Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Nov 1.
Published in final edited form as: Hypertension. 2021 Sep 13;78(5):1414–1422. doi: 10.1161/HYPERTENSIONAHA.121.17661

Historical and Recent Changes in Maternal Mortality Due to Hypertensive Disorders in the United States, 1979 to 2018

Cande V Ananth 1, Justin S Brandt 1, Jennifer Hill 1, Hillary L Graham 1, Sonal Grover 1, Meike Schuster 1, Haylea S Patrick 1, KS Joseph 1
PMCID: PMC9218546  NIHMSID: NIHMS1814393  PMID: 34510912

Abstract

We evaluated the contributions of maternal age, year of death (period), and year of birth (cohort) on trends in hypertension-related maternal deaths in the United States. We undertook a sequential time series analysis of 155 710 441 live births and 3287 hypertension-related maternal deaths in the United States, 1979 to 2018. Trends in pregnancy-related mortality rate (maternal mortality rate [MMR]) due to chronic hypertension, gestational hypertension, and preeclampsia/eclampsia, were examined. MMR was defined as death during pregnancy or within 42 days postpartum due to hypertension. Trends in overall and race-specific hypertension-related MMR based on age, period, and birth cohort were evaluated based on weighted Poisson models. Trends were also adjusted for secular changes in obesity rates and corrected for potential death misclassification. During the 40-year period, the overall hypertension-related MMR was 2.1 per 100 000 live births, with MMR being almost 4-fold higher among Black compared with White women (5.4 [n=1396] versus 1.4 [n=1747] per 100 000 live births). Advancing age was associated with a sharp increase in MMR at ≥15 years among Black women and at ≥25 years among White women. Birth cohort was also associated with increasing MMR. Preeclampsia/eclampsia-related MMR declined annually by 2.6% (95% CI, 2.2–2.9), but chronic hypertension–related MMR increased annually by 9.2% (95% CI, 7.9–10.6). The decline in MMR was attenuated when adjusted for increasing obesity rates. The temporal burden of hypertension-related MMR in the United States has increased substantially for chronic hypertension–associated MMR and decreased for preeclampsia/eclampsia-associated MMR. Nevertheless, deaths from hypertension continue to contribute substantially to maternal deaths.

Keywords: continental population groups, eclampsia, live birth, maternal mortality, preeclampsia

The World Health Organization1 reported that maternal mortality rates (MMRs) in the United States increased from 12 to 28 per 100 000 live births between 1990 and 2013. However, recent reports from the National Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention show that this reported increase was an artifact of maternal mortality surveillance.2–4 The NCHS reports, based on careful recoding and reanalysis of death certificates, showed that MMR in the United States increased following the introduction of a pregnancy checkbox on death certificates in 2003 and its subsequent staggered adoption by the states. Another notable finding was that the effect of the checkbox was cause-of-death dependent; maternal deaths due to nonspecific causes increased substantially, while those due to specific pregnancy-related causes, such as hypertensive disorders, edema, and proteinuria, were unimpacted.2–4

Worldwide, hypertensive complications during pregnancy are implicated in over 50 000 maternal deaths annually.5 In the United States, about 8% to 10% of all pregnancies are complicated by some form of hypertension,6 with severe preeclampsia7 and chronic hypertension8 showing a steady temporal increase over the last 4 decades. Deconstructing the changing profile of maternal deaths due to hypertensive disorders in relation to biological determinants and environmental factors requires an examination of historical trends in an age-period-cohort (APC) framework. MMRs are age dependent, and the steep increase in mortality with advancing age at childbirth has implications for populations where women increasingly delay child-bearing.9 Age effects may be impacted by extraneous factors including specific medical therapies, and mortality rates at a given age can change substantially over time. Such secular variation in age-specific mortality (a period effect) may also be influenced by ascertainment and classification of maternal deaths. Birth cohorts (ie, cohorts defined on the year of birth of women) may be characterized by cumulative exposures to social conditions, lifestyle (eg, obesity), and behavioral factors (eg, smoking or drug use), and such determinants can also impact maternal mortality.

We undertook this study to disentangle the independent contributions of maternal age, year of death (period), and year of birth (birth cohort) on trends in hypertension-related MMR in the United States. We hypothesized that secular increases in obesity could explain, at least partly, any recent temporal increase in hypertension-related MMR.

METHODS

Data Sharing

The data utilized in this study are in the public domain in deidentified form and can be accessed at https://www.cdc.gov/nchs/data_access/vitalstatsonline.htm.

We designed a sequential time series analysis to examine temporal changes in MMR between 1979 and 2018 in the United States. The mortality data assembled by the NCHS includes all deaths in 50 states and Washington, DC, are derived from standard death certificates. We used data on all US live births (ascertained from birth certificates) between 1979 and 2018, which were also assembled by the NCHS and correspond to data on live births in the 50 states plus Washington, DC. Since these data were fully deidentified and publicly available, we did not require ethics approval. The study followed the The Strengthening the Reporting of Observational Studies inn Epidemiology reporting guidelines for cross-sectional studies.

Maternal Deaths

We restricted the study to women aged 15 to 49 years. We first identified deaths related to pregnancy and childbirth, captured in the primary cause-of-death field, and an additional 14 (1979–1998) or 20 (1999–2018) cause-of-death fields on the death certificate. These data were coded based on the International Classification of Diseases (ICD), Ninth (from 1979–1998) and Tenth (from 1999–2018) Revisions, pregnancy-related deaths were identified as those with codes 630–679 in ICD, Ninth Revision, and O chapter codes except O96 and O97 in ICD, Tenth Revision (Table S1 in the Data Supplement).10 We further restricted the analysis to deaths that had hypertensive conditions, specifically chronic hypertension, gestational hypertension, and preeclampsia/eclampsia, listed in the cause-of-death fields. These pregnancy-related maternal deaths and associated rates are hereafter referred to as hypertension-related MMR.

