Introduction
Extensive research supporting the Developmental Origins of Health and Disease (DOHaD) hypothesis has shown that fetal adaptation to negative psychosocial, lifestyle, and environmental exposures during pregnancy can have both short- and long-term health impacts on the offspring. Reference Lacagnina1 Several health metrics measured during pregnancy and at the time of delivery, including inappropriate gestational weight gain (GWG) and infant birth weight extremes (small (SGA) and large for gestational age (LGA)), can be indicative of adverse in utero exposures and fetal programming events as well as higher risk for future chronic diseases in adulthood. Reference Hack, Klein and Taylor2–Reference Scifres4 Food insecurity (FI), defined as the inadequate or insecure access to food due to financial constraints, Reference Li, Fafard St-Germain and Tarasuk5 and heightened stress during pregnancy, were both exacerbated for many pregnant individuals during the COVID-19 pandemic. Reference Azevedo, Morais and Silva6–Reference Pagès, Gorgui and Wang8 Both factors are recognized as contributing factors to adverse developmental programming events. Reference Coussons-Read9,Reference Bell, Nguyen and Andreae10
FI, which expands beyond the direct access to nutrition food, is related to various indicators of social and economic disadvantages such as low socioeconomic status, single-parent households, renting rather than owning a home, and identifying with a minority group (such as Black or Indigenous). Reference Li, Fafard St-Germain and Tarasuk5,Reference Fafard St-Germain and Tarasuk11 The latest United Nations State of Food Security and Nutrition report revealed that roughly 2.33 billion people were in a state of moderate to severe FI worldwide in 2024. 12 In Canada alone, approximately one in four individuals (25.5%), including 2.5 million children experienced household FI in 2024. 13 This marks a significant increase from pre-pandemic prevalence, when approximately one-in-eight households faced FI (2017–2018). Reference Tarasuk and Mitchell14 The experience of FI is particularly consequential when it occurs during pregnancy, given the crucial need for optimal nutrition during this critical window of development, to accommodate the changing metabolic requirements of both mother and offspring. Reference Buchanan, Meillère, Jessop, Costantini and Marasco15 FI during pregnancy is associated with an increased risk of inappropriate fetal growth profiles, birth defects, and stunted offspring growth Reference Carmichael, Yang, Herring, Abrams and Shaw16–Reference Karbin, Khorramrouz and Farkhani18 and with increased risk of diabetes and coronary heart disease later in life. Reference Roseboom, Painter, van Abeelen, Veenendaal and de Rooij19 For mothers, FI during pregnancy has been associated with hypertensive disorders of pregnancy, inappropriate GWG, gestational diabetes, as well as generalized anxiety disorders and major depressive episodes. Reference Hoseini, Kazemi, Alimoradi, Oskoei, Alijani and Zolghadr20–Reference Laraia, Siega-Riz and Gundersen22
Maternal stress is a well-documented gestational exposure independently associated with adverse maternal and infant health outcomes. Reference Coussons-Read9 Heightened maternal stress during pregnancy has been shown to be associated with preterm birth, SGA newborns, delayed motor and mental child development, as well as postpartum depression, and poor quality of sleep. Reference Alves, Cecatti and Souza23 Studying the long-term consequences of a 1998 ice storm in Canada, which left individuals without electricity for up to 40 days during winter, revealed significant associations between exposure to gestational stress and various outcomes in offspring, including increased cortisol levels and heightened Hypothalamic-pituitary-adrenal axis reactivity, Reference Yong Ping, Laplante, Elgbeili, Jones, Brunet and King24 elevated childhood body mass index (BMI) and adiposity, Reference Liu, Dancause, Elgbeili, Laplante and King25 externalizing behaviors and aggression, Reference Nguyen, Jones and Elgbeili26,Reference Jones, Dufoix and Laplante27 and characteristics of the autism phenotype in young adulthood. Reference Li, Laplante, Elgbeili and King28
The COVID-19 pandemic was a major traumatic event that disrupted numerous aspects of daily life while disproportionately affecting vulnerable populations, including pregnant and food-insecure individuals. Reference Azevedo, Morais and Silva6 Among Canadian families with young children, the COVID-19 pandemic has led to reduced food resources, increased FI, economic instability, and stress levels. Reference Carroll, Sadowski and Laila29 Pregnant individuals were disproportionately affected by both the direct impact of COVID-19 infection, having more risk of preeclampsia, gestational diabetes, preterm birth, low birth weight and stillbirth, Reference Wei, Bilodeau-Bertrand, Liu and Auger30 and indirect social consequences, including reduced healthcare access, limited social support and increased maternal anxiety and depression linked to economic insecurity. Reference Joo and Liu31,Reference Thayer and Gildner32 In fact, Shaw-Churchill et al. (2025) reported that perinatal healthcare experiences and measures of social support and coping, were more predictive of pandemic-related pregnancy stress than demographic characteristics in a Canadian cohort. Reference Shaw-Churchill and Phillips33
