Child and Maternal Mortality

in Islamic Countries

1988. This approach is adapted from the method first developed by Graham et al. (1989)

estimate adult mortality indirectly based on the survivorship of siblings. In the DHS, a series of

questions was asked of a sample of adult respondents regarding their sisters born to the same

mother, including 1) how many sisters were ever-married; 2) the vital status of these sisters;

and 3) if dead, did they die while pregnant, during childbirth, or during the six weeks following

delivery. The reporting of deaths covers a large interval time, and as a result the indirect

sisterhood method generates a maternal mortality estimate for a period centered around l0-I2

years before the survey, which is a major disadvantage. The DHS surveys currently use the direct

sisterhood method, which collects additional data to allow for the calculation of person-years of

exposure time necessary for the calculation of age-specific mortality rates. Both the indirect and

direct methods include all pregnancy-related deaths as maternal deaths. However, the direct

method also requires and assumes that the respondent can report accurately on the age of their

living siblings and the age at death and years since death for their dead siblings.

The standard DHS surveys predominately rely on the direct sisterhood method of measuring

maternal mortality. Selected special studies, dedicated to maternal morbidity and mortality,

utilized the

direct household method

. Some key examples are the Bangladesh Maternal Mortality

and Health Care Survey in 2001 and 2010, the Afghanistan Mortality Survey 2010, and the 2015

SUPAS (sample registration survey) in Indonesia. The direct household method incorporates

questions into an existing household survey with a sample population large enough to provide

reliable estimates. Key informant respondents are asked about deaths that occurred in the

household during a reference period prior to the survey (usually three years). For deaths of

women of reproductive age, there are further questions related to the timing of death to

determine if the death was pregnancy-related. There are several advantages of this method: the

place of death is known, which provides an opportunity to estimate MMR for the subnational

level or the geographical region; household characteristics of the deceased women are known,

which may be utilized for assessing disparity in maternal mortality by household related risk

factors (e.g., economic status). In some countries, a follow-up survey is often conducted to

ascertain the causes and contexts of maternal deaths using verbal autopsy tools, which provide

an opportunity to differentiate between true maternal deaths and pregnancy-related deaths.

In Chapter-III of the case-country studies, we present the MMR estimates of BMMS and SUPAS.

These surveys have the advantage of estimating MMR for subgroup of population (e.g., by age

group, education level of women, socioeconomic status and regions).

We use both the Maternal Mortality Ratio (MMR) and Maternal Mortality Rate (MMRate) in this

report. They are defined below.

The MMR is defined as the number of maternal deaths per 100,000 live births:

ℎ

100,000 =