Statistical Analysis

Through an APC modeling approach, we examined changes in hypertension-related MMR in the United States between 1979 and 2018 overall and separately among Black and White women. We began by first examining graphical plots of MMR: (1) MMR examined by age (observations within each period connected) depicting cross-sectional age-specific rates; (2) MMR versus age, observations within each birth cohort connected, depicting longitudinal age-specific rates; (3) MMR by period, observations within each age-class connected; and (4) MMR by cohort, observations within each age-class connected. These analyses helped guide the parametrization of the age, period, and cohorts in the APC framework. APC modeling was based on Poisson regression with MMR modeled as the number of maternal deaths cross-classified by age, period, and cohort (all in single years), with the number of live births as an offset (denominator) in the APC cross-classification.

To overcome the identifiability problem in an APC analysis (any two of the effects is sufficient to determine the third leading to collinearity11), we imposed a series of constraints. We first modeled MMR as a function of maternal age, followed by an overall linear trend in MMR (sum of the linear component of period and cohort effects); this estimate is referred to as the drift parameter (ie, annual percent change in MMR). We then assessed deviations from linearity that can be uniquely attributed to the period and cohort effects, referred to as the curvature effect (ie, if the effect of period or cohort on MMR is nonlinear). Models were fit sequentially beginning with age only and then adding the drift parameter and subsequently the period and cohort parameters. APC effects were modeled as smooth functions based on natural splines with 6 knots each.

Impact of Obesity on Maternal Mortality Trends

We examined how secular changes in rates of obesity may have influenced the APC effects on hypertension-related MMR. This was performed because obesity is a strong risk factor for maternal complications, including those related to hypertensive diseases, and the prevalence of obesity has been increasing.12 We extracted prevalence rates of obesity from the National Health and Nutrition Examination Survey data by period and birth cohorts, where available.13 We undertook a hierarchical APC analysis14 based on a weighted Poisson model to evaluate changes in MMR following adjustment for obesity rates. Yearly obesity rates and maternal age were modeled as categorical fixed effects; categorical period variables were included as random effect terms.

Sensitivity Analysis

Given the potential for misclassification of deaths with hypertensive disease as the cause, we undertook a probabilistic bias analysis to correct for differential outcome misclassification. This analysis was based on conservative estimates of sensitivity and specificity for maternal deaths: sensitivity ranging from 0.65 to 0.99 for the 2018 period and 0.80 to 0.99 for the 1980 reference period and specificity ranging from 0.98 to 1.00 for both periods. We replicated the analysis 100 000× for various combinations of sensitivity and specificity parameters based on a uniform distribution and reported the misclassification-adjusted median rate ratio (RR) and 95% CI.

The APC analysis was performed in R implemented in the Epi package15 in RStudio (version 1.2).16 Hierarchical APC models were fit in SAS (version 9.4; SAS Institute, Cary, NC). The probabilistic bias analysis was implemented in RStudio based on the episensr() package.17

RESULTS

During the 40-year period (1979–2018), there were 155 710 441 live births and 3287 (2.1 per 100 000 live births) hypertension-related maternal deaths. Hypertension-related MMRs were 5.4 (n=1396) and 1.4 (n=1747) per 100 000 live births among Black and White women, respectively. The mean (SD) maternal age was 3 years lower among women (27.0 [6.0] years) who did not die compared with women who died (30.0 [7.4] years). Annual hypertension-related MMRs were stable between 1979 and 2018, but MMRs were 2- to 4-fold higher among Black than White women (Figure S1). The numbers of pregnancy-related deaths with hypertension as the underlying cause and corresponding MMR by maternal age and period are shown in Table 1. Hypertension-related MMR showed a steady increase with advancing maternal age within each period, with the MMR being the highest among women aged 45 to 49 years.

Table 1.

Number of Live Births and Hypertension-Related MMRs (per 100 000 Live Births) by Maternal Age, Period, and Maternal Birth Cohorts: United States 1979 to 2018

Period of death Maternal age, y Central birth cohort
15–19 20–24 25–29 30–34 35–39 40–44 45–49
1087 (92.0) 1937
20 862 (9.6) 1167 (85.7) 1942
1 27 183 (7.1) 28 391 (7.0) 1 640 (61.0) 1947
502 051 (4.6) 214 660 (3.7) 48 665 (10.3) 2728 (36.6) 1952
1 017 915 (2.4) 697 298 (1.7) 317 959 (3.5) 67 314 (7.4) 4355 (23.0) 1957
1 125 164 (1.7) 1 202 858 (0.9) 887 1 12 (2.4) 384 128 (3.1) 90 124 (4.4) 6132 (16.3) 1962
506 591 (3.9) 1 142 457 (0.6) 1 278 669 (1.2) 905 511 (1.4) 452 575 (3.8) 104 815 (10.5) 7182 (69.6) 1967
1980 Inline graphic 467 761 (1.5) 1 094 950 (2.5) 1 064 615 (1.3) 930 499 (2.5) 484 004 (3.7) 107 266 (9.3) 8228 (121.5) 1972
1985 Inline graphic 522 204 (2.5) 966 306 (1.9) 1 089 179 (2.3) 952 787 (1.8) 465 948 (3.0) 1 12 232 (9.8) 1977
1990 Inline graphic 500 1 12 (0.6) 1 018 922 (1.3) 1 133 747 (2.1) 964 440 (2.5) 529 565 (4.2) 1982
1995 Inline graphic 469 423 (1.1) 1 041 847 (1.0) 1 136 047 (1.4) 1 097 846 (1.5) 1987
2000 Inline graphic 415 226 (1.2) 953 064 (1.3) 1 155 359 (1.8) 1992
2005 Inline graphic 368 165 (1.4) 852 149 (0.9) 1997
2010 Inline graphic 230 167 (0.9) 2002
2015 Inline graphic
Open in a new tab

Table entries denote the total number of live births (MMRs: expressed per 100 000 live births). Period of death refers to the year when women died, and central maternal birth cohort refers to the mid-year (in a 5-y interval) when women were born. Among women who died in 2015 (period) aged 15 to 19 y, the MMR was 0.9 per 100 000 live births. These women would have been born between 1995 and 2000 (central maternal birth cohort 2002). MMR indicates maternal mortality rate.

