Research and Professional Literature to Inform Practice, May/June 2025

California has been a national leader in efforts to reduce maternal deaths. It was one of the first states to organize a maternal mortality review committee, and the California Maternal Quality Care Collaborative has actively promoted statewide quality improvement projects aimed at making labor and birth safer. An analysis of pregnancy-related deaths (See Box 1) in California from 2002 through 2007 found that obstetric hemorrhage was a major cause and that uterine atony was the primary cause of death from obstetric hemorrhage. Following this finding, a major statewide quality improvement process occurred from 2015 to 2016, aimed at reducing morbidity following obstetric hemorrhage. The current study was designed to assess deaths related to obstetric hemorrhage after this quality improvement program and to identify further opportunities for improvement. Studies of maternal mortality differentiate between pregnancy-associated and pregnancy-related deaths. Pregnancy-associated deaths are those that occur during pregnancy or within one year of the end of pregnancy regardless of pregnancy outcome and regardless of whether or not the death was from pregnancy causes. A pregnancy-related death is one that occurs during pregnancy or within one year of the end of pregnancy regardless of pregnancy outcome that is from causes related to pregnancy or treatment received during pregnancy. A death from a ruptured ectopic pregnancy is pregnancy-associated and pregnancy-related. A death of a person from a motor vehicle accident 6 months postpartum is pregnancy-associated but not pregnancy-related, because the pregnancy was not a cause of the death. The California Pregnancy-Associated Mortality Review Hemorrhage Committee is a multidisciplinary group of 10 individuals who met over a period of 6 months to perform in-depth reviews of all deaths from obstetric hemorrhage in California from 2014 to 2018. They reviewed case summaries compiled by nurse-abstractors, who summarized data from vital records, coroner or autopsy reports, and prenatal and hospital records. The committee discussed each case in-depth to determine cause of death, contributing factors to the death, and opportunities for improving care. In California between 2014 and 2018, there were 2,409,732 live births, with 328 pregnancy-related deaths, 49 of which were due to obstetric hemorrhage. Unlike during the years 2002 to 2008, placenta accreta spectrum was the largest underlying cause of hemorrhage deaths (32.7%). Uterine atony and intra-abdominal bleeding each accounted for 20.4% of deaths, and ruptured ectopic pregnancies (12.2%), placental abruptions (8.2%), and uterine ruptures (6.1%) each accounted for smaller numbers. Risk factors significantly associated with deaths from hemorrhage included age over 35 years, being born in China, having public insurance, preterm birth, and birth at a hospital with fewer than 1200 births per year. Only 5 of the 49 deaths occurred in first pregnancies. Deaths from obstetric hemorrhage occurred in 0.61 per 100,000 live births for vaginal births but in 2.48 and 5.22 per 100,000 live births for primary and repeat cesareans, respectively. No deaths were associated with births in freestanding birth centers or at home with a skilled attendant, but there were 3 deaths associated with unattended home births. In a relatively small sample such as this, it is not possible to tease out the relationships between all the risk factors to determine which are causal. Older age may be an independent risk factor for hemorrhage death, or it may simply be associated with higher parity and greater likelihood of having had a previous cesarean birth. Cesarean birth may itself increase risk of death from hemorrhage, but is also related with occurrence of placenta accreta, which may be the underlying cause. Precise causation is not established, but the analysis does identify people at risk, allowing providers to target interventions. The committee deemed that the majority (66.7%) of hemorrhage deaths were highly preventable. Deaths from uterine atony and intra-abdominal bleeding were most likely to be identified as preventable. The most common clinician, facility, and system-level factors identified as contributing to deaths were delayed response to hemorrhage (77.8%), delayed recognition (72.2%), and insufficient quantities of blood products (52.8%). The most common patient level factors included anemia (22.2%) and presence of other chronic conditions (19.4%). The Centers for Disease Control and Prevention encourages state maternal mortality review committees to consider whether discrimination contributed to pregnancy-related deaths. In this review, the committee members determined that discrimination was likely involved with 12 (24.5%) of hemorrhage deaths. Discrimination was attributed to client race, ethnicity, income or insurance status, language or immigration status, or substance use disorder. Examples of discrimination included failure of staff to follow-up on family concerns in a timely fashion, inadequate assessment caused by lack of translation services, and attributing symptoms to substance abuse despite the presence of negative toxicology screens. The committee members concluded that the statewide quality improvement program aimed at reducing hemorrhage deaths was successful at decreasing deaths