Psychological Effects of COVID-19 on Pregnant Women and New Mothers Living in a US Hotspot

Purpose : This study investigated COVID-19 related psychological distress among expectant and new mothers, with and without infection, in metropolitan New York. It also examined the trajectories of participants’ distress during pregnancy and postpartum, and the moderating effect of socioeconomic status (SES). Methods: An online survey was conducted April through June 2020 among expectant and new mothers with infants (<12 months) (N=642). Associations between infection status and psychological symptoms, suicidal ideation, and substance use were examined. Changes in distress related to COVID-19 infection and SES were then examined. Results: We found elevated anxiety and depression symptoms among infected compared to uninfected women. Similarly, infected, compared to uninfected women, had elevated risk for suicidal thoughts (quite often, AOR=3.97, sometimes AOR=13.2), and for substance use [alcohol (AOR=3.30); tobacco (AOR=4.54); cannabis (AOR=7.01); heroin (AOR=7.09); cocaine (AOR=10.05)]. Differences in trajectories of distress across pregnancy between the two groups were significant. Among infected women, distress was consistently high throughout. Among uninfected women, it started low and intensified toward the end of pregnancy. SES further moderated the impact of infection on distress. During earlier trimesters, infected/low SES women had greatest, and uninfected/high SES women had lowest, levels of distress. Their trajectories converged nearing childbirth. Conclusions: New and expectant mothers, especially those infected, have suffered substantial psychological distress due to the pandemic. Moreover, SES moderated the trajectory of distress. Infected women who also had low SES experienced the highest distress levels among all groups. Mitigating strategies are imperative to alleviate this distress.


Introduction
On March 11, 2020, the World Health Organization officially announced the Coronavirus Disease 2019 (COVID-19) to be a pandemic.Measures such as quarantine, social isolation, daycare/ school closures, and "work-from-home" initiatives taken by local and national authorities in the USA profoundly disrupted daily life.
COVID-19 presents an unprecedented challenge for maintaining psychological health.Excessive alcohol consumption [1] and an increase in self-harm [2] have been reported.COVID-19 stress hits vulnerable populations (i.e., expecting and new mothers) hard.Although pregnancy is thought of as a happy time, it is also a stressful, vulnerable period [3,4].Women have to adapt to biological and emotional changes, financial and social pressures, restrictions on time for pre/postnatal care, and to a new role as a mother.These factors combined, elevate the risk of depression during this period [5,6].Recent research has reported COVID-19 elevated psychological distress, worry, and fear [7] and elevated anxiety and depression symptomatology [8,9] among new and expecting mothers.A pilot study (n=31) of pregnant women in the U.S. reported high depressive symptoms and moderate to severe anxiety related to COVID-19 [10].Concerns regarding COVID-infection or exposure may exacerbate the distress and mental health problems in expectant and new mothers [11,12].
Given the unprecedented level of suffering in the New York metropolitan area, the first epicenter of COVID-19 in the USA, we utilized this opportunity to evaluate the psychological consequences in a population already known to have increased vulnerability to psychological distress and whose psychological health has direct consequences not only to themselves but also to the well-being of their infants [13].To do this, we employed online surveys to assess perceived risk of exposure, negative emotions, and thoughts and behaviors concerning COVID-19 among pregnant and new mothers.We expected that infection status would influence all of these factors.We also examined socioeconomic status (SES), expecting it to have a differential impact and further hypothesized psychological distress may be magnified among women with low SES, who have fewer resources to buffer the impact of the pandemic [14] and to cope with the restrictions and unknowns of COVID-19.

