The COVID in the Context of Pregnancy, Infancy and Parenting (CoCoPIP) Study: protocol for a longitudinal study of parental mental health, social interactions, physical growth, and cognitive development of infants during the pandemic

The secondary impact of COVID-19 on pregnant women

The COVID-19 pandemic has been the biggest public health emergency for over a century, necessitating extreme measures at a societal level to mitigate against death and prevent acute health services from being overwhelmed. However, resulting from this have been a number of secondary consequences ((i.e., increased caregiving demands for children and family members; isolation from family and community due to social distancing; job loss; financial hardship and increased interpersonal stressors or relationship violence [8]), having a disproportionate and significant impact on women of childbearing age [4]. In the UK, the impact on this group of women has also been exacerbated by National Health Service (NHS) guidance that was produced in response to the national lockdown restrictions [9–11], in which hospital-based midwifery services placed limitations on partners being present during ultrasound visits and birth. In addition, most community-based services were discontinued, other than antenatal contact and new baby visits, all of which were required to be provided virtually unless otherwise indicated, with all other contacts being assessed and stratified according to vulnerability or clinical need (e.g., maternal mental health).

These changes not only affected the capacity of practitioners to support women during the perinatal period at a time of significantly heightened stress/distress [6] but also resulted in significant regional variations in access to healthcare and advice for expectant mothers across the UK. The changes have been associated with (i) a fourfold increase in stillbirths attributed to lack of preventive antenatal care [12], (ii) birthing partners denied access to the hospital for the birth or asked to leave immediately following the birth, and (iii) limited access to babies admitted to neonatal intensive care. The NHS has also reported a reluctance on the part of parents to attend postnatal GP checks, due to parental attitudes related to COVID-19 infection (Institute of Health Visiting).

The impact of COVID-19 on parental mental health and parent-infant interaction

Several online surveys conducted during the first national lockdown indicated that there was a significant increase in antenatal anxiety both in terms of pandemic-related pregnancy stress associated with feeling unprepared for birth due to the pandemic, and stress related to fears of perinatal COVID-19 infection, with one large US survey (n = 4451) showing that around 30% of pregnant women experienced both types of stress [13]. Another US survey (n = 2740) that examined wider sources of stress showed that more than half of women reported increased stress about food running out (59.2%, n = 1622), losing a job or household income (63.7%, n = 1745), or loss of childcare (56.3%, n = 1543). More than a third reported increased stress about conflict between household members (37.5%, n = 1028); and 93% (n = 2556) reported increased stress about getting infected with COVID-19 [14].

A number of online cross-sectional surveys found significantly increased rates of anxiety and depression, based on the use of self-report standardised measures (e.g., Edinburgh Postnatal Depression Scale (EPDS); Hospital Anxiety and Depression Scale (HADS)). For example, a cross-sectional survey of 1,987 pregnant women in Canada in April 2020, found substantially elevated anxiety and depression symptoms, compared to similar pre-pandemic pregnancy cohorts; 37% reported clinically relevant symptoms of depression, and 57% reported clinically relevant symptoms of anxiety [15]. A second Canadian study found that a cohort of pregnant women, who were recruited during the COVID-19 pandemic, were twice as likely to present clinically significant levels of depressive and anxiety symptoms compared with a cohort of pregnant women recruited prior to the pandemic [16]. Early evidence in the UK similarly suggests that the impact on the mental health of pregnant women has been significant with heightened anxiety and depression being reported during the national lockdown (levels of mental distress rising from 18.9% (2018-19) to 27.3% in April 2020, one month into the national lockdown) [4].

This is of concern because there is consistent evidence to suggest that anxiety and depression in pregnancy can have a long-term impact on child development. For example, recent systematic reviews found that antenatal anxiety is associated with a range of adverse perinatal outcomes, including for example, premature delivery and low birthweight [17], in addition to a range of negative child outcomes that can persist into late adolescence, including an increased risk of child behaviour problems [18].

Early pre- and postnatal experiences in the context of COVID-19 and associated societal restrictions

The limited evidence on the impact of the pandemic and lockdowns on postnatal depression, suggest a similar picture, with around half of mothers caring for babies born during 2020 in the UK, reporting feeling down, lonely and worried, with mental health symptoms exacerbated in mothers who travel to work, had a baby born prematurely or were from a lower-income household [19]. One Australian study that examined all online perinatal support forum posts related to COVID-19, from women between January 27 to May 12, 2020, showed that the content was predominantly negative, with around 63% being very or moderately negative. Negative words that were frequently used in the 831 posts included: “worried” (n=165, 19.9%), “risk” (n = 143, 17.2%), “anxiety” (n = 98, 11.8%), “concerns” (n = 74, 8.8%), and “stress” (n = 69, 8.3%) [20].

