In both the United States and Europe, the classification of pregnancy-related mortality using the International Classification of Diseases (ICD) cause-of-death codes leads to a significant underreporting, researchers report in the journal Obstetrics and Gynecology.

“To define effective strategies to reduce maternal mortality, it is crucial to accurately determine both the level and the causes of mortality,” said Lead investigator Dr. Catherine Deneux-Tharaux.

“Using a standardized enhanced method,” she added, “we found higher rates of maternal mortality than those officially reported in all study regions, and profiles of causes of mortality that greatly differed between regions.”

Deneux-Tharaux of Maternite Hopital Tenon, Paris and colleagues examined the details of 404 pregnancy-associated deaths that had taken place in France, Finland and in the states of Massachusetts and North Carolina in the U.S. The “enhanced” method they used to identify these cases “included the use of all data available from the death certificate as well as computerized linkage of births and deaths registers.”

Using the ICD cause-of-death codes alone led to underreporting that varied from 22 percent in France to 93 percent in Massachusetts. The researchers also found that inconsistencies made even these comparisons difficult.

Leading causes varied by region with cardiovascular conditions being most common in Massachusetts. Non-cardiovascular problems were the leading mortality cause in Finland, while hemorrhage was the most common cause in France and North Carolina.

Given the limitations of the ICD approach, Deneux-Tharaux called for “using linkage of vital records in the routine surveillance of maternal mortality, and defining customized strategies to understand why a cause is more prevalent than another in a given region.”

In an accompanying editorial, Dr. Vern L. Katz agrees and points out that “there are many valuable lessons from this report.”

SOURCE: Obstetrics and Gynecology, October 2005.