Maternal mortality is difficult to measure. Vital registration and health information systems in most developing countries are weak, and thus, cannot provide an accurate assessment of maternal mortality. Even estimates derived from complete vital registration systems, such as those in developed countries, suffer from misclassification and underreporting of maternal deaths.

Because maternal mortality is a relatively rare event, large sample sizes are needed if household surveys are used. This is very costly and may still result in estimates with large confidence intervals. To reduce sample size requirements, the sisterhood method measures maternal mortality by asking respondents about the survivorship of sisters. While this method reduces sample size requirements, it produces estimates covering some 7-12 years before the survey, which renders data problematic for monitoring progress or observing the impact of interventions. The direct sisterhood method asks respondents to provide date of death, which permits the calculation of more recent estimates, but even then the reference period tends to refer to 0-6 years before the survey.

The ability to generate country, regional, and global estimates with higher precision and accuracy would be greatly facilitated if country civil registration systems were further improved. This improvement would reduce the need to conduct special maternal mortality studies (which are time-consuming, expensive, and of limited use in monitoring trends).

In addition, owing to the very large confidence limits around maternal mortality estimates, trends in maternal mortality should be interpreted with caution. It is recommended that process indicators, such as attendance by skilled health personnel at delivery and use of health facilities for delivery, be considered in assessing progress towards the reduction in maternal mortality.

The maternal mortality ratio should not be confused with the maternal mortality rate (the number of maternal deaths in a population divided by the number of women of reproductive age). The maternal mortality rate captures the likelihood of both becoming pregnant and dying during pregnancy or the puerperium (six weeks after delivery).

Project Countries

Tajikistan

Comparisons

Country Indicators

Differences

Why is there a difference?

International

Country