The COVID-19 pandemic exacerbated FI and maternal stress among vulnerable pregnant populations, likely compounding the risk of adverse pregnancy outcomes through multiple pathways of disadvantage. Despite growing awareness of these issues, there remains a critical gap in understanding how pandemic-related stressors and social determinants of health interact to influence maternal and infant health outcomes, key indicators of fetal development and long-term disease risk. The purpose of this study was to identify relevant social determinants of health associated with pandemic-related FI and heightened maternal stress and to explore whether these factors were associated with maternal GWG and newborn birth weight in a Canadian pregnant cohort.
Methods
Study population and design
This retrospective cohort survey involved 273 Canadian participants of varying socioeconomic status who gave birth during the COVID-19 pandemic (March 2020 to March 2023). Participants were recruited through purposive sampling from November 2022 to April 2023 using study advertisements on social media platforms (Facebook, Twitter, and Instagram), and targeted recruitment strategies to ensure the inclusion of food-insecure participants. Institutions serving vulnerable populations across Canada were contacted by email, mainly partners from the Canada Prenatal Nutrition Program (CPNP), 34 as well as members of our research network, community health centers and food banks. Participants were eligible if they had given birth to a live singleton infant and resided in Canada during their pregnancy. Individuals below 18 years of age and those who did not understand either French or English were excluded. Eligible participants who completed the questionnaire received a $25 gift card. All completed surveys were systematically screened to ensure data validity and to exclude fraudulent (bot) respondents (see Supplementary Material Fig 1 for details on screening and filtering strategy). Ethics approval was obtained from the University of Ottawa’s Health Sciences and Science Research Ethics Board (H-07-22-8071).
Development of the online survey
The online survey was developed using existing or adapted questions from previously validated questionnaires, as detailed below. The survey aimed to collect information on social determinants of health, household FI, pandemic-related maternal stress, and self-reported maternal and newborn health outcomes. Participants answered between 100 and 159 questions, depending on their responses to filtered questions. Data was collected and managed using REDCap electronic data capture tool hosted on the secured server of Montfort Hospital. 35
-
1. Social Determinants of Health: Maternal social, economic, and demographic characteristics were collected using questions from the validated Canadian Community Health Survey (CCHS), 36 Canadian Maternity Experience Survey (MES), 37 Electronic Maternal Health Survey (EMat Health Survey), Reference Ockenden, Gunnell and Giles38 and the Survey on Maternal Health. 39 The range of questions covered themes such as sexual orientation, gender identity, ethnicity, household income, marital status, parental structure, place of birth and residence, education level, number of children in the household, and whether pregnancy was planned.
-
2. Household FI: FI during pregnancy was measured using the Canadian Household Food Security Survey Module (HFSSM), an 18-item validated questionnaire designed to assess household food security in the last 12 months. Reference Canada40 The HFSSM is divided into a 10-item adult food security scale and an 8-item child food security scale, if children under 18 years old are living in the household. The HFSSM focuses on self-reports of inadequate or uncertain access to food and compromised eating habits owing to financial constraints. Health Canada’s method was used to classify participants into a gradient ranging from food-secure, marginally insecure, moderately insecure, to severely insecure, as described by Tarasuk et al. 2015. Reference Tarasuk, Cheng, De Oliveira, Dachner, Gundersen and Kurdyak41 For this study, the HFSSM was adapted by asking study participants to consider their pregnancy period rather than the standard past 12 months while answering questions about their household food security experiences. In addition to the HFSSM questions, participants were asked three specific questions to capture pandemic-related changes in food quantity, quality, and accessibility: 1) “Did the amount of food consumed (e.g., cutting portions or number of meals) change because of the pandemic?” 2) “Did the quality of the food eaten (e.g., types of food you would normally like/expect to eat) change because of the pandemic?,” and 3) “Did your access to food (e.g., being able to go to the grocery store or have your order delivered) change because of the pandemic?”