Procedure and Participants
The study was set-up with the online platform Qualtrics in conjunction with the Completely Automated Public Turing Test (CAPTCHA), which ensured data integrity by eliminating robotic participation.Participants were recruited through fliers at OB-GYN clinics and social-media platforms between April and June 2020.Interested individuals answered online eligibility questions.Criteria included living in the NY metropolitan area and being pregnant or having a child 12 months-old or younger.After eligibility was confirmed, 3 CAPTCHA verification challenge responses were required to proceed to digitally signing consent.Of 744 respondents, 99 did not meet screening criteria and 3 had missing responses, leaving 642 participants.The study was approved by the IRB at Queens College, CUNY.
Online questionnaires were used to assess stress, psychological symptoms, substance use, distress related to COVID-19, and COVID-19 infection status.Questions regarding distress related to COVID-19 were asked at different stages of pregnancy (1 st , 2 nd , 3 rd trimester, and postpartum): postpartum mothers who had been pregnant during the pandemic responded to questions from earlier trimesters

Cross Sectional Analysis
First, univariate analyses (Chi-Square for dichotomous and analysis of variance for continuous outcomes) were conducted, followed by multivariate analyses (logistic regression for dichotomous and analysis of co-variance for continuous outcomes).Covariates included age, marital status, race, and parity.

Longitudinal Analysis
We used a generalizing estimating equation (GEE) to evaluate the influence of COVID-19 infection, time (1 st , 2 nd , 3 rd trimesters, and postpartum), and the interaction between infection status and time on the distress score at each time point and an overall difference in trajectory.This was followed by HLM, which estimates both withinperson longitudinal and between-person effects [19].The within-person model mapped the trajectory of distress at four time points.All models in the analysis were corrected for non-normal distributions of level 2 residuals by applying the full maximum likelihood estimation (MLE) with robust standard errors to incorporate the missing data imputation [20].The interaction evaluated the magnitude of the moderating effect of SES on the influence of COVID-19 infection.
Given that little information was available on changes in distress among expecting and new mothers in the early pandemic, we chose to let the data determine the best-fit model.We began by testing the trajectory of changes without predictors and covariates.As both linear and curvilinear models could be significant, tests of relative model fit were computed by comparing deviance statistics to choose the bestfit model.Random effects were included in the intercept and change coefficients.Time was centered on the intercept that represented distress in the 3 rd trimester.After determining the best-fit model, we examined whether infection status was a significant predictor for the distress score, and then examined the joint effects of infection and SES with an additional interaction between the two.

Missing Data
Distress scores include a choice of NA, which was treated as system missing (missing as they should).Longitudinal analyses apply MLE, using available data to yield parameter estimates for the missing time points for a total distress score (within-subject variability), but not for predictor variables that explain between-subject variability [19].

Infected Compared to Uninfected Women
Stress and Psychological Functioning: Infected, compared to uninfected, women reported greater levels of anxiety (100.10 vs. 93.40,p=.00003) and depression (17.05 vs. 13.25,p<.00001) symptoms, but did not differ on perceived stress.
Influence of COVID-19 infection, time (1 st , 2 nd , 3 rd trimesters pregnancy, and postpartum), and the interaction between the two on the level of distress GEE was used to evaluate the influence of COVID-19 infection, time (1 st , 2 nd , 3 rd trimesters, and postpartum), and interaction between infection status and time on the level of distress at each time point, and an overall difference in trajectory.Results showed a significant timeeffect (p<.001), infection-effect (p=.008), and trajectory difference (i.e., interaction) between infected and uninfected women (p<.001).

Longitudinal changes across pregnancy and postnatal period in distress by COVID-19 infection
We first examined each distress measure as a function of the intercept plus the linear/quadratic effect of time without predictors and covariates (Panel A).We found that a curvilinear model was the best-fit to explain the trajectory of distress (Table 4).After choosing the best-fit model, Model 1 examined the trajectory of distress with infection status.Then, we evaluated the model with infection, SES, and the interaction between the two (Model 2) to determine whether the effect of infection was moderated by SES (Figure 1).

Model 1. With only infection status
There was a significant difference in distress (t=2.62,p=.009) between infected and uninfected women at the intercept (3 rd trimester).
Panel B shows the patterns of change in distress over time by infection status.Among infected women, the level of distress increased slightly throughout the study period.Among uninfected women, the level started low but increased throughout pregnancy and exceeded the level of infected women after childbirth (t=-5.58,p<.001).

