Data for Maternal Mortality

Maternal Figures present a webinar on data sources for maternal mortality, with a focus on Nigeria.

Transcript

1. Data for Maternal Mortality Webinar March 28, 2020
2. Ashley Okwuosa Project Lead Leads the fact-checking of the interventions, interfaces with our partner organizations, and supervises the qualitative research. Chuma Asuzu Responsible for data research and product manager for the database. 2 / 25 Team #MaternalFigures
3. An introduction to statistical terms and maternal health metrics Maternal Mortality Ratio - definition and sources When to use which source Call for research participants Webinar Overview 3 / 25 #MaternalFigures
4. Population: The total number of people in the group under study. Sample Size: the number of completed responses a survey receives. Margin of error: the percentage that tells how well the survey results reflects the views of the overall population. Confidence level: A percentage that reveals how confident you can be that the population would select an answer within a certain range. For example, a 95% confidence level means that you can be 95% certain the results lie between x and y numbers. Correlation: a statement about the relationship between two variables Causation: a statement that implies that an event is the result of another Statistical Terms 4 / 25 #MaternalFigures
5. Data helps show the patterns in a population. 5 / 25 #MaternalFigures
6. Exclusive breastfeeding (%) Skilled birth attendance (%) Antenatal care (%) In-facility delivery rate (%) Under-5 mortality (%) Percentage of underweight children under-5 Maternal Mortality Ratio Relevant maternal and child health metrics 6 / 25 #MaternalFigures
7. The number of maternal deaths per 100,000 live births. A maternal death (as cited in International Classification of Disease or ICD-10, [WHO, 1992]) is the death of a woman while pregnant or within 42 days of termination of pregnancy, irrespective of the duration and the site of the pregnancy, and can stem from any cause related to or aggravated by the pregnancy or its management, but not from accidental or incidental causes. It is simply calculated using the formula: All maternal deaths occurring within a reference period (usually 1 year) x 100,000 ______________________________________________________ Total # of live births occurring within the reference period Maternal Mortality Ratio 7 / 25 #MaternalFigures
8. Many maternal deaths are unrecorded Delivery not in hospitals Reason for death not determined or unknown Lack of skilled birth attendants at the delivery Many live births are also unrecorded, but at a smaller scale comparatively Nigeria does not have an up-to-date civic registration and vital statistics (CRVS) database 8 / 25 Difficulty calculating the MMR #MaternalFigures
9. One SDG goal is to reduce the global MMR to less than 70 per 100,000 live births by 2030. In one of WHO's recent documents, Strategies toward ending preventable maternal mortality (EPMM), there is also a target that no country should have an MMR greater than 140 per 100,000 live births. One of the biggest challenges is having a robust CRVS system. Why is the maternal mortality ratio important? 9 / 25 #MaternalFigures
10. National Demographic Health Survey (NDHS) World Health Organization, The World Bank, UNFPA The Institute for Health Metrics and Evaluation That depends on who you ask. There are three frequently cited sources. A survey carried out using the DHS methodology, implemented by the National Planning Commission and the National Ministry of Health, with technical assistance from ICF who administer the DHS programme funded by USAID. Global estimates published every few years to show a trend over the last 15 years using three main variables. Global estimates published every few years to show a trend from 1990 using all the available surveys and data about and from the country. So, what is Nigeria's MMR? 10 / 25 #MaternalFigures
11. 11 / 25 Each source tells a different figure. So, what is Nigeria's MMR? #MaternalFigures
12. It's a large survey carried out nationwide in geographically-mapped enumeration areas (EAs). For the 2018 survey, 42,000 households were targeted as the sample size. The objective of the survey is to provide reliable estimates of fertility levels and preferences, awareness and use of family planning methods, breastfeeding practices, nutritional status of mothers and young children, early childhood mortality and maternal mortality, maternal and child health, knowledge and behaviour regarding HIV/AIDS and other sexually transmitted infections (STIs), female genital cutting, fistula, and domestic violence. The maternal mortality ratio is sampled using a method called the sisterhood method. 1 National Demographic Health Survey 12 / 25 #MaternalFigures