-
3. Pandemic-Related Maternal Stress: Maternal stress during the COVID-19 pandemic was assessed using the Pandemic-Related Pregnancy Stress Scale (PREPS), a validated questionnaire developed to assess prenatal maternal stress and coping. Reference Preis, Mahaffey and Lobel42 The PREPS asked participants to report their general worries related to pandemic preparation for birth and postpartum (PREPS-Preparedness), fear of maternal or infant COVID-19 infection (PREPS-Infection), and aspects of the pandemic that were perceived as positive (PREPS-Positive appraisal). The score for each PREPS subscale was calculated as the mean response of the items using a 5-point Likert scale, with higher scores indicating higher levels of prenatal stress or positive appraisal.
-
4. Maternal and infant Health Outcomes: Maternal and infant health outcomes included self-reported pre-pregnancy height and weight, pre-delivery weight, infant birth weight, sex, and gestational age at delivery. Maternal GWG was calculated and categorized into three groups (inadequate, adequate, or excessive) according to the 2009 IOM guidelines 43 and was analyzed as inadequate GWG, excessive GWG or as a binary outcome (within recommended and outside recommended GWG (inadequate + excessive)). Infant birth weights were normalized by fetal sex and gestational age at delivery and analyzed as birth weight percentiles derived for the Canadian population. 44 Birth weight centiles were categorized as appropriate for gestational age (AGA between 10th and 90th percentile), SGA (< 10th percentile), LGA (> 90th percentile), or as a binary outcome of AGA vs. inappropriate for gestational age (IGA = SGA + LGA).
Data analysis
Statistical analyses were performed using SAS software (version 9.4; SAS Institute Inc., Cary, NC, USA). Descriptive statistics (e.g., means, frequencies, and percentages) were used to characterize the study population. The chi-square test for categorical variables and Fisher’s exact test were used to compare the categorical social, economic, and demographic characteristics between food-secure and food-insecure participants (Table 1), as well as for the pandemic-related changes in food quantity, quality and accessibility (Table 2). The independent samples t-test (two groups) and ANOVA (more than two groups) were used to compare the mean differences in stress scores between groups of different social, economic and demographic characteristics (Table 3). Tukey’s post hoc test was used to assess significant differences between the pair of group means when using ANOVA. Univariate logistic regression was used to examine the associations between two exposures – FI and heightened stress (both treated as continuous exposures) – and binary outcomes of GWG and birth weight (Table 4). Potential interaction effects between FI and stress were also examined. In general, no significant associations were found between potential confounding variables (i.e., participants’ characteristics), exposures, and outcomes of interest; therefore, the regression models were not further adjusted for these variables. However, additional models were run adjusting for pre-pregnancy BMI, maternal health conditions, and maternal age to ensure the robustness of the findings and to account for factors commonly considered in perinatal research. Effect sizes were estimated using Cohen’s d for continuous variables and Cramer’s V for categorical variables. Reference Kallogjeri and Piccirillo45 Statistical significance was defined as a two-sided p-value < 0.05.
Table 1. Sociodemographic characteristics by household food security status
Values are presented as frequencies (n) and percentages (%). P-values are from chi-square tests and Fisher’s exact test, comparing food-insecure and food-secure participants. Statistically significant values are bolded.
a sample size n = 258 of 273 due to missing data
b sample size n = 253 of 273 due to missing data
c sample size n = 267 of 273 due to missing data
d sample size n = 272 of 273 due to missing data
Table 2. Pandemic-related changes in food quantity, quality, and accessibility according to household food security status
P-values are from chi-square tests, comparing food-insecure and food-secure participants. Statistically significant values are bolded.
a sample size n = 269 of 273 due to missing data
b sample size n = 272 of 273 due to missing data
Table 3. Sociodemographic characteristics associated with pandemic-related preparedness stress, COVID-19-infection stress and positive appraisal PREPS scores
Values are means ± SEM. P-values were obtained using two-sample t-tests and ANOVA. Tukey’s post hoc test was used to assess significant differences between the pair of group means when using ANOVA. Statistically significant values are bolded.
