Maternal depression symptoms experienced by the participants in the past week were assessed using the Edinburgh Postpartum Depression Scale (EPDS) [71], a self-report questionnaire with possible scores ranging from 0 to 30, where higher scores indicate more severe symptoms. A cut-off score ≥13 is used to identify individuals with clinically concerning depression symptoms [71]. Scores ≥13 during pregnancy have a sensitivity and specificity of 100% and 87%, respectively, for classifying major depression, and a positive predictive value of 33 [72]. This scale has been validated for both prenatal and postnatal assessment [73,74].

General anxiety symptoms experienced by the participants in the past week were assessed using the 7-item Patient-Reported Outcomes Measurement Information System (PROMIS) Anxiety–Adult Short Form [75]. We follow the standard practice of converting raw scores to T-scores using the US general population norms; possible scores range from 36.3 to 82.7, with a mean score of 50 (SD 10). T-scores in the range of 60-69.9 are indicative of moderately elevated anxiety symptoms, and scores ≥70 are considered indicative of severely elevated anxiety symptoms [76].

Pregnancy-related anxiety symptoms (ie, worries about the health of the baby, birth, and caring for a new baby) experienced by the participants were assessed using the 10-item Pregnancy Related Anxiety Questionnaire (PRAQ) [77] in the English-language survey and the French translation of the 10-item Pregnancy Related Anxiety Questionnaire–Revised 2 (PRAQ-R2) in the French-language survey [78]. Possible scores on the PRAQ range from 10 to 40 and those on the PRAQ-R2 range from 10 to 50. Both questionnaires have acceptable internal consistency (Cronbach α=.80-.81) [78,79], and their validity is supported by extensive use in relation to both maternal and infant outcomes. There are no cut-off scores for these scales, but previous treatment studies have used a median split to define groups with higher versus lower pregnancy-related anxiety symptoms [80], with higher scores indicating more severe symptoms.

Anger experienced by the participants in the past week was assessed using the 5-item PROMIS Short Form v1.1–Anger 5a [75]. As with other PROMIS measures, raw scores are converted to T-scores. T-scores range from 32.9 to 82.9; the mean score was 50 (SD 10), with scores in the range of 60-69.9 indicative of moderately elevated anger and scores ≥70 indicative of severely elevated anger [76].

The Distress Thermometer was used to measure the overall level of subjective distress experienced by the participants in the previous week, on a visual analog scale of 0 (“Not distressed”) to 10 (“Extremely distressed”) [81]. A cut-off score ≥4 is typically used to signify clinically concerning distress [82]. The Distress Thermometer has demonstrated good validity and temporal stability [83,84].

Participants’ subjective experience of psychological stress over the past month was assessed with the widely used 10-item Perceived Stress Scale (PSS) [85]. The PSS measures the degree to which participants appraise their lives as unpredictable, uncontrollable, and overloaded [86]. Scores range from 0 to 40, with higher scores indicating greater perceived stress. Reliability and validity of the PSS have been supported across multiple studies [85].

COVID-19 represents a novel exposure for which there are no previously developed measures. Nevertheless, previous work on objective measures of hardship resulting from exposure to natural disasters provided a principled and systematic framework for developing a new measure [87]. For instance, work by King and Laplante [87] has identified four major components of disaster exposures, which were adapted to the COVID-19 context: scope, loss, threat, and change. Scope refers to the duration and intensity of the hardship, with the former referring to the amount of time for which major aspects of participants’ lives were disrupted and the latter focusing on the number of individuals within participants’ communities who were similarly affected. Loss refers to financial, social, and physical losses experienced as a result of the pandemic. For example, the loss of employment, savings, closures of schools, and daycares represent relevant losses. Threat refers to physical and health-related consequences of exposure to the stressor. For example, being infected with SARS-CoV-2 or hospitalization of a close friend with COVID-19 represent threats to self and others. Change captures the adjustments to daily living, prenatal care, work, and social interaction caused by the COVID-19 pandemic. Relevant changes include working from home, altering a birth plan, and reductions in physical activity or diet quality. The timeframe for these items was the previous month or the previous questionnaire (ie, up to three months).

