Trends and Racial/Ethnic, Socioeconomic, and Geographic Disparities in Maternal Mortality from Indirect Obstetric Causes in the United States, 1999-2017

1. Introduction

Although the leading causes of maternal deaths in the United States are hemorrhage, pregnancy-related hypertension, embolism, and obstetric infection, maternal deaths due to indirect obstetric causes, including chronic medical conditions such as cardiovascular disease (CVD), diabetes, respiratory disorders, and mental health conditions, have increased substantially in the past two decades.1,2 Globally, including the United States, 25-30% of all maternal deaths are attributed to these indirect causes.3-5 Recent trends and socioeconomic and demographic disparities in US maternal mortality from indirect causes have not been well studied, although the proportionate share of various indirect causes to overall mortality and racial/ethnic patterns have been examined using death certificate-based maternal mortality and Pregnancy Mortality Surveillance System data.6-9 To our knowledge, none of the US studies have analyzed inequalities in maternal mortality from indirect causes by a wide range of social determinants, including maternal education, marital status, nativity/immigrant status, place of residence, geographic region, and area deprivation. Analyzing disparities according to these social determinants is important in identifying sociodemographic groups and disadvantaged communities that may be at increased risk of maternal morbidity and mortality and which can benefit from targeted social and public heath interventions.

To address the aforementioned gaps in research, we analyzed racial/ethnic, socioeconomic, demographic, rural-urban, and regional variations in US maternal mortality from indirect obstetric causes and specific chronic conditions (ICD-10 codes O98-O99) occurring during or within 42 days after the end of pregnancy.1,2,10

2. Methods

We used 1999-2017 data from the National Vital Statistics System, the national mortality database.1,2,10 Maternal deaths in this study, are defined in accordance with the WHO as those related to or aggravated by pregnancy or pregnancy management and which occur during or within 42 days after the end of pregnancy.1,2,10 This official definition of maternal mortality differs from the definition of pregnancy-related mortality, which includes deaths during or within 1 year of pregnancy.8,9

Rates of maternal mortality from indirect causes were computed annually between 1999 and 2017 for all races, Whites and Blacks. Maternal deaths and rates for Asian and Pacific Islanders (APIs), American Indians and Alaska Natives (AIANs) and Hispanics, maternal education, marital status, nativity/immigrant status, and maternal age were derived using the pooled 5-year mortality and natality data for the period 2013-2017 because of the small numbers of maternal deaths annually and to produce statistically robust estimates for these demographic groups. The numbers of births by these sociodemographic characteristics, obtained from the national natality files, served as the denominators for computing maternal mortality rates.11

Area-based socioeconomic patterns in maternal mortality were derived by linking the 2008-2012 census-based county-level deprivation index to the race- and county-specific mortality statistics from 2013 to 2017.12-15 We used previously developed factor-based area deprivation index (ADI), developed by Singh and colleagues, using data from the 2008-2012 American Community Survey.12-15 The ADI consisted of 21 census-based socioeconomic indicators, which may be viewed as broadly representing educational opportunities, labor force skills, economic, and housing conditions in a given county.12-15 Selected indicators of education, occupation, wealth, income distribution, unemployment rate, poverty rate, and housing quality were used to construct these indices by factor and principal components analyses.12-15 The factor loadings (correlations of indicators with the ADI) for the 2008-2012 county index varied from 0.91 for 150% of the poverty rate to 0.23 for home ownership rate.12,15 To compute maternal mortality rates by deprivation level, we used the weighted population quintile distribution of the deprivation index that classified all 3,141 US counties into 5 groups of approximately equal population size.12-15 The groups thus created ranged from being the most-deprived (first quintile) to the least-disadvantaged (fifth quintile) population groups.12-15 To simplify analysis and for statistical robustness, we combined 2nd, 3rd, and 4th quintiles as the middle deprivation category. Each of the 3,141 counties in the mortality database was assigned one of these 3 deprivation quintiles.

Log-linear regression models were used to estimate annual rates of change in maternal mortality from indirect causes.13-15 Specifically, the logarithm of mortality rates were modeled as a linear function of time (calendar year), which yielded annual exponential rates of change in mortality rates.13-15 Disparities in mortality by social factors and deprivation level were described by rate ratios (relative risks=RR) and rate differences (absolute inequalities), which were tested for statistical significance at the 0.05 level.

