Maternal Mortality and Morbidity Study

[Table of Contents]

---

Research Activities

Chair:	Linda Turner
Members:	Linda Bartlett, Maureen Heaman, Robert Kinch, Michael Kramer, Robert Liston, Sylvie Marcoux, Konia Trouton, Shi Wu Wen

Purpose

---

To assess the magnitude and further our knowledge of the epidemiology of maternal mortality and serious pregnancy-related morbidity in Canada.

Relevance to the CPSS

---

The risks of death and serious morbidity related to pregnancy have declined considerably in the past 50 years, partly as a result of improvements in maternal care, but also because of reduced fertility and reductions in the number of births occurring at the extremes of maternal age. Nevertheless, there is still a need to monitor secular trends, regional differences in maternal mortality rates and serious pregnancy-related morbidity, so that any emerging problems that are identifiable on a population level can be addressed.

It is important to develop a capacity within the CPSS to monitor maternal deaths on a national level. The rules and weights that have been developed through this study (i.e. the rules used to link records and the probabilistic weights assigned to records to determine how likely they are to match records in another file) will facilitate future record linkages to ascertain maternal mortality through the expanded definitions.

Definitions of Maternal Mortality

---

1. ICD-9 definition:¹ A maternal death is defined as the death of a woman while pregnant or within 42 days of the termination of pregnancy, irrespective of the duration and the site of the pregnancy, from any cause related to or aggravated by the pregnancy or its management but not from accidental or incidental causes.^c

c.    This is the standard definition of maternal mortality.

    Maternal deaths should be subdivided into two groups

• Direct obstetric deaths: those resulting from obstetric complications of the pregnant state (pregnancy, labour and puerperium); from interventions, omissions or incorrect treatment; or from a chain of events resulting from any of the above.
• Indirect obstetric deaths: those resulting from previous existing disease or disease that developed during pregnancy and that were not due to direct obstetric causes but were aggravated by the physiologic effects of pregnancy.

2. ICD-10 definition:² The definition of "maternal death" is the same as the ICD-9 definition. Included, however, are two additional methods of defining maternal death. These are:

• Late maternal deaths: deaths from direct or indirect obstetric causes more than 42 days but less than one year after the termination of pregnancy.
• Pregnancy-related deaths: deaths during pregnancy or within 42 days of the termination of pregnancy, irrespective of the cause.

The primary purpose of the expansions beyond the traditional ICD-9 definition of maternal mortality is to provide greater assurance that all deaths that are truly "maternal deaths" - that is, caused or aggravated by pregnancy or its management - are captured. Expanding consideration to include deaths of women from any cause within a year of a pregnancy outcome provides a greater likelihood of ascertaining those that may have been miscoded as unrelated to pregnancy on the death certificate or that occurred later than the usual 42-day cut-off.

Extending the period of consideration to a year post-pregnancy outcome and including deaths from other causes also allow ascertainment of cause-specific death rates among women following a pregnancy outcome. These can then be compared with cause-specific death rates among women who were not known to have been pregnant within the preceding year.

Study Objectives

---

1. To determine the extent of under-reporting of maternal mortality in Canada using the ICD-9 definition.

2. To determine the number of maternal deaths due to direct and indirect obstetric causes that occur between 42 and 365 days following a pregnancy outcome (late maternal deaths).

3. To determine cause-specific non-pregnancy-related death rates of women within 42 days of a pregnancy outcome (pregnancy-related) and within a year of a pregnancy outcome (pregnancy-associated). Cause-specific pregnancy-associated rates will be compared with cause-specific death rates among women not known to have been pregnant within the preceding year.

4. To explore approaches to defining and quantifying serious pregnancy-related morbidity nationally using the CIHI admission/separation records.

5. To document secular trends and/or regional differences in indicators of serious pregnancy-related morbidity using the CIHI admission/separation records.

Activities

---

A record linkage has been completed by Statistics Canada to address the first three objectives of this study. This has been done according to Statistics Canada's Record Linkage policy and the agreements with the provincial/territorial registrars of vital statistics using the Generalized Record Linkage System. Briefly, this process involved using the Canadian Mortality Database for the years 1987 through 1993 to create a file that included all deaths of women between the ages of 10 and 50 in those years. This was linked to the Canadian Birth Database for the years 1987 through 1992. In addition to the linkage of women's deaths to live births and stillbirths, a file was created containing records of deaths of women between the ages of 10 and 50 for which the cause of  death code was in the pregnancy chapter of ICD-9 but the record did not link to a live birth or a stillbirth.

A pilot phase of this study was completed early in 1997. This phase involved an initial assessment of three strategies to gather information to meet the study objectives. A record linkage of births in 1989 and 1990 to deaths of women of reproductive age in 1989, 1990 and 1991 was completed in 1996. Linkage strategies were subsequently refined, and a database and data entry system were developed.

The linkage has now been completed for the remaining years. Currently in progress is a review of all death certificates of women who can be linked to a live birth or a stillbirth and who died within a year of the pregnancy outcome, but whose cause-of-death code was not related to the pregnancy chapter of ICD-9, or to cancer or intentional/unintentional injury. Each of these cases is being reviewed by two physicians independently and judged as belonging to one of four categories: directly related to pregnancy, indirectly related to pregnancy, not related to pregnancy or not enough information given to make a judgment. A third physician will review cases about which there is disagreement.

Additionally, a pilot record linkage involving the CIHI database was completed for 1992. Its purpose was to determine whether this strategy could be used to ascertain additional pregnancy-related deaths that occurred in hospital following pregnancies that did not result in a live birth or a stillbirth.

An additional investigation of maternal deaths among First Nations women is being conducted in conjunction with the Medical Services Branch of Health Canada.  For this study, First Nations women were identified by linking the Indian Register File obtained from Indian and Northern Affairs Canada to the Maternal Mortality Study Group Database, the version of the record linkage file for the main study that incorporates updated information obtained from the provinces, physician reviews, and supplementary scans of the death certificates.

Maternal deaths among First Nations women will be described in a separate report.

Time Frame

---

Analyses to address the first three study objectives will be completed by the end of 1998. Subgroups have been formed to address specific morbidity issues. These groups will determine strategies of data gathering and analysis over the course of the coming year.

References

---

1. World Health Organization. International classification of diseases, 1975 revision. Geneva: World Health Organization, Vol 1, 1977:764.

2. World Health Organization. International statistical classification of diseases and related health problems, tenth revision. Geneva: World Health Organization, Vol 1, 1992:1238.

[Table of Contents] [Previous] [Next]