†Significantly different from $125,000–$149,999; ‡significantly different from $100,000–$124,999; * significantly different from $50,000– $74,999; **significantly different from 1 child; ⁑significantly different from no children; ⸆significantly different from 24–29 years old.
a sample size n = 264 due to missing data
b sample size n = 252 due to missing data
c sample size n = 244 due to missing data
d sample size n = 258 due to missing data
e sample size n = 263 due to missing data
Table 4. Likelihood of gestational weight gain and birth weight extremes in relation to maternal FI and PREPS stress scores
Values are odds ratios and 95% confidence intervals (OR (95%CI)) with PREPS scores analyzed as a continuous variable. GWG = gestational weight gain, IGA = inappropriate gestational age, LGA = large for gestational age, Ref=reference group SGA = small for gestational age. P-values were obtained using univariate logistic regressions. Statistically significant values are bolded.
Sample size n = 208 due to missing data for the GWG categories.
Sample size n = 145 due to missing data for the birth weight categories.
Results
FI and social determinants of health
The sociodemographic characteristics of the entire study cohort, stratified according to FI status, are shown in Table 1. The majority of pregnant participants were in a two-parent household (86.1%), in their mid-twenties and early thirties (78.0%), house owners (63.9%), had an annual household income of less than $100,000 (63.6%), had a high school diploma or higher (93.4%) and came from central provinces of Canada (Ontario and Quebec, 65.2%). There was an over-representation of individuals from the LGBTQ2S+ (16.1%), Indigenous (18.5%), and visible minority (20.2%) groups in the sample.
More than one half of the households surveyed (55.7%) reported FI during the COVID-19 pandemic. When FI was stratified by severity, approximately 15% of households were marginally FI, 30% were moderately FI, and 11% were severely FI. Participants from food-secure and food-insecure households differed significantly in parental structure, age groups, sexual orientation, housing status, total household income, employment status, and the number of children in the households (Table 1; p ≤ 0.009). Only 31 participants (11.4%) were born outside of Canada, of whom six had immigrated within the past five years; among these, one was food-secure and five were food-insecure. Cramer’s V values indicated small to medium effect sizes for most significant differences (0.16–0.27), except for employment (0.31), and household income (0.46), which showed larger effects.
In response to the COVID-19 pandemic, nearly half of all participants, regardless of food security status, reported a reduction in daily portion size or meal frequency (42%) and a change in food quality (49%), while slightly more than half (61%) reported decreased food accessibility (Table 2). However, FI participants were significantly more affected: they were seven times more likely to reduce portion sizes or meal frequency (OR 7.29, 95% CI: 4.14–12.8), more than four times as likely to report changes in food quality (OR 4.5, 95% CI: 2.70–7.62), and over three times as likely to experience reduced food accessibility during the pandemic (OR 3.56, 95% CI: 2.13–5.94). Medium to large effect sizes (Cramer’s V = 0.30–0.44) were observed between food security status and reported pandemic-related changes in food quantity, quality, and accessibility.
Maternal stress and social determinants of health
Maternal stress and coping during the pandemic were assessed using the PREPS subscales (preparedness, infection, and positive appraisal) (Table 3). Approximately one-third (33.7%), and one-fifth (19.7%) of the study participants reported a score of at least 4 (out of 5) on the PREPS-Infection and PREPS-Preparedness subscales, respectively, indicating heightened levels of pandemic-related pregnancy stress. Conversely, only 13% of study participants scored at least 4 (out of 5) on the PREPS- Positive Appraisal subscale, indicating that most participants did not perceive the COVID-19 pandemic as a positive event during their pregnancy.