The separate (unadjusted) associations of age on hypertension-related MMR showed a sharp linear increase in hypertension-related MMR among Black women aged ≥15 years and a sharp increase among White women aged ≥25 years (Figure S2, left). Compared with rates in 1980 (reference period), the age-adjusted RR for hypertension-related MMR declined up to 1990s and plateaued thereafter (Figure S2, middle). Compared with the 1960 reference birth cohort, the RRs for hypertension-related MMR declined in subsequent birth cohorts although CIs were wide (Figure S2, right).

Age-Period-Cohort Models

The results of the APC model for hypertension-related MMR overall and separately among Black and White women are shown in Figure 1. The model showed that advancing age was strongly associated with increasing MMR at age ≥15 years and ≥25 years among Black women and White women, respectively. Age-specific rates were substantially higher in Black compared with White women (Table 2). Compared with 1980, the RRs for hypertension-related MMR decreased up to 1990 and stabilized thereafter; this pattern was similar among Black and White women. Maternal birth cohorts were also associated with MMR: compared with the 1960 reference cohort, MMRs increased marginally overall and among Black and White women in the 1970 birth cohort and declined thereafter. The overall fit of these models was satisfactory (Table S2).

Figure 1. Age-period-cohort trends in overall and race-specific maternal mortality with hypertensive conditions as the cause in the United States, 1979 to 2018.

Figure 1.

Open in a new tab

Maternal mortality rate (MMR; per 100 000 live births) is shown in relation to maternal age on the y axis on the left. The MMR ratio is shown in relation to maternal birth cohort (with the 1960 maternal birth year as the reference) and period (with 1980 as the reference) on the right axis. The shaded bands denote 95% CI. The yearly change in MMR was −0.3% (95% CI, −0.6 to −0.0) for all women (solid line), −0.4% (95% CI, −0.9 to −0.1) for White women (dashed line), and 0.0% (95% CI, −0.5 to 0.5) for Black women (dotted line). Advancing maternal age was strongly associated with increasing MMR after age 25 y, and age-specific rates were substantially higher in Black compared with White women. Compared with 1980, the RRs for hypertension-related MMR showed that the MMR declined up to 1990 and stabilized thereafter. The MMR increased marginally and among White and Black women in the 1970 birth cohort and declined thereafter. RR indicates rate ratio.

Table 2.

Hypertension-Related MMRs (per 100 000 Live Births) for Selected Maternal Ages, Periods, and Maternal Birth Cohorts: United States 1979 to 2018

All women White women Black women
Maternal age, y MMR (95% CI) per 100 000 live births
 15 1.5 (1.1–1.9) 1.1 (0.8–1.6) 2.1 (1.4–3.1)
 20 1.3 (1.1–1.6) 1.0 (0.8–1.2) 2.6 (2.0–3.4)
 25 1.3 (1.1–1.5) 0.9 (0.7–1.1) 3.4 (2.6–4.5)
 30 1.7 (1.4–2.0) 1.2 (0.9–1.5) 6.3 (4.8–8.4)
 35 2.3 (2.0–2.8) 1.7 (1.3–2.1) 8.1 (6.1–10.9)
 40 6.3 (5.2–7.6) 4.6 (3.6–5.9) 20.6 (15.4–27.7)
 45 22.7 (17.9–28.9) 17.4 (12.6–24.0) 69.3 (47.8–100.4)
 49 63.3 (45.8–87.6) 50.2 (32.3–78.1) 182.4 (111.6–298.3)
Year of death Adjusted rate ratio (95% CI) of MMR
 1980 1.00 (reference) 1.00 (reference) 1.00 (reference)
 1990 0.70 (0.58–0.84) 0.69 (0.53–0.89) 0.65 (0.48–0.88)
 2000 0.78 (0.66–0.92) 0.78 (0.63–0.98) 0.76 (0.58–0.99)
 2010 0.80 (0.68–0.95) 0.75 (0.60–0.94) 0.83 (0.64–1.09)
 2018 0.86 (0.70–1.05) 0.83 (0.63–1.10) 0.89 (0.65–1.22)
Birth cohort, y Adjusted rate ratio (95% CI) of MMR
 1930 1.33 (0.84–2.10) 1.27 (0.68–2.35) 1.08 (0.54–2.16)
 1940 1.19 (0.89–1.60) 1.17 (0.79–1.73) 1.04 (0.66–1.63)
 1950 1.07 (0.94–1.22) 1.07 (0.90–1.27) 0.99 (0.81–1.23)
 1960 1.00 (reference) 1.00 (reference) 1.00 (reference)
 1970 1.28 (1.13–1.46) 1.22 (1.03–1.45) 1.47 (1.19–1.83)
 1980 1.17 (1.05–1.31) 1.13 (0.97–1.32) 1.20 (1.01–1.42)
 1990 0.90 (0.80–1.01) 0.89 (0.74–1.05) 0.88 (0.75–1.04)
 2000 0.69 (0.52–0.91) 0.69 (0.46–1.04) 0.63 (0.41–0.97)
Open in a new tab

The MMRs by age are for the reference period 1980 and are not adjusted for other factors. The rate ratios for period and birth cohorts are adjusted for the effects of maternal age. MMR indicates maternal mortality rate; and y, years.

