Spatiotemporal trends in maternal deaths and mortality surveillance

By Pooja Toshniwal PahariaReviewed by Danielle Ellis, B.Sc.Oct 9 2024

Findings suggest misclassified maternal deaths and overestimated maternal mortality in some US states.

A recent study in The Lancet Regional Health Americas examined spatiotemporal patterns in maternal mortality ratios (MMRs) across the United States (US).

Background

The US maternal death rate has been a source of contention owing to rising rates, discrepancies among races and ethnicities, and state-level variations in MMRs. The National Vital Statistics System forecasts a threefold rise from 1999 to 2021, raising worries about the safety of pregnancy and delivery. However, a rising body of literature undermines this assumption and raises concerns about maternal mortality surveillance.

The pregnant checkbox on death certificates may misclassify maternal fatalities. The National Vital Statistics System has probably reported exaggerated MMRs. This contradicting information has created doubt, leaving it unclear if the CDC's stated high and increasing MMRs are accurate or simply surveillance artifacts.

About the study

In the present study, researchers studied US trends in specific-cause maternal mortality ratios to understand the precision of reported MMRs.

The researchers analyzed maternal fatalities in the Centers for Disease Control and Prevention (CDC) records between 1999 and 2021. They quantified state- and national-level changes in MMR values due to specific and potentially incidental causes. They also quantified MMR changes in California (low MMR) and Texas (high MMR). Between 2018 and 2021, MMR values (for every 100,000 living individuals) below 20 denoted low MMR, and those ≥26 denoted high MMR.

Fatalities analyzed included those resulting from specific causes and unclear or potentially incidental ones. Specific causes included pre-eclampsia, placental disorders, eclampsia, antepartum hemorrhage, obstructed labor, uterine rupture/other obstetric trauma, postpartum hemorrhage, puerperal sepsis, and amniotic fluid embolism. Less-specific causes were chronic hypertension, cardiovascular disease, malignant tumors, other pregnancy-associated conditions, and erroneous pregnancy checkbox ticking.

The CDC's Wide-ranging Online Data for Epidemiologic Research (WONDER) databases provided state-level mortality data. Researchers ascertained maternal deaths according to the National Center for Health Statistics and the National Vital Statistics System. Maternal deaths included those with a pregnancy-related cause. The researchers used the tenth revision of the International Classification of Diseases (ICD-10) codes to determine the cause. They compared MMR values by race and ethnicity. In the sensitivity analyses, the researchers analyzed direct obstetric fatalities in states with low MMR versus high MMR and California vs. Texas. They also used alternative causes for potentially incidental deaths.

Results

The study included 1,543 maternal fatalities from 1999 to 2002 and 3,478 maternal deaths from 2018 to 2021. Between 1999-2002 and 2018-2021, MMRs increased by 144%. Temporal increases in MMRs resulted from small decreases in mortality from specified obstetric reasons and huge increases from less-specific causes, such as malignant tumors and chronic hypertension.

MMR values rose from 9.6 in 1999 and 2002 to 24 between 2018 and 2021. The rise in MMR was less in states with low MMR values (from 7.8 to 14) than in those with high MMRs (from 11 to 31). MMRs in California were 9.0 from 1999 to 2002 and 10 from 2018 to 2021. The MMRs in Texas for the comparable times were 8.7 and 28, respectively.

The team found similarly lowered MMRs for particular obstetric causes in states with low and high MMR values. The rise in MMR values from potentially incidental or unclear causes was less in states with low MMRs than those with high MMRs [MMR ratio (RR) of 5.6 versus 7.1] and for California versus Texas (RR of 1.7 versus 11). The team discovered a lesser difference in malignant tumor-related MMRs in California than in Texas (RR of 1.2 versus 91).

MMRs increased threefold in non-Hispanic Whites between 1999-2002 and 2018-2021 and twofold among non-Hispanic Blacks and Hispanic females. Women of all four racial/ethnic groups had similar relative rises in MMR values from unclear causes, while non-Hispanic Blacks had a much higher rise.

Direct obstetric fatalities decreased similarly, although not significantly, in states with low versus high MMRs and California versus Texas. Using alternate causes of mortality, researchers found comparable, significant temporal elevations in states with low MMR versus high MMRs and California versus Texas.

Conclusions

Based on the findings, the spatiotemporal trends of specific-cause MMRs indicate that maternal fatalities were misclassified and exaggerated due to the pregnancy checkbox in a few US states.

The similar decreases in MMR values from specific obstetric reasons and differential elevations in the MMR values from potentially incidental reasons, malignant tumors, and chronic hypertension in states with low versus high MMRs and California versus Texas indicate significantly misclassified potentially incidental and non-maternal fatalities in a few states.

The findings suggest that states with more frequent use of the gestation checkbox without proper verification reported more maternal fatalities from unclear causes. The findings call for a comprehensive re-evaluation of the pregnant checkbox on death certificates to get more exact estimates.