Introduction

Death by suicide in the perinatal period (pregnancy or the first 12 months postpartum) is a tragedy that can be difficult to imagine. Beyond the loss of life due to perinatal suicide, the effects on family and community are profound and have a lasting impact. It can be challenging to imagine that someone would take their own life during a period that is generally viewed as one of the happiest events in life. Yet death by suicide is a leading cause of maternal mortality and accounts for about 20% of postpartum deaths [1]. Rates of perinatal death by suicide remain high even as maternal mortality due to more commonly recognized causes such as sepsis and hemorrhage decline.

Maternal mortality is a marker of population health and reducing maternal mortality rates is a priority for countries worldwide. It is estimated that between 2000 and 2017, there was a 38% reduction in global maternal mortality ratio (defined as the number of maternal deaths during a given time period per 100,000 live births during the same time period) [2]. Despite this drop across the world, in the USA, the maternal mortality ratio increased by 26.6% between 2000 and 2014 [3]. Although part of this increase is attributed to improved methods of identifying maternal deaths, maternal mortality rates in the USA are the highest among developed countries [3], and suicide is a leading cause of maternal mortality [4•].

Box 1 summarizes commonly used definitions regarding maternal mortality. The World Health Organization (WHO) definitions of maternal death and late maternal death [5] differ from the US Centers for Disease Control and Prevention (CDC) definitions of pregnancy-related deaths and pregnancy-associated deaths [6]. State Maternal Mortality Review Committees or Panels (MMRCs) track pregnancy-associated deaths and pregnancy-related deaths [7].

While there is robust data surrounding medical conditions such as postpartum hemorrhage or hypertension as they relate to maternal mortality, research surrounding suicide and maternal mortality has been relatively limited. This is partly because until recently, deaths related to behavioral health were not considered to be pregnancy-related [8], and not counted towards maternal mortality rates. In 2019, a comprehensive review by Mangla et al. focused on maternal mortality due to suicide and overdose [9••]. In this paper, we assess recently published literature not included in that review, and with a focus only on suicide. We identified key words related to suicidality during pregnancy and postpartum in PUBMED. The search terms used were pregnancy, postpartum, peripartum, antepartum, maternal mortality, suicide, self-injurious behavior, and self-harm, yielding an initial 1995 results. After confining our search to literature published in English from 2018 to 2021, and removing duplicates, we were left with 483 results. We aimed to include articles that discussed suicide, suicidal ideation, and suicide attempts in pregnancy and up to 1 year postpartum. We included case reports, original research, and reviews. On reviewing title and abstract, we narrowed our search to 118 articles and further narrowed to 105 articles after reviewing full-text articles (Table 1). Bibliography search of included articles yielded an additional five relevant articles for a total of 110 articles reviewed. For each of the articles reviewed, we extracted information on prevalence, risk factors, outcomes, prevention, and interventions for perinatal suicide.

Table 1 Summary of all included studies
Full size table

Prevalence of Perinatal Suicide and Perinatal Suicidality

Measurement and Reporting of Perinatal Suicide

Accurate measurement of the prevalence of perinatal death by suicide is challenging for several reasons [10, 11]. Suicide is a rare outcome that is difficult to measure in any population, and additional considerations in measuring perinatal suicide include lack of standardization of terms that describe the timing and cause of maternal deaths, and a lack of uniformity in methods of ascertaining maternal deaths [12]. The broadest definition for maternal mortality is “death in pregnancy or the first year postpartum.” However, up until 1999, the International Classification of Diseases (ICD), ninth revision, definition included deaths during pregnancy and only up to 42 days postpartum as a maternal death. There is also a lack of standardization in how the term “postpartum” is used. It is variously defined as starting from the time of delivery date, registration of a live birth, or registration of fetal death, leading to further variation in reported rates [13].

