For much of the 20th century the risk of maternal death decreased dramatically, yet no change in the risk of maternal death has occurred since 1982.1 One reason for the lack of recent progress might be the increase in the proportion of pregnancies among older women, who consistently demonstrate a greater risk of maternal death than their younger counterparts.2–9 Pregnancy rates for women 35–39 years old increased 74% between 1976 and 1997, with a concomitant increase of 38% for women 40 years and older.10,11 Women 35–39 years old have a twofold to threefold higher risk of pregnancy-related death than women in their twenties, and the risk is even more dramatic for women 40 years and older.12
The last examination of maternal deaths among mature women in the United States was conducted before the inception of the Centers for Disease Control and Prevention's Pregnancy Mortality Surveillance System in 1987.2 Although the traditional problem of undercounting maternal deaths remains, the Pregnancy Mortality Surveillance System takes advantage of enhanced methods for identifying deaths within 1 year of pregnancy and thus represents the most complete surveillance of these events. We undertook a descriptive analysis of pregnancy-related death among women 35 years and older using this large, population-based data set. In this analysis we describe pregnancy-related deaths among women 35 years and older for the United States population by demographic and obstetric characteristics and cause of death, and we compare the risk of death for these women to that of a younger group.
MATERIALS AND METHODS
Beginning in 1987, the Division of Reproductive Health at the Centers for Disease Control and Prevention collected data on deaths caused by pregnancy and its complications. Health departments in the 50 states, the District of Columbia, and New York City voluntarily provide copies of death certificates with individual identities masked for all pregnancy-related deaths. Linked birth or fetal death certificates for deaths after a live birth or stillbirth are also made available when possible. Since 1991, states were asked to send death and available birth certificates for all deaths that occurred during or within 1 year of the end of pregnancy, regardless of the cause of death. After review of all information available, clinically experienced epidemiologists code a cause of death, associated obstetric conditions, and the outcome of pregnancy.
Deaths were considered pregnancy-related if they occurred during pregnancy or within 1 year of pregnancy termination and resulted from complications of pregnancy itself, a chain of events initiated by pregnancy, or aggravation of an unrelated event by the physiologic effects of pregnancy. The pregnancy-related mortality ratio was defined as the number of pregnancy-related deaths per 100,000 live births. Pregnancy-related deaths were identified from the Pregnancy Mortality Surveillance System for the years 1991–1997, and these deaths served as numerators for calculating the pregnancy-related mortality ratio. The denominators, live births, were obtained from public-use natality tapes created by the Centers for Disease Control and Prevention's National Center for Health Statistics.13
We categorized women 35 years and older into the following age groups: 35–39 years and 40 years and older. Maternal race was categorized as white, black, or other (nonwhite, nonblack). Because parity is not reported on natality tapes, we used live birth order (the number of live births, including the index pregnancy) as a proxy for parity and categorized it as 1, 2–4, and 5 or more. Prenatal care was classified according to trimester of entry or as no prenatal care. Maternal education was categorized as less than 12, 12, or more than 12 years. Because information about live birth order was available only on the birth certificates, and prenatal care data were missing on over one third of fetal death certificates, we confined analyses of these variables to women who died after a live birth. Cause of death was categorized into the following eight groups: hemorrhage, infection, embolism (thrombotic and amniotic fluid), hypertensive disorders of pregnancy, cardiomyopathy, anesthesia, cerebrovascular accident, or other medical conditions. The latter category is largely composed of diseases and conditions that predated and were aggravated by the pregnancy. Pregnancy outcomes were categorized as live birth, stillbirth, undelivered, abortion, ectopic, gestational trophoblastic neoplasia, and unknown.
Because black women have almost four times the risk of pregnancy-related death as white women do,12 we stratified the analysis by race. The category designated “other races” was omitted from stratified analyses, as the small size and heterogeneous composition of this group precluded meaningful interpretation. We calculated risk ratios for pregnancy-related death in older women by dividing the pregnancy-related mortality ratios for each age group by the pregnancy-related mortality ratio for women aged 25–29, a group with a low risk of pregnancy-related death. Hence, risk ratios measure the excess risk of death for older women compared with younger women. Cause of death distributions were determined for older women and compared with those for women 25–29 years old. The cause-specific pregnancy-related mortality ratio for each older age group was compared with that of women 25–29 years old by calculating risk ratios for each cause of death. The categories of hemorrhage deaths and other medical conditions in the Pregnancy Mortality Surveillance System are heterogeneous groups of diagnoses and etiologies. Therefore, we performed subanalyses to look at the risk of death for specific causes within these categories.
