Skip to main content
NCBI home page
PLOS One logoLink to PLOS One
. 2023 Apr 17;18(4):e0284578. doi: 10.1371/journal.pone.0284578

Maternal pandemic-related stress during pregnancy associates with infants’ socio-cognitive development at 12 months: A longitudinal multi-centric study

Sarah Nazzari 1,‡, Serena Grumi 2,‡, Giacomo Biasucci 3, Lidia Decembrino 4, Elisa Fazzi 5,6, Roberta Giacchero 7, Maria Luisa Magnani 4, Renata Nacinovich 8,9, Barbara Scelsa 10, Arsenio Spinillo 11, Elena Capelli 2, Elisa Roberti 2, Livio Provenzi 1,2,*; on behalf of the MOM-COPE Study Group
Editor: Claudine Irles12
PMCID: PMC10109481  PMID: 37068062

Abstract

Background

Prenatal maternal stress is a key risk factor for infants’ development. Previous research has highlighted consequences for infants’ socio-emotional and cognitive outcomes, but less is known for what regards socio-cognitive development. In this study, we report on the effects of maternal prenatal stress related to the COVID-19 pandemic on 12-month-old infants’ behavioral markers of socio-cognitive development.

Methods

Ninety infants and their mothers provided complete longitudinal data from birth to 12 months. At birth, mothers reported on pandemic-related stress during pregnancy. At infants’ 12-month-age, a remote mother-infant interaction was videotaped: after an initial 2-min face-to-face episode, the experimenter remotely played a series of four auditory stimuli (2 human and 2 non-human sounds). The auditory stimuli sequence was counterbalanced among participants and each sound was repeated three times every 10 seconds (Exposure, 30 seconds) while mothers were instructed not to interact with their infants and to display a neutral still-face expression. Infants’ orienting, communication, and pointing toward the auditory source was coded micro-analytically and a socio-cognitive score (SCS) was obtained by means of a principal component analysis.

Results

Infants equally oriented to human and non-human auditory stimuli. All infants oriented toward the sound during the Exposure episode, 80% exhibited any communication directed to the auditory source, and 48% showed at least one pointing toward the sound. Mothers who reported greater prenatal pandemic-related stress had infants with higher probability of showing no communication, t = 2.14 (p = .035), or pointing, t = 1.93 (p = .057). A significant and negative linear association was found between maternal prenatal pandemic-related stress and infants’ SCS at 12 months, R2 = .07 (p = .010), while adjusting for potential confounders.

Conclusions

This study suggests that prenatal maternal stress during the COVID-19 pandemic might have increased the risk of an altered socio-cognitive development in infants as assessed through an observational paradigm at 12 months. Special preventive attention should be devoted to infants born during the pandemic.

Introduction

Pregnancy is a period of complex development during which the fast-paced maturation of multiple neurophysiological and neurobehavioral domains is highly expectant on and susceptible to environmental stimuli [1, 2]. During this sensitive period, the stress experienced by pregnant women might be embedded into the progressing neurophysiological and neurobehavioral phenotype of the fetus [3, 4], contributing the long-lasting programming of infants’ development possibly via neuroendocrine [5, 6], inflammatory [7], and epigenetic mechanisms [8, 9]. Infants of women reporting higher levels of stress during pregnancy have been found to show altered brain connectivity [10, 11] and to be at greater risk for a series of negative outcomes, including socio-emotional and stress dysregulation [12], cognitive deficit or delay [13], altered neurodevelopment [14], and even mental and physical health problems later in life [15, 16].

Meta-analytic evidence of studies conducted during the COVID-19 pandemic highlighted that pregnant women experienced high levels of distress during this global traumatic experience [17] and that this might associate with stress-related biological alterations for both mothers [18, 19] and infants [4, 20]. A US-based study reported that pregnant women who experienced high levels of pandemic-related stress had also adverse perinatal outcomes [21]. Notably, independent research studies are concordant in suggesting that pregnancy during the pandemic per se, but not exposure to SARS-CoV-2 infection, might significantly raise the risk for altered developmental outcomes [22, 23]. For example, higher risk of neurodevelopmental delay at age 6 months in infants born during the pandemic was observed in a large cohort study in New York City and the effect was not significantly associated with the direct exposure to COVID-19 infection [23]. Moreover, the study by Imboden et al. [24] reported a slight decrease in the communication domain scores among 12-month-old infants compared to a cohort born before the pandemic. These initial findings suggest that the negative impact of the pandemic is specifically linked to the prenatal exposure to the healthcare emergency and maternal stress. Noteworthy, Sperber et al. [25] did not report any significant effects of the pandemic on maternal mental health and infants’ developmental outcomes among mother-infant dyads that experienced the onset of the pandemic during their first year of life. In this perspective, pregnancy seems to represent a specific window of susceptibility for the effects of maternal pandemic-related stress on infants’ development. Understanding pregnant women experiences of stress during the COVID-19 pandemic and assessing its effects during the first thousand days is critical to inform timely preventive interventions to protect infants’ socio-emotional and socio-cognitive development [26].

Socio-cognitive behaviors include the capacity to orient, point, and communicate toward a third object and are meant to reach a mature stage toward the end of the twelfth month of life [27]. They are generally considered as half-way between socio-emotional skills–e.g., approach toward other humans, prosocial and collaborative behaviors–and cognitive abilities–e.g., recognizing others as separated, understanding others’ goals, sharing a specific interest for an object in the environment [28]. For instance, by the end of the first year of life, infants show clear emergence of declarative pointing as a way to highlight their interest for something place in distal positions, to signal their state of mind to another interactive agent, thus organizing triadic exchanges and shared attention states that are key precursors of later theory of mind development [28–30]. Socio-cognitive behaviors can be easily observed and coded using videotaped of mother-infant interaction at this age, as they rely on a clear combination of gestures (e.g., pointing and reaching-out in the direction of the stimulus) and looking behaviors (e.g., orienting and fixating for appropriately prolonged time a specific target) [31]. These behaviors are critical indicators of potentially altered trajectories that might be, at least partly, programmed by prenatal exposure to adversities and that can lead to problems in psychological adjustment later in life [31]. Nevertheless, the majority of studies investigating the association between prenatal maternal stress and infant’s development focused on indices of either cognitive [32, 33] or socio-emotional [34, 35] development, whereas the effect of prenatal maternal stress on infants’ socio-cognitive development is largely unexplored.

In the present study, we investigated the association between maternal pandemic-related prenatal stress (PRS) and infants’ socio-cognitive behaviors at 12 months. To this aim, we assessed specific socio-cognitive behaviors–i.e., orienting, pointing, and active communication–toward auditory human and non-human stimuli in 12-month-old infants born during the COVID-19 pandemic. Specifically, socio-cognitive behaviors were assessed in terms of their rate (how many infants displayed them at least once) and frequency (how often infants displayed them). Higher levels of PRS were hypothesized to associate with a lower rate and frequency of socio-cognitive behaviors in infants.

Methods

Participants

This study is part of the multi-centric and longitudinal Measuring the Outcomes of Maternal COVID-19-related Prenatal Exposure (MOM-COPE) research project. The complete protocol of the project is reported elsewhere [36]. The original sample of 320 mother-child dyads were enrolled between May 2020 and February 2021 from eight neonatal units in Northern Italy. Subjects were considered eligible to the study if mothers were at least 18 years old, in the absence of prenatal and perinatal diseases or injuries, if delivery was at term (i.e., from 37 + 0 to 41–6 weeks of gestation), and if mothers tested negative for SARS-CoV-2 at delivery. Here we report on a subset of mother-child dyads (N = 91) that provided complete data for perinatal and 12-month assessments. Dyads from this sample did not significantly differ from those who withdrawn in terms of stress experienced during pregnancy and socio-demographic characteristics, except for gestational age (t(318) = -2.687, p = .008; M = 39.90; SD 1.05 for the final sample; M = 39.54, SD 1.09 for the participants who dropped out) and maternal education (t(317) = -2.124, p = .034; M = 15; SD 3.35 for the final sample; M = 15.86, SD 2.91 for the participants who dropped out). Complete sociodemographic characteristics of the participants who dropped out from the study as well as differences with the final study sample are reported in Supplementary Table 1 in S1 File.