Anxiety and depression in the postnatal period have been shown to affect the development of the infant because of the impact on the mother’s interactions with her baby. For example, depressed mothers have been shown to be less sensitively attuned to infants, and less affirming and more negating of their experiences with their infant [21]. Babies of depressed mothers can exhibit deficits in their interpersonal functioning, such as less affective sharing, lower rates of interactive behaviour, poorer concentration, increased negative responses with strangers, and reduced secure attachment at 12 and 18 months [22,23]. Children of these women are also 42% more likely to experience depression by age 16 [24].

There is now increasing evidence regarding the impact of the mother’s bond with her unborn baby, including her mental representations of the baby in pregnancy, in terms of an association with their interactions postnatally [25] including the infant’s attachment status [26]. However, there is currently limited evidence regarding the impact of pandemic-related social restrictions on this prenatal relationship. While a number of studies have found that the pandemic has had an impact on bonding with the baby in the postnatal period [27] and on breastfeeding [28], no studies to date have explicitly examined the impact on the parent-child interaction. Furthermore, concerns have been raised regarding infants limited exposure to infant peers or to diverse social partners other than household members or members of the public wearing masks, in terms of the impact on infant looking and communicative bids during social interactions [29]. Whilst ongoing research examines how the pandemic and the above-mentioned points are affecting children aged 8-36 months, there remains a critical lack of study of development during the period of development referred to as the “baby blind-spot” (from pregnancy to 2 years old age) [29,30].

There remain more questions empirical answers at the present time regarding how ‘stay at home’ orders, lack of access to social support from family members and pandemic-specific stressors might have affected expectant parents and those caring for babies during the national lockdowns. Concerns have been raised about women at risk for domestic violence, without suitable shelter and higher incidence of unemployment due to their role in the retail, caregiving and hospitality workforce [31,32]. It is also important to take into consideration other mediating influences on infant development including poverty [33], maternal education [34], marital discord [35], single parenthood and ethnicity [36].

Overall, little is currently known about the impact of COVID-19 guidance and restrictions on the long-term development of the child. To address this gap in our knowledge, this study has two main aims: i) to examine how parent’s experiences of the social, medical, and financial changes during the pandemic have impacted pre and postnatal parental mental health and parent-infant social interaction; and (ii) to investigate the extent to which these COVID related changes in parental pre and postnatal mental health and social interaction are associated with fetal and infant development.

Eligibility criteria and recruitment strategies

Eligibility criteria for the study is expectant parents (at any stage of pregnancy) or parents of an infant between the ages of 0-6 months. Either parent can take part, with questionnaires being adapted to the parents’ status as mother or father. The study is open to parents who had a baby shortly before and during the first period of lockdown in the UK (23^rd March 2020) as well as continuing to collect information from parents during the current, and future changes in COVID-related health and societal restrictions.

For optimal national representation across the UK, recruitment strategies include (i) targeting NHS antenatal classes and National Childbirth Trust (NCT) groups identified nationally, with an emphasis on areas of low socio-economic status (SES) using the government indices of multiple deprivation (IMD) and rural areas without access to NCT groups, (ii) partnering with NHS/National Institute for Health Research (NIHR) collaborative sites and charity and policy group partners (e.g., The Brazelton Centre UK, Centre for Health and the Public Interest) to widen knowledge of the survey, (iii) posting online via social media platforms (e.g., Twitter) and public sharing to facilitate snowball sampling and (iv) targeting populations experiencing increased local lockdown measures as, and when, COVID-19 rates and related policy change across the UK. Whilst recruitment efforts have been focused on the UK, the survey is currently open to all expectant and new families worldwide. Participation in the survey is incentivized using the offer of a chance to win a £100 digital gift card (on receipt participants are able to select from either an Amazon^® or one4all^® gift card). A prize is drawn for every 100 participants who complete the survey, giving a 1/100 chance of winning at each time point that they complete.

Power Calculation

To ensure sufficient power for the study, statistical power calculations (G^*Power adapted for regression) based on three outcomes, up to three predictors and four co-variables, estimated a minimum sample size for Hypothesis 1 of n = 400 (small effect, f² = 0.02). Statistical power calculations were based on a study of sociodemographic control variables, traumatic event impact scale and pregnancy-specific anxiety [37]. In the same manner, for Hypothesis 2, a minimum sample size of n = 800 (small effect, n = 400 infants x 2 postnatal timepoints) is required, based on ongoing analyses by our group on parent-infant social interaction data [38]. For Hypothesis 3 and 4, we will minimise data loss using post-hoc assignment of families to an accelerated longitudinal design - this requires a minimum sample size of n = 500 (small effect) using a study on acute disasters, parental mental health and infant development [39]. With timing and cross lag accounted for, and attrition rate of 30% - assuming current pattern of 80% of parents’ consenting to be contacted again (as indicated by our pilot survey) – a minimum cohort sample of n = 1500 is required.

