The goal of this study is to generate unique information to guide improvements on interventions to reduce maternal and newborn mortality as well as prevent stillbirths.

The objective of this study is to determine the burden, timing, and causes of maternal deaths, stillbirths and neonatal deaths.

This will be an observational study where data will be collected retrospectively in the context of the ongoing study in Haryana, India.

Women of reproductive age living in the study area have already been enumerated for the parent studies. Active surveillance is being conducted for identifying pregnancies and deaths among women of reproductive age in the population through 3 monthly home visitations.

Verbal autopsies will be conducted for all deaths of women in the reproductive age, including those who died in pregnancy, childbirth and up to 42-60 days after childbirth. Verbal autopsy interviews will also be conducted for all stillbirths.

RATIONAL & BACKGROUND INFORMATION

Among the Millennium Development Goals, those that aim to reach significant reductions in maternal and child mortality have been proving the hardest to achieve.

Most maternal and neonatal deaths are caused by a few conditions, and majority of them are preventable with currently available, effective interventions. About a third of stillbirths occur in the intrapartum period, integrally related to the availability and quality of essential and emergency childbirth services. Effective interventions to reduce maternal deaths, stillbirths and newborn and child deaths reach less than half the pregnant women, mothers, newborns and children who need them. Information on mortality and morbidity as well as health care delivery is largely available only from surveys, which have inherent methodological issues that limit validity. Vital registration information and reliable information from the health system is missing in the same countries where the burden of mortality is the highest and health care is least accessible. This leads to underestimates of the magnitude of the problem with negative implications for the prioritization of interventions to scale-up.

The BMGF and WHO are supporting a set of large intervention trials that will follow cohorts of pregnant women and newborns to document the impact of the interventions. These trials provide a unique opportunity to address additional questions relevant for improving maternal, newborn and child health (MNCH) programs and for addressing the above-mentioned gaps in data.

STUDY GOALS AND OBJECTIVES Goal

The study will be conducted as a part of the large ongoing parent study on newborn health research in Haryana, India titled "Efficacy of neonatal vitamin A supplementation in improving child survival in Haryana, India: generation of evidence necessary for informing global policy" to better understand the epidemiology of maternal mortality, stillbirths and neonatal deaths.

Objective

The current study is collecting information on the burden, timing and causes of neonatal deaths and an estimate of maternal deaths and still births. The proposed additional study presents a unique opportunity to estimate population-based burden, timing, and causes of maternal deaths and stillbirths.

Study Design

This will be an observational study where data will be collected retrospectively.

Study Population

This study will be conducted in the same study areas and population as the parent study. The participants in this study will be pregnant women identified through surveillance among all women of reproductive age in the population where the parent study is being conducted, and their newborn infants. Some women not eligible for enrollment into the parent study due to loss of their babies or live babies with exclusion criteria or those who do not give consent for the parent study or women who died during pregnancy or child birth may still be included in the AMANHI study. Information on women refusing to take part in the parent studies is being documented. This information will be used for analysis exploring whether this group has a different risk of maternal or newborn mortality. Only the control group in the parent study will be considered for analysis of neonatal deaths in this observational study.

Maternal mortality, pregnancy outcomes and consequent neonatal mortality will be measured among women identified as pregnant during surveillance and followed-up until 42-60 days after delivery.

The period of data collection for this objective will be 12-24 months.

Sample Size Estimates

Maternal mortality, stillbirth rates and neonatal mortality will be estimated for all pregnant women, mothers and newborns from the time that the site meets the minimum criteria for inclusion.

The verbal autopsies will be filled retrospectively for a window period of 1 year for deaths in women and for still births.

Verbal autopsies for neonatal deaths are already being filled in parent study.

METHODOLOGY

Description of study site:

The parent study, Neonatal Vitamin A supplementation trial, is being conducted in 16 clusters, of which 13 are catchment areas of Primary Health Centres in Faridabad and Palwal districts in the state of Haryana and 3 are urban slum clusters in Faridabad district.

Study Procedures

Active surveillance is being conducted for identifying pregnancies and deaths among women of reproductive age in the population through 3 monthly home visitations.

At the time of identification of pregnancy, information is recorded on the details of the pregnancy including last menstrual period. For women whose infants are enrolled into the study, information on vital status of the mother is obtained from the 3 monthly follow up visit and the pregnancy surveillance register retrospectively as well as prospectively for those infants whose 3 month follow up visits are still due.

For women whose infants are not enrolled due to exclusion criteria or women who had still births or abortions, visits are made at 42 to 60 days to record information on vital status of the women.

