Field test results of the motherhood method to measure maternal mortality

Mahesh K. Maskey1, Kedar P. Baral 2, Rajani Shah3 , Bhagawan D. Shrestha4, Janet Lang5, & Kenneth J. Rothman6

1Nepal Public Health Foundation, Kathmandu, 2 Patan Academy of Health sciences,

3 CTEVT, Bharatpur, 4 Plan Nepal, Nepal 5 Watson Institute for International studies, Brown University, Providence RI,

USA and 6 RTI Health solutions, Research Triangle Park, NC & Boston University school of Public Health, USA

Indian J Med Res 133, January 2011, pp 64-69

Presented by

Dr. Fredrick Stephen

P.G in community Medicine

About The Journal

• Scope : Technical and clinical studies related to health, ethical and social issues in field

of biomedical research

• Frequency : Monthly, in two volumes and 12 issues per year

• Indexed by : Caspur, CNKI (China National Knowledge Infrastructure), EBSCO

Publishing's Electronic Databases, Google Scholar, Index Copernicus, Index Medicus for

South-East Asia Region, Indian Science Abstracts, IndMed, MEDLINE/Index Medicus,

National Science Library, Open JGate, PubMed, Pubmed Central, Science Citation

Index, Science Citation Index Expanded, Journal Ranking, SCOLOAR, SCOPUS, Ulrich's

International Periodical Directory, Web of Science

• Impact factor : 2.061

• Editor : Dr Anju Sharma

Introduction

• The current estimate of global maternal deaths is 3429001*

• Over the past decade, reduction in maternal deaths has attained a high priority in

global health movements

• MDG5 has set a target of reducing the maternal mortality ratio by 75 % between 1990

and 2015.

• The most widely used measure of maternal mortality is the maternal mortality ratio

• Developing countries - vital registration of medically-certified births and deaths is non-

existent / incomplete, validity or feasibility of other purely records-based approaches is

questionable

* Hogan MC, Foreman KJ, Naghavi M, Ahn SY, Mengru W, Makela SM, et al. Maternal mortality for 181 countries, 1980-2008: a systematic analysis of progress towards Millennium Development Goal 5. Lancet. 2010;375:1609–23. [PubMed]

Aims & Objectives

• To test a community-based method (the motherhood method) and measure maternal and child mortality in a developing country setting.

Maternal Mortality Estimation

Maternal Mortality

Empirical Methods

Analytical Methods

Routine opportunity

Measurements

Special opportunity

Measurements

Main ApproachComposite Approach

Routine Opportunity Empirical

Measurements

Special OpportunityEmpirical Measurements

• Vital

Registration

System

• Census

• Facility Based

Health Services

Records

• Sisterhood

Methods

• Demographic

Surveillance

Systems

Empirical Methods

Analytical Methods

Main Approach

• Birth & death

Record Linkage

• Capture

Recapture

• Statistical

Modelling used

by UN systems

Composite Approach

• ReproductiveAge Mortality

Study (RAMOS)

• Motherhood Method

Reproductive Age Mortality study (RAMOS)

• It uses multiple sources such as records from hospital, police, public-health

department and vital data registries to identify and investigate the cause of

deaths for each woman of reproductive age in a defined population

• Interviews with household members and health care providers provide a basis

to classify the deaths as maternal or otherwise.

• Is considered to be the most complete estimation of maternal mortality

• Complex, because information regarding the number of births must come

from separate sources

Sisterhood Method

• Was originally developed during the late 1980’s

• The approach was designed to overcome the problem of large sample

sizes and thus reduce costs

• It is an indirect measurement technique of the kind frequently used to

measure a variety of demographic parameters (such as child or adult

mortality), which has been adapted for the measurement of maternal

mortality.

• The method reduces sample size requirements because it obtains

information by interviewing respondents about the survival of all their

adult sisters.

When to Use

• When there is no reliable estimate of the level of maternal mortality

• An approximate level of maternal mortality needed for advocacy purposes and to draw attention to the problem

• Poor Resources

• A starting point is needed for more detailed follow-up of maternal deaths identified during the recent past.

Not appropriate for

Measuring progress towards safe motherhood in the short term

Evaluating programme impact

Comparing geographic areas or studying trends

Allocating resources.

Approximate Sample Size Requirements for Indirect Applications of the Sisterhood Method

(*) Assuming 2 adults available for interviewing per household(**) Maternal mortality ratios from about 200 per 100 000 live births for "intermediate" level to more than 1000 for "extremely high" level.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

Time Location of Estimates from Indirect Sisterhood Method

Reports cover deaths occurring over a large interval time, hence results generate an overall estimate of maternal mortality for a point centred around 10 to 12 years before the survey

Method Strengths Limitations

Original indirect method

Four simple questions can be added to ongoing household survey

Minimal time requirements Minimal sample size requirements Simple calculations to estimate

ratios Additional information can be

gathered on place/time/cause of death

Care needed in the use understanding of the questions

Provides retrospective estimate (10-12 years prior to the survey)

Not appropriate for use in settings with high levels of migration

Not appropriate for use in settings with declining or low fertility (TFR<3)

Direct Method

Can be added to ongoing multipurpose household survey

Smaller sample size requirements than household surveys but larger than indirect methods

Can be used to provide more recent estimates than the indirect method

Relatively inexpensive

Data collection more complex & longer than indirect method

Not appropriate for use in settings with high levels of migration

Not appropriate for use in settings with low fertility (TFR<3)

Not appropriate for monitoring in the short term

Number of respondents needed to establish a maternal mortality ratio of 300 per 1,00,000 live births correct to within 20%

MMR Indirect Method

Direct Method Household Survey

300 4,000* 5,000* 50,000**

* Adult respondents** Births

THE ORIGINAL SISTERHOOD METHOD'S FOUR QUESTIONS

1. How many sisters (born to the same mother) have you ever had who were ever-married (including those who are now dead)?