Model 2. With infection, SES and interaction of the two
Infection status (t-ratio=11.52,p<.001) and SES (t-ratio=4.81,p<.001) predicted a significantly different distress level at the intercept (3 rd trimester).The interaction was also significant (t-ratio=-6.64,p<.001), indicating that SES moderated the effect of infection.Panel C shows the trajectory of distress by infection status and SES.Both infection status and SES had significant effects in predicting the linear change (t-ratio=-6.50,p<.001; t-ratio=-10.14,p<.001, respectively) and curvilinear changes in distress (t-ratio=-3.35,p<.001; t-ratio=-10.11,p<.001, respectively).The interaction was significantly different for both linear (t-ratio=5.86,p<.001) and curvilinear changes (t-ratio=6.02,p<.001).Specifically, SES has a differential impact on the effect of infection on the trajectory of distress.Among uninfected women, those with low SES had a substantially greater level of distress approaching their 3 rd trimester, relative to women with high SES, whereas women with high SES had a lower level of distress toward the end of the 3 rd trimester, but it continued to increase postpartum.

Discussion
This study examined how the COVID-19 pandemic impacted the mental health of new and expectant mothers living in metropolitan New York.Consistent with early reports [7][8][9][10], we found alarmingly high levels of mental health problems, especially among infected mothers.Further, we observed different trajectories of distress levels across pregnancy based upon infection status and SES.Infected compared to uninfected women had higher levels of distress in early pregnancy, but for both groups, distress levels reached their height toward the end of pregnancy.Notably, during the 1 st and 2 nd trimesters, infected expecting mothers who also had low SES were the most vulnerable group and had the highest levels of distress.
Mental health consequences of the pandemic were worse for infected than for uninfected women.Forty participants had confirmed COVID-19 and 4 had suspected infection.Infected participants had higher levels of anxiety and depression symptoms, and suicidal ideation.There was a 4-fold increase in having suicidal thoughts "quite often" and an over 13-fold increase in having them "sometimes" among infected women.Importantly, the prevalence of substance use among infected women was also markedly higher (infected/uninfected -tobacco: 70.0%/17.4%;cannabis: 65.0%/11.6%;alcohol: 72.5%/35.7%).Given that we lack pre-pandemic substance use data, we are unable to conclude that this disturbing pattern of substance use results from the stress of infection, as opposed to riskier COVID-19 behavior among substance users leading to higher infection rates, or possibly to underresourced and marginalized communities having higher rates of both.Future research is required to clarify the relationship between substance use and infection in expecting and new mothers.
We hypothesized that infection status had different effects on the trajectory of distress across pregnancy and the postpartum period.We observed that infected, relative to uninfected mothers, had substantially higher levels of distress in early pregnancy, and the level of distress remained relatively unchanged throughout, reaching its height in the postpartum period.This has been similarly observed in non-COVID studies on the progression of anxiety and depression symptoms during pregnancy [5,25].Notably, while the distress level in uninfected women started low in early pregnancy, it rose at a greater rate toward the end of the pregnancy, exceeding the level among infected women in the early postpartum period.Although speculative, these findings may suggest that mothers, regardless of infection status, become more concerned as they approach their delivery date and the consideration of the health of their newborns becomes more immediate.Furthermore, we simultaneously examined infection status and SES on the trajectory of COVID-19 distress during pregnancy and postpartum.Although both factors influenced change in COVID-19 distress, SES status moderated the trajectory specifically among infected women.In the earlier trimesters, distress level for low SES participants was greater than for high SES participants.It gradually  increased, peaking around the 2 nd trimester, and then decreased again.
On the other hand, women with high SES started with lower distress levels which increased gradually at the same steady rate throughout.High SES participants who were uninfected by COVID-19 had the lowest levels of distress in their 1 st trimester.Distress levels decreased to their lowest point in the second trimester but rose again toward postpartum.The level of COVID-19 related distress appeared to be elevated in all mothers after their babies were born.