Ethical considerations

The study was approved by the Ethics Committees of Pavia (Italy) and the participating hospitals. All mothers provided written informed consent to participate in the study.

Procedures and measures

Socio-demographic (maternal age, maternal education level) and neonatal (gestational age, birth weight, head circumference, neonatal length, Apgar score at minute 5) variables were obtained from medical records. Within 48 hours from delivery, mothers reported on their prenatal pandemic-related stress by filling in a self-report questionnaire (see S1 File). The questionnaire included six 5-point Likert scale items (1, not at all; 5, very much) on the emotional stress response to the COVID-19 emergency and an average maternal prenatal pandemic-related stress (PRS) score was obtained. The Cronbach’s α of this scale was .834. Direct (own positivity with or without symptoms) or indirect (positivity, hospitalization or death of a relative or significant others) physical exposure to the SARS-CoV-2 virus was tracked using dichotomous items [36]. Only two women reported testing positive to the COVID-19 during pregnancy.

At infants’ 12-month-age, mothers and infants participated from home to a remote videotaped face-to-face interaction. The remote observational procedure was structured according to recommendations from previous studies [37]. The online session was planned according to parents’ availability and the videorecording started only when the infant was in a quiet alert state. Prior to the interaction, mothers were instructed to set the smartphone or the tablet in a position that allowed the experimenters to have a full side-view of both the interactive partners. For the entire duration of the interaction, both the mother and the experimenter switched-off their screen; the experimenter also muted her mic while sharing PC reproduced sound. The interactive task lasted 6 minutes (Fig 1). Mother and infant interacted face-to-face with no use of toys or pacifier for two minutes. Subsequently, one of four prerecorded auditory stimuli was played three times every 10 seconds (Exposure episode). Two auditory stimuli were human sounds–i.e., “Ciao” (Italian for “Hello”) and “Che bello!” (Italian for “How nice!”)–whereas the other two auditory stimuli were non-human–i.e., water and mixer sound. All infants were exposed to the entire set of four auditory stimuli, nonetheless the order of exposure was counterbalanced among participants to avoid a sequence bias. During each Exposure episode, the mother maintained a still and neutral facial expression. After each 30-sec Exposure episode mother and infant resumed the face-to-face interaction (Reprise episode). Overall, a total of 102 mother-infant were recorded, but 11 videos were removed from the study because the task was not performed as required (in most of the cases the infant was crying in the highchair and was held by the mother for part of the task). All the videos had a frame and a video output quality that made them suitable for video-coding. For the purposes of the present study, specific infants’ socio-cognitive behaviors–i.e., orienting, communication, and pointing–were micro-analytically coded during the Exposure episodes. Orienting was coded as any occurrence of infants’ face and/or gaze clearly directed toward the source of the auditory stimuli. Communication was coded as any behavior of the infant clearly directed as communicating toward the source of the auditory stimuli, including waving bye with the hand, smiling, vocal production while looking at the sound source, etc. Pointing was coded as any clear gesture made with the index finger toward the source of the auditory stimuli. Two scores were obtained for each socio-cognitive behavior: a dichotomous index of display (1) or non-display (0) for the entire duration of the Exposure episodes and a continuous index (percentage of Exposure time). Two main trained coders, blind to all other study data, rated the videos independently. A subset of randomly selected videos (approximately 10%) were independently coded by a third independent coder and the percentage of inter-rater agreement was 99.48%.

Fig 1. Task overview. Note.

Fig 1

Open in a new tab

The order of the stimuli during the Exposure episodes was counterbalanced among subjects, keeping the alternation between human and non-human sounds. As the task was conducted with Italian mother-infant dyads, the human sounds were “Ciao” (for “Hello!”) and “Che bello!” (for “How nice!”).

Plan of analysis

Data reduction

Two Principal Component Analysis (PCA) were used to reduce the complexity of behavioral data. All the socio-cognitive continuous indexes separately obtained for human (i.e., orienting, communication, pointing) and non-human (i.e., orienting, communication, pointing) auditory stimuli were included. A non-constrained Varimax solution with criterion eigenvalue greater than 1 revealed a three-component structure (see Table 1). The principal components (PCs) clearly clustered together each socio-cognitive behavior independently from the nature (human or non-human) of the auditory stimuli, thus yielding three principal components: PC-ORI (orienting, 27.6% variance), PC-COM (communication, 23.3% variance), and PC-POI (pointing, 27.5% variance). A further constrained one-factor PCA solution with Varimax rotation was carried to test the presence of an over-arching socio-cognitive score (SCS). The constrained solution revealed one PC explaining 41.3% of total variance (Table 2).

Table 1. Non-constrained Varimax solution for socio-cognitive behaviors principal component analysis (N = 91).
Component
1, PC-ORI 2, PC-POI 3, PC-COM Uniqueness
Orienting to non-human sound 0.908     0.162
Orienting to human sound 0.836     0.270
Pointing to human sound   0.886   0.204
Pointing to non-human sound   0.870   0.185
Communication to human sound     0.920 0.151
Communication to non-human sound     0.703 0.322
Open in a new tab

Note. Loadings below .40 are hidden.

Table 2. Constrained Varimax solution for socio-cognitive behaviors principal component analysis (N = 91).
Component
  1, SCS Uniqueness
Pointing to non-human sound 0.734 0.462
Communication to non-human sound 0.720 0.481
Pointing to human sound 0.639 0.592
Orienting to human sound 0.634 0.598
Orienting to non-human sound 0.626 0.608
Communication to human sound 0.469 0.780
Open in a new tab

Note. Loadings below .40 are hidden.

Main analyses

Separate independent-sample t-tests were used to assess the presence of statistically significant differences in maternal prenatal PRS between infants who showed or not each of the socio-cognitive behaviors (i.e., orienting, communication, pointing). Bivariate Pearson’s correlations were used to assess the presence of statistically significant linear associations between maternal prenatal PRS and infants’ socio-cognitive development (i.e., PC-ORI, PC-COM, PC-POI, SCS). The Benjamini-Hochberg algorithm was used to adjust for multiple comparison bias. Lastly, separate hierarchical regression models were used to further test the effect of PRS on infants’ socio-cognitive development continuous measures while adjusting for potential infant and maternal confounders. All the analyses were performed using Jamovi 2.2.5.0 for Windows.

Results

The study sample characteristics are reported in Table 3. For full description of the original sample of 320 dyads see Supplementary Table 1 in S1 File. The sample was balanced for infants’ sex (44 females, 48.4%, 47 males, 51.6%). All infants oriented toward the sound during the Exposure episode, 80% exhibited any communication directed to the auditory source, and 48% showed at least one pointing toward the sound. The distribution of infants’ continuous socio-cognitive measures (i.e., PC-ORI, PC-COM, PC-POI, and SCS) is reported in Fig 2.

Table 3. Sample description (N = 91).

  Mean SD Minimum Maximum
Gestational age (weeks) 39.901 1.055 37.00 42.0
Birth weight (grams) 3382.593 421.426 2480.00 4435.0
Head circumference (cm) 34.247 1.313 30.00 39.0
Neonatal length (cm) 50.593 2.015 46.00 56.0
Apgar at minute 5 9.879 0.360 8.00 10.0
Maternal age (years) 33.154 4.681 18.00 46.0
Maternal education (years of study) 15.857 2.912 8.00 22.0
Open in a new tab

Fig 2. Distribution of infants’ socio-cognitive development continuous measures: Orienting to auditory source, PC-ORI; pointing to auditory source, PC-POI; communication to auditory source, PC-COM; cumulative socio-cognitive score, SCS.