The APC analysis for MMR due to preeclampsia/eclampsia showed a strong age effect among women aged ≥25 years (Figure 2). Compared with 1980, MMR associated with preeclampsia/eclampsia showed a steady temporal decline, with an average annual decline in MMR of 2.6% (95% CI, 2.0–3.2) and 2.5% (95% CI, 2.0–2.9) among Black and White women, respectively. No appreciable cohort effects were evident.

Figure 2. Age-period-cohort trends in overall and race-specific maternal mortality with preeclampsia/eclampsia as the cause in the United States, 1979 to 2018.

Figure 2.

Open in a new tab

Maternal mortality rate (MMR; per 100 000 live births) is shown in relation to maternal age on the y axis on the left. The MMR ratio is shown in relation to maternal birth cohort (with the 1960 maternal birth year as the reference) and period (with 1980 as the reference) on the right axis. The shaded bands denote the 95% CI. The yearly change in MMR was −2.6% (95% CI, −2.9 to −2.2) for all women (solid line), −2.5% (95% CI, −2.9 to −2.0) for White women (dashed line), and −2.6% (95% CI, −3.2 to −2.0) for Black women (dotted line). The APC analysis shows a strong linear increase in preeclampsia/eclampsia-related MMR among Black women aged ≥30 y and among White women aged ≥35 y. Compared with MMR in 1980, RRs for mortality declined, on average, by 2.5% and 2.6% per year among White and Black women, respectively. No appreciable birth cohort effects were evident. RR indicates rate ratio.

Changes in MMRs associated with chronic hypertension showed a sharp linear increase in MMR with advancing maternal age among both Black and White women (Figure 3). While no cohort effects were seen, the MMR showed a sharp temporal increase: MMR increased annually by 9.1% (95% CI, 7.1–11.1) among Black women and by 9.4% (95% CI, 7.4–11.4) among White women.

Figure 3. Age-period-cohort trends in overall and race-specific maternal mortality with chronic hypertension as the cause in the United States, 1979 to 2018.

Figure 3.

Open in a new tab

Maternal mortality rate (MMR; per 100 000 live births) is shown in relation to maternal age on the y axis on the left. The MMR ratio is shown in relation to maternal birth cohort (with the 1960 maternal birth year as the reference) and period (with 1980 as the reference) on the right axis. The shaded bands denote 95% CI. The yearly change in MMR was 9.2% (95% CI, 7.9–10.6) for all women (solid line), 9.4% (95% CI, 7.4–11.4) for White women (dashed line), and 9.1% (95% CI, 7.1–11.1) for Black women (dotted line). Chronic hypertension–related MMR showed a sharp linear increase with advancing maternal age, with MMRs being higher among Black women compared with White women. Compared with 1980, MMR increased by 9.4% and 9.1% per year among White and Black women, respectively. These analyses showed no appreciable effects due to maternal birth cohorts. RR indicates rate ratio.

Multilevel models that included cohort-specific obesity rates did not converge and were dropped from further consideration. Multilevel models showed that period changes in hypertension-related MMR remained unaffected by period-specific increase in obesity rates (RR, 1.00 [95% CI, 0.99–1.01]; Figure S3). However, the period change in preeclampsia/eclampsia-related MMR (RR, 0.96 [95% CI, 0.95–0.97]) was attenuated after adjustment for period-specific increase in obesity rates MMR (Figure S4). In other words, the observed period effect showing a 66% decline in preeclampsia/eclampsia-related MMR between 1980 and 2018 (RR, 0.34 [95% CI, 0.28–0.42]; Table S3) was reduced to a 4% decline (RR, 0.96 [95% CI, 0.95–0.97]) after adjusting for increasing obesity rates over the same period. Models for chronic hypertension–related MMR did not converge.

Sensitivity Analysis for Misclassification Bias

The probabilistic bias analysis to evaluate potential mis-classification of maternal deaths indicated that the overall period-specific RRs for MMR in 2018 compared with 1980 were 0.73 (95% CI, 0.18–3.21) for any hypertension and 0.14 (95% CI, 0.03–0.27) for preeclampsia/eclampsia. For chronic hypertension–related MMRs, the RR for chronic hypertension–specific MMRs comparing 2018 with 1980 (reference) was 20.66 (95% CI, 16.31–25.00) following correction for misclassification.

DISCUSSION

This population-based study, which included all hypertension-related pregnancy–related deaths in the United States over 4 decades, yielded 4 important findings. First, advancing maternal age is associated with a sharp increase in hypertension-related MMR. While effects due to birth cohorts were largely absent, the MMR associated with preeclampsia/eclampsia showed a temporal decline, while that due to chronic hypertension increased. Second, a strong racial disparity in MMR between Black and White women was evident. Third, adjustments for secular changes in rates of obesity appear to have played an important role in shaping the recent temporal changes in hypertension-related MMR, particularly the preeclampsia/eclampsia-related MMR. Fourth, correction for outcome misclassification resulted in stronger period changes in hypertension-related MMR.

Studies have shown temporal increases in the prevalence of chronic hypertension8 and severe preeclampsia7 in the United States. Although overall hypertension-related MMRs have been stable during the 40-year study period in this study, chronic hypertension–related MMRs have increased while preeclampsia/eclampsia-associated MMRs have declined. Risk factors for chronic hypertension include advanced maternal age and obesity. As women in the United States have deferred childbirth to a later age,18 the frequency of pregnancies among women with advanced maternal age, especially women ≥40 years of age, has increased.19 Advanced age, in turn, is associated with increased risk of chronic hypertension and is an established risk factor for maternal deaths.20,21

Obesity rates have surged, reaching the status of an epidemic,22,23 and is associated with adverse outcomes in all stages of pregnancy,24–26 from conception to the puerperium. In a meta-analysis, approximately one-quarter of pregnancy complications were attributable to maternal overweight/obesity.27 It is likely that obesity interacts with advanced maternal age, thereby increasing the prevalence of hypertensive complications.