Increased maternal stress scores – both for preparedness and COVID-19 infection – were significantly associated with single-parenth household (vs. two-parent), LGBTQ2S+ identity (vs. heterosexuals), and unplanned pregnancies (vs. planned; Table 3; p < 0.05). Both maternal stress scores (preparedness and COVID-19 infection) were also significantly associated with household income and number of children (Table 3; p ≤ 0.04). Tukey’s post hoc tests showed that participants in the highest income bracket (≥$ 150,000) had significantly lower preparedness stress scores than those earning $100,000–$124,999 (p = 0.02) and those earning$125,000–$149,999 (p = 0.05) . In turn, those earning $100,000–124,999$ and $125,000–$149,999 reported significantly higher stress score than those in the $50, 000–$74,999 bracket (p = 0.008 and p = 0.02, respectively). No significant differences in infection stress scores were found among income subgroup. Additionally, participants with two or more children reported higher preparedness (p = 0.01) and infection (p = 0.03) stress scores compared to those with one child, but scores did not differ significantly from those with no children (Table 3).
Perceived positive appraisal of the COVID-19 pandemic also varied significantly by age groups, total household income, and number of children (p < 0.05; Table 3). Tukey’s post hoc comparisons showed that participants aged ≥35 had significantly lower positive appraisal scores than those aged 24–29 years (p = 0.05). Similarly, participants in the highest income bracket ( ≥$150,000) had lower positive appraisal scores compared to those earning $125,000–$149,999 (p = 0.02) and $50,000–$74,999 (p = 0.03). Finally, participants with one child reported higher positive appraisal than those with no children (p = 0.005), with no significant difference observed between those with two or more children (Table 3). Significant differences were generally of small to moderate magnitude (Cohen’s d = 0.18–0.45), with larger effects observed for sexual orientation (0.52 for infection and 0.71 for preparedness).
FI, maternal stress, and their impact on maternal–infant health outcomes
The majority (77%) of respondents reported GWG values outside the recommended guidelines, with 31% reporting inadequate GWG and 46% reporting excessive GWG. Overall, one third of the study participants (32%) reported having an IGA infant, with 17% reporting SGA birth weights (< 10th percentile), and 15% reporting LGA birth weights ( >90th percentile).
The likelihood of having outside recommended GWG and birth weight extremes when experiencing FI and maternal stress are displayed in Table 4. FI was not associated with the odds of outside recommended GWG (p = 0.88) nor with increased odds of having an infant with inappropriate birth weights (p = 0.36) (Table 4). Heightened maternal stress scores related to fear of COVID-19 infection and pandemic-related perinatal preparedness were not associated with GWG outside the recommended ranges (p = 0.21) and (p = 0.41), respectively (Table 4). However, when stratified by categories of GWG, higher maternal COVID-19 infection stress score was significantly associated with a 30% decrease in the likelihood of inadequate GWG (p = 0.02). No associations were observed between pre/postnatal preparedness stress or positive appraisal scores and outside recommended GWG categories. Adjusting for pre-pregnancy BMI, maternal health conditions (e.g., gestational diabetes, preeclampsia, and COVID-19 infection), and maternal age slightly widened the confidence interval for inadequate GWG (OR = 0.71, 95% CI = 0.45–1.06, p = 0.09).
Heightened maternal stress scores related to COVID-19 infection and pandemic perinatal preparedness were significantly associated with an increased likelihood of 1.50 (p = 0.02) and 1.60 (p = 0.01) of having IGA infants, respectively (Table 4). When stratified by categories of SGA or LGA (compared to non-SGA or non-LGA, respectively), no significant differences were observed between groups. Adjustment for pre-pregnancy BMI, maternal health conditions (e.g., gestational diabetes, preeclampsia, and COVID-19 infection), and maternal age did not materially change the estimates for IGA. There was no association between increased positive appraisal of the COVID-19 pandemic and IGA. We then tested the combined effect of FI and PREPS stress scores on GWG outside the recommended ranges and IGA infants but no significant interaction were observed (data not shown).
