Abstract
OBJECTIVE:
Reports from the National Center for Health Statistics suggest that maternal mortality rates in the United States increased after the introduction of the pregnancy status checkbox on the 2003 revised US standard death certificate; however, this increase was because of an artifact of surveillance.1–3 This study aimed to evaluate the trends in cardiovascular disease (CVD)-related pregnancy-associated mortality in the United States and assess the impact of ascertainment by death certificate type using national- and state-level data.
STUDY DESIGN:
We conducted a cross-sectional analysis of all live births and CVD-related pregnancy-associated deaths in the United States (1999–2018) based on the National Vital Statistics System data. The primary outcome was CVD-related pregnancy-associated deaths identified using the International Classification of Diseases, Tenth Revision, codes. To account for the effect of the implementation of the 2003 revised US standard death certificate on reported pregnancy-associated mortality ratios (PMRs), we used state-level indicators at the individual level to determine how many deaths were recorded each year according to the standard US death certificate (1989 revision) vs the 2003 revised version. We examined changes in 42-day PMRs (expressed per 100,000 live births) between 1999 and 2018 by death certificate type. The trends in PMRs were quantified based on rate ratios (RRs) with 95% confidence intervals (CIs), derived from log-linear regression models with a Poisson distribution (using a “log” link function with robust variance estimation).4
RESULTS:
Throughout the study period (1999–2018), there were 80,802,690 live births and 238 CVD-related deaths in the United States. The overall CVD-related PMR increased from 0.24 to 0.91 per 100,000 live births (adjusted RR, 3.32; 95% CI, 1.99–5.54). The overall CVD-related PMRs were 0.20 per 100,000 live births using the 1989 version and 0.38 per 100,000 live births using the 2003 version (Table). However, when we adjusted for death certificate type, there was a reduction in the overall risk between 1999 to 2000 and 2017 to 2018 (RR, 0.55; 95% CI, 0.12–2.48). Analyses stratified by CVD subtype, including pulmonary heart disease and hypertensive heart disease, are shown in the Table.
CONCLUSION:
In this large population-based cross-sectional analysis, we identified an increase in the CVD-related PMR in the United States from 1999 to 2018 using national data. However, in the analyses that adjusted for death certificate type, we found that the use of the 2003 revised US standard death certificate resulted in RRs that showed no difference in mortality rates throughout the study period. Despite the wide 95% CIs (highlighting imprecision in the estimates), this analysis suggests that the certificate version may be an effect modifier. We believe that the increased number of deaths captured by the 2003 revised US standard death certificate reflects a true increase in mortality rather than an artifact from misclassification. Although the study has some limitations, including its retrospective design based on data that lack specific clinical details regarding each death, this analysis suggests that the observed time trend of an increasing CVD-related PMR is not an artifact of the 2003 checkbox revision but a real trend that warrants attention.
TABLE.
CVD-related pregnancy-associated mortality based on death certificate revision: United States, 1999 to 2018
| Pregnancy-associated deaths and mortality ratios | ||||
| Maternal characteristics | Live births: number (%) | CVD-related pregnancy-associated deaths | ||
| Number of deaths | Ratio (95% CI a ) per 100,000 live births | |||
| All women | 80,802,690 (100) | 238 | 0.30 (0.29–0.30) | |
| 1989 revisionb | ||||
| All CVD-related | 36,835,279 (46) | 72 | 0.20 (0.15–0.25) | |
| Pulmonary heart disease | 26 | 0.07 (0.05–0.10) | ||
| Hypertensive heart disease | 27 | 0.07 (0.05–0.10) | ||
| 2003 revisionc | ||||
| All CVD-related | 43,967,173 (54) | 166 | 0.38 (0.32–0.44) | |
| Pulmonary heart disease | 62 | 0.14 (0.11–0.18) | ||
| Hypertensive heart disease | 64 | 0.15 (0.11–0.19) | ||
| Changes in CVD-related pregnancy-associated mortality | ||||
| Year | All CVD-related disease d | Pulmonary heart disease d | ||
| Adjusted e : RR (95% CI) | Adjusted f : RR (95% CI) | Adjusted e : RR (95% CI) | Adjusted f : RR (95% CI) | |
| 1999–2000 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| 2009–2010 | 1.10 (0.59–2.03) | 0.24 (0.56–1.03) | 2.88 (0.93–8.95) | 1.26 (0.23–6.91) |
| 2017–2018 | 3.32 (1.99–5.54) | 0.55 (0.12–2.48) | 6.15 (2.13–17.75) | 2.18 (0.36–13.22) |
CI, confidence interval; CVD, cardiovascular disease; RR, rate ratio.
The 95% CI estimates were derived on the basis of the Clopper-Pearson exact method
The 1989 revision refers to the use of the previous version of the US standard death certificate
The 2003 revision refers to the use of the 2003 revised version of the US standard death certificate
All Table entries below denote RRs and 95% CIs, derived from Poisson log-linear regression models with robust variance
The RRs were adjusted for maternal age and race and ethnicity
The RRs were adjusted for maternal age, race and ethnicity, and death certificate version use.
Acknowledgments
The authors report receiving no support from any organization for the submitted work, no financial relationship with any organization that might have an interest in the submitted work in the previous 3 years, and no other relationship or activity that could seem to have influenced the submitted work.
No financial support was necessary for the preparation of this manuscript or acquiring data. National Heart, Lung, and Blood Institute (grant number R01-HL150065) and the National Institute for Environmental Health Sciences (grant number R01-ES033190), National Institute for Environmental Health Sciences.
Footnotes
The authors report no conflict of interest.
Contributor Information
Kimberly Bodenlos, Division of Maternal-Fetal Medicine, Department of Obstetrics, Gynecology, and Reproductive Sciences, Rutgers Robert Wood Johnson Medical School, New Brunswick, NJ.
Justin S. Brandt, Division of Maternal-Fetal Medicine, Department of Obstetrics, Gynecology, and Reproductive Sciences, Rutgers Robert Wood Johnson Medical School, New Brunswick, NJ.
Hillary L. Graham, Division of Epidemiology and Biostatistics, Department of Obstetrics, Gynecology, and Reproductive Sciences, Rutgers Robert Wood Johnson Medical School, New Brunswick, NJ.
Meike Schuster, Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Jefferson Abington Health, Abington PA.
Cande V. Ananth, Division of Epidemiology and Biostatistics, Department of Obstetrics, Gynecology, and Reproductive Sciences, Rutgers Robert Wood Johnson Medical School, 125 Paterson St., New Brunswick, NJ 08901, Department of Medicine, Rutgers Robert Wood Johnson Medical School, New Brunswick, NJ, Cardiovascular Institute of New Jersey, Rutgers Robert Wood Johnson Medical School, New Brunswick, NJ, Department of Biostatistics and Epidemiology, Rutgers School of Public Health, Piscataway, NJ, Environmental and Occupational Health Sciences Institute, Rutgers Robert Wood Johnson Medical School, New Brunswick, NJ.
REFERENCES
- 1.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]
- 2.Callaghan WM. Overview of maternal mortality in the United States. Semin Perinatol 2012;36:2–6. [DOI] [PubMed] [Google Scholar]
- 3.Joseph KS, Lisonkova S, Muraca GM, et al. Factors underlying the temporal increase in maternal mortality in the United States. Obstet Gynecol 2017;129:91–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Spiegelman D, Hertzmark E. Easy SAS calculations for risk or prevalence ratios and differences. Am J Epidemiol 2005;162:199–200. [DOI] [PubMed] [Google Scholar]