Fig 2

Open in a new tab

Infants who did not display communication toward to the auditory source had mothers who reported greater prenatal PRS, t(89) = 2.14, p = .035, compared to communicating counterparts (Fig 3). Infants who did not display pointing toward to the auditory source had mothers who tended to report greater prenatal PRS, t(89) = 1.93, p = .057, compared to pointing counterparts (Fig 3). Significant linear associations emerged for maternal prenatal PRS and infants’ PC-COM, r = -.26, p = .015 (Fig 4), as well as with infants’ SCS, r = -.29, p = .006 (Fig 5). No significant associations emerged for PRS with both infants’ PC-ORI and PC-POI.

Fig 3. Maternal pandemic-related stress (PRS) during pregnancy by infants’ display of communication and pointing toward auditory stimuli at 12-month-age.

Fig 3

Open in a new tab

Fig 4. Linear association between maternal pandemic-related stress (PRS) during pregnancy and infants’ orienting (PC-ORI), pointing (PC-POI), communication (PC-COM) at 12-month-age.

Fig 4

Open in a new tab

Note. All outcome variables on the y-axis are standardized (mean = 0, sd = 1). In regression model predicting PC-ORI, standardized estimate for PRS = -0.09, p = 0.41; In regression model predicting PC-POI, standardized estimate for PRS = -0.16, p = 0.16; In regression model predicting PC-COM, standardized estimate for PRS = -0.26, p = 0.02.

Fig 5. Linear association between maternal pandemic-related stress (PRS) during pregnancy and infants’ over-arching socio-cognitive score (SCS) at 12-month-age.

Fig 5

Open in a new tab

Note. Infant SCS scores are standardized (mean = 0, sd = 1). Standardized estimate for PRS = -0.28, p = 0.01.

The hierarchical regression models included: infant sex, gestational age, and birth weight at step 1; maternal age and maternal educational level at step 2; PRS at step 3. Final models included all variables tested at Step 1, 2 and 3. When testing the model on infants’ SCS (see Table 4A), a significant ΔR2 = .07 (p = .010) was obtained only for step 3 which highlighted a significant effect of PRS on infants’ socio-cognitive development at 12 months, while controlling for potential confounders. When looking at specific socio-cognitive scores, the model was significant for infants’ PC-COM only at step 3, yielding a ΔR2 = .06 (p = .017) and a significant effect of PRS on infants’ PC-COM at 12 months (see Table 4D). The model was not significant when testing infants’ PC-ORI (see Table 4A) and PC-POI (see Table 4C).

Table 4. Hierarchical regression models testing the effect of maternal PRS on infants’ socio-cognitive development at 12 months (N = 91).

A. SCS
Step R2 ΔR2 Cross-step sig. Predictors Standardized coefficient 95% Confidence Interval Predictor sig.
1 0.02 Sex 0.24 [-0.17,0.66] 0.25
Gestational age -0.02 [-0.26,0.23] 0.89
Birth weight 0.06 [-0.17,0.30] 0.59
2 0.03 0.01 0.57 Maternal age -0.02 [-0.24,0.20] 0.88
Maternal education 0.08 [-0.13,0,30] 0.44
 3 0.10 0.07 0.01 PRS -0.28 [-0.49,-0.07] 0.01
B. PC-ORI
Step R2 ΔR2 Cross-step sig. Predictors Standardized coefficient 95% Confidence Interval Predictor sig.
1 0.05 Sex 0.41 [-0.02,0.84] 0.06
Gestational age 0.03 [-0.22,0.28] 0.78
Birth weight -0.04 [-0.28,0.20] 0.75
2 0.06 0.01 0.66 Maternal age -0.11 [-0.12,0.33] 0.34
Maternal education -0.01 [-0.24,0,21] 0.9
 3 0.07 0.01 0.41 PRS -0.09 [-0.31,0.13] 0.41
C. PC-POI
Step R2 ΔR2 Cross-step sig. Predictors Standardized coefficient 95% Confidence Interval Predictor sig.
1 < .01 Sex -0.02 [-0.45,0.42] 0.93
Gestational age -0.01 [-0.26,0.24] 0.92
Birth weight -0.03 [-0.28,0.22] 0.78
2 0.01 0.01 0.81 Maternal age -0.04 [-0.27,0.19] 0.72
Maternal education 0.04 [-0.19,0,27] 0.73
 3 0.03 0.02 0.16 PRS -0.16 [-0.38,0.06] 0.16
D. PC-COM
Step R2 ΔR2 Cross-step sig. Predictors Standardized coefficient 95% Confidence Interval Predictor sig.
1 0.04 Sex 0.03 [-0.39,0.44] 0.90
Gestational age -0.06 [-0.29,0.18] 0.65
Birth weight 0.22 [-0.02,0.45] 0.07
2 0.07 0.03 0.22 Maternal age -0.11 [-0.33,0.11] 0.32
Maternal education 0.14 [-0.08,0,35] 0.21
 3 0.13 0.06 0.02 PRS -0.26 [-0.46,-0.05] 0.02
Open in a new tab

Note. SCS, socio-cognitive score; PC-ORI, orienting; PC-POI, pointing; PC-COM, communication; PRS, pandemic-related stress.

Discussion

The present study is among the first to suggest the presence of a long-lasting association between maternal stress related to the experience of the pandemic during pregnancy and infants’ socio-cognitive development at 12 months of age. While all the infants enrolled in this study showed appropriate orienting to auditory stimuli source during the interactive task, the majority of them (80%) also showed active communications directed to the sound while only half of the sample (48%) displayed index finger pointing gesture. These findings confirm that deictic gesture is still developing at the end of the first year of life and they extend previous evidence from visual stimuli [28, 38] to auditory ones [39].

Notably, variability in socio-cognitive behaviors appeared to be at least partially associated with maternal stress exposure during pregnancy: those infants who did not show pointing and active communications were also exposed to higher levels of prenatal pandemic-related stress. Evidence on the long-lasting effects of prenatal exposure to the COVID-19 pandemic on infant development across the first postnatal year is still scarce. We previously reported on the association between PRS and infant temperament at 3 months of age [22, 40]. Furthermore, prenatal maternal distress during the pandemic was associated with infant brain connectivity at 3 months of age [41] and with poorer socio-emotional development at 2 months of age [42]. Consistently, infants born during the pandemic had higher risk of neurodevelopmental delays at 6 months [23] and lower communication scores at 12 months [24], as compared with historical cohorts born before the pandemic. Current findings suggest that the potential negative effects of PRS per se–even in absence of antenatal exposure to SARS-CoV-2 infection–on offspring’s development are likely to persist until 12 months after childbirth. Furthermore, while previous research on the effects of maternal prenatal stress mainly reported on emotional [43, 44] and cognitive [13, 45] infant outcomes, here we suggest that a precocious marker of altered developmental trajectory as consequence of adversity exposure during pregnancy might be traced in the early delay or deficit in specific socio-cognitive behaviors. Mechanisms underlying the observed associations are still unknown, but likely involved neuroendocrine, epigenetic, and environmental pathways. For example, pandemic-related stressors have been associated with greater maternal hair cortisol levels [19], thus suggesting a significant activation of maternal biological stress response systems during the pandemic, possibly leading to greater fetal exposure to maternal stress hormones and affecting later development. Furthermore, infants antenatally exposed to higher levels of PRS were found to have higher levels of SLC6A4 gene methylation at birth, which, in turn, was associated with later infant’s temperament [40]. Additionally, higher levels of DNA methylation of stress-related genes were previously reported in infants prenatally exposed to the COVID-19 pandemic [46], thus suggesting that epigenetic mechanisms might play a role in mediating the embedding of the prenatal exposure to the pandemic on infants’ development. Lastly, maternal postnatal distress was found to partially mediate the effects of maternal PRS respectively on infants’ socio-emotional development at 2 months [42] and regulatory capacity at 3 months [22]. Likewise, the quality of mother-infant postnatal relationship might play a mediating or moderating [20, 47] role in the observed associations. Noteworthy, Hane and colleagues [48] recently showed that infants of emotionally connected dyads showed more approach-seeking behaviors and social engagement as well as a specific autonomic activation patterns as compared to those from non-connected dyads during the face-to-face still-face task. Future studies would benefit from the inclusion of physiological measures along with behavioral markers in order to elucidate whether the observed differences in children’s socio-cognitive outcomes possibly reflect differences in autonomic responding, in line with the novel hypothesis of an autonomic socio-emotional “reflex” pathway [49]. This would contribute shedding light on the mechanisms underlying the effects of PRS on infant development.