Another factor to consider in determining the prevalence of perinatal suicide is how maternal deaths are ascertained. The US Centers for Disease Control and Prevention (CDC) uses ICD codes listed in medical records to track maternal deaths. They also use information listed in the death certificate, which has included a pregnancy status checkbox (indicates whether the decedent was or was not pregnant within 1 year of death) since 2003. However, the pregnancy checkbox alone does not address the issue of errors in measurement, and women who died by accidental or incidental means between 2001 and 2008 were not accurately identified as pregnant or postpartum in almost 50% of cases despite adoption of the pregnancy check box on the death certificate [9••].

Most importantly, until recently, deaths associated with suicide and overdose were not included in the “direct” maternal deaths counts [14]. The International Classification of Diseases for Maternal Mortality (ICD-MM) started classifying all suicides in pregnancy and up to 12 months postpartum as direct obstetric deaths only in 2012 [15]. Lommerse et al. examined the effect of the 2012 change in ICD 10 MM and found that reclassifying deaths by suicide as direct deaths did not change maternal mortality rates significantly when limited to 42 days postpartum but did significantly increase rates when extended to 1 year postpartum [16]. Thirteen to 36% of maternal deaths were attributed to suicide when the postpartum period was extended to 1 year postpartum [16].

There are many ways to increase the accuracy of measurement of perinatal suicide prevalence, for example, increasing autopsy rates and using postmortem pregnancy tests [17] and “database autopsy” [18] in which national hospitalization records are used to create a sequential narrative to identify the cause of maternal death. Data gathered from multiple sources such as death certificates, reports of postmortem examinations linked to health administrative data for hospital admissions, and outpatient mental health codes [13] can increase accuracy of prevalence rates and also capture the multiple and variable clinical contacts that can occur through pregnancy and postpartum or shed light on socioeconomic factors which contribute to risk for perinatal mental health problems such as interpersonal violence, housing stability, or poverty [13].

In the USA, the Preventing Maternal Deaths Act was passed in 2018 to standardize the definitions of maternal mortality and fund state MMRCs to better track maternal death [19]. With more states forming MMRCs and standardizing the way in which maternal deaths are measured and reported, it is hoped that we will have more accurate reports of maternal mortality in the years to come.

Prevalence Rates of Perinatal Suicide

The most recent reports on maternal mortality in the USA by the CDC indicate that 754 women died of maternal causes in the USA for a maternal mortality rate of 20.1 deaths per 100,000 live births, more than half of which occurred in the postpartum period [20]. There are severe and concerning racial and ethnic disparities in maternal mortality rates, and the maternal death ratio for Black women is 2.5 times the ratio for White women and three times the ratio for Hispanic women [21]. With regard to suicide as a cause of maternal mortality, perinatal suicide rates in the USA are estimated at 1.6 to 4.5 per 100,000 live births [22]. This compares to 5.3 to 5.5 per 100,000 nonperinatal women aged 10–54 years [23]. Global perinatal suicide rates range between 1.27 and 3.7 per 100,000 live births [22]. Israel reports a lower rate of 0.43 per 100,000 live births, or 3.6% of postpartum mortality [24], reasons for this being religion as a protective factor, or under-reporting of suicide due to mental health stigma. Trost et al. [4•] in reviewing pregnancy-related mental health deaths (including suicides and accidental overdoses) in 14 states, found that, 11% of pregnancy-related deaths were due to mental health conditions, and 63% of pregnancy-related mental health deaths were by suicide. Most (86%) of these deaths occurred in White people, which was a significant difference from the race and ethnicity distribution of pregnancy-related deaths from other causes [4•]. However, racial and ethnic differences in maternal mortality due to suicide can be difficult to evaluate; for example, American Indian/Alaska (AI/AN) maternal mortality data for homicide and suicide often include small samples and may categorize AI/AN maternal deaths in an ‘‘Other’’ race/ethnicity category [25]. Furthermore, an evaluation of suicidal behaviors (thoughts, plans, and attempts) among 7479 pregnant women using the National Survey on Drug Use and Health found that all racial/ethnic groups of women in the third trimester were less likely to be suicidal relative to black non-Hispanic women [26•].

The percent of maternal deaths attributable to suicide varies between states, ranging from 4% in Philadelphia, 5% in Texas and New York, 7% in Virginia and Illinois, to 13% in Colorado [9••].