Discussion
The primary aim of this study was to identify the social determinants of health associated with pandemic-related FI and heightened maternal stress, and to explore the relationships between these adverse gestational exposures and the maternal–infant health outcomes – maternal GWG and newborn birth weight – in a Canadian pregnant cohort. We found that psychosocial and economic factors, including parental structure, sexual orientation, household income, number of children in the household and pregnancy planning, were common determinants of both FI and heightened stress. While higher scores of pandemic-related pregnancy stress (preparedness and infection) were associated with the likelihood of inappropriate birth weight for gestational age, but not with GWG outside the recommended range. No associations were observed between FI and maternal/infant health outcomes.
FI during pregnancy is a public health concern that poses significant risks for both maternal and fetal health outcomes. Reference Choi, Crimmins and Ailshire46 More than half of our Canadian pregnant sample (55.7%) experienced FI during the pandemic, more than twice the national Canadian average in 2024 (25.5%) 13 and higher than the the US estimates during the COVID-19 pandemic (11 to 21% among pregnant and postpartum women) Reference Shriver, Eagleton, Hosseinzadeh, Buehler, Wideman and Leerkes47,Reference Ujah, Olaore, Ogbu and Kirby48 ; however, FI thresholds differ between the two countries, which may partly explain these differences. These discrepancies may be the result of oversampling of high-risk populations from the targeted study advertisement at food banks, community centers, and programs that focus on vulnerable and food-insecure populations (notably partners from the CPNP). Our sample had an over-representation of Indigenous and visible minority groups who face up to two times greater risk of FI in Canada. 13 The financial incentive for study completion may also have attracted more economically vulnerable individuals with a higher prevalence of FI. Interestingly, FI was reported even among the higher-income households (earning $100,000 to $149,000 CAD), consistent with previous Canadian findings that certain factors (e.g., renting, a single-parent status, lower educational attainment, larger household size, and the presence of chronic conditions) can increase FI risk despite higher income. Reference Olabiyi and McIntyre49 Similarly, we observed that some high-income households experiencing FI had a high number of children, lived in an urban area, and/or were single-parent, which may help explain their vulnerability to FI despite their high income (data not shown). Our findings further revealed that even food-secure households experienced disruptions in the access, quality, and quantity of food during COVID-19, indicating that the COVID-19 pandemic’s impact extended beyond traditionally food-insecure populations.
Although previous systematic reviews and meta-analyses of studies conducted before and during the pandemic have linked FI to both inadequate and excessive GWG in mothers, Reference Arzhang, Ramezan and Borazjani50 as well as low birth weight infants, Reference de Freitas Rocha, de Souza de Morais, Azevedo, de Castro Morais and Priore51 our study did not find significant associations. This lack of association may be due to the absence of a temporal dimension to measure FI, which could have captured fluctuations of FI during the COVID-19 pandemic, resulting from distinct measures and lockdowns implemented by each Canadian province and territory. Some participants may have experienced short-term FI (e.g., temporary supermarket shortages), while others faced prolonged FI, leading to heterogeneous exposures and outcomes. Furthermore, some participants reporting FI may have remained nutrient- and/or calorie-sufficient, possibly mitigating adverse effect. Additionally, some of our FI participants, being fairly educated, coming from mostly middle- and high-income households, with a culturally and geographically diverse background, might have had variations in dietary practices, access to prenatal care, and social support during the pandemic, potentially mitigating negative maternal–infant health outcomes.