Interestingly, the present study further supports the hypothesis of a linear association between prenatal stress and infants’ early socio-cognitive development at 12 months. This association remained significant even after controlling for potential infants and maternal confounders. The effect was specifically evident for infants’ tendency to produce active communications (e.g., waving by, smiling, vocalizing) toward an auditory stimulus while interacting with their mother. A significant relationship between prenatal stress exposure and child outcomes has often been reported [50, 51]. The current findings extend this evidence by showing that even the association between PRS and infants’ socio-cognitive development behaves in a linear fashion, with important clinical implications. Assessment of maternal psychological well-being during the pandemic is fundamental as PRS is likely to have a small though detectable impact even at subclinical levels. Furthermore, formal and informal support should be offered to expectant families, even more during a pandemic, in order to protect maternal and infant perinatal mental health.

The study has limitations. First, the sample size is relatively low which also depends on the longitudinal and observational nature of the project which, especially during the pandemic time, has resulted in remarkable sample attrition (only 28.44% of the dyads of the original sample took part in the observational task planned at 12 months). Additionally, potential systematic patterns of missingness limit generalizability of the current findings. In particular, while small differences in gestational age and maternal education were found among dyads who attended the follow-up and dyads who withdrew, only variables collected at birth were available for comparison among the two groups, thus possible selective attrition on unobserved variables remained an issue. Second, we adapted previously available tasks to assess infants’ socio-cognitive behavior remotely by developing a new observational procedure. The accumulating experiences during the pandemic period about remote data collection [37] informed the present procedure and allowed us to obtain a high number of videos suitable for remote coding. However, the findings derived from the remote collection in a home setting are not directly comparable to those from lab settings and should be considered descriptive and not clinically informative. Third, the prenatal stress questionnaire was developed ad hoc for this study prioritizing sensitivity to the specific and unprecedented nature of COVID-19 emergency over measure standardization. Lastly, the correlational design prevents us from establishing causal links.

Conclusions

The present study suggests that infants who were born during the pandemic might be at risk for an altered trajectory in socio-cognitive development and that this risk might be at least partially associated with the extent of maternal pandemic-related prenatal stress. These findings add to previous reports on socio-emotional and cognitive developmental risk in infants’ born during the pandemic. As such, we suggest that this population should be specifically followed-up to inform smarter preventive care during and after the pandemic period.

Supporting information

S1 File

(DOCX)

Click here for additional data file. (18.2KB, docx)

Acknowledgments

The authors are thankful to the colleagues of the MOM-COPE Study Group: G. Bensi, R. Bonini, R. Borgatti, A. Cavallini, R. Falcone, B. Gardella, G. Kullmann, V. Manfredini, F. Masoni, S. Orcesi, D. Pantaleo, G. Pettenati, B. Pietra, C. Pisoni, F. Prefumo, V. Spartà.

Data Availability

The raw data related to the present publication are available from the Zenodo repository (url: 10.5281/zenodo.7516250).

Funding Statement

This study is supported by funds from the Italian Ministry of Health (Cinque per Mille, 2017; Ricerca Corrente 2022) and Fondazione Roche Italia (Fondazione Roche per la Ricerca Indipendente 2020) to author LP. The funders had no role in study design, data collection, analysis, decision to publish, or preparation of the manuscript.

References

Decision Letter 0

Claudine Irles

14 Nov 2022

PONE-D-22-23782Maternal pandemic-related stress during pregnancy associates with infants’ socio-cognitive development at 12 months: A longitudinal multi-centric studyPLOS ONE

Dear Dr. Provenzi,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

 Please provide in the revised manuscript where the data may be found (doi). In the original submission the authors only stated: "All the raw data related to the present publication are available from the XXX database (doi: XXX)."

Please submit your revised manuscript by Dec 29 2022 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

  • A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Claudine Irles, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

3. One of the noted authors is a group or consortium “MOM-COPE Study Group”. In addition to naming the author group, please list the individual authors and affiliations within this group in the acknowledgments section of your manuscript. Please also indicate clearly a lead author for this group along with a contact email address.

4. Please upload a copy of Supplementary File S1 which you refer to in your text on page 5 and 6.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: the manuscript technically sound, and do the data support the conclusions. I am not sure but I am assuming the data has been made available. the manuscript is well written in a way that makes the subject very understandable.

Reviewer #2: The present study evaluates the associations between prenatal maternal stress during the pandemic and infant socio-cognitive functioning at 12-months postpartum. The authors find that infants of mothers who reported higher pandemic-related stress during pregnancy were less likely to point and demonstrate communication when presented with auditory stimuli. The study is clearly written, and has the potential to make a valuable contribution to the literature.

However, several points should be addressed. Overall, I take issue with the analytic approach (t-tests and correlations) and the cherry picking of results (frequency of socio-cognitive behaviors are not reported or interpreted, only the presence of them). Critically, the content of the pandemic stress measure is not presently available (the manuscript says it is available in the Supplement, but I do not have access to it). My recommendation for publication relies heavily on the availability of these items, since it is not a validated scale.

Intro

• The authors cite evidence related to increases in depression during the pandemic, but the main predictor in this paper is perceived stress. Clinical symptoms of mental health disorders are different from perceived stress levels, and we should avoid conflating the two constructs.

Method

• Need more information about how the domains of socio-cognitive functioning were coded. Were there multiple coders, and how was reliability assessed?

• The authors mention there was “remarkable sample attrition,” thus they only report results on infants with complete data. What was the original sample size, and what are the patterns of missing data? It is likely that large amounts of attrition will bias results. The authors should also consider using some method of imputation to reduce bias.

• Limitations of the remote interaction task should be considered. Are there any other published studies that have coded a remote parent-child interaction task? This is a novel method, and any evidence linking remote recordings to in-person recordings should be cited. Additionally, I would appreciate more details on the remote interaction (overall quality of the recordings, percent of recordings that were unusable, instructions given to the parent, etc) as this may be of use to future researchers.

Results

• An eligibility criteria for the study required mothers to test negative for COVID at delivery. The authors cite important evidence by Shuffrey et al, which reports neurodevelopmental delays of infants born to mothers who reported a COVID infection during pregnancy. Therefore, the authors should conduct a robustness check on mothers who tested positive for COVID during pregnancy with infant SCS scores (frequency & rate) if possible as a supplemental analysis

• The authors collected data on not just the occurrence, but also the frequency by which infants displayed these pointing/communication behaviors during the auditory task. These results are missing from the manuscript

• The authors test for correlations between potential confounds but do not report the results of these tests. Why not just control for confounds using linear regression? Even if the covariates are not significantly related to either the predictor or outcome variable, they are theoretically important to the outcome and accounting for them will result in a more precise estimate. I recommend the authors use logistic regression for the dichotomous indicators to obtain odds ratios (and linear regression with standardized predictor variables for the continuous outcomes), and include covariates for the demographic characteristics listed

• Did you account for age of the infant at time of the assessment? This should also be included as a covariate, as there might be differences in developmental timing of pointing/communication abilities by age

• What is the socioeconomic spread of the sample? Maternal education is reported, but is there any further information such as income?

