductive health outcomes ever (miscarriage, stillbirth, and abortion) as well as maternal complications for the index pregnancy (early and/or prolonged labor complications, other complications during pregnancy or delivery) were assessed using multivariable logistic regression, adjusting for demographics and fertility history of the mother. Models were then stratified by wealth index to determine whether observed associations were stronger for poorer versus wealthier women.ResultsIPV was reported by 45% of women in the sample. A history of miscarriage, stillbirth, and abortion was reported by 8.7, 4.6, and 1.3% of the sample, respectively. More than one in 10 women (10.7%) reported labor complications during the last pregnancy, and 16.3% reported other complications during pregnancy or delivery. Adjusted regressions revealed significant associations between IPV and miscarriage (AOR = 1.35, 95% CI = 1.11–1.65) and stillbirth (AOR = 1.36, 95% CI = 1.02–1.82) ever, as well as with labor complications (AOR = 1.27, 95% CI = 1.04–1.54) and other pregnancy/delivery complications (AOR = 1.68, 95% CI = 1.42–1.99). Women in the poorest quartile (Quartile 1) saw no associations between IPV and miscarriage (Quartile 1 AOR = 0.98, 95% CI = 0.67–1.45) or stillbirth (Quartile 1 AOR = 1.17, 95% CI = 0.69–1.98), whereas women in the higher wealth quartile (Quartile 3) did see associations between IPV and miscarriage (Quartile 3 AOR = 1.55, 95% CI = 1.07, 2.25) and stillbirth (Quartile 3 AOR = 1.79, 95% CI = 1.04, 3.08).DiscussionIPV is highly prevalent in Bihar and is associated with increased risk for miscarriage, stillbirth, and maternal health complications. Associations between IPV and miscarriage and stillbirth do not hold true for the poorest women, possibly because other risks attached to poverty and deprivation may be greater contributors.