Prevalence of Suicidality and Suicide Attempts

In considering suicide as a cause of maternal mortality, it is also important to understand the prevalence of suicidal ideation and suicide attempts (sometimes together labeled as suicidal behavior) and self-harm, as these are major risk factors for suicide [27]. However, many reports of prevalence of self-harm are based on hospital discharge diagnoses which may not be accurate or comprehensive and do not capture those who self-harm but are not admitted to the hospital.

Prevalence of suicidal ideation in the perinatal period ranges from 2 to 5% among women seeking obstetrical care [28–32], 10% among women veterans between the third trimester and 6 weeks postpartum [33], 5 to 14% among perinatal women with depression or in mental health treatment [22, 34], 8% among perinatal women living with HIV [35], and 9% among those screening positive for postpartum depression among mothers of babies admitted to a neonatal intensive care unit (NICU) [36]. One cross-sectional study from Ghana reported a high rate of SI of 14.1% among women attending a maternal child health program [37]. A prospective study of 202 women attending prenatal care found that rates of SI decreased from 13.9% in the antenatal period to 6% in the postnatal period [38]. Among racially and ethnically diverse women enrolled in a Women Infant and Children (WIC) clinic, the prevalence of suicidal ideation was almost 5% and 13 times higher among women with depression [39]. In a town along the US-Mexico border, among women living in poverty with depressive symptoms, women > 35 years had higher odds of reporting thoughts of self-harm than younger women [40]. Among migrant and refugee perinatal women living on the Thailand-Myanmar border, 5% (30/568) experienced suicidal ideation [41]. Pregnant women had significantly higher rates of depressive symptoms and were more likely to have thoughts of self-harm after the onset of the COVID-19 pandemic compared to before [42], with rates as high as 20% reported [43].

A systematic review of 14 studies including 6,406,245 pregnant and postpartum women found that the pooled worldwide prevalence of suicide attempts was 680 per 100,000 (95% confidence interval 0.10–4.69%) during pregnancy and 210 per 100,000 (95% confidence interval 0.01–3.21%) during the first year postpartum [44]. Among pregnant and postpartum adolescents in Bangladesh, 6.5% (61/940) reported suicide attempts in the past 12 months, with the majority (88.5%) of the attempts occurring in the first postpartum year [45]. Among 475 women in their third trimester, 0.8% reported self-harm in the current pregnancy [46]. Prevalence of self-harm is as high as 20% among perinatal women with severe mental illness [47], and Vigod et al. found the rate of self-harm in the postpartum period to be 1.11 per 1000 births [48].

Recent studies have addressed the issue of under-reporting and inaccurate ascertainment of suicidal ideation and attempts by using natural language processing in electronic medical records [49]. Using these methods, the prevalence of suicidal ideation and attempts is 515.87 per 100,000 women. In addition to diagnostic codes related to pregnancy or delivery, including codifiable concepts such as “suicidal behavior” from clinical notes increased the sensitivity.

Is Pregnancy Protective of Suicide?

Just as it is challenging to accurately measure the prevalence of suicide in the perinatal period, it is also difficult to determine with certainty how the prevalence of suicide during pregnancy and postpartum compares to the prevalence of suicide among non-perinatal individuals. Comparator groups vary widely across studies, with the age range for “childbearing age” ranging from 10 up to 54 years in different studies [13].

Some studies report that suicide attempt rates are lower in pregnancy compared to pre-pregnancy and postpartum. Older studies report the rate of suicide in the postnatal period to be six times lower than expected in a matched female population [50]. In a cohort study comparing 45,362 women who experienced live birth between 2011 and 2014 to 139,705 age-matched women with no pregnancies during the same period, Mota et al. found lower rates of suicide attempts during pregnancy compared to pre-pregnancy and compared to the postpartum period (0.06% compared to 0.12%). [51]. This study used physician visits and hospitalization for mental health concerns and suicide to calculate prevalence and is possible that rather than reflecting lower symptoms during pregnancy, the findings may reflect decreased mental health service treatment utilization.