2. How many of these ever-married sisters are alive now?

3. How many of these ever-married sisters are dead?

4.How many of these dead sisters died while they were pregnantor during childbirth, or during the six weeks after the end of pregnancy?

THE "DIRECT" SISTERHOOD METHOD'S QUESTIONS

1. How many children did your mother give birth to?

2. How many of these births did your mother have before you were born?

3. What was the name given to your oldest (next oldest) brother or sister?

4. Is (NAME) male or female?

5. Is (NAME) still alive?

6. How old in (NAME)?

7. In what year did (NAME) die? OR How many yeas ago did (NAME) die?

8. How old was (NAME) when she died?

For dead sisters only:

9. Was (NAME) pregnant when she died?

10.Did (NAME) die during childbirth

11.Did (NAME) die within two months after the end of pregnancy or childbirth?

Motherhood Method• It is a direct technique for deriving local population-based estimates of

maternal mortality.

• The method involves estimating the same information within a

geographic area as would be collected in a census, but without visiting

every household.

• It is a targeted census of births and deaths within a defined study period.

• Evolutionary variant of the Participatory Community Survey method - for

neonatal tetanus and the perinatal mortality rate in rural Nepal

• It shares features with the Boerma and Mati’s “networking” approach of

eliciting maternal deaths and

• MIMF (Maternal death from Informants and Maternal

death Follow-on review)

• It differs, however, in eliciting deaths through group discussion of

listed mothers and community health care providers.

• To implement the method, the local health volunteers assist in

facilitating group discussions related to maternal and child health.

• Information on total births and maternal death during pregnancy,

childbirth or puerperium is elicited through immunization

registries, group discussions (FGD), peer memory, memory aids

and verbal autopsy.

Materials and Methods

• Motherhood method was pretested in a small, relatively well-off community of

about 8000 population (MMR 140/100000)

• The method was tested in a larger sample of 15161 births in the Bara district of

Nepal

• The sample size was expected to provide estimate of MMR within 30 % o

margin of error.

• This study employed the pregnant women group (PWG) approach

• The PWG comprised 7-15 pregnant women living in the same village or wards.

• They met once a month to discuss issues related to mother and child health.

The female community health volunteers (FCHVs) facilitated these meetings

Materials & Methods

Bara District, Nepal (Total predicted population for 2005=6,15,933)[1,30,578 women of reproductive age (15-49)]

Divided into 7 sectors

7 VDC* with population of 6000 were selected

Total 49 VDC selected

* - Village Development Committees

• Study Period : 2 yr from 17th July 2003 to 16 July 2005

• Information : Births, Maternal Death, Infant Death & PWG status

• Survey Period : 12 weeks

• The data were checked every day for omissions and errors and

corrected in the

field by revisits when necessary.

• Sub-sample of 49 wards was randomly selected, one from each VDC, to

conduct a census to validate the information obtained from the

motherhood method.

• Training : 2 days training was provided to supervisors and enumerators

to enable them to elicit required information from BCG and TT

vaccination registries and from the group discussion.

• Limitations of BCG registries were partially compensated for by

augmenting the list from TT vaccination registries.

• FGD : The typical group comprised 10-15 mothers and the local health

workers

• Deliveries outside the study period were excluded from the list of

counted pregnancies

• Each group had four data collectors with 3 enumerators and one

supervisor.

• On an average one group took five days to cover one VDC.

• The total cost of the evaluation was $ 10,896

• It was found that doing a census was 10 times more costly than

collecting data from motherhood method

(per unit cost $ 50.5 and $ 4.4)

Table 1 Total Births and Deaths by Study Groups, Bara District

RESULTS

Table 2: Mortality Indices of Bara District Compared with National Estimates

Results

• The results compared well with national data.

• A comparison with the census results in 49 wards showed 100 per

cent agreement with MM in detecting maternal and child deaths.

• There was about a 0.25 per cent under-reporting of births.

• The maternal, infant, neonatal and perinatal indicators in PWG

women were lower than the non-PWG women and the national

statistics.

Discussion

• Motherhood method demonstrated - MMR can be directly measured if

the BCG and TT vaccination registers are in place and local health

workers are properly mobilized and supervised for data collection

• Proper motivation of community key informants, health volunteers,

and mobilizers is crucial for the accuracy of data

• The motherhood method can be applied in a time and cost-efficient

manner to measure and monitor the progress in the reduction of

maternal and child deaths

• It can give current estimates of maternal mortality as well as averages

over the past few years

Discussion

• It appears to be particularly well-suited in measuring and

monitoring programmes in sub-national regions and districts

• Group discussion counteracted the disinclination of mothers

to talk about the death of their child, and enhanced collective

memory for recalling details

• The motherhood method appears to be effective regarding

problems induced by migration.

• The group discussion could elicit which mothers migrated to

the village to live or came to their mother’s home for delivery

Limitations

• Requires proper training of field assistants to moderate the group

discussion among mothers and health volunteers

• Motivation of key community informants and health volunteers is crucial

to the accuracy of data, and mothers need to be aware of the need for

accuracy

• The effort in collecting data depends on the duration of the study period,

the longer the study period, the greater the potential for inaccurate

recall

Limitations

• Reporting of maternal deaths in early pregnancy and those

related to abortion as non-maternal deaths may occur.

• Maternal Deaths related to “Hidden pregnancy” among teens

can be missed.

• The method would need further adaptation to measure births

and deaths in urban areas.

THANK YOU