This pattern of change is inconsistent with a recent pre-COVID study of 186 women which demonstrated that anxiety was the highest during the first trimester, decreased as the pregnancy progressed into the second trimester, and remained low during the third [26].Although the reason for these different patterns of distress between a prepandemic and a pandemic group remain in the realm of speculation, they strengthen the evidence for significant effects of COVID, and its accompanying uncertainties, on elevating women's levels of distress just before childbirth and during the early postpartum period when their children are very vulnerable.Our findings suggest that healthcare providers and policy makers need to offer additional resources that address new mothers' concerns and anxiety about safeguarding their newborns from the possibility of infection.
There are several limitations in the current study.First, it was conducted when New York was at a COVID-19 peak and under lockdown.Although face-to-face interviews were preferable, circumstances necessitated online self-report.Second, women participated at different pregnancy stages, resulting in retrospective responses for periods earlier in pregnancy.This opens the possibility of some recall bias due to retrospective data ascertainment.Third, some asymptomatically infected participants may have unknowingly been classified as uninfected; however, this study was intended to evaluate the psychological, not biomedical, consequences of COVID-19.Fourth, there is a relatively low prevalence of infected women in our study population.Fifth, we did not collect data on all social and financial factors potentially affecting COVID-19 distress (e.g., unemployment, non-COVID stressful life-events, access to healthcare, social support).Consequently, statistical analyses do not include these potentially important factors.Sixth, although our participants' racial distribution is closest to that of Staten Island, where the largest proportion of participants live, it is disproportionately Caucasian, married, and well-educated.Given that Caucasian women with higher SES were overrepresented in this study and that financial and racial minorities are considerably disadvantaged with higher COVID-19 infection rates [27,28], our findings suggest an alarming picture for those experiencing increased vulnerability due to socioeconomic privation and racial discrimination.Future in-depth studies, that include a broader range of demographic factors and target financial and racial inequality, will help bridge the gap in our understanding of the sociopsychological vulnerabilities in new and expectant mothers.
Despite these limitations, our study contributes a deeper understanding of the serious impact of the COVID-19 pandemic on the mental health of expecting and new mothers.Highlighting the effect of COVID-19 on these women has implications beyond the current pandemic.Our study addresses what may be a COVID-19 mental health crisis among expectant and new mothers and suggests that these women-especially if they are infected and are classified as low SES-are suffering considerable mental health consequences.It is essential that services such as psychological support and stress reduction, be made available for women of reproductive age and their families in order to prevent potential long-term consequences in their ability to care for their newborns.It is equally important for policy makers, obstetricians, and pediatricians to create an infrastructure to assist pregnant women and their families when confronting COVID infection, in order to mitigate risks to themselves and their children in-utero.Finally, it is important to adapt longitudinal studies such as this one to evaluate the multi-faceted long-term effects of the COVID pandemic on maternal mental health outcome, as well as those of their infants and partners.Policy changes, with state and federal support, are urgently needed to alleviate the high level of distress among this population.
Citation: Nomura Y, Kittler P, Taveras S, et al.Psychological Effects of COVID-19 on Pregnant Women and New Mothers Living in a US Hotspot.J Gynec Obstet 2022; 5:038.Demographic and Psychosocial Characteristics of the participantsArea of residence, N (%)a There are missing data among n=5 participants on suicidal thoughts STRESS,
in stress and emotions, and substance use among women infected and not infected with COVID-19 OR=odds ratio; AOR=adjusted odds ratio; CI=confidence interval a. Adjusted model includes race, parity, age, and education of participants.Citation: Nomura Y, Kittler P, Taveras S, et al.Psychological Effects of COVID-19 on Pregnant Women and New Mothers Living in a US Hotspot.J Gynec Obstet 2022; 5:038.

Table 3 .
The effects of COVID infection on feeling and distress toward COVID-19 pandemic at different gestational periods (1 st trimester, 2 nd trimester, 3 rd trimester, and within 1 year postpartum).

st TM 2 nd TM 3 rd TM postpartum Statistics
GEE was used to assess the effects of COVD-19 infection on a linear change.Marital status, race, parity, and age of the participants were controlled.The interaction term (INFECTION x TIME) shows the difference in the linear change of the distress scores over time between infected and uninfected participants.1 st TM = first trimester, 2 nd TM = second trimester, and 3 rd TM = third trimester.M= mean; SE= standard error

Table 4 .
Linear and quadratic change in feelings and distress toward COVID-19 pandemic during different gestational periods (1 st trimester, 2 nd trimester, 3 rd trimester, and within 1 year postpartum).