3. Results

3.1. Trends in Maternal Mortality from Indirect Causes

Maternal mortality from indirect causes showed an upward trend during 1999-2017, with the rate increasing by 11.2% (95% CI=9.0%-13.5%) annually for the overall population, 12.9% (95% CI=10.5%-15.3%) for non-Hispanic Whites, and 9.4% (95% CI=7.2%-11.6%) for non-Hispanic Blacks (Figure 1). During 1999-2017, non-Hispanic Black women had 2-3 times higher mortality from indirect causes than non-Hispanic White women. The proportion of all maternal deaths due to indirect causes rose from 12.0% in 1999 to 26.9% in 2017 (Figure 2). Of 12,135 maternal deaths that occurred during 1999-2017, 3184 deaths were from indirect causes, including 173 deaths from maternal infectious and parasitic diseases, 3011 deaths from all chronic conditions, 993 deaths from CVD, 178 deaths from mental disorders, 191 deaths from respiratory diseases, 48 deaths from digestive diseases, 26 deaths from endocrine/metabolic disorders, and 30 deaths from HIV (Table 1). Maternal mortality from CVD increased sharply over time, from 0.40 per 100,000 live births deaths in 1999 to 1.82 in 2017 (Figure 1). During 1999-2017, the average annual rate of increase in maternal mortality was 8.8% (95% CI=7.0%-10.6%) for CVD, 11.3% (95% CI=8.9%-13.7%) for all chronic conditions combined, and 10.0% (95% CI=7.2%-13.0%) for infectious and parasitic diseases (Figures 1 and 2). Trends from other causes are not shown because of small numbers of deaths.

Table 1 Maternal deaths and mortality rates per 100,000 live births due to specific indirect obstetric causes, (ICD-10 codes: O98-O99), United States, 1999-2017 (number of live births =76,972,334)

Cause of death (ICD-10 codes)	Maternal	Maternal
	Deaths	Mortality Rate
All indirect causes (O98-O99)	3,184	4.14
Maternal infectious and parasitic diseases (O98)	173	0.22
Tuberculosis (O98.0)	9	N/A
Syphilis, Gonorrhea, and other sexually transmitted infections (O98.1-O98.3)	2	N/A
Viral hepatitis (O98.4)	5	N/A
Other viral diseases (O98.5)	12	0.02
Protozoal diseases (O98.6)	0	N/A
HIV (O98.7)	30	0.04
Other maternal infectious and parasitic diseases (O98.8)	115	0.15
Unspecified maternal infectious and parasitic diseases (O98.9)	0	N/A
Chronic diseases (O99)	3,011	3.91
Anemia (O99.0)	22	0.03
Blood & immune disorders (O99.1)	24	0.03
Endocrine and metabolic disorders (O99.2)	26	0.03
Mental disorders and diseases of the nervous system (O99.3)	178	0.23
Cardiovascular diseases (O99.4)	993	1.29
Respiratory diseases (O99.5)	191	0.25
Digestive system diseases (O99.6)	48	0.06
Skin and subcutaneous disorders (O99.7)	5	N/A
Other specified diseases and conditions (O99.8)	1,524	1.98

Source: Derived from the National Mortality and Natality Detail Files, 1999-2017 Maternal mortality rates based on fewer than 10 deaths are not shown because of high variability associated with those rates. N/A = not available

Close

Figure 1

Trends in maternal mortality from indirect obstetric causes (ICD-10 Codes: O98-O99), United States, 1999-2017

Source: Data derived from the National Vital Statistics System.

Export to PPT

Close

Figure 2

Trends in maternal mortality from infections (ICD-10 code O98) and chronic diseases (ICD-10 code O99), and proportion of all maternal deaths due to indirect obstetric causes (ICD-10 codes: O98-O99), United States, 1999-2017

Source: Data derived from the National Vital Statistics System

Rates of maternal mortality from infections are three-year moving average rates and are per 1,000,000 (million) live births. Rates of maternal mortality from chronic diseases are annual rates and are per 100,000 live births.

Export to PPT

4. Discussion

Using national vital statistics mortality data from 1999 to 2017, we found a rising trend in US maternal mortality from indirect causes, including CVD and all chronic conditions combined. The proportion of all maternal deaths due to indirect causes increased from 12.0% in 1999 to 26.9% in 2017, with maternal mortality from CVD increasing almost 5-fold during this time period. Our study shows marked disparities in maternal mortality from indirect causes by a wide array of social determinants, including maternal education, marital status, immigrant status, place of residence, geographic region, and area deprivation, which, although shown previously for overall maternal mortality and pregnancy-related mortality8,9,16, had not been documented for mortality from indirect causes to the best of our knowledge. Analysis of a broad range of disparities is important in showing the potential for reductions in US maternal mortality that can be achieved by implementing effective policies and interventions among high-risk groups and communities.