• In observing Figure 2, there appears to be an outlier (high PC-COM score). Are these results robust to outliers? This should be reported in the supplement, given the small sample

Discussion:

• The discussion begins with the sentence “the present study is among the first to suggest the presence of a long-lasting effect of maternal stress…”. This terminology is again used in the next paragraph “…socio-cognitive behaviors appeared to be at least partially affected by maternal stress exposure during pregnancy.” This implies causation.

• This study would be helpful to cite in either the intro or discussion – it is not correlational, which provides more compelling evidence: Imboden, A., Sobczak, B. K., & Griffin, V. (2021). The impact of the COVID-19 pandemic on infant and toddler development. Journal of the American Association of Nurse Practitioners. https://doi.org/10.1097/JXX.0000000000000653

• This recent study also is a natural experiment: Sperber, J. F., Hart, E. R., Troller-Renfree, S. V., Watts, T. W., & Noble, K. G. (2022). The effect of the COVID-19 pandemic on infant development and maternal mental health in the first 2 years of life. Infancy. https://doi.org/10.1111/infa.12511

• I challenge the authors to remove terminology such as “suboptimal” development, as it assigns value judgements to developmental trajectories. It is possible (and there is a large body of literature arguing this) that developmental differences may be “suboptimal” according to traditional Western contexts, but may be adaptive and advantageous in adverse contexts. I encourage the writers to remove value judgements of developmental trajectories, and explore this idea further – how might socio-cognitive difference due to early stress exposure be adaptive in certain contexts?

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Attachment

Submitted filename: PlosOne review (Final).docx

Click here for additional data file. (14.2KB, docx)
Attachment

Submitted filename: R&R covid provenzi.docx

Click here for additional data file. (17.3KB, docx)
PLoS One. 2023 Apr 17;18(4):e0284578. doi: 10.1371/journal.pone.0284578.r002

Author response to Decision Letter 0

9 Jan 2023

REVIEWER #1

1) The manuscript technically sound, and do the data support the conclusions. I am not sure but I am assuming the data has been made available. The manuscript is well written in a way that makes the subject very understandable. In this study, the authors have assessed possible effects of COVID related prenatal maternal stress on infant socio-cognitive development. They found a significant negative linear association between maternal stress and the Infant’s socio-cognitive score at 12 months and conclude that findings suggest preventive measures should be taken during the pandemic. The authors are to be congratulated on a very interesting paper. The study is well-designed, the manuscript is well-written and the findings add important evidence about the impact of COVID on human health. This paper fits the scope, and meets the criteria for publication in Plos One and in my opinion should be accepted.

REPLY. We wish to thank Reviewer 1 for appreciating this work and for highlighting its implications. The raw data will be made available on permanent third-party archive (Zenodo). We added in text a data availability statement to clarify this aspect.

2) I do have some suggestions that I think could make the paper even more interesting. Introduction: The still-face is a well-validated paradigm designed to measure how an infant responds to a maternal stressor. The author’s have introduced a novel auditory stimulus during the still-face segment, coded orienting behaviors of the infant, and interpreted the results according to accepted orienting theory, which focuses on the cortically controlled orienting “response” phenomenon. The author’s might want to cite a recent paper that extensively reviews the orienting reflex theory, from Pavlov to present (Ludwig and Welch 2022 Frontiers in Psych) and concludes by identifying a novel autonomic socioemotional “reflex” (ASR) pathway. To that point, a recent still-face study coded orienting as a socioemotional reflex behavior pertaining to the mother-infant “relationship” – validating a new mother-infant emotional connection construct (see Hane AA et al. 2016 Act Paediatr).

REPLY. We thank Reviewer 1 for these suggestions. The suggested references have now been included in the manuscript to broaden the discussion about possible mechanisms underlying the observed associations in this study.

3) Discussion: The authors might want to refer to other studies that showed that maternal stress before, during and after pregnancy, whatever the cause, has profound effects on the development and lifelong function of the infants neurocognitive development. (Sharma etal 2022 Clin Epigenetics). The authors might want to cite some studies that looked at perinatal stress that emerged after 9-11. (Levendosky et al 2021 BMJ Open) (Van Sjelghem etal 2022 Early Hum Dev), which are similar to those emerging from Pandemic stress.

REPLY. Thank you for these suggestions, we have added in the introduction and discussion the reference to the study by Sharma et al. (2022) to strengthen the rationale about how prenatal adversities may be embedded into the progressing neurophysiological and neurobehavioral phenotype of the fetus.

4) The authors note in Discussion; “The effect was specifically evident for infants’ tendency to produce active communications (e.g., waving by, smiling, vocalizing) toward an auditory stimulus while interacting with their mother.” This observation deserves a paragraph to consider possible explanations of the orienting reflex data in this study. For instance, ASR theory predicts that those infants emotionally connected to the mother will display approach behaviors (i.e., attend to, orient to, etc.), while those not emotionally connected would display lessor or avoidant orienting behaviors. While mother-infant connection was not assessed in this study, it would be interesting in future studies to test the ASR hypothesis by conducting post-hoc analyses of mother-infant video interactions collected in this study using the Welch Emotional Connection Screen (WECS) to assess infant orienting to the mother and their emotional connection. Going forward, should the authors do further longitudinal follow-up studies, autonomic measures, such as HR and vagal tone could provide clinically relevant information from this study.

REPLY. We thank Reviewer 1 for making this point. Following also comment 2, we have now broadened the discussion of our findings to include possible environmental mechanisms encompassing the quality of mother-infant relationship that might underline the observed associations.

REVIEWER #2:

The present study evaluates the associations between prenatal maternal stress during the pandemic and infant socio-cognitive functioning at 12-months postpartum. The authors find that infants of mothers who reported higher pandemic-related stress during pregnancy were less likely to point and demonstrate communication when presented with auditory stimuli. The study is clearly written, and has the potential to make a valuable contribution to the literature.

REPLY. We wish to thank Reviewer 2 for carefully reading our manuscript and for his/her useful comments.

However, several points should be addressed. Overall, I take issue with the analytic approach (t-tests and correlations) and the cherry picking of results (frequency of socio-cognitive behaviors are not reported or interpreted, only the presence of them). Critically, the content of the pandemic stress measure is not presently available (the manuscript says it is available in the Supplement, but I do not have access to it). My recommendation for publication relies heavily on the availability of these items, since it is not a validated scale.

REPLY. We thank the Reviewer for raising these points. We addressed the Reviewer’s comments to the analytical approach in response to comments number 6 and 7.

Regarding to the ad-hoc pandemic-related stress measures, we regret that the reviewer did not have access to the supplementary material that we submitted together with the manuscript. We report also here the list of the ad hoc-items developed to assess physical exposure to COVID-19 (A) and pandemic related stress (B) during pregnancy. As noted by the Reviewer 2, the pandemic-related stress scale was specifically designed to assessed maternal stress during the COVID-19 pandemic and thus was not previously validated. Nevertheless, as now reported in the manuscript, the scale has a good internal consistency (Cronbach’s α =.834) and was already used in many previous published studies, such as:

• Grumi, Provenzi, Accorsi et al. (2021) https://doi.org/10.3389/fpsyt.2021.716488

• Provenzi, Grumi, Altieri, et al. (2021) https://doi.org/10.1017/S0954579421000766

• Provenzi, Mambretti, Villa, et al (2021) https://doi.org/10.1038/s41598-021-95053-z

• Roberti, Giacchero, Grumi et al. (2022) https://doi.org/10.1007/s10995-022-03540-0

• Nazzari, Grumi, Mambretti et al. (2022) https://doi.org/10.1038/s41398-022-02160-0