While the rates of both chronic hypertension8 and preeclampsia/eclampsia7 have increased in the United States, there has been a concurrent reduction in MMR due to preeclampsia/eclampsia. The reasons for improving preeclampsia/eclampsia-related MMR are many but likely reflect improved management of blood pressure, use of magnesium sulphate for severe preeclampsia and eclampsia, early recognition of Hemolysis, Elevated Liver enzymes and Low Platelets syndrome, and advances in critical care.28,29 Preeclampsia/eclampsia management is also associated with increasing rates of planned delivery,30 which allows for medical optimization and delivery before progression of disease to end organ damage. The steady decline in preeclampsia/eclampsia-related MMR between 1980 and 2018 may reflect large improvements in the management of women with these serious and life-threatening pregnancy complications. The increasing use of low-dose aspirin to prevent preeclampsia31 among women with risk factors for preeclampsia (chronic hypertension, advanced maternal age, and a body mass index ≥30 kg/m2) is likely to further accelerate this welcome trend.

In contrast, the optimal management of chronic hypertension in pregnancy remains uncertain. Current strategies involve baseline assessment for subclinical disease, such as renal and cardiac assessments in the first trimester, as well as monitoring for and early recognition of associated complications, such as fetal growth restriction and superimposed preeclampsia.32 Optimal antepartum blood pressure management remains uncertain as experts continue to debate the risks and benefits of more aggressive antihypertensive management among patients with mild disease.33 The results of the study underscore the need for improved strategies to actively manage chronic hypertension rather than responding to its profound sequelae.

The evolving changes in the diagnostic criteria for pre-eclampsia and chronic hypertension, as well as changes in the obstetric management for preeclampsia,34–38 including decisions to intervene to end the pregnancy to optimize maternal-fetal health, impacted trends in MMR. The finding that adjustment for obesity attenuated the period effect on preeclampsia/eclampsia MMR was unexpected and deserves further scrutiny. It is possible that temporal increases in the early delivery of obese women with preeclampsia played in a role in reducing maternal deaths from preeclampsia/eclampsia.

The disparate trends in chronic hypertension and preeclampsia/eclampsia-related MMRs are noteworthy. Our study showed a 3-fold decline in preeclampsia/eclampsia-related MMR over the 40-year period and a 15-fold increase in MMR associated with chronic hypertension. One reason that may partly explain the large temporal increase in maternal deaths from chronic hypertension is that nonmaternal deaths from chronic hypertension were increasingly being misclassified as maternal deaths due to pregnancy checkbox errors.38 False positive pregnancy checkbox errors were identified as the cause for the artifactual increase in reported maternal mortality in the United States in recent years.2–4 A small misclassification in nonmaternal deaths from chronic hypertension could explain the large rise chronic hypertension–related MMR given the relative rarity of maternal deaths from chronic hypertension.38

There is growing awareness about the magnitude of medical disparities leading to disproportionate risk of severe maternal morbidity and mortality in the United States.39–41 This study reinforces the importance of this issue, showing substantially higher hypertension-related MMR among Black compared with White women. This disparity may reflect social inequities, poor access to care, and other manifestations of systemic racism in the American health care system. The stark disparity in MMR rates by race needs urgent attention.

It is estimated that two-third (63%) of maternal deaths are preventable.39,42 There have been great efforts to reduce MMR in the United States, including standardizing prenatal care with quality improvement initiatives and implementation of checklists and bundles43–46; our study suggests that we are making progress. The overall MMR in the United States in 2018 was 17.4 per 100 000 live births,2,47 and our study shows that hypertension-related MMR contributed 2.2 maternal deaths per 100 000 live births to this figure. Despite the United Nations Millennium Development Goals to reduce MMR between 1990 and 2015 by 75%, MMR in the United States was essentially unchanged between 1999 and 2017.4 While the management of preeclampsia/eclampsia has improved substantially, this study also underscores the need to better identify and treat women with chronic hypertension and to target associated risk factors.

Strengths and Limitations of the Study

This APC analysis provides insights that address the extent to which each of these time-related factors influenced temporal changes in MMR. With almost 156 million live births and 3300 maternal deaths with hypertension as the cause, this is perhaps one of the largest studies to date on temporal changes in MMR over the last 4 decades in the United States. Adjustment of MMR trends for period-specific increase in rates of obesity, which appeared to account for the declining preeclampsia/eclampsia-associated MMR, is important. This finding suggests that the steady decline in preeclampsia/eclampsia-related MMR may be, at least partly, related to increasing rates of obesity observed during the study period.

A few important limitations of the study merit discussion. Although recent NCHS reports demonstrate that the pregnancy checkbox resulted in a gross misclassification of maternal deaths, the pregnancy check box did not affect hypertension as a specific cause of death.2–4 Another limitation includes transition between ICD, Ninth and Tenth Revision, coding as well as changes in the diagnostic criteria for preeclampsia. The extent to which the APC effects interact with each other remains unknown. Finally, the National Health and Nutrition Examination Survey data only sample women of Black, White, or Mexican American race/ethnicity, so there is some uncertainty regarding the generalizability of findings that pertain to adjustment for obesity trends. These limitations, however, are unlikely to account for the powerful trends in MMRs associated with hypertensive disorders.