For the years 1991–1997, 554 pregnancy-related deaths among women aged 35–39 years and 211 deaths among women 40 years and older were reported to the Pregnancy Mortality Surveillance System. Women aged 35–39 years and 40 years and older had fewer deaths associated with abortive outcomes (3.3% and 1.4%, respectively) compared with women 25–29 years old (4.5%). Otherwise, pregnancy outcomes associated with deaths for older women were similar to those of younger women, with nearly 60% of deaths occurring after a live birth. For each race, women aged 35–39 years had a higher pregnancy-related mortality ratio than younger women did, and women aged 40 years and older had an even higher pregnancy-related mortality ratio (Table 1). Overall, the increased risk of death for women 35–39 years and 40 years and older was similar for black and white women.
In general, within each stratum of live birth order, prenatal care, and education, the pregnancy-related mortality ratio was higher for older women compared with younger women (Table 2). Within most strata of live birth order, level of prenatal care, and level of education, the excess risk of death for women 40 years and older was approximately twofold greater than the excess risk for women aged 35–39 years.
Hemorrhage, embolisms, other medical conditions, and hypertensive disorders of pregnancy were the major causes of death for all women 35 years and older (Figure 1). Except for deaths from anesthesia, where the pregnancy-related mortality ratio was low for all women, the risk of death increased with age for all causes of death. The greatest excess risks of death among older white women, particularly those 40 years and older, were due to hemorrhage, cardiomyopathy, embolisms, and other medical conditions. Among black women, the excess risks for older women were greatest for hypertensive disorders of pregnancy, cerebrovascular accidents, infections, and other medical conditions (Table 3).
Among the specific causes of hemorrhage deaths, three causes (ectopic pregnancy, uterine atony, and abnormalities of placentation such as placenta accreta, increta, or percreta) accounted for 55% of the deaths. Although white women 35–39 years old were 2.7 times more likely than younger women to die overall from hemorrhage (Table 3) they were four times more likely to die of ectopic pregnancy. White women 40 years old and older were 7.9 times more likely to die overall from hemorrhage but 15 times more likely to die of abnormalities of placentation. Whereas black women aged 35–39 were 2.2 times more likely to die overall from hemorrhage than were black women aged 25–29, they were 8.7 times as likely to die of abnormalities of placentation. Black women 40 years and older were 5.5 times more likely to die of hemorrhage but were 28.5 and 22.8 times more likely to die of abnormalities of placentation and uterine atony, respectively.
Cardiovascular deaths accounted for 39% of deaths from other medical conditions. White women 35–39 years and 40 years and older were 2.7 and 5.5 times more likely to die of other medical conditions, respectively, than their younger counterparts but 3.9 and 8.2 times more likely to die of cardiovascular disease, respectively. A similar comparison for black women found that although women 40 years and older were 6.2 times more likely to die of other medical conditions, they were 10.6 times more likely to die of cardiovascular disease than 25–29-year-old black women.
This analysis extends the observation of an increased risk of death among older women12 by exploring other risk factors for pregnancy-related death, including examining specific causes of death. This is the only population-based investigation of deaths among women 35 years and older since that done by Buehler et al in 1986,2 and it is the first in-depth analysis of pregnancy-related deaths among older women since the inception of the Pregnancy Mortality Surveillance System. Compared with women 25–29 years old, women aged 35 years and older had an increased risk of pregnancy-related death, regardless of parity, prenatal care, or education. Among both blacks and whites, the risk of death for all pregnancy-related causes was higher for older women than for women 25–29 years old.
During the period 1974–1978, the pregnancy-related mortality ratios among women 35 years and older were 41.1 for white women, 154.5 for black women, and 43.3 for women of other races.2 The corresponding pregnancy-related mortality ratios for 1991–1997 were 16.6, 84.9, and 26.6, respectively. Although the data from the earlier study used methodology similar to that of the Pregnancy Mortality Surveillance System, this preceded the time when some states linked death certificates for women of reproductive age with birth and fetal death certificates, a strategy shown to increase the identification of maternal deaths.14,15 Because of ongoing changes in the identification of pregnancy-related deaths, pregnancy-related mortality ratios over time are not directly comparable. However, because the identification of pregnancy-related deaths improved in the 1990s,12 undercounting of deaths was likely to be greater in 1974–1978 than 1991–1997. Therefore, the decline in the risk of death for older women is probably even greater than the difference in the reported pregnancy-related mortality ratios over time. This decline might be explained by improvements in clinical care or changes in the risk profile of older women. Unfortunately, our data do not allow that level of analysis.