Other studies have found higher rates of suicide and suicide attempts in the perinatal period. An analysis of data from 7479 pregnant women in the 2009–2018 National Survey on Drug Use and Health [26•] found the prevalence of suicidal behaviors (including ideation, planning and attempt) to be 3.4%. Prevalence was higher in the first trimester (4.4%) compared to the second/third trimesters (2.9%). Among those with suicidal behaviors, 63.0% had ideation, 18.9% planned suicide, and 18.1% attempted suicide. A retrospective chart review of female patients hospitalized over a period of 1 year found that 33% (9 out of 27)  were pregnant at the time of the suicide attempt [52].

Most studies which attempted to calculate comparative rates of suicide in the perinatal period used non-perinatal women as comparators. To account for any protective effect of motherhood itself, Lysell et al. [53] conducted a well-designed nested case control study of all women who had given birth in Sweden between 1974 and 2009. They used proximity to delivery as the explanatory variable to estimate risk of suicide among mothers. They found only weak negative association between childbirth during the preceding year and suicide, suggesting that previous studies may have overestimated the protective effect of recent delivery on suicide. Severe mental disorder after delivery and a history of self-harm were particularly noteworthy risk factors for suicide in the postpartum year. There is some indication that women who die by suicide during the peripartum period use violent and lethal means more frequently than non-perinatal women, perhaps pointing to the higher levels of distress and psychopathology in these women [54].

The postpartum period appears to be a higher risk period than pregnancy for suicide as two-thirds to three-quarters of all maternal deaths by suicide occur between 6 weeks and 1 year postpartum [16]. To summarize, while there are challenges in study design, it appears that any protective effect of pregnancy and postpartum against suicide, if present, is less than previously suggested, especially among those with severe postpartum mental disorder and history of self-harm.

Trends in Perinatal Suicide Rates

The rate of maternal mortality in the USA has increased from 9.8 per 100,000 live births in 2000 to 21.5 in 2014, a situation unique among high-income countries. Some part of this increase can be explained by improvements in detection such as that due to the addition of the pregnancy checkbox [55]: about 80% of the reported increase in maternal mortality between 2000 and 2014 could be due to the improved reporting [3]. However, even after correcting for improved reporting, the adjusted average MMR across 48 US states is still estimated to have risen by 27% from 18.8 to 23.8 per 100,000 live births from 2000 to 2014.

While perinatal suicide is more common among older non-Hispanic women, there is a disturbing trend of increase (nearly tripling among childbearing people between 2006 and 2017) in suicidality (including suicidal ideation and/or intentional self-harm reported in the diagnosis field among a large commercially insured population) among younger and non-Hispanic Black women [56]. The greatest increases were found among Black, low-income, and younger individuals, and those with comorbid anxiety and depression or serious mental illness. It is of course not possible to state with certainty whether these data reflect actual increases in perinatal suicidality rates or increases in detection and reporting of suicidality. In addition, this study only included women with commercial insurance and the findings may not be generalizable to perinatal individuals insured by Medicaid.

Health Service Utilization Patterns

In a recent review of suicidal behaviors, Meurk et al. [13] comment on the health service utilization behaviors associated with suicide in the perinatal period, highlighting a study in which women who died of suicide were more likely to have had contact with healthcare services between delivery and death [13, 57]. Additionally, they note that peripartum women who died by suicide were more likely to have had a mental health contact (in primary care or in specialty care) in the year prior to their death than peripartum women who did not die.

Risk Factors for Perinatal Suicidality

Demographics

Risk factors for perinatal suicide and suicidal behavior are similar to risk factors in the general population. In the general population, risk factors include younger age, limited education, being unmarried, history of childhood abuse, intimate partner violence, and psychiatric comorbidity [58].

In perinatal populations, younger age, being unmarried, and marital dissatisfaction are risk factors [13, 26, 59, 60, 63], while social support and cohabitation with partner are associated with reduced odds of suicidality [61, 62]. In a study of 762 pregnant Ethiopian women, poor social support was the only variable to have been significantly correlated with suicide attempt during pregnancy [64].

