Methods for identifying pregnancy-associated deaths: population-based data from Finland 1987-2000

Area of Science:

• Public Health
• Epidemiology
• Health Informatics

Background:

• Accurate identification of pregnancy-associated deaths is vital for maternal mortality surveillance.
• Traditional death certificate data alone may underestimate maternal and pregnancy-related mortality.
• National health care registers offer a potential solution for comprehensive data linkage.

Purpose of the Study:

• To evaluate the effectiveness of linking national health care registers with death certificate data for identifying pregnancy-associated deaths.
• To assess the impact of data linkages on the completeness of maternal death ascertainment.
• To determine the proportion of pregnancy-associated deaths missed by death certificates alone.

Main Methods:

• Utilized data from Finland (1987-2000) for women aged 15-49 years.
• Linked Cause-of-Death Register data with Medical Birth Register, Register on Induced Abortions, and Hospital Discharge Register.
• Reviewed death certificates of identified women to confirm pregnancy status and coding.

Main Results:

• Data linkages identified 405 out of 419 (96.7%) pregnancy-associated deaths.
• Without data linkages, 73% of all pregnancy-associated deaths would have been missed.
• The underestimation was even higher for deaths following induced and spontaneous abortions.

Conclusions:

• Data linkages to national health care registers significantly enhance the identification of maternal and pregnancy-associated deaths compared to death certificates alone.
• Including data on pregnancies not ending in live births in linkages is recommended for more accurate mortality statistics.
• This approach provides a more complete picture of maternal mortality for public health initiatives.