Conclusions

Over the last 4 decades in the United States, deaths due to chronic hypertension increased, whereas deaths due to preeclampsia/eclampsia decreased. The trends in MMR highlight the profound effect of maternal age and obesity in driving the increasing prevalence of chronic hypertension. While management of the most severe complications of preeclampsia/eclampsia has substantially improved, strategies are needed to reduce overall hypertension-related MMRs by targeting improvements in the women’s lifecourse,48,49 such as advanced maternal age, obesity, and racial inequities in care. These issues deserve urgent public health attention.

Supplementary Material

Supplemental data

Novelty and Significance.

What Is New?

  • During the 40-year period (1979–2018), the overall hypertension-related maternal mortality rate (MMR) was 2.1 per 100 000 live births.

  • Yearly hypertension-related MMRs were stable between 1979 and 2018, with MMRs being almost 4-fold higher among Black compared with White women.

  • There was a strong age effect with a sharp increase in MMR with advancing age. Maternal birth cohorts were also associated with MMR as women born after 1990 had higher MMR compared with those born in 1960.

  • Preeclampsia/eclampsia-related MMRs showed a temporal decline while those due to chronic hypertension increased dramatically. These patterns were similar for Black and White women, although risks were higher for Black women.

  • Adjustments for the secular increase in obesity rates appear to attenuate the temporal decline in preeclampsia/eclampsia-related MMR.

What Is Relevant?

  • The temporal burden of hypertension-related MMR in the United States has increased substantially for chronic hypertension–associated MMR and decreased for preeclampsia/eclampsia-associated MMR.

  • Deaths from preeclampsia/eclampsia continue to contribute significantly to pregnancy-related mortality.

Summary

The recent decline in MMR in the United States was deemed an artifact of increased surveillance. This study shows that the temporal burden of hypertension-related MMRs in the United States has increased substantially for chronic hypertension–associated MMR and decreased for preeclampsia/eclampsia-associated MMR, although deaths from preeclampsia/eclampsia continue to contribute substantially to pregnancy-related deaths. While we are better at managing the most severe complications of preeclampsia/eclampsia, strategies to reduce overall hypertension-related MMR should target improvements in the women’s lifecourse, such as advanced maternal age, obesity, and racial inequities in care.

Perspectives.

This large, population-based epidemiological study examines temporal trends in hypertension-related MMRs over 4 decades in the United States. Maternal deaths due to chronic hypertension substantially increased while deaths due to preeclampsia/eclampsia decreased. The study critically underscores the need (1) to develop targeted prenatal interventions, including tight blood pressure control and efforts to reduce body mass index, to ameliorate rates of hypertensive conditions before and during pregnancy and (2) to address the prevailing and concerning race disparity in maternal deaths with hypertension as the cause. The increase in maternal deaths due to chronic hypertension is concerning and deserves careful and immediate attention.

Sources of Funding

C.V. Ananth and H. Graham are supported, in part, by a grant (R01-HL150065) from the National Heart, Lung, and Blood Institute (NHLBI), National Institutes of Health, awarded to C.V. Ananth. The work of K.S. Joseph is supported by the BC Children’s Hospital Research Institute.

Nonstandard Abbreviations and Acronyms

APC

age-period-cohort

ICD

International Classification of Diseases

MMR

maternal mortality rate

NCHS

National Center for Health Statistics

RR

rate ratio

Footnotes

Presented at the 40th Annual Pregnancy Meeting of the Society for Maternal-Fetal Medicine, Grapevine, TX, February 5–10, 2020.

Disclosures

None.