from hemorrhage due to uterine atony, but additional efforts may further reduce deaths. Health care facilities need to quickly identify and respond to symptoms of hemorrhage, with aggressive management of resuscitation and blood products replacement. The shift to more deaths caused by placenta accreta spectrum was concerning, and members thought the change was associated with the increasing percentage of cesarean births. They made strong recommendations that quality improvement programs should focus on preventing primary cesarean births and encouraging trials of labor after cesarean birth. Providers should consider referring clients with prenatally diagnosed placental abnormalities to high-volume hospitals for birth. They also recommend that providers treat anemia during pregnancy more proactively, to decrease the likelihood of death should hemorrhage occur. It is difficult to determine from the written health record how discrimination may have influenced outcomes, so the committee identification of discrimination in a quarter of deaths is likely an undercount. To address discrimination, the committee recommends that facilities expand use of qualified interpreters and enact antiracist policies and practices. An observational study cannot prove causality, and the small number of deaths makes it difficult to determine precise relationships between all the factors associated with mortality. However, the large, 5-year, population-based sample provides a strong basis to support the effectiveness of quality improvement programs to prevent mortality from hemorrhage due to uterine atony. Midwives can continue to encourage health care facilities to improve care for obstetric hemorrhage, using resources such as the Alliance for Innovation on Maternal Health Obstetric Hemorrhage Safety Bundle (https://saferbirth.org/revised-obstetric-hemorrhage-patient-safety-bundle), as well as work for cesarean prevention to reduce occurrence of placenta accrete syndrome. All providers must work to reduce implicit bias and discrimination in health care. Attention-deficit hyperactivity disorder (ADHD) affects approximately 3% to 4% of pregnant people, frequently coexisting with other neurodevelopmental disorders or psychological illnesses.1 In adults, ADHD is characterized by inattention, disorganization, and/or hyperactive or impulsive behavior.2 Research shows a correlation between presence of ADHD and various negative reproductive outcomes. People with ADHD are more likely to experience early and unplanned pregnancies. In one study, those with coexisting ADHD and depression who discontinued ADHD medication during pregnancy had a higher increase in depressive symptoms than those remaining on their medications. Parental distress is higher in new parents with ADHD. Although the reason ADHD affects pregnancy outcomes is uncertain, people with ADHD may find many pregnancy tasks that require executive functioning to be challenging, including making and keeping many health care appointments, completing paperwork, and preparing the home for the newborn.1 Medication treatment for ADHD could potentially improve pregnancy outcomes by decreasing impulsive behavior and assisting the pregnant person with adhering to appointment schedules and healthy behaviors. Most adults, however, do not receive medical treatment for ADHD, and little is known about use of medication for ADHD during pregnancy. The goal of the current study was to determine the proportion of pregnant people using medication to treat ADHD prior to pregnancy who continued medication throughout the pregnancy and the postpartum period. The researchers conducted a retrospective cohort study using data from the Merative MarketScan Commercial Database from 2011 to 2021. This is a large, national database of insurance claims information for privately insured people. The researchers used International Classification of Diseases, 10th Revision codes to identify people diagnosed with ADHD who filled prescriptions for ADHD medications consistently during the 6 months prior to pregnancy, and assessed whether they filled prescriptions throughout the pregnancy and in the postpartum period. The researchers identified 3676 people who used ADHD medications continuously prior to pregnancy. Only 7.8% of these people filled their prescriptions continuously throughout the pregnancy, with 45.2% stopping as early as the first trimester. The majority (57.5%) did not resume filling prescriptions in the 90 days postpartum. The authors concluded that many people with ADHD discontinue their medication during a vulnerable period when medication use could benefit them. This study benefits from use of a large, national database. It has many limitations. This data set includes only clients with private insurance; patterns of use may be different among people with public insurance. There is no information about severity of ADHD symptoms, nor how the clients’ pregnancy care providers counseled them. The 11-year time of the study provides the possibility that patterns of use may have changed over time. Most medications prescribed for ADHD have relatively safe profiles for use in pregnancy and lactation.1 Given this safety record and the challenges pregnant and people and new parents face, continuing medication may be a good choice for many people. Shared decision-making, preferably in the preconception period, could help clients make informed