• Provenzi, Villa, Mambretti et al. (2022) https://doi.org/10.3389/fpsyt.2022.950455

A. Physical exposure (response: Yes, No)

During pregnancy…

1 I tested positive for COVID-19

2 I had symptoms reminiscent of COVID-19

3 I had contacts with relative or friends who tested positive for COVID-19

4 I live in a high contagion zone (e.g., red zone)

5 I had contacts with relatives or friends who live in a high contagion zone (e.g., red zone)

6 One of my relatives or friends was hospitalized due to COVID-19 infection

7 One of my relative or friends died with COVID-19

B. Pandemic related stress (Response: 5-point Liker scale)

During pregnancy…

1= not at all; 2= slightly; 3= Moderately; 4= Very much; 5= Extremely;

1 How much worried were you about the risk of COVID-19 infection?

2 How much did you feel that your pregnancy was at risk due to COVID-19 pandemic?

3 How much did you fear for your health?

4 How much did you fear for your baby's health?

5 How much did you feel that you were losing confidence in your health?

6 How much did you feel you had lost faith in medicine?

Intro

1) The authors cite evidence related to increases in depression during the pandemic, but the main predictor in this paper is perceived stress. Clinical symptoms of mental health disorders are different from perceived stress levels, and we should avoid conflating the two constructs.

REPLY. To avoid misunderstanding, we removed from the introduction section references about depression and replaced them with evidence about prenatal maternal distress during the pandemic (Yan, Ding, & Guo, 2020).

Method

2) Need more information about how the domains of socio-cognitive functioning were coded. Were there multiple coders, and how was reliability assessed?

REPLY. We thank the Reviewer for highlighting that more information on the coding procedure was needed. Two main coders were trained on the coding procedure described in the manuscript by the PI, were blind to all other study data and rated the videos independently. A subset of randomly selected videos (approximately 10%) were independently coded by two independent coders and the percentage of inter-rater agreement was 99.48%. We have added these details in the text.

3) The authors mention there was “remarkable sample attrition,” thus they only report results on infants with complete data. What was the original sample size, and what are the patterns of missing data? It is likely that large amounts of attrition will bias results. The authors should also consider using some method of imputation to reduce bias.

REPLY. The initial sample size at childbirth was n= 320 mother-child dyads, but only n=91 dyads took part in the observational task planned at 12 months, with an attrition rate of 71.56% (we added this details in the text). All infants were full-term births without clinical conditions. Dyads of the final sample of the present study did not significantly differ from those who dropped out in terms of stress experienced during pregnancy and socio-demographic characteristics, except for gestational age (M= 39.90; SD 1.05 for the final sample; M= 39.54, SD 1.09 for the participants who dropped out) and maternal education (M= 15; SD 3.35 for the final sample; M= 15.86, SD 2.91 for the participants who dropped out):

• Pandemic related stress t(318)= -.489, p=.625

• Gestational age t(318)= -2.687, p=.008

• Birth weight t(318)= -1.123, p= .262

• Infants’ sex χ2(1)= .053, p= .817

• Apgar score (min 5)= t(317)= -.232, p= .816

• Maternal age t(316)= .418, p=.676

• Maternal education t(317)= -2.124, p= .034

• Maternal job χ2(1)= .002, p= .967

These details have now been included in the methods section and acknowledged in the limitations. Lastly, we did not consider using any methods of imputation given that among dyads who took part in the observational task there was no missing data.

4) Limitations of the remote interaction task should be considered. Are there any other published studies that have coded a remote parent-child interaction task? This is a novel method, and any evidence linking remote recordings to in-person recordings should be cited. If not, this is an important limitation to acknowledge, as the home environment is different from a lab environment (and, there might be connectivity issues with coding interactions remotely, etc). Additionally, I would appreciate more details on the remote interaction (overall quality of the recordings, percent of recordings that were unusable, instructions given to the parent, etc) as this may be of use to future researchers.

REPLY. The reviewer is right, this is a novel method that implies specific strengths and limitations. It allowed us to collect observational data during the pandemic, but it will be helpful also in the post-pandemic period to maximize the participation in longitudinal studies for families living far from labs/hospitals. Moreover, it allows to collect more ecological data, observing the infants’ behaviors in a context that is familiar for them. However, it includes also specific challenges to take care of as: verifying the quality of internet connection; engaging the mother in positioning the smartphone/tablet in the right position to have the full side-view of both the interactive partners; giving instructions to the mother to set the smartphone/tablet with a black background so as not to distract the child.

Overall, our procedure was consistent with recommendations by Shin et al. (2021) doi: 10.3389/fpsyg.2021.703822 for behavioral remote data collection in the home setting. And this allowed us to collect a high number of video suitable for coding. Indeed, a total of 102 tasks were recorded, but only 91 were adapt for coding and were used for the present study. All videos had good output quality which made them suitable for coding. A total of 11 videos were removed from the study because the task was not performed as expected: in most cases the infant was crying in the highchair and was held by the mother for part of the task; in one case another child entered in the room during the recording.

A validation study of this remote auditory task is being submitted to another journal. We have added details in the manuscript and presented limitations in the discussion section.

Results

5) An eligibility criteria for the study required mothers to test negative for COVID at delivery. The authors cite important evidence by Shuffrey et al, which reports neurodevelopmental delays of infants born to mothers who reported a COVID infection during pregnancy. Therefore, the authors should conduct a robustness check on mothers who tested positive for COVID during pregnancy with infant SCS scores (frequency & rate) if possible as a supplemental analysis.

REPLY. We think that there was a misunderstanding regarding our reference to Shuffrey’s evidence: his results suggested that birth during the pandemic, but not in utero exposure to maternal SARS-CoV-2 infection, was associated with differences in neurodevelopment at age 6 months. Unfortunately, we cannot try to replicate this result in our study about early socio-cognitive development. Indeed, as now specified also in the manuscript, in our final sample only two mothers reported testing positive to COVID-19 during pregnancy. Thus, it is not possible to perform supplemental analysis to control the possible specific influence of the COVID infection. However, as the current results are based on a sample of infants not exposed to the infection during pregnancy, strengthen the hypothesis that prenatal pandemic-related stress per se, rather than the infection, may be associated to the infants’ early socio-cognitive development.

6) The authors collected data on not just the occurrence, but also the frequency by which infants displayed these pointing/communication behaviors during the auditory task. These results are missing from the manuscript

REPLY. The descriptive data for the dichotomous occurrence (no = 0, yes = 1) is reported at the beginning of the Results section. We have now included a Figure (new Figure 1) reporting the distribution of the continuous (frequency) measure of pointing and communication. We agree with the reviewer that this adds to the clarity of the manuscript.

7) The authors test for correlations between potential confounds but do not report the results of these tests. Why not just control for confounds using linear regression? Even if the covariates are not significantly related to either the predictor or outcome variable, they are theoretically important to the outcome and accounting for them will result in a more precise estimate. I recommend the authors use logistic regression for the dichotomous indicators to obtain odds ratios (and linear regression with standardized predictor variables for the continuous outcomes), and include covariates for the demographic characteristics listed

REPLY. We thank the Reviewer for this suggestion. We have now included separate hierarchical regression models for each of socio-cognitive outcomes (i.e., SCS, PC-ORI, PC-POI, PC-COM), adjusting for potential infants (step 1) and maternal (step 2) confounders. We have adjusted the discussion accordingly. Also we revised all the Figures to make them clearer for the reader.

8) Did you account for age of the infant at time of the assessment? This should also be included as a covariate, as there might be differences in developmental timing of pointing/communication abilities by age.

REPLY. All infants were tested at 12 months ± 14 days. Specifically, mean age at testing was 376 (sd = 10, min = 332, max = 400): the youngest infant was 10 months and 27 days old, whereas the oldest was 1 year, 1 month and 3 days old. As such, considering the very narrow age range we did not expect any significant effect of age on the observed outcomes.

