Prenatal origins of suicide mortality: A prospective cohort study in the United States

Abstract

Most suicide research focuses on acute precipitants and is conducted in high-risk populations. Yet, vulnerability to suicide is likely established years prior to its occurrence. In this study, we aimed to investigate the risk of suicide mortality conferred by prenatal sociodemographic and pregnancy-related factors. Offspring of participants (N = 49,853) of the Collaborative Perinatal Project, a U.S. population-based cohort of pregnancies enrolled between 1959 and 1966, were linked to the U.S. National Death Index to determine their vital status by the end 2016. We examined associations between sociodemographic factors during pregnancy, pregnancy complications, labor and delivery complications, and neonatal complications with suicide death coded according to ICD-9/10 criteria. By the end of 2016, 3,555 participants had died. Of these, 288 (214 males, 74 females) died by suicide (incidence rate = 15.6 per 100,000 person-years, 95% Confidence Interval [CI] = 13.9–17.5). In adjusted models, male sex (Hazard Ratio [HR] = 2.98, CI: 2.26–3.93), White race (HR = 2.14, CI = 1.63–2.83), low parental education (HR = 2.23, CI = 1.38–3.62), manual parental occupation (HR = 1.38, CI = 1.05–1.82), being a younger sibling (HR = 1.52, CI = 1.10–2.11), higher rates of pregnancy complications (HR = 2.36, CI = 1.08–5.16), and smoking during pregnancy (HR = 1,28, CI = 0.99–1.66) were independently associated with suicide risk, whereas birth and neonatal complications were not. Consistent with the developmental origins of psychiatric disorders, vulnerability to suicide mortality is established early in development. Both sociodemographic and pregnancy factors play a role in this risk, which underscores the importance of considering life course approaches to suicide prevention, possibly including provision of high-quality prenatal care, and alleviating the socioeconomic burdens of mothers and families.

Children's cognitive performance and suicide risk through middle adulthood

“…Table 1 shows the distribution of neurocognitive, sociodemographic, and prenatal factors for suicide decedents and all other participants, the corresponding incidence rates of suicide per 100,000 person-years, and the proportion with missing data. As reported previously (Vidal-Ribas, Govender, Sundaram, et al, 2022), suicide rates were higher for white offspring, males, and for offspring with higher scores on the sociodemographic disadvantage index and the Perinatal Risk Index. For intelligence scores, suicide rates were lower for children with standardized scores above 115.…”

“…learning processes or socioemotional skills), parent-child interactions, and access to stimulating environments have an impact on language development, including literacy (Pace, Luo, Hirsh-Pasek, & Golinkoff, 2017). In the current study, the strength of the associations of spelling and reading skills with suicide risk was attenuated when adjusting for early sociodemographic and prenatal factors previously linked to suicide among CPP offspring (Vidal-Ribas, Govender, Sundaram, et al, 2022). In previous studies in children and adolescents, the associations between neurocognition and suicide risk were due to early life socioeconomic and perinatal factors (Osler et al, 2008;Richard-Devantoy et al, 2021).…”

“…We determined the vital status of CPP offspring through the end of 2016 using a probabilistic linkage of the cohort to death certificates recorded in the National Death Index (NDI). As described previously (Vidal‐Ribas, Govender, Sundaram, et al., 2022), the linkage compared the identifying information of study participants (first name, middle initial, last name, father's surname for females, birth day, birth month, birth year, sex, race, and state of birth) to the information recorded on death certificates in the NDI between 1979 and 2016 (NCHS, 2013). For each potential match, a probabilistic score was generated that quantified the likelihood that the identifying information from the death certificate and the study record belonged to the same person (Fellegi & Sunter, 1969; Rogot, Sorlie, & Johnson, 1986).…”

“…A socioeconomic disadvantage index was created based on early life factors that have been associated with a higher risk of suicide mortality in the current sample (Vidal‐Ribas, Govender, Sundaram, et al., 2022) including low parental education (high‐school graduate or less [≤12 years] vs. more than high school [>12 years]), manual parental occupation (vs. non‐manual), and third, fourth or higher birth order (vs. first or only child), with higher scores indicating higher disadvantage. Sex of the child and race were also included in the analyses.…”

“…It is unclear whether the associations found in prior studies between children's cognitive performance and suicide mortality are independent of other early life risk factors for suicide. For example, we have previously found that parents' education and occupation, and prenatal factors including birth order and pregnancy complications, were associated with a higher risk of suicide mortality (Vidal‐Ribas, Govender, Sundaram, Perlis, & Gilman, 2022). These factors are associated with cognitive ability in children (Bradley & Corwyn, 2002; Camp, Broman, Nichols, & Leff, 1998; Keim & Pruitt, 2012; Seidman et al., 2000) and adults (Jefferis, Power, & Hertzman, 2002).…”