Maternal Mortality - Methodology Resources

OBJECTIVE: Available maternal mortality statistics do not allow valid international comparisons. Our objective was to uniformly measure underreporting of mortality from pregnancy in official statistics from selected regions within the U.S. and Europe, and to provide comparable revised profiles of pregnancy-related mortality.
METHODS: We developed a standardized enhanced method to uniformly identify and classify pregnancy-associated deaths from 2 U.S. states, Massachusetts and North Carolina, and 2 European countries, Finland and France, for the years 1999-2000. Identification method included the use of all data available from the death certificate as well as computerized linkage of births and deaths registers. All cases were reviewed and classified by an international panel of experts.
RESULTS: Four-hundred-and-four pregnancy-associated deaths were identified and reviewed. Underestimation of mortality causally related to pregnancy based on International Classification of Diseases cause-of-death codes alone varied from 22% in France to 93% in Massachusetts. Underreporting was greater in the regions with lower initial maternal mortality ratios. The distribution of causes of pregnancy-related mortality was specific to each region. The leading causes of death were cardiovascular conditions in Massachusetts; hemorrhage, pregnancy-induced hypertension, and peripartum cardiomyopathy in North Carolina; noncardiovascular medical conditions in Finland; and hemorrhage in France.
CONCLUSION: This study shows the limitations of maternal mortality statistics based on International Classification of Diseases cause-of-death codes alone. Linkage of births and deaths registers should routinely be used in the ascertainment of pregnancy-related deaths. In addition, extension of the definition of a maternal death should be considered. Beyond pregnancy-related mortality ratios, considering the specific distribution of causes-of-death is important to define prevention strategies.

OBJECTIVE: The purpose of this study was to measure the reliability of a model that defines preventability in maternal morbidity and death.
STUDY DESIGN: One hundred cases of serious morbidity and death among peripartum women were reviewed by 2 independent groups of medical experts to identify potentially preventable provider or system events that may have led to the progression of illness.
RESULTS: Seventy-seven percent of the cases had concordant findings in both groups regarding the identification of any preventable events. Interrater agreement, which was measured by Cohen's Kappa, was 0.49, which suggests moderate to good reliability. Cases with multiple comorbidities, poor documentation, or preventable patient factors tended to have higher levels of disagreement.
CONCLUSION: There was high agreement between the 2 groups regarding the identification of preventable events that impact maternal morbidity and death. The reliability of this model for the assessment of preventability is an important step for improvement in obstetric and medical care.

BACKGROUND: Maternal mortality, as a largely avoidable cause of death, is an important focus of international development efforts, and a target for Millennium Development Goal (MDG) 5. However, data weaknesses have made monitoring progress problematic. In 2006, a new maternal mortality working group was established to develop improved estimation methods and make new estimates of maternal mortality for 2005, and to analyse trends in maternal mortality since 1990.
METHODS: We developed and used a range of methods, depending on the type of data available, to produce comparable country, regional, and global estimates of maternal mortality ratios for 2005 and to assess trends between 1990 and 2005.
FINDINGS: We estimate that there were 535,900 maternal deaths in 2005, corresponding to a maternal mortality ratio of 402 (uncertainty bounds 216-654) deaths per 100,000 livebirths. Most maternal deaths in 2005 were concentrated in sub-Saharan Africa (270,500, 50%) and Asia (240,600, 45%). For all countries with data, there was a decrease of 2.5% per year in the maternal mortality ratio between 1990 and 2005 (p<0.0001); however, there was no evidence of a significant reduction in maternal mortality ratios in sub-Saharan Africa in the same period.
INTERPRETATION: Although some regions have shown some progress since 1990 in reducing maternal deaths, maternal mortality ratios in sub-Saharan Africa have remained very high, with little evidence of improvement in the past 15 years. To achieve MDG5 targets by 2015 will require sustained and urgent emphasis on improved pregnancy and delivery care throughout the developing world.