choices about discontinuing ADHD medications, changing to a medication with an optimal safety profile, lowering their dosages, or continuing on their current medications. Health care providers can give reassuring information on safety of these medications, and support clients in their choices. In the United States, although pregnant people of all races and ethnicities express similar intention to breastfeed, Black and Hispanic people are less likely than White people to achieve their breastfeeding goals. Over the past 30 years, the Baby-Friendly Hospital Initiative (BFHI) has successfully increased the quality of breastfeeding support that postpartum dyads receive in the hospital. Its “Ten Steps to Successful Breastfeeding” describe evidence-based clinical practices that support breastfeeding, which have been adopted by many hospitals and are considered the gold standard of breastfeeding support. The aim of the current study was to determine whether postpartum clients of different races and ethnicities reported receiving different BFHI supportive practices while in the hospital after birth. This study is an analysis of data from the National Pregnancy Risk Assessment Measurement System Survey (See Box 2). This national survey contains a number of questions on breastfeeding, and the researchers were able to map 8 of the 10 Steps to Successful Breastfeeding to specific questions on the survey. The sample included people with a full-term, singleton, live birth in US hospitals between 2016 and 2019. Race and ethnicity were self-reported. To explore the potential role of language barriers, people who reported their ethnicity as Hispanic but answered the survey in English were designated “Hispanic-English speaking” and those who reported their ethnicity as Hispanic and answered the survey in Spanish were considered “Hispanic-Spanish speaking.” The National Pregnancy Risk Assessment Measurement System Survey is a survey administered continuously by the Centers for Disease Control and Prevention in coordination with state health departments. It surveys people at 2 to 6 months postpartum about their experiences during pregnancy and the postpartum period. By contacting a national sample, using birth certificate information, the researchers are able to construct a mathematical sample that is broadly representative of the entire population of the United States. The researchers analyzed results from 28,032 respondents. The majority (56.2%) identified as White non-Hispanic. Smaller numbers identified as Black non-Hispanic (13.3%) and Hispanic-English speaking (11.6%). Les than 10% identified as Hispanic-Spanish speaking, Asian or Pacific Islander non-Hispanic, American Indian or Alaska Native non-Hispanic, or mixed race non-Hispanic. The vast majority of respondents reported receiving some of the 10 steps: receiving information about breastfeeding, being provided information on breastfeeding from staff, rooming-in, and being encouraged to engage in cue-based infant-driven feeding. Other steps were experienced less frequently. About half of respondents reported their babies received a pacifier while in the hospital, in contradiction to the 10 steps. Although more than half of White non-Hispanic respondents reported that their babies only received breast milk in the hospital, this was true for only about one-third of Black non-Hispanic, Spanish-speaking Hispanic, and Asian or Pacific Islander respondents. After multivariable analysis to account for different risk characteristics, non-Hispanic White respondents were more likely to report skin-to-skin contact and breastfeeding in the first hour after birth and feeding only breastmilk. Spanish-speaking Hispanic participants reported 32% lower odds of having been provided information on breastfeeding by staff and 43% lower odds of having received a phone number for breastfeeding help after discharge. There was no difference in receipt of key clinical practices between American Indian/Alaska Native non-Hispanic respondents and non-Hispanic White clients, which the authors attribute to universal adoption of the BFHI by hospitals in the Indian Health Service, spearheaded by then First Lady Michele Obama. The authors conclude that Black Non-Hispanic and Spanish-speaking Hispanic postpartum people receive ideal breastfeeding care in the hospital at only half the rate of White non-Hispanic clients. This disparity in receipt of ideal care may be related to the disparity in breastfeeding outcomes. The authors comment that people often attribute health care disparities to social or economic factors, and indeed, one explanation for this disparity in the receipt of care may be that hospitals that serve larger proportion of non-Hispanic Black individuals may in fact provide lower quality of care. However, it is also possible that health care providers give different quality of care to different groups of people. They may hold implicit biases about the desire or ability of members of minoritized groups to breastfeed and, thus, provide less information and support. Providers may also give non-English speakers less education and informational resources. The authors conclude that ensuring equitable breastfeeding education and support in the hospital is necessary to address racial disparities in breastfeeding outcomes. For planned cesarean birth with spinal anesthesia, the current standard practice for preoperative