Expert Q&A: Measuring and informing policy to address gaps in maternal mental health care

Kara Zivin, professor of psychiatry and obstetrics and gynecology, discusses her research and efforts to inform policy to improve mental health services for pregnant and postpartum women.

Despite overall improvements in maternal health across the globe, the maternal mortality rate in the United States has been steadily rising and remains the worst among developed countries.

According to a Centers for Disease Control and Prevention (CDC) report released in January 2020, the U.S. had 17.4 maternal deaths per 100,000 live births in 2018, corresponding to an estimated 658 lives lost. A recent World Health Organization (WHO) report showed the U.S maternal mortality rate increased by 25% from 2000 to 2015, while the global rate had decreased.

Yet, the WHO definition of maternal mortality excludes suicide and overdose, meaning that these already sobering statistics underestimate the societal burden of maternal mortality and mental illness.

“What we measure drives what we do,” says Kara Zivin, Ph.D., M.S., M.A., a professor of psychiatry and obstetrics and gynecology at U-M Medical School and a researcher with the U-M Institute for Healthcare Policy & Innovation (IHPI). “By not including suicide and overdose in the maternal mortality rate, we are missing out on a critical opportunity to better understand and mitigate perinatal mental disorders and suicidality -- a growing public health crisis.”

Zivin and her colleagues study maternal mental illness and suicidality trends in the U.S. to address this lack of data. Her recent work, developed in collaboration with several other IHPI researchers, includes:

A new study that shows the prevalence of suicidal ideation and self-harm occurring in the year preceding or following birth increased substantially over a twelve-year period, published in JAMA Psychiatry.
A study published in BMC Women’s Health which found that the prevalence of both perinatal mood and anxiety disorders and serious mental illness among delivering women increased substantially over the past decade.
Published in the American Journal of Public Health, a study examining the societal costs of not treating perinatal mood and anxiety disorders determined that maternal mental illnesses cost over $14 billion per year in the U.S. in 2017.

In this Q&A, she discusses her research and her efforts to inform policy to improve mental health services for pregnant and postpartum women.

What inspired your research on maternal mental health and mortality?
Zivin: Throughout my career as a mental health services researcher, I have focused on studying access to care, treatment, quality of care, and the impacts of various policies on people with mental disorders. I focused primarily on examining mental health care among older adults and veterans with mental illness for several years. Then, I became pregnant and experienced a difficult case of perinatal depression, anxiety, and insomnia during my pregnancy and afterward. After struggling to talk about my experience publicly for a long time, I started writing and examining maternal mental health through a research lens, which ultimately led me to focus on this issue as part of my broader work.
Your recent body of research shows a steady increase in the prevalence of perinatal mood disorders, serious mental illness, and suicidal ideation among delivering women in the U.S. What factors may have contributed to these increases?
Zivin: I do not have a single answer to this question, as a wide array of complex factors influenced our findings. One issue my team and I highlighted in our JAMA Psychiatry paper is a recent shift in diagnosis codes used to document mental disorders in patient records, which could have led to the increase in recorded cases. Other factors to consider include increases in screening and greater public awareness and discussion about mental illnesses between providers and patients. Of course, societal, interpersonal, financial, genetic, and other factors also influence the prevalence and patterns of illness. Altogether, it remains difficult to pinpoint how much these factors contribute to this increase versus an actual increase in mental disorders among pregnant and postpartum women.
Much of your work focuses on collecting data to inform mental health policy. What more do you think should be done on a policy level to address maternal mental illness and mortality, as well as disparities among populations?
Zivin: One policy option would involve creating more financial incentives to encourage providers to offer mental health services to pregnant and postpartum women. Policymakers could also implement or extend quality metrics so that providers and health plans change their behaviors to focus more on maternal mental health. Unfortunately, mental health care may not always be fully addressed during visits with some providers and within certain health plans. Financial incentives represent an effective tool in driving more screening and follow-up for patients.

Moreover, data shows that Black, Native American, and Alaska Native women have two to four times the risk of maternal mortality than White women, reflecting widespread systemic racism and the need for policies and programs to address racial disparities.
Despite the WHO not including suicide or drug overdose in its definition of maternal mortality, how can we better measure maternal suicide?
Zivin: One of the potential reasons that the WHO does not include suicide or overdose in its maternal mortality rate is inconsistent data reporting. In the U.S., many changes have taken place regarding reporting criteria on the time period and nature of maternal mortality, including how long after delivery a death counts as a maternal death. For instance, the WHO defines maternal death as “the death of a woman while pregnant or within 42 days of termination of pregnancy,” whereas some experts in the U.S. have extended this definition to within a year of the end of a pregnancy.

Death certificates also have a pregnancy checkbox, which not every medical examiner uses consistently, meaning that someone could die without documentation of pregnancy. Practices also vary widely by state -- most states have their own independently operating maternal mortality review committees, multidisciplinary panels of health experts that perform comprehensive reviews of deaths among pregnant and postpartum women.

Switching to a centralized system of data collection for maternal suicide data during pregnancy and up to a year postpartum through the CDC or another national organization would improve the measurement of the problem and support data-informed policy solutions.
What more should doctors and clinicians do to support pregnant and postpartum women with mental illness?
Zivin: First, we need to do more mental health screening. The American College of Obstetricians and Gynecologists recommends that obstetrician–gynecologists and other obstetric care providers screen patients at least once during the perinatal period for depression and anxiety symptoms using a standardized, validated tool, yet patients could benefit from more frequent screening both during pregnancy and postpartum, as symptoms can change over time.

Next, we need to ensure connections to follow-up care. It is not enough to diagnose someone with depression or anxiety -- providers need to make sure they develop a care plan and connect the patient with the appropriate resources. In part, this depends on where a woman gets screened. For example, if a new mother brings her child to a pediatrician’s visit and the pediatrician happens to screen the mom for depression and identifies her as having a depression diagnosis, the pediatrician will not treat the mother, so they need to direct her to a therapist or her personal care provider. Such coordination among different types of care providers and systematic approaches to mental health care will influence the mother and the child’s well-being.
What do you have in store for future research?
Zivin: In addition to examining the impact of federal health policy changes on treatment for perinatal mood and anxiety disorders, we also plan to focus on perinatal substance use disorders and how they affect expecting women. Specifically, we will explore the use of opioids and the impact of mothers’ substance use on infants. We will also examine impact of policies on maternal morbidity and mortality. I look forward to continuing and expanding this critical research.