Miscarriages/abortions, stillbirths, live births, neonatal deaths and maternal deaths are recorded. Verbal autopsies will be conducted for all deaths of women in the reproductive age, including those who died in pregnancy, childbirth and up to 42-60 days after childbirth for still births and neonatla death. Sites will use a standardized verbal autopsy questionnaire. Specially trained workers will conduct verbal autopsies within 3 months of reporting of a death. For deaths that have already been reported since the initiation of the parent studies, verbal autopsies can be done within 12 months of death (WHO verbal autopsy standards, 2007).

The verbal autopsy instruments have been pre-tested and used extensively in many settings. Nevertheless, the site will translate the questionnaire and pretest it to ensure that the translated questions capture the intended information. Any changes based on pre-tests will be made prior to implementation.

Safety Considerations

This is an observational study in which pregnant women are being identified and followed up until the end of the postnatal period to identify stillbirths and maternal and neonatal deaths as part of the parent studies. A verbal autopsy interview will be conducted for all of these deaths. There are no direct safety concerns.

Ethical Considerations

Ethical approval has been obtained from WHO and SAS ERC.

Informed Consent

Informed consent is already being requested from each household to conduct reproductive surveillance in order to identify pregnant women and periodic follow up. This consent is verbal and the informed consent form is signed by the member of the research team obtaining consent to confirm that he/she has accurately read the consent form to the potential participant/head of household, the individual has had the opportunity to ask questions and has consented freely. The informed consent forms for verbal autopsies will be confirmed with a signature or thumbprint by the participant. In the absence of a signature, a witness (other than the member of the research team obtaining consent) will be asked to sign and counter signed by the member of the research team obtaining consent.

Confidentiality

All identifier information will be delinked and identifier information will be stored under lock and key after the analysis is complete.

Data Management

Information on a core set of variables will be collected with standard definitions used across all sites. The WHO will provide a set of range and consistency checks that must be applied to these variables, although the exact procedures used to carry out these checks will be left to the individual sites.

Electronic data capture or double data entry by separate data entry specialists will be performed onsite. Alternatively, scanning of forms using special TeleForms software will be performed. The site will be responsible for initial cleaning of the data, including running range and consistency checks as well as periodic reviews of distributions and identification of outliers. The study site will resolve any inconsistencies within their database in consultation with their field data collection team and with field verification if needed.

Data on the core set of variables from the study site will be sent every 3 months to a central data repository established for the project at the WHO. The WHO team will run another set of range and consistency checks including checking of consistency of data quality across sites. Any inconsistencies or queries will be provided to the study site, which will be expected to check and address the list of queries and resubmit data. Cleaned data from all sites will be pooled and stored in a SQL database at the WHO.

Statistical Analysis

All verbal autopsies will be analyzed using reliable and valid computer-based methods including the Random Forests, Inter-VA and Tariff methods. The site will conduct the analysis of their verbal autopsy data in order to determine the causes of deaths for stillbirths and neonatal and maternal deaths. A data analysis workshop will be held at the WHO in Geneva in the second quarter of 2014 to conduct a pooled analysis of the data.

Quality Assurance

Principal investigators will provide monthly progress reports during the entire study period and will participate in regular telephone conferences with WHO staff.

Field supervisors will check all forms completed by field workers before data entry. Field supervisors will observe 5% of randomly selected field worker verbal autopsies. Data cleaning quality assurance will be performed using consistency and range checks both at the study sites and at the WHO data coordination centre as described above in the data management section.

The WHO Department of Maternal, Newborn, Child and Adolescent Health will be responsible for monitoring the study in collaboration with selected site investigators. WHO staff and/or site investigators will perform site monitoring visits to each site at least every 6 months to assess the quality of study implementation and data collection using a standardized site monitoring checklist. The site visit will lead to the provision of recommendations for study improvement. More intensive visits are planned for the first six months of study implementation to ensure prompt problem identification and resolution.

Inclusion Criteria:

• All deaths

Exclusion Criteria:

• Non consent for verbal autopsy

All deaths of women in reproductive age group and all still births in specified window period for whom the respondents have given consent to provide information.

This study will be conducted in the same study areas and population as the parent study. The participants in this study will be pregnant women identified through surveillance among all women of reproductive age in the population where the parent study is being conducted, and their newborn infants. The parent study is not conducting an intervention related to maternal health or stillbirths. Therefore, the results of the AMANHI study related to maternal and stillbirth outcomes are not likely to be influenced by what happens in the parent study. However, only the control group in the parent study will be considered for analysis of neonatal deaths in this observational study.