Hypertension is available at www.ahajournals.org/journal/hyp

REFERENCES

  • 1.WHO, UNICEF, UNFPA, World Bank Group and the United Nations Population Division. Maternal Mortality: Levels and Trends 2000 to 2017. 2019. Accessed 23, October 2020. https://www.who.int/reproductivehealth/publications/maternal-mortality-2000-2017/en/
  • 2.Hoyert DL, Miniño AM. Maternal mortality in the United States: changes in coding, publication, and data release, 2018. Natl Vital Stat Rep. 2020;69:1–18. [PubMed] [Google Scholar]
  • 3.Hoyert DL, Uddin SFG, Minino AM. Evaluation of the pregnancy status checkbox on the identification of maternal deaths. Natl Vital Stat Rep. 2020;69:1–25 [PubMed] [Google Scholar]
  • 4.Rossen LM, Womack LS, Hoyert DL, Anderson RN, Uddin SFG. The impact of the pregnancy checkbox and misclassification on maternal mortality trends in the united states, 1999–2017. Vital & health statistics. Series 3, Analytical and epidemiological studies. U.S. Dept. of Health and Human Services, Public Health Service, National Center for Health Statistics. 2020:1–61. [PubMed] [Google Scholar]
  • 5.Ghulmiyyah L, Sibai B. Maternal mortality from preeclampsia/eclampsia. Semin Perinatol. 2012;36:56–59. doi: 10.1053/j.semperi.2011.09.011 [DOI] [PubMed] [Google Scholar]
  • 6.Hutcheon JA, Lisonkova S, Joseph KS. Epidemiology of pre-eclampsia and the other hypertensive disorders of pregnancy. Best Pract Res Clin Obstet Gynaecol. 2011;25:391–403. doi: 10.1016/j.bpobgyn.2011.01.006 [DOI] [PubMed] [Google Scholar]
  • 7.Ananth CV, Keyes KM, Wapner RJ. Pre-eclampsia rates in the united states, 1980-2010: age-period-cohort analysis. BMJ. 2013;347:f6564. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Ananth CV, Duzyj CM, Yadava S, Schwebel M, Tita ATN, Joseph KS. Changes in the prevalence of chronic hypertension in pregnancy, United States, 1970 to 2010. Hypertension. 2019;74:1089–1095. doi: 10.1161/HYPERTENSIONAHA.119.12968 [DOI] [PubMed] [Google Scholar]
  • 9.Davis NL, Hoyert DL, Goodman DA, Hirai AH, Callaghan WM. Contribution of maternal age and pregnancy checkbox on maternal mortality ratios in the United States, 1978-2012. Am J Obstet Gynecol. 2017;217:352 e351–352 e357. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Centers for Disease Control and Prevention. Pregnancy mortality surveillance system. Accessed 16 October 2019. https://www.cdc.gov/reproductivehealth/maternal-mortality/pregnancy-mortality-surveillance-system.htm
  • 11.Holford TR. Analysing the temporal effects of age, period and cohort. Stat Methods Med Res. 1992;1:317–337. doi: 10.1177/096228029200100306 [DOI] [PubMed] [Google Scholar]
  • 12.Nelson DB, Moniz MH, Davis MM. Population-level factors associated with maternal mortality in the United States, 1997-2012. BMC Public Health. 2018;18:1007. doi: 10.1186/s12889-018-5935-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey. 2019. Accessed July 7, 2018. https://www.cdc.gov/nchs/nhanes/index.htm
  • 14.Yang Y, Land K Chapter 7: mixed effects models: hierarchical APC-cross-classified random effects models (HAPC-CCREM), part 1: the basics. In: Yang Y, Land K, eds. Age-period-cohort analysis: new models, methods, and empirical applications. Chapman & Hall/CRC Interdisciplinary Statistics; 2013. [Google Scholar]
  • 15.Carstensen B, Plummer M, Laara E, Hills M. Epi: A package for statistical analysis in epidemiology (R package version 1.1.34). 2012. August 12, 2015. https://cran.r-project.org/web/packages/Epi/Epi.pdf
  • 16.RStudio Team. Rstudio: Integrated development for R. RStudio Inc., Boston, MA. 2015. Accessed November 6, 2019. http://www.rstudio.com/ [Google Scholar]
  • 17.Haine D Basic sensitivity analysis of epidemiological results. 2018. Accessed November 6, 2019. https://cran.rproject.org/web/packages/episensr/episensr.pdf
  • 18.Mathews TJ, Hamilton BE. Mean age of mother, 1970-2000. Natl Vital Stat Rep. 2002;51:1–13. [PubMed] [Google Scholar]
  • 19.MacDorman MF, Declercq E, Thoma ME. Trends in maternal mortality by sociodemographic characteristics and cause of death in 27 states and the District of Columbia. Obstet Gynecol. 2017;129:811–818. doi: 10.1097/AOG.0000000000001968 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Fitzpatrick KE, Tuffnell D, Kurinczuk JJ, Knight M. Pregnancy at very advanced maternal age: a UK population-based cohort study. BJOG. 2017;124:1097–1106. doi: 10.1111/1471-0528.14269 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Waldenström U, Cnattingius S, Vixner L, Norman M. Advanced maternal age increases the risk of very preterm birth, irrespective of parity: a population-based register study. BJOG. 2017;124:1235–1244. doi: 10.1111/1471-0528.14368 [DOI] [PubMed] [Google Scholar]
  • 22.Mitchell NS, Catenacci VA, Wyatt HR, Hill JO. Obesity: overview of an epidemic. Psychiatr Clin North Am. 2011;34:717–732. doi: 10.1016/j.psc.2011.08.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Hales CM, Fryar CD, Carroll MD, Freedman DS, Ogden CL. Trends in obesity and severe obesity prevalence in US youth and adults by sex and age, 2007-2008 to 2015-2016. JAMA. 2018;319:1723–1725. doi: 10.1001/jama.2018.3060 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Galtier-Dereure F, Boegner C, Bringer J. Obesity and pregnancy: complications and cost. Am J Clin Nutr. 2000;71(5 suppl):1242S–1248S. doi: 10.1093/ajcn/71.5.1242s [DOI] [PubMed] [Google Scholar]
  • 25.Lisonkova S, Muraca GM, Potts J, Liauw J, Chan WS, Skoll A, Lim KI. Association between prepregnancy body mass index and severe maternal morbidity. JAMA. 2017;318:1777–1786. doi: 10.1001/jama.2017.16191 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Caughey AB. Prepregnancy obesity and severe maternal morbidity: what can be done? JAMA 2017;318:1765–1766. doi: 10.1001/jama.2017.16189 [DOI] [PubMed] [Google Scholar]