9) What is the socioeconomic spread of the sample? Maternal education is reported, but is there any further information such as income?

REPLY. Unfortunately we did not have any detailed information about the parents’ income. While we did not have any precise assessment of the socioeconomic status of the participants, we can confirm that the sample was a low-risk community sample.

10) In observing Figure 2, there appears to be an outlier (high PC-COM score). Are these results robust to outliers? This should be reported in the supplement, given the small sample

REPLY. We excluded the outlier and repeated the analyses. Findings – in terms of statistical significance – did not differ, so we keep the complete set of 91 dyads.

Discussion:

11) The discussion begins with the sentence “the present study is among the first to suggest the presence of a long-lasting effect of maternal stress…”. This terminology is again used in the next paragraph “…socio-cognitive behaviors appeared to be at least partially affected by maternal stress exposure during pregnancy.” This implies causation.

REPLY. We agree with the reviewer about the fact that the longitudinal nature of our data is no guarantee for drawing valid causal inferences, excluding all competing explanations. We rephrased these sentences in terms of association.

12) This study would be helpful to cite in either the intro or discussion – it is not correlational, which provides more compelling evidence: Imboden, A., Sobczak, B. K., & Griffin, V. (2021). The impact of the COVID-19 pandemic on infant and toddler development. Journal of the American Association of Nurse Practitioners. https://doi.org/10.1097/JXX.0000000000000653

REPLY. Thank you for the suggestion, we have cited this study both in the introduction and discussion.

13) This recent study also is a natural experiment: Sperber, J. F., Hart, E. R., Troller-Renfree, S. V., Watts, T. W., & Noble, K. G. (2022). The effect of the COVID-19 pandemic on infant development and maternal mental health in the first 2 years of life. Infancy. https://doi.org/10.1111/infa.12511

REPLY. Thank you for the suggestion, we have cited this study in the introduction.

14) I challenge the authors to remove terminology such as “suboptimal” development, as it assigns value judgements to developmental trajectories. It is possible (and there is a large body of literature arguing this) that developmental differences may be “suboptimal” according to traditional Western contexts, but may be adaptive and advantageous in adverse contexts. I encourage the writers to remove value judgements of developmental trajectories, and explore this idea further – how might socio-cognitive difference due to early stress exposure be adaptive in certain contexts?

REPLY. We thank the reviewer for this input, we have rephrased that terminology accordingly.

Attachment

Submitted filename: momcope_sociocogn_replies.docx

Click here for additional data file. (32.1KB, docx)

Decision Letter 1

Claudine Irles

17 Feb 2023

PONE-D-22-23782R1Maternal pandemic-related stress during pregnancy associates with infants’ socio-cognitive development at 12 months: A longitudinal multi-centric studyPLOS ONE

Dear Dr. Provenzi,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Apr 03 2023 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

  • A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Claudine Irles, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: All comments have been addressed

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: Yes

Reviewer #3: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

Reviewer #3: No

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: The authors did an excellent job responding to reviewer concerns. The current manuscript is well-written, presents compelling evidence, and will make a valuable contribution to the literature. I am happy to recommend this paper for publication. A few small notes:

Though statistically significant differences were found in maternal education and gestational age between those who attritted from the study and those who participated in the 12-month assessment, the magnitude of this difference is small (and ultimately, inconsequential). I think the authors should acknowledge this and present it as such using the mean differences as evidence. Rather, the bigger concern is that there are unobserved difference between the hundreds of dyads that attritted from the study and the follow-up sample (this is quite likely). I request the authors raise this concern in the limitation section, rather than the statistically significant differences found in the t-test.

Small typo found on the first line of page 11: “thought” should be “though”

Figure 4 is a bit overwhelming to the eye. I recommend adding sub-titles above each scatterplot regarding which infant outcome is being represented (rather than only depending on the Y-axis) and limiting only 2 infant outcomes to a page. Additionally, I recommend putting the standardized coefficient and p-value from the regression analysis under each figure to contextualize the figure for the reader. Also specify for all figures when variables are standardized.

Reviewer #3: This is a highly interesting and relevant manuscript relating pandemic related stress to socio-cognitive development in 12 month old infants. Many of the comments raised in the previous version have been well addressed. As I did not perform the previous review I have a few relatively minor comments.

1. The authors argue that they found a "fairly lnear or dose response relationship...", however, the regression models that they ran did not really test whether the relationships are linear. To do this the investigators would need to evaluate either a quadratic term for the PRS measure or perform spline analyses. This is important because the shape of the dose response curve matters when policy makers are considering the availability of 'formal and informal' support (although such support is always good). Indeed, inspecting figure 4 leads me to conjecture that (a) for some of the outcomes such as SCS and PC-ORI the relationships are not linear and (b) the relationships may differ for male and female infants.

2. The large degree of missing data (ie loss to follow up) is troubling, especially since follow up is related to important confounders such as maternal education and gestational age. This could be better handled by performing inverse probability weighting which would account for some of the potential bias due to differential loss to follow up.

3. Please use more descriptive headers for the tables. For example, in tables 1 and 2 should say something about the sample size, and whether this was an exploratory or confirmatory PCA. Table 3 should also have the sample size and indicate if this is the larger sample or the analytic sample. further this table should give the data to evaluate the loss to follow up, i.e. compare those who were followed at 12 months to those who were not.

4. Please indicate whether successive steps include the variables in the previous step. That would make sense, but it is not clear the way the table is formatted.

5. Because the scores on the socio-cognitive development measures were not normally distributed there is a chance that extreme outliers may influence the results. Did the authors also consider the log transform of these variables, which would normalize the distribution.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

Reviewer #3: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2023 Apr 17;18(4):e0284578. doi: 10.1371/journal.pone.0284578.r004

Author response to Decision Letter 1

7 Mar 2023

Article ID: PONE-D-22-23782R1

Title: Maternal pandemic-related stress during pregnancy associates with infants’ socio-cognitive development at 12 months: A longitudinal multi-centric study

Replies to comments from reviewers

Reviewer #2

1. The authors did an excellent job responding to reviewer concerns. The current manuscript is well-written, presents compelling evidence, and will make a valuable contribution to the literature. I am happy to recommend this paper for publication.

Reply. Thanks for appreciating our revision.

2. A few small notes: Though statistically significant differences were found in maternal education and gestational age between those who attritted from the study and those who participated in the 12-month assessment, the magnitude of this difference is small (and ultimately, inconsequential). I think the authors should acknowledge this and present it as such using the mean differences as evidence. Rather, the bigger concern is that there are unobserved difference between the hundreds of dyads that attritted from the study and the follow-up sample (this is quite likely). I request the authors raise this concern in the limitation section, rather than the statistically significant differences found in the t-test.

Reply. Thanks for this suggestion, we have modified the text as follows:

“Additionally, potential systematic patterns of missingness limit generalizability of the current findings. In particular, while small differences in gestational age and maternal education were found among dyads who attended the follow-up and dyads who withdrew, only variables collected at birth were available for comparison among the two groups, thus possible selective attrition on unobserved variables remained an issue.”

3. Small typo found on the first line of page 11: “thought” should be “though”

Reply. Thank you again for careful reading of our manuscript. We have fixed the typo.

4. Figure 4 is a bit overwhelming to the eye. I recommend adding sub-titles above each scatterplot regarding which infant outcome is being represented (rather than only depending on the Y-axis) and limiting only 2 infant outcomes to a page. Additionally, I recommend putting the standardized coefficient and p-value from the regression analysis under each figure to contextualize the figure for the reader. Also specify for all figures when variables are standardized.

Reply. We agree with Rev #2. Figure 4 has been split into two new figures: the new Fig 4 reports the scatterplots for ORI, POI, COM; the new Fig 5 reports the scatterplot for the cumulative SCS. Statistics are reported in the text – anyway, they are also replicated in the Table legend now. These legends also highlight that all outcome variables on the y-axis are standardized.