Specific causes of death among older women merit further discussion. The large risk ratios we calculated for death from abnormalities of placentation are consistent with the association between increasing maternal age and the risk of placenta previa16 as well as a high cesarean rate. The occurrence of placenta previa in a woman with a prior uterine scar is a strong risk factor for placenta accreta, increta, or percreta, conditions that can result in life-threatening hemorrhage, especially if not detected antenatally.17 Moreover, the risk of placenta percreta in the presence of placenta previa increases with the number of previous cesarean deliveries.18 Maintaining a high index of suspicion for this problem in this high-risk group, utilizing antenatal detection methods, and developing a multidisciplinary surgical plan may decrease the mortality rate from this complication.17
Chronic hypertension prior to pregnancy may contribute to the increased risk of death from hypertensive disorders of pregnancy and cerebrovascular accidents among older black women. Grimes and Gross19 found chronic hypertension among 22.5% of gravid black women older than 35 years. Women with chronic hypertension for at least 4 years or who have a diastolic blood pressure of at least 100 mm Hg are at increased risk of developing preeclampsia,20 and case fatality rates for preeclampsia and eclampsia are greater for black women than white women.21 Therefore, older black women represent a particularly high-risk group for the hypertensive disorders of pregnancy, and they bear a disproportionate risk of mortality compared with their younger counterparts and similarly aged white women. Further, the high prevalence of chronic hypertension among older black women places them at greater risk of cerebrovascular accidents.22–24
Women in the later reproductive years may represent a group that merits more attention to cardiovascular disease than has traditionally been given. Between 1989 and 1998 there was a 21% increase in the rate of sudden cardiac deaths among women aged 35–44 years,25 and among pregnant women myocardial infarction occurs most commonly in older women.26 The contribution of cardiovascular disease to deaths among older women that we found in this analysis is likely to become increasingly important as more older women go through pregnancy.
The Pregnancy Mortality Surveillance System data for this analysis included those variables recorded on available vital records. For the approximately 40% of deaths that did not have a live birth outcome, the death certificate was the sole source of information and obstetric variables were not available. Therefore, we were unable to present a comprehensive analysis of factors associated with pregnancy-related deaths among older women. Moreover, the path from initial event to death is frequently complex, and vital records do not contain information with sufficient nuance to accurately understand the clinical course. Even assigning a cause of death may, in some cases, be problematic. However, there is no reason to believe that these data limitations differ by age. The Pregnancy Mortality Surveillance System attempts to capture all pregnancy-related deaths, but these events remain rare. Therefore, stratification by demographic and obstetric characteristics and cause of death results in comparisons of small numbers of women and possibly unstable ratios.
This analysis also did not allow us to determine whether older pregnant women have a greater risk of death because they are more likely to have a potentially fatal condition or because they are more likely to die when they do present with such a condition. The prevalence of pregnancy complications, including gestational diabetes, placenta previa, postpartum hemorrhage, cesarean, pulmonary embolism, and prolonged hospital stay, is higher among older women.27 A greater risk of preeclampsia among multiparous women aged 35 years and older than among younger multiparas was found in a large hospital-based study.28 However, the lack of population-based information about the prevalence of potentially fatal maternal morbidity hinders our ability to partition risk into its two components, prevalence and case-fatality rate.
Despite the decrease in risk over the past 20 years, women 35 years and older remain at increased risk of pregnancy-related death from all causes. Moreover, disparities in risk persist between white women and black women. To obtain a more complete picture of how such deaths in this increasing group of pregnant women can be decreased, the comprehensive review of maternal deaths should be a core public health function in all states. Such analysis will shed light on the medical and nonmedical factors related to death from pregnancy and associated complications. In addition, when viewed as a sentinel event, each pregnancy-related death represents many women who experience significant short- and long-term morbidity related to pregnancy. Understanding the events surrounding these outcomes may provide evidence for interventions and policies in clinical medicine and health care systems that will result in improving the experience of pregnancy for all women.
1. Centers for Disease Control and Prevention. Maternal mortality–United States, 1982–1996. MMWR Morb Mortal Wkly Rep 1998;47:705–7.
2. Buehler JW, Kaunitz AM, Hogue CJR, Hughes JM, Smith JC, Rochat RW. Maternal mortality in women aged 35 years or older: United States. JAMA 1986;255:53–7.