oral intake is nothing from midnight to time of surgery. This conservative policy is theorized to prevent aspiration of gastric contents should general anesthesia be needed. However, previous studies have found low incidence rates of clinically significant aspiration during general anesthesia for cesarean birth. One study found an aspiration rate of 0.009%, none of which led to death.1 In this randomized controlled trial, researchers evaluated the impact of free water access up until admission to the operating room as compared with fasting from midnight on perioperative vomiting and maternal satisfaction in planned cesarean births under spinal anesthesia. Secondary outcomes were feeling of thirst, capillary blood glucose level, preoperative intravenous hydration, ketonuria, and intraoperative vasopressor doses. The trial was conducted at the University of Malaya Medical Centre obstetric unit and was a randomized, controlled, parallel-design clinical trial. Inclusion criteria included planned cesarean birth under spinal anesthesia, age greater than 17 years, term gestation, singleton pregnancy, and reassuring fetal heart rate tracing. Clients were excluded if they had a body mass index greater than 40, had placental pathology, anticipated intensive care unit admission, were in labor, were on insulin therapy, or were SARS-CoV-2 positive. Between October 2020 and May 2022, researchers randomized 504 total participants to preoperative unlimited water access (252) or fasting from midnight (252). Participants in the intervention group were permitted light meals until 2 am and free water access until operating room admission. The primary outcome of vomiting was defined as vomiting up to 6 hours after completion of cesarean birth. Client satisfaction was measured using an 11-point visual numerical rating scale with the statement, “Rate your experience of your preoperative oral intake care experience.” A score of 0 meant completely dissatisfied, and a score of 10 meant completely satisfied. In the intervention group, vomiting rates were significantly lower (3.6%) as compared with the control group (9.5%) (P = .007). Concurrently, median client satisfaction scores were significantly higher in the intervention group (9, interquartile range [IQR] 8-10) compared with the control (5, IQR 3-7) (P < .001). For secondary outcomes, compared with the control group, the intervention group reported less thirst (27.4% v 53.2%) (P < .001), preoperative intravenous fluid hydration was reduced (19.4% v 30.2%) (P = .005), and detection of ketones in catheterized urine was reduced (15.1% v 31.0%) (P < .001). The number of vasopressor doses to correct hypotension was 2.3 ± 1.7 compared with 2.7 ± 2.2 in the control (P = and the capillary glucose were not different between the intervention group and the control ± ± (P = There were no significant in secondary outcomes and The of this study sample data and due to This study was potentially to It was also conducted at a facility, reducing The researchers concluded that for planned cesarean birth under spinal anesthesia, preoperative free access to water as compared with fasting from midnight significantly reduced perioperative vomiting rates and increased maternal satisfaction In the intervention group, secondary of thirst, capillary blood glucose preoperative intravenous hydration, ketonuria, and intraoperative vasopressor doses were significantly and participants were more likely to recommend their oral intake to a Midwives can use these study to within their for preoperative oral intake a more experience for their clients. 17 states have completely or after 6 of pregnancy, some with and some for fetal The of this study was to determine the between and mortality and to determine if the has been different in different The study used data from birth and death for all states and the of from through The authors considered the states that between and 2022, as the states with and of death rates is because these rates change over study of deaths over the past is by the occurrence of the which had a on death mortality in the United until 2020 and has To account for this the authors used a to the between an and an outcome when the outcome is measured as a in this case number of deaths per year. They used the in states to the mortality rates have been in states with if the had not is by the largest of the states with an the researchers the including and The researchers that states with had a higher incidence of mortality than they have had the to an deaths per live births. The majority of deaths occurred in Black had a larger increase in deaths than White with an higher rate than be Other racial and groups had deaths to make an The largest increase was in deaths due to but there was also a smaller increase in death due to other causes The authors conclude that increase with the among vulnerable that have the rates of mortality. The increase in mortality due to fetal is but deaths from other causes are more difficult to The researchers that this is due to the of on vulnerable Health care providers need to the that have on reproductive The Committee on the for or United States, at October with to The age for receiving has from years to The committee recommends that people and should receive 2 or more doses of the for is or more the on was changed to that of health care providers is not and specific of have changed for The is at