Study Measures

The online survey is logic-dependent and adaptive, only showing questions relevant to the parent’s current situation (e.g., antenatal or with an infant of 2 months of age) in relation to the following six time points: the second and third trimester of pregnancy; infant aged 0-3 months, infant aged 3-6 months; and toddler aged 12- and 18-months. The following data is collected: (i) parental mental health and attitudes, (ii) healthcare access and support during pregnancy and birth, (iii), fetal physical development and infant social and cognitive development, (iv) direct impact of COVID-19 on daily lives and lastly (v) developmental outcomes in infants born during the pandemic. Table 1 provides an overview of the measures used timepoints and Table 2 provides a detailed summary of the measures used within the survey.

Follow-ups and reminders

Participants are invited to take part in a follow-up survey at the end of the initial survey. Where they consent to this, they are contacted via email containing a link to the separate online survey. The follow-up survey has been condensed to include follow-up questionnaires only (see Table 1 and Figure 2 for participant follow-up flowchart). The appropriate time for follow-up is calculated based on the ages (infant or fetal gestation) provided by the participant at initial recruitment (see Figure 2 for project timeline).

• Download figure
• Open in new tab

Figure 2:

Flow diagram of Covid in the Context of Pregnancy, Infancy and Parenting (CoCoPIP) study follow up participation.

A participant can join the study at any of the above 3 underlined timepoints

^*Participants are only followed up at these two timepoints if they have participated in at least two previous timepoints.

^{* *} Projected participant follow-up completion dates.

Quality Control

Ongoing quality control is evaluated bi-weekly. All data is checked for accuracy and invalid data is removed (e.g., one parent responded to the question “what is the gender of your child” as “clownfish” and completed no other question within the survey - this participant was removed from the dataset). Study data is collected and managed using Research Electronic Data Capture (REDCap^®) tools hosted at the University of Cambridge [54]. REDCap^® is a secure, web-based software platform designed to support data capture for research studies, providing 1) an intuitive interface for validated data capture; 2) audit trails for tracking data manipulation and export procedures; 3) automated export procedures for seamless data downloads to common statistical packages; and 4) procedures for data integration and interoperability with external sources. Personal data (e.g., caregiver DOB, email address) is stored securely within a password encrypted electronic databased isolated from the research data. Access to the data is fully audited to ensure data security is governed by a management team and in compliance with ethical guidelines.

Analysis plan

Overall, we aim to identify which stress-associated moderators (i.e., loss of income, COVID-19 illness, local access to ante/postnatal support) impact significantly on parental mental health, and in turn, infant development.

To address Hypothesis 1 (see Figure 1), a combination of quantitative and qualitative analyses will be undertaken. Structural equation modelling (SEM) and hypothesis-driven regressions will explore how multiple aspects of pre-and post-natal family support (social, financial and health) are associated with stress (maternal and paternal) and wider mental health. An inductive approach to data analysis will be undertaken [55] for the open-ended qualitative data, and we will use NVivo (QSR International Pty Ltd) to code the data. Following this, several approaches including thematic [56], sentiment and context content analysis will be undertaken using a natural language processing (NLP) approach [57] (machine learning) to identify forms of social, medical and financial support in relation to the valence of parental attitudes. Regression analyses will then be conducted to understand the directional relationship between resulting latent factors (qualitative responses and quantitative data) and parental mental health.

To address Hypothesis 2, regressions will be used to explore how COVID-19 related restrictions interact with the frequency and forms of social interaction (i.e., face-to-face, digital, at distance) that the infant has, and whether this varies across the first six months of life.

To address Hypothesis 3, regressions will be used to explore the influence of maternal mental health longitudinally on the developing offspring across pre-to post-natal life: from fetal (12-weeks/20 weeks gestational age) to 18 months of age. Standardised z-scores will be created from the collected fetal growth measurements (i.e., head circumference, femur length and abdominal circumference), which will then be transformed into a composite score accounting for gestational age and/or estimated fetal weight at time of scan to be used for analysis. Z-scores will also be computed and used where appropriate within the analysis (e.g., infant-toddler sensory profile).

To address Hypothesis 4, outputs from Hypothesis 1 and 2 (impact of COVID-19 on parental mental health and infant social interactions) will be explored in relation to longitudinal social and cognitive development (e.g., language, motor sensory and the early emergence of developmental conditions) of the infant/toddler across the 0-18 months of life using SEM, full information maximum likelihood to account for missingness, and regression modelling.