Changes and disruptions to prenatal and postnatal care received by the participants were assessed using a series of questions tailored to our study relating to the way that prenatal care was delivered, cancellations, ability to bring partner to appointments, and changes to birth plans. Participants are also asked to evaluate the impact of these changes on the quality of care they and their baby have received in the past three months.

The degree to which participants feel that COVID-19 was a threat to their health or the health of their baby at any time during the pandemic was assessed by three items developed for the study: (1) “How much do (did) you think your life is (was) in danger during the COVID-19 pandemic?” (2) “How much do (did) you think your baby’s life is (was) in danger at any time during the COVID-19 pandemic?” and (3) “How much are you worried that exposure to the COVID-19 virus will harm your baby?” Responses were scored on a 100-point sliding scale, with the left anchor indicating 0 points (“Not at all”); the middle anchor, 50 points (“Somewhat”); and the right anchor, 100 points (“Very much so”).

Adverse childhood experiences (ACEs) of the participants were assessed using a 10-item measure [88] of early-life adversity (age 0-18 years) across three domains: abuse (emotional, physical, and sexual), neglect (physical and emotional), and exposure to household dysfunction (domestic violence, substance abuse, mental illness, parental separation or divorce, and incarcerated household member). The ACEs questionnaire is widely used and has demonstrated good reliability and internal consistency (Cronbach α=.81) [89], as well as adequate test-retest reliability (weighted κ=0.64) [90]. The occurrence of individual ACEs is summed to create the ACE score with a potential range of 0-10. Based on previous work that has examined dose-response relationships, participants were coded as having experienced 0, 1, 2, 3, or ≥4 ACEs [88,91,92].

Participants’ perception of the quality of social support received from their romantic partner over the previous three months was assessed using the 35-item Social Support Effectiveness Questionnaire (SSEQ) [79]. Within the domains of emotional, informational, and task support, participants were asked to rate their experience over the past three months on a 5-point scale: (1) how well the quantity of support received from their partner matched the amount they wanted, (2) whether they wished the support had been different somehow, (3) how skillful their partner was at providing support, (4) how often it was difficult to solicit support, and (5) whether their partner offered support without being asked. Participants also rated the extent to which they perceived their partner’s support as negatively infringing upon their self-efficacy or self-esteem. Internal consistency of the SSEQ is strong (Cronbach α=.87), and it has previously been used to distinguish levels of social support in samples of pregnant individuals [79,93,94]. Total scores can range from 0 to 80, with higher scores indicating more effective support.

Perceived social support was also assessed using the 12-item Interpersonal Support Evaluation List (ISEL) [95,96], to determine appraisal (eg, advice or problem solving), belongingness (eg, shared experiences), and tangible support (eg, help with daily chores) over the past three months. Here, we focus on the total score, which is the sum of the three subscales. Scores range from 12 to 48, with a higher score indicating greater perceived support.

Feelings of social isolation were assessed using the following item: “During the COVID-19 pandemic, I have felt more alone than usual.” Participants’ responses were on a 100-point sliding scale, with the left anchor indicating 0 points (“Not at all”); the middle anchor, 50 points (“Somewhat”); and the right anchor, 100 points (“Very much so”).

Participants’ perception of partner relationship quality was assessed using the 4-item Couple Satisfaction Index (CSI-4) [97]. Scores range from 0 to 21, with higher scores indicating higher levels of relationship satisfaction. CSI-4 scores below 13.5 suggest notable relationship dissatisfaction. Changes in relationship quality as a function of the pandemic were assessed for relationships with partner, children, and close friends and family by using items generated for the study. For each relationship type, participants were asked how the COVID-19 pandemic affected their relationship. Responses were recorded on a 100-point sliding scale, with the left anchor indicating 0 points (“It has strained our relationship”); the middle anchor, 50 points (“Not much has changed”); and the right anchor, 100 points (“It has brought us closer together’”).

Experiences of physical harm in the 12 months prior to pregnancy and during the current pregnancy were assessed using two items adapted from the Pregnancy Risk Assessment Monitoring System (PRAMS) [98]. These items were queried with regard to the 12-month periods before pregnancy, since the last survey, and after giving birth.