Studies have also suggested that risk may vary by timeline during the perinatal period. An analysis of data from 2009 to 2018 National Survey on Drug Use and Health found that suicidal behavior was more prevalent in the first trimester [26]. Likewise, another study found that intentional poisoning was more common in the first and second trimester relative to the third trimester [65]. Most suicides are late maternal deaths, occurring between 43 and 365 days after the end of pregnancy [66].

Stressors including financial instability and sickness of the new baby have also been associated with increased suicidal behavior [67]. This finding was also supported by a qualitative study of suicidal behavior in adolescent mothers [68]. Among women living with HIV, perinatal depression, anxiety, undisclosed HIV status, HIV stigma, and unplanned pregnancy are associated with suicidal ideation [35, 69, 70].

An observational study comparing postpartum depression and suicidality following cesarian deliveries found increased odds of PPD, suicidal ideation and self-inflicted injury among women who had the cesarian delivery under general anesthesia as compared to neuraxial anesthesia. These findings have not been replicated [71].

Most recently, Trost et al. [4] found that, among pregnancy-related mental health deaths (including suicide and overdose), three-quarters had a history of depression, and more than two-thirds had past or current substance use. They also found that 63% of pregnancy-related mental health deaths were covered by Medicaid during prenatal care or at the time of delivery.

History of Suicidal Ideation and Behavior

Lifetime history of suicidal ideation and behavior are significantly correlated with suicidality during current pregnancy [59]. This finding is replicated in several studies and is similar to the risk conferred by a history of suicidal ideation and behavior in general populations [61].

Comorbid Psychiatric Diagnosis

Having a psychiatric diagnosis is a strong risk factor for suicidality in the perinatal period [72, 73]. Women diagnosed with a postpartum mental disorder have a 6.2 times higher risk for self-harm compared to mothers without mental disorders [74]. Depression is one of the most common diagnoses among perinatal patients who report suicidal ideation [75] or who attempt or complete suicide [62]. For example, in a Japanese study examining suicide attempts during pregnancy, depression was the most common diagnosis, seen in 15.4% of those who made an attempt [76]. This finding was replicated in other studies, in which MDD was found in between 30% [77] and 50% [66] pregnant patients who died by suicide. Furthermore, a recent major depressive episode (within the past 12 months) is associated with 4.9 times higher odds of exhibiting suicidal behavior compared to women who did not experience a recent major depressive episode [26]. Among women with hearing loss, postpartum depression increased the risk of suicidal thoughts as measured by the suicidal ideation attributes scale, but the presence of social support decreased the risk [78]. Depressive symptom severity in perinatal women is positively correlated with suicide score; however, while a decrease in depression severity (as measured by HDRS) was associated with decrease in suicide score, an increase in HDRS score was not correlated with an increase in suicide score [79].

In addition to depression, adjustment disorder may also contribute to increased risk of suicide. In a study of a population in an Irish perinatal mental health service, suicidal ideation or behavior was more common in the group with a diagnosis of adjustment disorder compared with those diagnosed with a depressive disorder, although this difference was not statistically significant [80]. Of note, the population of participants endorsing suicidal ideation or behaviors was relatively small in this study, with only 23 of the total 154 subjects expressing suicidal ideation and 26 demonstrating suicidal behaviors. However, pending larger studies, adjustment disorder can be considered a risk factor for perinatal suicide. Other diagnoses, such as anxiety, bipolar disorder, schizophrenia, post-traumatic stress disorder (PTSD), and personality disorders have been associated with suicide attempts as well [66, 76, 81, 82]. Finally, preliminary evidence indicates that self-reported emotional dysregulation [83] and poor sleep quality [84–86] are associated with antenatal self-injurious thoughts and behaviors.

Comorbid Substance Use

In a study of perinatal suicides in Queensland, Australia, alcohol consumption before death and history of illicit drug use were found in 42% of cases [66]. Pregnant women with alcohol abuse were 3.7 times more likely to feel suicidal compared to those without alcohol abuse [26], and cannabis use during pregnancy and breastfeeding is associated with thoughts of self-harm [87]. Opioid overdose as a cause of maternal mortality has been reviewed in detail elsewhere [9].