OBJECTIVES: I studied the extent to which maternal deaths are underreported on death certificates.
METHODS: We collected data on maternal deaths from death certificates, linkage of death certificates with birth and fetal death records, and review of medical examiner records.
RESULTS: Thirty-eight percent of maternal deaths were unreported on death certificates. Half or more deaths were unreported for women who were undelivered at the time of death, experienced a fetal death or therapeutic abortion, died more than a week after delivery, or died as a result of a cardiovascular disorder.
CONCLUSIONS: The number of maternal deaths is substantially underestimated when death certificates alone are used to identify deaths, and it is unlikely that the Healthy People 2010 objective of reducing the maternal mortality rate to no more than 3.3 deaths per 100000 live births by 2010 can be achieved. Increasing numbers of births to older women and multiple-gestation pregnancies are likely to complicate efforts to reduce maternal mortality.

OBJECTIVE: This report presents data on U.S. deaths to pregnant or recently pregnant women, summarizes long-term processing issues, and examines recent changes affecting the data and the impact of the changes on the statistics for these women.
METHODS: This report presents descriptive tabulations of information reported on death certificates that are completed by funeral directors, attending physicians, medical examiners, and coroners. The original records are filed in the state registration offices. Statistical information is compiled into a national database through the Vital Statistics Cooperative Program of the Centers for Disease Control and Prevention's (CDC) National Center for Health Statistics (NCHS). Causes of death are processed in accordance with the International Classification of Diseases (ICD).
RESULTS: Maternal mortality fluctuates from year to year but was 12.1 deaths per 100,000 live births in 2003. The implementation of the International Classification of Diseases, Tenth Revision (ICD-10) in 1999 resulted in about a 13 percent increase in the number of deaths identified as maternal deaths between 1998 and 1999. The rate increased again between 2002 and 2003 after a separate pregnancy question became a standard item on the U.S. Standard Certificate of Death. The adoption of a standard separate question on pregnancy facilitates the identification of late maternal deaths.
CONCLUSION: Maternal deaths increased with the introduction of the ICD-10 and with changes associated with the addition of a separate pregnancy status question on the U.S. Standard Certificate of Death. These changes may result in better identification of maternal deaths.

OBJECTIVE: To estimate mortality ratios for all reported pregnancy deaths in the United States, 1999–2005, and to estimate the effect of the 1999 implementation of International Classification of Diseases, Tenth Revision (ICD- 10) and adoption of the U.S. Standard Certificate of Death, 2003 Revision, on the ascertainment of deaths resulting from pregnancy. METHODS: We combined information on pregnancy deaths from the National Vital Statistics System and the Pregnancy Mortality Surveillance System to estimate maternal (during or within 42 days of pregnancy) and pregnancyrelated (during or within 1 year of pregnancy) mortality ratios (deaths per 100,000 live births). Data for 1995–1997, 1999–2002, and 2003–2005 were compared in order to estimate the effects of the change to ICD-10 and the inclusion of a pregnancy checkbox on the death certificate. RESULTS: The maternal mortality ratio increased significantly from 11.6 in 1995–1997 to 13.1 for 1999–2002 and 15.3 in 2003–2005; the pregnancy-related mortality ratio increased significantly from 12.6 to 14.7 and 18.1 during the same periods. Vital statistics identified significantly more indirect maternal deaths in 2002–2005 than in 1999–2002. Between 2002 and 2005, mortality ratios increased significantly among 19 states using the revised death certificate with a pregnancy checkbox; ratios did not increase in states without a checkbox. CONCLUSION: Changes in ICD-10 and the 2003 revision of the death certificate increased ascertainment of pregnancy deaths. The changes may also have contributed to misclassification of some deaths as maternal in the vital statistics system. Combining data from both systems estimates higher pregnancy mortality ratios than from either system individually. (Obstet Gynecol 2011;118:104–10) DOI: 10.1097/AOG.0b013e31821fd49d LEVEL OF EVIDENCE: II

WHO on-line site for ICD-10 codes. This link is to the pregnancy section which defines all the terms used to define conditions for maternal mortality reporting.