Participants reported their physical activity using a modified form of the Godin-Shephard Leisure-Time Exercise Questionnaire (GLTEQ) in a typical week over the past month [99]. Participants reported the number of days per week in which they engaged in mild (eg, light walking), moderate (eg, brisk walking), and strenuous (eg, running) exercise of more than 15 minutes. Definitions of each category of physical activity were provided. The total score was calculated per standard scoring procedure for GLTEQ, by multiplying episodes of mild exercise by 3, moderate exercise by 5, and strenuous exercise by 9. Participants with scores below 14 are considered sedentary, those with scores ranging from 14 to 23 are considered moderately active, and those with scores equal to or more than 24 are considered active. An additional item was included asking participants how their level of physical activity changed because of the COVID-19 pandemic. Responses were recorded on a 5-point Likert-type scale with the following options: “Substantially decreased” (1 point), “Somewhat decreased” (2 points), “No change” (3 points), “Somewhat increased” (4 points), and “Substantially increased” (5 points).

Sleep disturbance and impairment in the past week were assessed using the 4-item PROMIS Sleep Disturbance–Short Form 4a and the 4-item PROMIS Sleep-Related Impairment–Short Form 4a [100]. As with other PROMIS measures, raw scores are converted to T-scores. T-scores ranging from 60 to 69.9 are considered moderate problems, and scores ≥70 are considered severely elevated. Sleep duration (ie, hours of sleep per night) was assessed with a single item from the Pittsburgh Sleep Quality Index [101].

Changes in diet and eating patterns were assessed using a questionnaire developed for the study to determine how participants’ diet during the COVID-19 pandemic differed from their prepandemic diet and the reasons for the change. Dietary changes were assessed in the following categories: fresh vegetables or fruits, dairy, meats, canned or dried foods, fast foods, take-out or home delivery, and sweets or snacks. For each category, participants responded on a 5-point Likert-type scale, with the following options: “I eat much more” (1 point), “I eat more” (2 points), “I eat about the same” (3 points), “I eat less” (4 points), and “I eat much less” (5 points). If participants reported a change in their diet (ie, did not select “I eat about the same” option), then they were asked about the reason for the change with the following response options: “Can no longer afford,” “Can’t go grocery shopping frequently,” “Can spend more time cooking and preparing food,” “Change in craving,” and “Other (specify below)” (scores: Yes=1, No=0). Although changes in craving were not expected as a direct result of the pandemic, they are commonly reported during pregnancy [102]; it was therefore important to disambiguate such pregnancy-related changes in diet from changes that are more directly related to the pandemic.

Substance use prior to (ie, in the 12 months before pregnancy) and during the current pregnancy (ie, in the past month) were assessed using a self-report measure separately for alcohol, cannabis, tobacco, and illicit drugs. Participants were asked how many days per week they consumed each substance and how many drinks or products per day they typically consumed. Participants were also asked whether they changed their use patterns in pregnancy and if they had, they were asked when they changed their use patterns.

Participants’ perceived ability to cope with stressful situations was assessed with the 2-item Connor-Davidson Resilience Scale (CD-RISC 2), which demonstrates good test-retest reliability and convergent and divergent validity [103]. Each item is rated on a 5-point Likert scale ranging from 0 (“Not true at all”) to 4 (“True all the time”); total scores can range from 0 to 8, with higher scores indicating more successful coping. We also included an open-ended question about coping: “People are responding to the pandemic in many ways. Can you tell us what things you are doing to cope with the COVID-19 pandemic?”

The 27-item Intolerance of Uncertainty Scale (IUS) was used to measure the extent to which participants believe uncertainty is stressful, upsetting, negative, unfair, and leads to the inability to take action [104,105]. Items are rated on a scale ranging from 1 (“Not at all characteristic of me”) to 5 (“Entirely characteristic of me”). Previous research shows that the IUS has excellent internal consistency (Cronbach α=.91) and good test-retest reliability (r=0.74), and it is highly correlated with symptoms of generalized anxiety disorder [104].

