Preventing maternal mortality

Definitions of maternal death

Over the past year Contemporary OB/GYN  has undertaken an intense examination of the common causes of the 700 maternal deaths that occur each year in the United States. I am very grateful for the extraordinary skill and dedication brought to this project by our series editor, Dr. Carolyn Zelop. This important topic has also attracted the attention of professional organizations, foundations, and state and federal governments.  Why is this subject so difficult to wrap our arms around? Well, for a start, there are many definitions of maternal mortality (see Table) and no universally accepted approach taken by individual states to measure its occurrence. In fact, because of uneven state uptake of the 2003 national death certificate, the Centers for Disease Control and Prevention (CDC) National Center for Health Statistics hasn’t published an official Maternal Mortality Ratio for the United States since 2007.  

The Global Burden of Disease group, using a sophisticated statistical algorithm, reported that the US maternal mortality ratio (MMR) increased from 16.9 per 100,000 livebirths in 1990 to 26.4 per 100,000 livebirths in 2015.¹ During that same period, the global MMR declined 30%. Among 31 nations in the Organization for Economic Cooperation and Development, the United States ranks 30th in maternal mortality with rates three times higher than Canada and the UK.² However, at least 80% and perhaps all of this increase can be attributed to improved ascertainment among states implementing the updated death certificate format which contains a detailed set of pregnancy questions.^2,3 While we can take some comfort in this epidemiological solace, none of us should be satisfied until every preventable maternal death is avoided. But this begs the question of exactly what percent of maternal deaths are preventable?

Pregnancy-related deaths

To answer that question, we need more detailed and accurate data. The CDC’s Pregnancy Mortality Surveillance System (PMSS) tracks both pregnancy-associated and pregnancy-related deaths (PRDs).⁴ The former is defined as any death of a woman during pregnancy or within the first postpartum year irrespective of whether the fatality is truly pregnancy-related. The latter includes all deaths of women during or within 1 year of pregnancy caused by a pregnancy complication or a chain of events initiated by pregnancy, or aggravation of an unrelated condition by physiologic effects of pregnancy. Thus, while all PRDs are pregnancy-associated, not all pregnancy-associated deaths are pregnancy-related.  

Because it is ascertained using both birth and death certificate data abstracted and analyzed by medical epidemiologists, the PRD rate is a far more accurate estimate of true US maternal mortality than the MMR.⁴ The CDC reports that the US pregnancy-related mortality ratio has increased from 7.2 deaths per 100,000 livebirths in 1987 to 18.0 deaths per 100,000 live births in 2014, although, as with the MMR, much of this increase is related to enhanced ascertainment and possible over-estimation.⁴ Importantly PRDs are tracked and analyzed by state maternal mortality review committees (MMRCs) to determine the exact causes of and contributing factors to maternal deaths and to generate estimate of preventability.   

The reports issued by those 35 states with MMRCs are extraordinary sources of detailed information on PRDs. Structured, detailed, and relevant data are collected using the CDC’s Maternal Mortality Review Information Application (MMRIA) from information gleaned from a variety of sources including birth and death certificates, newspaper stories, obituaries, and medical records. The MMRCs then opine as to the immediate causes of deaths, contributing factors and the percentage of deaths that are preventable.

Report from nine Maternal Mortality Review committees

This year, nine states (Colorado, Delaware, Georgia, Hawaii, Illinois, North Carolina, Ohio, South Carolina and Utah) reported on pooled PRD data developed through the CDC’s MMRIA and its predecessor, the Maternal Mortality Review Data System to describe lessons learned.⁵ Some of their findings will not come as a surprise to our readers. For example, half of all PRDs studied were caused by hemorrhage, cardiovascular and coronary conditions, cardiomyopathy, or infection. Nor should it be a surprise that significant racial disparities exist with non-Hispanic black women being three to four times more likely to suffer a PRD than non-Hispanic white women. 

The report indicates that the percentage of pregnancy-associated deaths that were pregnancy-related is 35% overall but that the PRD proportion of pregnancy-associated deaths increase with maternal age from 25.7% in women aged 25 to 29 years to 42.3% in women aged 35 to 44. Women in this older age group are more likely to die from cardiovascular disease, hemorrhage, and embolism than younger women while the latter are more likely to die of cardiomyopathy, infection, and mental health conditions.   Non-Hispanic black women have a higher percent of pregnancy-associated deaths that were pregnancy-related (48.2%) than either non-Hispanic whites (28.4%) or Hispanic women (30.2%). While African-American women are nearly twice as likely to die from (pre)eclampsia and embolism, non-Hispanic white women were nearly 10 times more likely to have a PRD linked to mental health conditions. 

Among women with PRDs, 38% occur during pregnancy, 45% within 42 days of a delivery or termination and 18% occur between 34 days and 1 year postpartum. Among PRDs occurring during pregnancy, hemorrhage and cardiovascular/coronary conditions each account for about 20% of deaths followed by embolism (9.2%) and mental health conditions (6.6%). The leading causes of PRDs within 42 days of birth/termination were infection (21.7%) followed by hemorrhage (12.4%), cardiovascular/coronary conditions (12.4%) and (pre)eclampsia (9.3%). Among PRDs occurring between 43 days to 1 year postpartum, cardiomyopathy (32.4%), mental health conditions (16.2%), and venous thromboembolism (VTE) (10.8%) were the most common causes. 