3. Hogberg U. Maternal deaths in Sweden, 1971–1980. Acta Obstet Gynecol Scand 1986;161–7.
4. Rochat RW, Koonin LM, Atrash HK, Jewett JF. Maternal mortality in the United States: Report from the maternal mortality collaborative. Obstet Gynecol 1988;72:91–7.
5. Dorfman SF. Maternal mortality in New York City, 1981–1983. Obstet Gynecol 1990;76:317–23.
6. Syverson CJ, Chavkin W, Atrash HK, Rochat RW, Sharp ES, King GE. Pregnancy-related mortality in New York City, 1980–1984: Causes of death and associated risk factors. Am J Obstet Gynecol 1991;164:603–8.
7. Hogberg U, Innala E, Sandstrom A. Maternal mortality in Sweden, 1980–1988. Obstet Gynecol 1994;84:240–4.
8. Berg CJ, Atrash HK, Koonin LK, Tucker M. Pregnancy-related mortality in the United States, 1987–1990. Obstet Gynecol 1996;88:161–7.
9. Hoyert DL, Danel I, Tully P. Maternal mortality, United States and Canada, 1982–1997. Birth 2000;27:4–11.
10. Ventura SJ, Mosher WD, Curtin SC, Abma JC, Henshaw S. Trends in pregnancies and pregnancy rates by outcome: Estimates for the United States, 1976–96. National Center for Health Statistics. Vital Health Stat 2000;21(56).
11. Ventura SJ, Mosher WD, Curtin SC, Abma JC, Henshaw S. Trends in pregnancy rates for the United States, 1976–97: An update. National vital statistics reports; vol. 49 no. 4. Hyattsville, Maryland: National Center for Health Statistics, 2001.
12. Berg CJ, Chang J, Callaghan WM, Whitehead SJ. Pregnancy-related mortality in the United States, 1991–1997. Obstet Gynecol 2003;101:289–96.
14. Jocums S, Mitchell EF, Entman SS, Piper JM. Monitoring maternal mortality using vital records linkage. Am J Prev Med 1995;11:75–8.
15. Horon IL, Cheng D. Enhanced surveillance for pregnancy-associated mortality–Maryland, 1993–1998. JAMA 2001;285:1455–9.
16. Ananth CV, Wilcox AJ, Savitz DA, Bowes WA, Luther ER. Effect of maternal age and parity on the risk of uteroplacental bleeding disorders in pregnancy. Obstet Gynecol 1996;88:511–6.
17. Hudon L, Belfort MA, Broome DR. Diagnosis and management of placenta percreta: A review. Obstet Gynecol Surv 1998;53:509–17.
18. Clark SL, Koonings PP, Phelan JP. Placenta previa/accreta and prior cesarean section. Obstet Gynecol 1985;66:89–92.
19. Grimes DA, Gross GK. Pregnancy outcomes in black women aged 35 and older. Obstet Gynecol 1981;58:614–20.
20. Sibai BM, Lindheimer M, Hauth J, Caritis S, Van Dorston P, Klebanoff M, et al. Risk factors for preeclampsia, abruptio placentae, and adverse neonatal outcomes among women with chronic hypertension. N Engl J Med 1998;339:667–71.
21. MacKay AP, Berg CJ, Atrash HK. Pregnancy-related mortality from preeclampsia and eclampsia. Obstet Gynecol 2001;97:533–8.
22. Lanska DJ, Kryscio RJ. Risk factors for peripartum and postpartum stroke and intracranial venous thrombosis. Stroke 2000;31:1274–82.
23. Teunissen LL, Rinkel GJE, Algra A, van Gijn J. Risk factors for subarachnoid hemorrhage. A systematic review. Stroke 1996;27:544–9.
24. Quan A, Kerlikowske K, Gueffier F, Boissel JP. Efficacy of treating hypertension in women. J Gen Intern Med 1999; 14:718–29.
25. Zheng Z, Croft JB, Giles WH, Mensah GA. Sudden cardiac death in the United States, 1989 to 1998. Circulation 2001;104:2158–63.
26. Roth A, Elkayam U. Acute myocardial infarction associated with pregnancy. Ann Intern Med 1996;125:751–62.
27. Jolly M, Sebire N, Robinson S, Regan L. The risks associated with pregnancy in women aged 35 years or older. Hum Reprod 2000;15:2433–7.
© 2003 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
28. Bobrowski RA, Bottoms SF. Underappreciated risks of the elderly multipara. Am J Obstet Gynecol 1995;172:1764–70.