Reviewer #3

This is a highly interesting and relevant manuscript relating pandemic related stress to socio-cognitive development in 12 month old infants. Many of the comments raised in the previous version have been well addressed. As I did not perform the previous review I have a few relatively minor comments.

Reply. Thanks for appreciating our work.

1. The authors argue that they found a "fairly linear or dose response relationship...", however, the regression models that they ran did not really test whether the relationships are linear. To do this the investigators would need to evaluate either a quadratic term for the PRS measure or perform spline analyses. This is important because the shape of the dose response curve matters when policy makers are considering the availability of 'formal and informal' support (although such support is always good). Indeed, inspecting figure 4 leads me to conjecture that (a) for some of the outcomes such as SCS and PC-ORI the relationships are not linear and (b) the relationships may differ for male and female infants.

Reply. We thank the Reviewer for making this point. It should be highlighted that the sentence “fairly linear or dose response relationship” was not referred to our findings, but to previous research – nonetheless, it has now been removed to avoid misinterpretations. At the same time, non-linearity of the observed associations was examined by the inclusion of a quadratic term for PRS within the regression models. Results for a quadratic effect of PRS on the study outcomes are as follows:

• No significant effect for ORI (as for linear testing)

• No significant effect for POI (as for linear testing)

• A tendency to statistical significancy for COM (R-squared = .10, p = .059; while the PRS linear term yielded a significant effect)

• A significant effect for SCS (R-squared = .09, p = .027; the linear effect had R-squared = .10, p = .017).

As such, we did not change our analytical plan. Nonetheless, adding the comparison between the linear and the quadratic association was not within the aims of the paper and we do not traced in previous literature enough support to make specific hypotheses on the shape of the linear or curvilinear association. Consistently, we opted to report on this linear association as it was tested as linear – of course, this does not rule out that in larger populations or in different settings the association might be curvilinear. For instance, few studies provided evidence that the association between maternal prenatal stress and children’s outcomes might be U-shaped, with mild to moderate levels of maternal distress during pregnancy actually promoting child development in healthy samples and being associated with more optimal outcomes, as compared to too little or too much prenatal maternal stress (e.g., Fernandes et al., 2014; DiPietro et al., 2006; Davis et al., 2017).

2. The large degree of missing data (ie loss to follow up) is troubling, especially since follow up is related to important confounders such as maternal education and gestational age. This could be better handled by performing inverse probability weighting which would account for some of the potential bias due to differential loss to follow up.

Reply. We thank the Reviewer for raising this issue. We agree that sample attrition is actually a significant issue in the current sample and is probably related to the pandemic time when recruitment occurs that limited participants’ engagement by allowing only remote data collection. Unfortunately, we only have information collected at birth about the dyads who withdrew from the study and, as also highlighted by Reviewer 2, while small differences were detected in terms of gestational age and maternal education among participants and not-participants to the follow-up phases, a number of unobserved differences are likely to exist, thus making our sample not representative of all observations. Thus, we do not feel appropriate to use inverse probability weighing in the current sample, as its effectiveness in “correcting” the selection bias highly depends on the availability of enough information, for the entire population, to predict the non-missingness probability (Oris et al., 2016; Narduzzi et al., 2014).

Following also comments from Reviewer 2, this has now been better acknowledged in the limitation section.

Oris, M., Roberts, C., Joye, D., & Stähli, M. E. (2016). Surveying human vulnerabilities across the life course: Balancing substantive and methodological challenges. In M. Oris, C. Roberts, D. Joye, & M. E. Stähli (Eds.), Surveying human vulnerabilities across the life course (pp. 1–25). Springer International Publishing. https://doi.org/10.1007/978-3-319-24157-9_1

Narduzzi, S., Golini, M. N., Porta, D., Stafoggia, M., & Forastiere, F. (2014). L'uso dell'Inverse probability weighting (IPW) nella valutazione e "correzione" del selection bias [Inverse probability weighting (IPW) for evaluating and "correcting" selection bias]. Epidemiologia e prevenzione, 38(5), 335–341.

3. Please use more descriptive headers for the tables. For example, in tables 1 and 2 should say something about the sample size, and whether this was an exploratory or confirmatory PCA. Table 3 should also have the sample size and indicate if this is the larger sample or the analytic sample. further this table should give the data to evaluate the loss to follow up, i.e. compare those who were followed at 12 months to those who were not.

Reply. Sample size has now been reported in all Table headers. Additionally, sociodemographic characteristics of the sample of dyads who dropped out from the study as well as differences with the study sample has now been summarized in Supplementary Table 1.

4. Please indicate whether successive steps include the variables in the previous step. That would make sense, but it is not clear the way the table is formatted.

Reply. Yes, the reviewer is correct. This has been added in the text.

5. Because the scores on the socio-cognitive development measures were not normally distributed there is a chance that extreme outliers may influence the results. Did the authors also consider the log transform of these variables, which would normalize the distribution.

Reply. This is a relevant issue. We indeed discussed about the possibility or opportunity to normalize the distribution. Nonetheless, it should be considered that these measures refer to specific precursors of socio-cognitive development that are emerging at the age (12-month) at which they were tested in this study. As such, it is expected that their distribution would be asymmetric as they are, reflecting the fact that some infants have already “mastered” these skills, while others are moving toward the same achievement. The distribution of these measures is consistent with this expectation. Indeed, we report here the asymmetry and kurtosis indexes for the socio-cognitive development measures:

Measure Asymmetry Kurtosis

ORI 0.41 -0.27

POI 2.26 4.94

COM 2.21 7.18

SCS 1.21 1.44

If we log-transform these measures, we obtain:

Measure Asymmetry Kurtosis

ORI 0.18 -0.37

POI 2.04 3.84

COM 1.72 4.27

SCS 0.95 0.78

These indexes are still problematic for some of the measures. Anyway, we ran the analyses using the log-transformed measures and we obtained comparable findings (we report here the final regression step):

• ORI: statistically non-significant

• COM: adjusted R-squared = .10, p = .042; PRS coefficient = -2.06, p = .042

• POI: statistically non-significant

• SCS: adjusted R-squared = .10, p = .016; PRS coefficient = -2.45, p = .016

In the light of these findings, we opted not to change our manuscript introducing the log-transformation.

Attachment

Submitted filename: pone-d-22-23782-r2_replies.docx

Click here for additional data file. (26.1KB, docx)

Decision Letter 2

Claudine Irles

4 Apr 2023

Maternal pandemic-related stress during pregnancy associates with infants’ socio-cognitive development at 12 months: A longitudinal multi-centric study

PONE-D-22-23782R2

Dear Dr. Provenzi,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Claudine Irles, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Acceptance letter

Claudine Irles

6 Apr 2023

PONE-D-22-23782R2

Maternal pandemic-related stress during pregnancy associates with infants’ socio-cognitive development at 12 months: A longitudinal multi-centric study

Dear Dr. Provenzi:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Claudine Irles

Academic Editor

PLOS ONE

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    S1 File

    (DOCX)

    Click here for additional data file. (18.2KB, docx)
    Attachment

    Submitted filename: PlosOne review (Final).docx

    Click here for additional data file. (14.2KB, docx)
    Attachment

    Submitted filename: R&R covid provenzi.docx

    Click here for additional data file. (17.3KB, docx)
    Attachment

    Submitted filename: momcope_sociocogn_replies.docx

    Click here for additional data file. (32.1KB, docx)
    Attachment

    Submitted filename: pone-d-22-23782-r2_replies.docx

    Click here for additional data file. (26.1KB, docx)

    Data Availability Statement

    The raw data related to the present publication are available from the Zenodo repository (url: 10.5281/zenodo.7516250).

    Articles from PLOS ONE are provided here courtesy of PLOS

    RESOURCES