Self-compassion, a trait level form of resilience, was assessed using the 12-item short form of the Self-Compassion Scale–Short Form (SCS-SF) [106]. The SCS measures the tendency to treat oneself with kindness and understanding rather than harsh self-judgment, to recognize imperfections and suffering as part of the human experience, and to have mindful awareness and tolerance of negative thoughts and feelings. Each item is rated on a 5-point Likert scale ranging from 1 (“Almost never”) to 5 (“Almost always”); total scores range from 10 to 60, with higher scores indicating more self-compassion. The SCS has adequate internal consistency (Cronbach α=.86) and strong correlation (r=0.97) with the long form [106].

Participants’ experiences of discrimination were assessed using the 5-item Everyday Discrimination Scale [107]. This questionnaire asks about day-to-day experiences of discrimination, including being treated differently than other people and feeling threatened or harassed. Responses items ranged from 0 (“Never”) to 6 (“Almost every day”). For items rated as more than “A few times a year,” participants were also asked a follow-up question about what they think is the main reason for these experiences.

Participants were asked if they planned to receive a COVID-19 vaccine, if they had received a COVID-19 vaccine (which vaccine, if yes), and the dates of the doses (or the gestational age if they were still pregnant).

The Ages and Stages Questionnaire, Third Edition (ASQ-3) [108] is a widely used parent-reported and norm-referenced developmental screening tool [109] that assesses delays in child development across five domains: communication, gross motor, fine motor, problem-solving, and personal adaptive skills. The ASQ-3 has been identified by the American Academy of Pediatrics as a high-quality tool for use in clinical practice to screen for delayed developmental milestones in children [110]. Sensitivity and specificity of the ASQ-3 are both 86% for distinguishing between children at risk for developmental delay and children not at risk. Parents rate each of the 30 items on a scale ranging from “Yes” (10 points), “Sometimes” (5 points), or “Not yet” (0 point) based on the infant’s current ability.

The Ages and Stages Questionnaire: Social-Emotional, Second Edition (ASQ:SE-2) [111] is a validated and widely used parent-report screening tool in 7 areas of socioemotional development: self-regulation, compliance, social communication, adaptive functioning, autonomy, affect, and interaction with people. A total score is calculated to index overall socioemotional problems. Parents rate each of the 27 items on a scale ranging from “Often or always” (0 point), “Sometimes” (5 points), or “Rarely or never” (10 points) based on the infant’s usual behavior.

The Infant Behavior Questionnaire–Revised Very Short Form (IBQ-R) [112,113] is a 37-item parent-report measure of infant temperament. Three broad dimensions of temperament are assessed: negative emotionality, regulatory capacity/orienting, and positive affectivity/surgency. Parents report their observations of specific infant behaviors in the past week using a 7-point Likert scale ranging from “Never” to “Always.” Each item also had a “Does not apply” option. Scores on each dimension can range from 1 to 7, with higher scores reflecting stronger evidence of each dimension. The IBQ-R has strong psychometric properties and is widely used in the child development literature [114,115].

Periods of persistent infant crying (defined as half an hour or more during which the baby would not settle) in the past week were assessed using several items from the Crying Patterns Questionnaire [116]. The validity of the Crying Patterns Questionnaire, relative to a cry diary, has been supported [117].

Parents’ report of their infant’s sleep quality in the past two weeks was assessed using the 19-item Brief Infant Sleep Questionnaire (BISQ)–Revised Short Form [118]. In addition to a total score, three subscale scores are calculated: infant sleep (5 items), parent perception (3 items), and parent sleep-related behaviors (11 items). The measure is scored using an age-based and norm-referenced system [119]. The total score and each subscale score are scaled from 0 to 100, with higher scores indicating better sleep quality, more positive perception of infant sleep, and parent behaviors that promote healthy and independent sleep. The BISQ is widely used, and its reliability and validity have been documented [120,121].

Based on previously published questionnaires [122,123], we assessed onset, duration, and proportion of breastfeeding and formula feeding by using a series of maternal-report questions that conform to the World Health Organization’s breastfeeding categories: exclusive breastfeeding, predominant breastfeeding, mixed feeding, and bottle feeding [124].