Experiencing COVID-19 while pregnant has been associated with increased stress, anxiety, and depression. Reference Moyer, Compton, Kaselitz and Muzik7,Reference Pagès, Gorgui and Wang8 In our Canadian cohort, 33.7% reported high infection-related stress and 19.7% reported high preparedness stress (both ≥4), higher than some countries (e.g., Germany and Italy) but lower than others (e.g., United States and Poland). Reference Preis, Mahaffey and Lobel42,Reference Ilska, Kołodziej-Zaleska, Brandt-Salmeri, Preis and Lobel52–Reference Schaal, La Marca-Ghaemmaghami, Preis, Mahaffey, Lobel and Castro54 This could likely be attributed to several factors, including demographic characteristics, as well as varying access to financial or social support and healthcare services. Similar to our study, Shaw-Churchill et al. (2025) also examined predictors of pandemic-related pregnancy stress in a Canadian pregnancy cohort. Although the only common significant predictor between both studies was income, their study reported that participants with middle income ($50,000–$79,999) experienced higher preparedness stress compared to those with a greater income ( >$150,000), Reference Shaw-Churchill and Phillips33 whereas we observed that those with a middle-higher income ($100,000–$149,999) reported higher preparedness than those with both middle ($50,000–$74,999) and higher ( ≥$150,000) income. The discrepancy between findings may be partly explained by differences in income distribution across cohorts; 63% of participants in Shaw-Churchill et al.’s study reported incomes above $100,000, compared with 32% in our sample, suggesting differing socioeconomic and living contexts between samples during the pandemic. Although Shaw-Churchill et al. did not investigate the role of FI and the relationship between pandemic-related stress and maternal or infant health outcomes, they showed that prenatal healthcare experiences, social support, and coping measures were stronger predictors of pandemic-related pregnancy stress, Reference Shaw-Churchill and Phillips33 reinforcing the complex and multidimensional nature of stress during pregnancy in the context of the COVID-19 pandemic.
The increased maternal stress could equally be attributed to elevated vaccine safety concerns or infant formula shortages that were being reported during the COVID-19 pandemic in North America. Reference Reifferscheid, Marfo, Assi, Dubé, MacDonald and Meyer55,Reference Sylvetsky, Hughes and Kuttamperoor56 In our cohort, increased pandemic-related pregnancy stress was associated with LGBTQ2+ identity, being a house renter, some middle- and higher-income groups and having two or more children. Notably, over one tenth of study participants (13%) reported high levels of positive appraisal of the pandemic, which, as suggested by Preis et al., may reflect adaptative coping strategies, personal growth, and resilience among pregnant individuals during the pandemic. Reference Preis, Mahaffey and Lobel42
Despite the high prevalence of pandemic-related maternal stress among our study participants, no significant association was observed globally with GWG outside the recommended range. However, when GWG was stratified into inadequate and excessive categories, increased infection-related stress was associated with a reduced likelihood of inadequate GWG. Previous studies have reported associations between elevated stress levels and GWG outside the recommended range. For instance, a North American study examining the impact of the initial COVID-19 lockdown found that this period was associated with lower-than-recommended weight gain during pregnancy. Reference Collins-Smith, Prasannan, Shan, Dori, Katzow and Blitz57 Similarly, four other observational studies conducted before the pandemic in the United States also found an association between heightened stress and inadequate GWG. Reference Kominiarek, Grobman and Adam58–Reference Orr, James and Miller61 In contrast, two studies conducted in the United States during the pandemic found associations between stress and excessive GWG. Reference Abdelwahab, de Voest and Metz62,Reference Cao, Sun and Danilack63 Variations in the effect of stress on inadequate versus excessive GWG across studies may partly be attributed to differences in the methods used to quantify stress as an adverse gestational exposure, as well as the type, timing, intensity, and duration of stress experienced during pregnancy. For example, Braig et al. as well as Zhu et al. both found that the association between stress and GWG was only observed when it was experienced specifically within the first trimester of pregnancy. Reference Zhu, Huang, Hao, Huang, Jiang and Tao64,Reference Braig, Logan, Reister, Rothenbacher and Genuneit65 Considering this observed time-specific effect, it is important to consider the temporal dimension of heightened stress and pandemic-related societal restrictions, which varied considerably by region of residence and may have differentially influenced GWG outcomes.
Our findings suggest an association between heightened maternal stress and infant birth weight extremes, which predict long-term health outcomes. Reference Hack, Klein and Taylor2,Reference Scifres4 This aligns with evidence from the 1998 Quebec ice storm in Canada that has documented associations between elevated maternal stress and several adverse infant health outcomes, including shorter gestational lengths and birth weight extremes. Reference Dancause, Laplante, Oremus, Fraser, Brunet and King66,Reference Mélançon, Bernard and Forest67 A large body of evidence has shown that disaster-related pregnancy experiences, may negatively impact infant health outcomes including fetal growth and birth weight profiles. Reference Lafortune, Laplante and Elgbeili68 A recent meta-analysis that examined the effect of the COVID-19 pandemic on neonatal birth weight between the pandemic and pre-pandemic periods, irrespective of maternal stress levels. Reference Yao, Li, Jiang, Ma and Wen69 The authors demonstrated a pandemic-associated increase in mean birth weight and decrease in very low birth weight infants, with no observed effect on low birth weight, SGA, LGA, and macrosomia. Reference Yao, Li, Jiang, Ma and Wen69 This underscores the need for future research that directly measures maternal stress and its associations with infant birth weight and other perinatal health outcomes across different types and timings of gestational stress exposure.