The report notes that three primary factors-systems of care, provider factors and patient factors-contribute to virtually all PRDs. For example, among deaths due to hemorrhage, deficient systems of care (e.g., inadequate training, unavailable personnel, absence of policies and procedures and lack of coordinated care) account for 36% of contributing factors. Provider factors such as missed or delayed diagnosis and ineffective treatment accounted for 31% of contributing factors while patient factors including a lack of knowledge of warning signs or failure to seek care accounted for 26% of contributing factors. The report concludes that addressing these factors would prevent 70% of hemorrhagic deaths. Overall, the authors estimate that 63.2% of PRDs could be prevented by addressing these and similar contributing factors. They further opine that 63.2% of PRDs occurring during pregnancy are potentially preventable while 66.7% of such deaths occurring within 42 days of birth/termination and 58.3% of deaths occurring between 43 days and 1 year postpartum are preventable. 

Interventions to prevent pregnancy-related deaths

The repetitive themes emerging from the report form the basis of recommendations to reduce PRDs. The authors conclude that the greatest benefits would accrue public policies addressing the social determinants of care, which is obviously the most difficult strategy to implement. The next largest benefit would result from ensuring that patients receive the appropriate level of care for their condition-analogous to the regionalization of perinatal care in neonatal intensive care units. 

Recommendations more readily under our control include improving:

• access to care and patient/provider communication 

• patient management of mental health conditions 

• communication and coordination of care among providers 

• standards of assessment, diagnosis, and treatment

• language translation (medical literacy)

• prevention initiatives, including substance abuse screening, prevention, and treatment programs

That such approaches work has been demonstrated by the California Maternal Quality Care Collaborative. Since its founding in 2006, the collaborative has used web-based care toolkits to help decrease maternal mortality by 57% from 16.0 to 7.3 deaths per 100,000 livebirths.⁶ In 2017, 211 hospitals participated, covering 95% of deliveries.  Nationally, the Alliance for Innovation on Maternal Health, (AIM) launched by the federal Health Resources Service Administration (HRSA) Maternal Child Health Bureau and led by the American College of Obstetricians and Gynecologists (ACOG) in collaboration with many other societies, seeks to reduce severe maternal morbidity and mortality using safety bundles.⁷ Currently more than 20 states and 800 hospitals participate.

In Washington, ACOG and other societies are supporting legislation designed to expand MMRCs to all 50 states. The Preventing Maternal Deaths Act of 2017, H.R.1318, and the Maternal Health Accountability Act of 2017, S.1112. would fund the Department of Health and Human Services (HHS) to make grants available to establish and coordinate a national MMRC network, accelerating the introduction of best practices. Happily, both have recently been passed by the House of Representatives and the Senate. 

Take-home message 

The United States has the dubious distinction of having among the highest rates of pregnancy-related deaths in the industrialized world. Even more distressing, we are the only such nation to have rising maternal mortality rates. The availability of MMRCs in every state along with the universal deployment of patient safety bundles such as those advocated by the AIM program are the surest and fastest approaches to reducing this national tragedy. Developing regionalized maternal care levels analogous to perinatal care levels would also likely reduce PRDs. Finally, because obesity is a major contributor to severe maternal morbidity and mortality from cardiac disease, VTE, depression, and (pre)eclampsia and confers many long-term adverse health sequelae, it should be a major focus of U.S. prevention efforts.⁸ While not all maternal deaths can be prevented by these actions, the opportunity to prevent nearly two-thirds of maternal deaths in this Nation must be seized upon.

References:

• GBD 2015 Maternal Mortality Collaborators. Global, regional, and national levels of maternal mortality, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet 2016; 388:1775-1812.

• MacDorman MF, Declercq E, Cabral H, Morton C. Recent Increases in the U.S. Maternal Mortality Rate: Disentangling Trends from Measurement Issues. Obstet Gynecol. 2016; 128(3):447-455.

• Joseph KS, Lisonkova S, Muraca GM, Razaz N, Sabr Y, Mehrabadi A, Schisterman EF. Factors Underlying the Temporal Increase in Maternal Mortality in the United States. Obstet Gynecol. 2017; 129(1):91-100. 

• https://www.cdc.gov/reproductivehealth/maternalinfanthealth/pregnancy-mortality-surveillance-system.htm (accessed 12/4/2018).

• Building U.S. Capacity to Review and Prevent Maternal Deaths. (2018). Report from nine maternal mortality review committees. http://reviewtoaction.org/Report_from_Nine_MMRCs (accessed 12/4/2018)

• California Maternal Quality Care Collaborative Annual Report 2017 https://www.cmqcc.org/news/cmqcc-releases-online-version-first-annual-report

• Alliance for Innovation on Maternal Health. https://safehealthcareforeverywoman.org/aim-states-systems-2/ (accessed 12/4/2018).

• 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(18):1777-1786.