13. The sisterhood method was developed by Wendy Graham in 1989 in order to overcome the problem of large sample sizes and reduce the cost of running large household surveys, particularly in developing countries without CRVS systems. It reduces sample size requirements because it obtains information by interviewing respondents about the survival of all their adult sisters. In settings with high levels of maternal mortality (over 500 maternal deaths per 100,000 live births), sample sizes needed can be of the order of 4,000 households or less. Hence, it is an indirect method. In 2018, the latest NDHS report, Nigeria's MMR was said to be 556 per 100,000 live births. 1 National Demographic Health Survey 13 / 25 #MaternalFigures
14. The likelihood of the siblings to know the true health status of their family members In Nigeria, many respondents have been noted to not know how to correctly fill the questionnaire (NDHS report, 1999) Wrong sibling histories (year of death) Mixing personal data with sibling data Skewed gender ratio showed underreporting of women and overreporting of men This method is noted to have some issues: 1 National Demographic Health Survey 14 / 25 #MaternalFigures
15. Every few years, the Maternal Mortality Estimation Inter-Agency Group (MMEIG) comprising of experts from the WHO, UNICEF, UNPD, UNFPA, World Bank, and some researchers from a couple of universities publish the Trends in Maternal Mortality report. This report details estimates, derived from data analysis, of the maternal mortality across the world. Before the estimates are published, WHO regional offices share the draft around the world for consultations and feedback from local experts. 2 WHO, World Bank, UNICEF Trends 15 / 25 #MaternalFigures
16. Gross Domestic Product - an economic indicator, using the World Bank's figures, tells the country's economic health General fertility rate - a risk metric, using UNPD's estimates proportion of births with skilled birth attendance - a process variable, using UNICEF's figures Their data analysis is based on three variables: These three variables are used to compute the proportion of deaths among women of reproductive age that are due to maternal causes, and then the MMR. In essence, the estimate is derived from other estimates. 2 WHO, World Bank, UNICEF Trends 16 / 25 #MaternalFigures
17. It uses the same variables for every country Since it has to be a trend for the analysis to work, every publication presents different values for some of the same years This method is noted to have some issues: 17 / 25 #MaternalFigures 2 WHO, World Bank, UNICEF Trends
18. The IHME has been publishing MMR data since 1990. Like the WHO they also publish trends, but they use put more variables into account in their General Burden of Disease (GBD) estimation tool. They input demographic surveys, regional reports, and published research. However, unlike the other two sources, the maternal mortality ratio they output is only one of the estimates generated by their model. It is not the primary output. 3 Institute for Health Metrics and Evaluation 18 / 25 #MaternalFigures
19. Source Comparison 19 / 25 #MaternalFigures NDHS Good for speaking about Nigerian demographic data Helpful when reporting about a federal government initiative WHO/World Bank/UNICEF Good when comparing countries published in the same trend Good when reporting about a donor-funded initiative Good when reporting on a health story concerning multiple African countries IHME Good when writing about disease causes, especially in comparison
20. Verbal and Social Autopsy (VASA) Survey 2016 National Urban Reproductive Health Initiative (NURHI) Women Health and Action Research Centre (WHARC) Other notable maternal health data sources 20 / 25 #MaternalFigures
21. What metrics matter when it comes to measuring success? Are there any conflicts of interest? Does the data help the story? What does the metric mean and is it composed of any variables? 21 / 25 #MaternalFigures Key Questions to ask of data
22. 22 / 25 Database of maternal health interventions in Nigeria Research and data support from the Brown Institute for Media Innovation Maternal Health reporting trainings and workshops Maternal Figures provides: #MaternalFigures How Maternal Figures can help you Email us: info@maternalfigures.com & read our blog at: blog.maternalfigures.com
23. 23 / 25 We are interested in recording the maternal deaths in a state or community over the period of one month, after the current COVID-19 scare is over. Will you be interested in going around your local government recording the live births and maternal deaths? Send us an email indicating your interest to maternalfigures@gmail.com or chuma@maternalfigures.com. #MaternalFigures Call for research participants
24. 24 / 25 research, publishing partnership advisory support grant funders, advisory support Partners & Advisors #MaternalFigures
25. #MaternalFigures Thank you for being here!