Strengths and limitations
Strengths of our study include a pan-Canadian cohort with broad cultural and social diversity spanning most of the pandemic (March 2020 to March 2023). The online format of the survey also allowed for anonymous participation, an important consideration when asking questions related to sensitive topics such as FI, GWG, and maternal mental health. Limitations should also be considered. Although FI was assessed using the validated Canadian HFSSM, Reference Canada40 this instrument was not designed to measure FI during pregnancy. The question addressing unintentional weight loss is, in most cases, not applicable in this context and could contribute to an underestimation of FI prevalence among pregnant individuals. Conversely, the purposive non-random sampling method, combined with the use of gift cards may have introduced selection bias and led to oversampling of more vulnerable groups of the pregnant population. This study also relied on a single, retrospective assessment of data which is subjected to recall biases, as participants might have incorrectly recalled some information between the time of actual experience and the time of questionnaire completion. Our measures lacked temporal detail (e.g., whether FI and stress varied across pregnancy stages), limiting interpretation. The exclusion of participants due to missing data, particularly those missing maternal obstetric outcomes data, led to a reduced sample size, particularly for birth outcome measures, which may have limited the statistical power. This study primarily reports univariate associations between maternal characteristics and perinatal outcomes. Although additional models adjusting for pre-pregnancy BMI, maternal health conditions, and maternal age yielded similar estimates for inadequate infant gestational age, the associations with maternal inadequate GWG were no longer statistically significant. Because pregnancy health and weight status may act as mediating factors, adjusting for these variables could attenuate part of the maternal stress–weight relationship. Therefore, caution should be exercised when attempting to generalize findings from this study to the broader Canadian pregnant population.
Conclusion
Adverse gestational exposures, such as FI and elevated maternal stress, were heightened during the COVID-19 pandemic, albeit experienced to varying degrees across the Canadian population. From a DOHaD perspective, understanding the population characteristics associated with these adverse stressors is essential for pinpointing pregnant individuals and their offspring who are most vulnerable to both short- and long-term health risks. This knowledge is critical not only for advancing public health research but also for informing targeted interventions – such as those funded by the CPNP – to more effectively address FI and maternal stress. Strengthening such programs may help build resilience among at-risk populations and enhance preparedness for future public health crises and environmental disruptions.
Supplementary materials
The supplementary material for this article can be found at https://doi.org/10.1017/S2040174425100317.
Acknowledgments
We would like to thank the research participants for their contribution and time to the study. We also acknowledge the contribution of Sagni Sumar and Aishwarya Pramudi Suresh for their assistance in filtering participants’ survey to identify and exclude fraudulent respondents.
Financial support
This research was supported by the University of Ottawa’s Faculty of Health Sciences’ Strategic Grant Research Support (#04130). S.B was the recipient of an Early Career Award (#ER16-12-210 ;Ontario Ministry of Research and Innovation and Science) which provided trainee support for L.K.
Competing interests
None.
Ethical standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national guidelines on human experimentation (Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans, Canada) and with the Helsinki Declaration of 1975, as revised in 2008, and has been approved by the University of Ottawa’s Health Sciences and Science Research Ethics Board (#H-07-22-8071). The privacy rights of our participants have been observed and their informed consent was obtained.
Declaration of generative AI and AI-assisted technologies in the writing process
During the preparation of this work, the Copilot under the University of Ottawa license was used to improve the readability and language of the manuscript. While using this AI tool, the authors wrote the first draft, and then reviewed and edited the AI-improved manuscript as needed and take full responsibility for the content of the published article.
Open access
