Participants
The study included 116 mother- infant dyads recruited from pediatric practices affiliated with children’s hospitals is Boston and Los Angeles. All available outcome data were analyzed and are described in detail in the results section. Participants were recruited if they met the following inclusion criteria: receiving early postnatal services at the designated primary care clinics, mothers aged 18 or older with an infant < 2 months, and birth weight greater than the 20th percentile (at least 2,500 grams and typically developing).
Exclusion criteria included infants who were born prematurely (< 36 weeks of gestation); infants who had congenital disorders (genetic, metabolic, malformations, or neurological disorder etc.); infants who experienced birth complications (fetal distress, requiring resuscitation at birth, meconium aspiration syndrome, etc.). Study visits took place at 2, 6, 9, and 12 months from 2016 to 2019. Ethical approval for the study was obtained from Institutional Review Boards (IRB) at both sites. Informed written consent was obtained from the mother for all mother- infant dyads who participated in the study.
Measures
Family demographic factors. The family demographic factors were collected through self-report from mothers at the 2- month visit. Demographic factors of interest were marital status, highest level of education completed, annual family income range, age, and racial identity (which we recognize as a social construct). Marital status was represented as single (single, widowed, separated, divorced) or partnered (married, cohabiting). Maternal education was assessed as seven categories ranging from 8th grade or less to M.D., Ph.D., J.D., or equivalent. The nine categories of annual family income ranged from <$5,000 to $100,000+. Maternal age was self-reported as age at the time of the 2-month visit. Participants were asked to select a racial category from a selection of 14 different identities which included the following options: White, Black or African American, American Indian or Alaska Native, Asian Indian, Chinese, Filipino, Japanese, Korean, Vietnamese, Other Asian, Native Hawaiian, Guamanian or Chomorro, Samoan, and Other Pacific Islander. Racial information collected about the study participants in Boston only included the infants, thus the race information analyzed in this study are of the race designations of the infants as reported by their mothers.
Maternal Mental Health. Two assessments of maternal mental health were utilized in this study. The Edinburgh Postnatal Depression Scale (EPDS) is a validated 10-item screening tool frequently used in the perinatal and postnatal setting to identify mothers with depression(31). Depressive symptoms are assessed using a four-tiered severity scale from over the past 7 days. Scores range from 0 to 30, with a score of 13 or greater indicating risk for major depression (31, 32). The EPDS was collected at all study visits (2, 6, 9, and 12 months postpartum) at both sites. The Patient Health Questionnaire 9 (PHQ9) is a 9-item assessment used to identify and monitor over time major depressive symptoms in the general population and in the postpartum period (33, 34). This assessment utilizes a four-tiered scale assessing the frequency of depressive symptoms over the past 2 weeks. The score ranges from 0 to 27, with a score of 10 or greater indicating elevated depressive symptoms. The PHQ9 was completed by the mothers at each research visit in LA and during the initial 2 month visit in Boston. Both EPDS and PHQ9 were assessed in this study to enable assessment of both anxiety and depressive symptoms (EPDS) and the constitutional symptoms of depression (PHQ9) (34).
Cumulative Risk Score
The cumulative risk score for maternal-infant early environment was calculated based on the cumulative risk modeling described by Evans et al.(29). Demographic factors including age, highest education attained, marital status, family income and race were standardized into risk categories and calculated as being present (1) or absent (0). Risk delineations for each demographic factor were calculated based on a participant occupying the lowest standard deviation of the sample mean (age<=25 years, income <$25,000) or occupying a demographic category previously described as having higher risk for infant morbidity and mortality (Black race, <= High School education, Single marital status) (29, 35, 36). Approximately thirty-percent of participants did not provide income data. For those participants with missing income data, risk stratifications were assigned using neighborhood poverty levels similarly to how socioeconomic status was extrapolated in Valdes et al. 2020 and the Internal Revenue Services’ census tract definition of low-income communities (LIC) by designating risk as residing in LIC communities (37-39). Mental health risk scaled scores were generated from previously described risk stratified scoring guidelines for PHQ9 and EPDS. Scores were standardized as low/no risk (0), medium risk (1) or high risk (2) (31-34). Total cumulative risk for each participant was calculated by summation of all demographic and mental health risk categories occupied to create an overall maternal-infant early environment cumulative risk score.
Cognitive scores. The primary outcome, infant cognition, was measured using The Mullen Scales of Early Learning (MSEL). The MSEL is a play-based developmental assessment that evaluates gross motor, visual reception, fine motor, expressive language, and receptive language (40). An early learning composite score (ELC) is generated from the assessments excluding gross motor (excluded per standard practice). This composite score is then standardized to compare cognitive developmental status from birth through sixty-eight months of age (40). Assessments were done at 2, 6, 9, and 12 months in Los Angeles and at 6 and 12 in Boston by trained members of the research team in either English or Spanish, based on the primary language spoken at home.
Oxidative Markers. The specific oxidative stress metabolite F2 -isoprostane (IsoPs) was the secondary measure of interest. F2 -isoprostane is one of several IsoPs products generated by reactive oxygen species peroxidation of arachadonic acid (17, 41). Infant urine samples were collected via bag specimen at all 4 study visits, aliquoted and stored frozen at –80oC within 1 hour of collection. Maternal blood samples (3x 3ml tubes=9ml total or 2 teaspoons) were collected by a trained phlebotomist at all 4 study visits. Urine and blood samples were shipped in batches, deidentified with a subject code, to the Eicosanoid Core Laboratory at Vanderbilt University for IsoPs analyses (42).
Statistical Analysis
Demographic characteristics of both mothers and infants were compared by site using chi-squared analysis for discrete categorical variables and independent sample t-test for continuous variables. Due to skewing in the data, EPDS and PHQ9 scores were compared using a Mann-Whitney U test. A priori α to distinguish differences between sites was set as 0.05.
To examine the MSEL developmental outcome at 6 and 12 months and the oxidative stress measure across the first year of life, univariate linear regression was utilized with each demographic and mental health variable along with the early maternal-infant cumulative risk score. Listwise deletion was used to account for missing data and only those with complete data for both independent and dependent variables were included in the models.
Linear Mixed Effect models (MLE) were applied to assess the association between oxidative stress markers (IsoPs) over the first year of life, measured at (2, 6, 9 & 12 months), and multiple risk factors. MLE models account for the within-individual correlation and made it possible to include patients with incomplete observations. The Pearson Correlation coefficient was used to evaluate the relationship between maternal and infant IsoPs levels and infant IsoPs levels and MSEL scores. All analyses were performed using SPSS Version 27.0 (IBM Corp, Armonk, NY, USA).