  • 27.Santos S, Voerman E, Amiano P, Barros H, Beilin LJ, Bergström A, Charles MA, Chatzi L, Chevrier C, Chrousos GP, et al. Impact of maternal body mass index and gestational weight gain on pregnancy complications: an individual participant data meta-analysis of European, North American and Australian cohorts. BJOG. 2019;126:984–995. doi: 10.1111/1471-0528.15661 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Clark SL, Hankins GD. Preventing maternal death: 10 clinical diamonds. Obstet Gynecol. 2012;119(2 Pt 1):360–364. doi: 10.1097/AOG.0b013e3182411907 [DOI] [PubMed] [Google Scholar]
  • 29.Tuffnell DJ, Jankowicz D, Lindow SW, Lyons G, Mason GC, Russell IF, Walker JJ; Yorkshire Obstetric Critical Care Group. Outcomes of severe pre-eclampsia/eclampsia in Yorkshire 1999/2003. BJOG. 2005;112:875–880. doi: 10.1111/j.1471-0528.2005.00565.x [DOI] [PubMed] [Google Scholar]
  • 30.Roberts CL, Algert CS, Morris JM, Ford JB. Increased planned delivery contributes to declining rates of pregnancy hypertension in Australia: a population-based record linkage study. BMJ Open. 2015;5:e009313. doi: 10.1136/bmjopen-2015-009313 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Roberge S, Bujold E, Nicolaides KH. Meta-analysis on the effect of aspirin use for prevention of preeclampsia on placental abruption and antepartum hemorrhage. Am J Obstet Gynecol. 2018;218:483–489. doi: 10.1016/j.ajog.2017.12.238 [DOI] [PubMed] [Google Scholar]
  • 32.ACOG practice bulletin no. 203: chronic hypertension in pregnancy. Obstet Gynecol. 2019;133:e26–e50. [DOI] [PubMed] [Google Scholar]
  • 33.Abalos E, Duley L, Steyn DW, Gialdini C. Antihypertensive drug therapy for mild to moderate hypertension during pregnancy. Cochrane Database Syst Rev. 2018;10:CD002252. doi: 10.1002/14651858.CD002252.pub4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.American College of Obstetricians and Gynecologists Task Force on Hypertension in Pregnancy. Hypertension in pregnancy. Report of the American College of Obstetricians and Gynecologists’ task force on hypertension in pregnancy. Obstet Gynecol. 2013;122:1122–1131. doi: 10.1097/01.AOG.0000437382.03963.88 [DOI] [PubMed] [Google Scholar]
  • 35.Bramham K, Parnell B, Nelson-Piercy C, Seed PT, Poston L, Chappell LC. Chronic hypertension and pregnancy outcomes: systematic review and meta-analysis. BMJ. 2014;348:g2301. doi: 10.1136/bmj.g2301 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Ananth CV, Vintzileos AM. Maternal-fetal conditions necessitating a medical intervention resulting in preterm birth. Am J Obstet Gynecol. 2006;195:1557–1563. doi: 10.1016/j.ajog.2006.05.021 [DOI] [PubMed] [Google Scholar]
  • 37.Joseph KS. Theory of obstetrics: an epidemiologic framework for justifying medically indicated early delivery. BMC Pregnancy Childbirth. 2007;7:4. doi: 10.1186/1471-2393-7-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Joseph KS, Demissie K, Kramer MS. Obstetric intervention, stillbirth, and preterm birth. Semin Perinatol. 2002;26:250–259. doi: 10.1053/sper.2002.34769 [DOI] [PubMed] [Google Scholar]
  • 39.Petersen EE, Davis NL, Goodman D, Cox S, Mayes N, Johnston E, Syverson C, Seed K, Shapiro-Mendoza CK, Callaghan WM, et al. Vital signs: pregnancy-related deaths, United States, 2011-2015, and strategies for prevention, 13 States, 2013-2017. MMWR Morb Mortal Wkly Rep. 2019;68:423–429. doi: 10.15585/mmwr.mm6818e1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Louis JM, Menard MK, Gee RE. Racial and ethnic disparities in maternal morbidity and mortality. Obstet Gynecol. 2015;125:690–694. doi: 10.1097/AOG.0000000000000704 [DOI] [PubMed] [Google Scholar]
  • 41.Metcalfe A, Wick J, Ronksley P. Racial disparities in comorbidity and severe maternal morbidity/mortality in the United States: an analysis of temporal trends. Acta Obstet Gynecol Scand. 2018;97:89–96. doi: 10.1111/aogs.13245 [DOI] [PubMed] [Google Scholar]
  • 42.Review to action. Report from nine maternal mortality review committees. Accessed November 13, 2002. http://reviewtoaction.org/Report_from_Nine_MMRCs
  • 43.D’Alton ME, Friedman AM, Bernstein PS, Brown HL, Callaghan WM, Clark SL, Grobman WA, Kilpatrick SJ, O’Keeffe DF, Montgomery DM, et al. Putting the “M” back in maternal-fetal medicine: a 5-year report card on a collaborative effort to address maternal morbidity and mortality in the United States. Am J Obstet Gynecol. 2019;221:311–317.e1. doi: 10.1016/j.ajog.2019.02.055 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Main EK. Reducing maternal mortality and severe maternal morbidity through state-based quality improvement initiatives. Clin Obstet Gynecol. 2018;61:319–331. doi: 10.1097/GRF.0000000000000361 [DOI] [PubMed] [Google Scholar]
  • 45.Main EK, Goffman D, Scavone BM, Low LK, Bingham D, Fontaine PL, Gorlin JB, Lagrew DC, Levy BS; National Partnership for Maternal Safety; Council on Patient Safety in Women’s Health Care. National partnership for maternal safety: consensus bundle on obstetric hemorrhage. Obstet Gynecol. 2015;126:155–162. doi: 10.1097/AOG.0000000000000869 [DOI] [PubMed] [Google Scholar]
  • 46.Kirkpatrick DH, Burkman RT. Does standardization of care through clinical guidelines improve outcomes and reduce medical liability? Obstet Gynecol. 2010;116:1022–1026. doi: 10.1097/AOG.0b013e3181f97c62 [DOI] [PubMed] [Google Scholar]
  • 47.Joseph KS, Boutin A, Lisonkova S, Muraca GM, Razaz N, John S, Mehrabadi A, Sabr Y, Ananth CV, Schisterman E. Maternal mortality in the United States: recent trends, current status, and future considerations. Obstet Gynecol. 2021;137:763–771. doi: 10.1097/AOG.0000000000004361 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Lu MC, Halfon N. Racial and ethnic disparities in birth outcomes: a life-course perspective. Matern Child Health J. 2003;7:13–30. doi: 10.1023/a:1022537516969 [DOI] [PubMed] [Google Scholar]
  • 49.Knight M, Joseph KS. Severe maternal morbidity and maternal mortality: a need for consensus on concepts and prevention efforts. Paediatr Perinat Epidemiol. 2020;34:377–378. doi: 10.1111/ppe.12676 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental data
NIHMS1814393-supplement-Supplemental_data.pdf (588.9KB, pdf)

RESOURCES