History of Trauma

A history of childhood abuse is strongly associated with perinatal suicidal behavior, accounting for 2.57 times increased odds of suicidal ideation, nearly threefold increased odds of suicide planning, and 2.43 times increased odds of suicide attempt [58]. A history of childhood trauma was associated with higher number of suicide attempts among low-income mothers in a home-visiting program [88]. The risk from experiencing depression and childhood abuse is additive: pregnant women with both childhood abuse and depression have increased risk of suicidal ideation compared to those with neither risk factor (OR = 17.78, 95% CI 7.20–43.92) [29, 89]. Those with three or more adverse childhood experiences (ACEs) are more likely than those with no ACEs to have suicidal ideation [90, 91]. Lifetime history of rape has also been associated with higher risk for suicidal behavior [67].

Intimate Partner Violence

Recent Intimate Partner Violence (IPV) is a common risk factor for perinatal suicide [92–94]. A study examining data from 17 states found that more than half of pregnancy-associated suicides involved intimate partner conflict [95]. Increased IPV during pregnancy was also associated with increased suicidal ideation and increased suicide attempts [96]. In a study from Ghana, current partner abuse was associated with 6.5 times increased odds for suicidal behaviors (OR = 6.5, 95% CI 1.14–37.05) [97]. In Brazil, among women of low income who reported postpartum suicidal ideation, 70% reported IPV during the postpartum period [98].

Stillbirth/Termination of Pregnancy

Pregnancy loss may be associated with increased risk of suicide. The prevalence of SI among women who screen positive for depression after a miscarriage was reported to be 33.1% [99]. A nested case–control study linking three nationwide population-based data sets in Taiwan found 485 cases of attempted and 350 cases completed suicide, and matched each case with ten controls. They found that the risk of completed suicide was higher in women who experienced a stillbirth [adjusted odds ratio (aOR) 5.2; 95% CI 1.77–15.32], miscarriage (aOR 3.81; 95% CI 2.81–5.15), or termination of pregnancy (aOR 3.12; 95% CI 1.77–5.5) than in those who had a live birth [81]. However, a small follow-up cohort study did not find any difference in suicidal ideation or depression diagnosis at 1 year among women with perinatal loss compared to those without [100]. In a Danish population cohort study, women who had abortions were more likely to have suicide attempts; however, the risk of nonfatal suicide attempt was the same in the year before the abortion compared to after the first abortion, which suggests that this risk was not attributable to the abortion itself [101].

Other Factors Associated with Perinatal Suicidality

Other studies have found factors such as bonding impairment with the baby, personality traits such as neuroticism and psychoticism, and hyperemesis gravidarum to be associated with perinatal SI [102–104]. However, these findings need to be replicated.

Outcomes of Non-fatal Suicide Attempts

Obstetric risks in pregnant women with suicidal behavior include increased risk of antepartum hemorrhage, placental abruption, postpartum hemorrhage, premature delivery, low birth weight [105], stillbirth, poor fetal growth, and fetal abnormalities [106]. Pregnant adolescents who are at risk of suicide during pregnancy have a nearly twice higher risk of giving birth prematurely [107]. Violent methods of suicide attempts are associated with higher risk of critical perinatal outcomes [76]. In a limited study of 10 case reports of suicide attempts during pregnancy by overdose, although some agents resulted in bleeding and pregnancy complications, and in one case spontaneous abortion, overall, these pregnancies resulted in a healthy live birth [108]. In three cases of intentional overdose via over-the-counter medication during pregnancy, two of the three pregnancies ended in fetal demise [109]. Infants of women with perinatal suicidality had lower scores in neuropsychological development at 4 months postpartum [61].

Prevention and Intervention Strategies

Prevention and intervention strategies recommended specifically to address the risk of perinatal suicide include screening for mental health problems and suicidality, treatment of underlying psychiatric conditions, and improving access to care by using telepsychiatry and integrated mental health treatment approaches.