Efforts to reduce maternal mortality in the United States should be focused on indirect obstetric causes, in addition to addressing direct complications of pregnancy and childbirth.3,5,8,9 While maternal mortality from direct obstetric causes such as hemorrhage, pregnancy-related hypertension, and embolism has declined during the past two decades, maternal deaths due to indirect causes, particularly from pre-existing medical conditions, including CVD and metabolic disorders have risen.1-4,8,9 Given the increasing relative importance of indirect maternal causes of death, increased attention should be given to the healthcare needs of pregnant women with pre-existing conditions and high-risk pregnancies.5,17 Tackling the rising prevalence of obesity, diabetes, chronic hypertension, kidney diseases, and cardiovascular conditions among women of reproductive age should become a priority.3,5

Trends and social inequalities in mortality from indirect causes reported here are generally consistent with those observed for maternal mortality overall (direct + indirect causes), pregnancy-related mortality, various risk factors, chronic medical conditions, and maternal complications including pre-pregnancy obesity, maternal diabetes, chronic hypertension, poor mental health, and cardiovascular conditions among women of reproductive age.8,9,16,18,19 For example, consistent with patterns in maternal mortality from indirect causes in our study, a recent study of the 2012-2014 US birth cohorts found that the risk of pre-pregnancy obesity was significantly higher among non-Hispanic Blacks, AIANs, non-Hispanic Whites, unmarried, US-born, women aged ≥35, women with less education, and rural residents.20 Mothers with less than a high school or with a high school diploma had more than twice the risk of pre-pregnancy obesity than those with a college degree, and US-born women had 59% higher risk than immigrant women20 – estimates similar to those reported in our study on maternal mortality from indirect causes.

Another recent US study showed marked disparities in maternal hypertension (consisting of both chronic hypertension and pregnancy-related hypertension) by various social determinants.19 AIANs, non-Hispanic Blacks, and Pacific Islanders had higher risks of maternal hypertension and chronic hypertension, whereas Asians, Hispanics, and non-Hispanic Whites had lower risks of maternal hypertension, including chronic hypertension.19 Increasing maternal age, lower education, US-born status, rural residence, pre-pregnancy obesity, excess weight gain during pregnancy, and gestational diabetes were all associated with increased risks of maternal hypertension.19 In addition, smoking before and during pregnancy was associated with significantly increased risk of chronic hypertension.19

Data on women of reproductive age (18-49 years) from the 2010-2014 US National Health Interview Survey indicate markedly higher prevalence of CVD, hypertension, diabetes, kidney disease, asthma, and serious psychological distress among non-Hispanic Black and AIAN women, women at older ages, those at lower education and income levels, unmarried (divorced/separated/never married) women, US-born women, current smokers, and women with obesity.18,19 Ethnic-minority and socially-disadvantaged groups in the US differ greatly from the majority, affluent groups in their social, physical, and living environments that might influence pre-existing chronic and medical conditions and related risks such as pre-pregnancy obesity, smoking, gestational diabetes, pregnancy-related hypertension, and weight gain during pregnancy.19 They have limited access to neighborhood amenities such as sidewalks, parks/playgrounds, green spaces, public transportation, and heathy, affordable foods that promote physical activity, healthy lifestyle, and healthy living.19,21 Many of the risk factors identified above are modifiable, and maternal deaths from most of the indirect causes are largely preventable. Several community-based federal programs have been successful in reducing rates of obesity, physical inactivity, smoking, and chronic hypertension and promoting healthy eating and weight management.19-21

5. Conclusions and Global Health Implications

While US maternal mortality from direct obstetric causes has declined during the past two decades, maternal deaths due to indirect causes, particularly from pre-existing medical conditions, including CVD, have increased.1,9,10,16 As is the case with overall maternal mortality16, socially and economically-disadvantaged demographic groups and communities experience 2-to-6 times higher risk of maternal mortality from indirect causes. There is a growing understanding that the contribution of indirect causes is misclassified and underestimated.5,17 Pre-pregnancy obesity, for example, increases the risk of both direct and indirect causes of maternal mortality, including chronic and gestational diabetes, pregnancy-related hypertension, pre-eclampsia and eclampsia, spontaneous abortion, hemorrhage, and uterine rupture.17,19 Yet, obesity is only considered an indirect or contributing condition rather than an underlying cause of maternal death.17 Monitoring social inequalities is important in that they represent the impact of underlying social determinants on both indirect and direct causes of maternal mortality.4,24 Understanding the complex interactions among social determinants, indirect causes, and more proximal direct causes of maternal deaths is important to reducing maternal mortality and improving maternal health.5,17