Screening for mental health problems, including suicidality, during the perinatal period is critical in efforts to help prevent suicide; however, universal screening is not yet widespread in practice. In a study of maternal suicides in Queensland, less than half of women who gave birth had records of screening for suicidal ideation, thus limiting the opportunities to identify those at risk [66]. In the USA, the American College of Obstetricians and Gynecologists currently recommends screening patients for depression at least once during the perinatal period using a validated tool [110]. The Edinburgh Postnatal Depression Scale (EPDS) is a validated tool that has been translated in 50 different languages, which consists of 10 questions, including one that asks specifically about thoughts of self-harm. The Patient Health Questionnaire-9 (PHQ-9) has also been validated in the perinatal population and similarly includes a question about passive or active suicidal ideation. If there is an affirmative response to either question above, clinicians should assess frequency and intensity of suicidal or self-harm thoughts, potential methods/plans, intent, reasons for living, and include assessment of suicide risk factors to assess overall suicide risk and determine appropriate level of clinical intervention (e.g., hospitalization, emergent psychiatric evaluation, ongoing outpatient care with pharmacotherapy and/or psychotherapy) [22]. For those with higher levels of risk, collateral information can be obtained from family or significant other who can also assist with means reduction from the patient’s home [22].

There are many opportunities to screen for mental health problems and for suicidality among reproductive-aged women as they frequently interact with healthcare services [111]. However, non-psychiatric provider may not be comfortable assessing next steps when a patient screens positive for suicidality. The use of specific screening tools for suicide, including the National Institute of Mental Health “Ask Suicide-Screening Questions” and Columbia Suicide Severity Rating Scale [112], can help non-mental health clinicians assess whether emergent psychiatric evaluation is needed.

While increased screening is a key component in helping identify women at risk of suicide, screening alone cannot mitigate the risk of suicide. In order to prevent and intervene on suicidality, patients must receive mental health treatment and/or interventions for suicidality. A systematic review of 41 studies found that almost three-fifths of women who screen positive for perinatal depression do not take up referral offers after screening [113]. A 15-year UK study found that perinatal women who died by suicide were half as likely to be receiving pharmacological, psychological, or any other treatment at the time of their death compared to non-perinatal women who died by suicide [114]. Thus, connecting women at risk for suicide to treatment is of utmost importance. The entire process in suicide prevention through early screening, assessment, monitoring, and intervention is important for all perinatal patients, with some recommendations calling for OB/GYNs to screen at every contact during the first postpartum year, using the EPDS or PHQ-9. Those who report suicidal ideation should receive a suicide risk assessment with discussion of lethal means restriction, involving a patient’s significant other or family member if possible. Pharmacotherapy can be considered for patients with mental health symptoms who have suicidal ideation, but who do not endorse plan or intent [22].

Telepsychiatry and integrated mental healthcare hold promise in bridging the gap between patients screening positive in the obstetrician’s office and getting connected with mental health services, including in rural areas or areas with limited resources [115]. Programs such as Massachusetts Child Psychiatry Access Project (MCPAP) for Moms and North Carolina Maternal Mental Health MATTERS (Making Access to Treatment, Evaluation, Resources, and Screening Better) Program help build primary care and obstetric providers’ capacity to treat perinatal depression though education, telephone perinatal psychiatric consultation, and care coordination [116]. These programs provide outreach and education for perinatal providers, screening toolkits for mental health disorders, a consultation line for providers, care coordination, and a perinatal telepsychiatry clinic [116]. Telepsychiatry and integrated care approaches may be able to reduce maternal mortality through improved access to perinatal mental health and substance use disorders services [117].

While provider-facing resources are important resources, patient-facing perinatal helplines may also help patients in crisis by increasing accessibility and decreasing barriers to mental healthcare. The Perinatal Anxiety & Depression Australia National Helpline (PANDA) in Australia identified one-third of callers as being at risk at intake; 73% of those at risk were deemed at risk due to their mental health status, with 40% endorsing suicidal thoughts and 7% endorsing self-harm [118, 119]. One intervention provided to 6.3% of callers during the study period was the ASIST (Applied Suicide Intervention Skills Training) Suicide Intervention Model [119].

Prompt recognition and treatment of mental health disorders that are associated with suicide and self-harm are an important consideration. For example, in women with postpartum psychosis, swift identification, treatment (ideally through inpatient care on a mother-infant unit), and management are needed to ensure safety and prevent suicide [22, 120]. Recognition and treatment of perinatal mental health disorders should extend beyond severe disorders such as psychosis to common mental disorders such as depression, which as we have seen, is an important risk factor for perinatal suicide [121].

There are few reports of treatments specifically geared toward treatment of perinatal suicidality. Antidepressants are the mainstay of treatments for moderate to severe perinatal depression, and adequate treatment of depression may be expected to reduce the risk of suicidal ideation associated with perinatal depression, although this has not yet been systematically studied. One study found no association between antidepressant treatment in pregnancy and reduced self-harm ideation postpartum as measured by the EPDS [122]. However, this finding has not been replicated and the study has several shortcomings such as retrospective report of antidepressant use, cross-sectional design, and possibility of unmeasured confounders. Medications such as lithium and clozapine have been shown to reduce the risk of suicide in general populations among those with bipolar disorder and schizophrenia [123, 124]. However, lithium and clozapine are not used frequently among pregnant or postpartum individuals given concern for teratogenicity and adverse fetal effects, and there are no studies of their use to reduce perinatal suicidality. We found one case report describing the use of esketamine for a woman who developed depression with psychotic features and attempted suicide at 4 months postpartum. The use of subcutaneous esketamine led to remission of her suicidal ideation [125]. Additionally, the use of prophylactic ketamine in women undergoing cesarean section was associated with lower incidence of postpartum depression even among those who experienced antenatal depression and suicidal ideation [126]. However, this study only included pregnant women undergoing caesarean section and excluded patients with known unstable psychiatric disorders, and it may not be generalizable to the perinatal population most at risk for suicide. Regarding psychotherapy, dialectical behavior therapy is often an approach used to treat suicidality in patients [127]. However, little research has been done to examine its effect in reducing suicidality in the perinatal population. Aspects unique to the perinatal period such as the mother-baby relationship may need additional assessment and intervention in the context of suicidality. Perinatal women with thoughts of self-harm reported higher levels of lack of affection and rejection towards the baby [128], and SI in the postpartum period has been associated with poor quality of mother-baby interactions [129]. There is a critical need for more research into interventions for suicidality in the perinatal period.

Several system-level interventions have recently focused on reducing perinatal suicidality. A narrative review describing initiatives to reduce maternal mortality in the USA, while not focused on mental health related deaths, describe many initiatives which apply to suicide prevention [130]. One example of a system-level intervention is the Preventing Maternal Deaths Act (H.R. 1318) which authorized federal funding for state MMRCs. Other important interventions include data collection through the Maternal Mortality Review Information Application (MMRIA) Data System, and telehealth access program funding through HRSA’s Screening and Treatment for Maternal Depression and Related Behavioral Disorders program.

Conclusions

Maternal deaths are an indicator of the overall public health of a country. While there is still work to be done to improve the measurement and reporting of maternal deaths, in the USA, the establishment of maternal mortality review panels in several states is an important first step toward obtaining uniform and high-quality data regarding perinatal deaths. There is also important ongoing research to develop ways to predict risk for perinatal suicide, for example, using natural language processing in clinical notes [49]. In the meantime, there are several interventions that can help reduce maternal mortality, including universal screening for perinatal depression and substance use disorder, and integrating mental health treatments into primary and prenatal care. In addition, clinicians should be aware of important risk factors for perinatal suicide such as mental health and substance use diagnoses, interpersonal violence, a history of abuse, and poor social support. MMRC recommendations can inform public health strategies to mitigate the risk for maternal mortality from various causes including suicide. Given that many individuals who attempt suicide or die from suicide in the perinatal period use health services, there are opportunities to identify and intervene for suicide risk, and this is an important step in reducing maternal mortality rates.