Abstract
Background Conflicting evidence exists on the impact of the COVID-19 pandemic restrictions on preterm birth (PTB) and stillbirth rates. We aimed to evaluate changes in PTB and stillbirth rates before and during the pandemic period and assess the potential effect modification of socioeconomic status (SES).
Methods Using the linked administrative health databases from Manitoba, Canada, we conducted a quasi-experimental study among all pregnant women, comparing 3.5 years pre-pandemic (1 October 2016 to 29 February 2020) to the first year of the pandemic (1 March 2020 to 31 March 2021). We used interrupted time series analysis using autoregressive integrated moving average models to assess the quarterly rates of PTB (<37 weeks) and stillbirths. We calculated the predicted trends based on pre-pandemic period data. Finally, we evaluated the lower and higher SES (average annual household income) using subgroup analysis and interaction models.
Results We examined 70,931 pregnancies in Manitoba during the study period. Following the implementation of COVID-19 restrictions in March 2020, there were no statistically significant changes in the rates of both PTB (p=0.094) and stillbirths (p=0.958). However, over the pandemic, the PTB rate significantly decreased as a rebound effect by 0.63% per quarter(p=0.005); whereas the stillbirth rate did not change significantly (p=0.878) compared to pre-pandemic period. During the first quarter of 2021, the absolute differences in the observed and expected PTB and stillbirth percentages were 2.05% and 0.04%, respectively. We observed a statistically significant effect modification by SES for PTB rates (p=0.047).
Conclusion While the onset of COVID-19 pandemic restrictions was not associated with significant effects on PTB and stillbirth rates, we observed a statistically significant rebound effect on PTB rates. The impact of COVID-19 on preterm birth was dependent on SES, with higher influence on families with lower SES. Further studies are needed to detect future trend changes during pandemic waves after 2021 and assess potential underlying mechanisms.
Introduction
Preterm birth (PTB) is considered one of the leading causes of neonatal deaths and can cause long-term complications among the survived infants.1–3 According to the World Health Organization, about 2 million stillbirths occur each year, and 40% of these stillbirths occur during childbirth.4 In the Canadian province of Manitoba, the preterm birth rate (7.8%) is similar to that of the overall Canadian rate (7.9%).5 However, Manitoba has a slightly higher stillbirth compared to the overall rate in Canada (6.0 compared to 4.6 per 1,000 births).5
Initially COVID-19 restrictions, such as virtual visits and social distancing, were considered the only available and effective means of preventing and reducing the spread of the virus6, especially among vulnerable populations, including pregnant women. However, these mitigation strategies influenced access to in-person health services, which can negatively impact maternal and neonatal health.7,8 The reduced access to perinatal services created a substantial obstacle for receiving optimal mother-infant care, thereby potentially increasing the risk of adverse pregnancy outcomes.9,10 The rate of adverse perinatal outcomes varied with lockdown strategies across countries; highlighting their potential impact on health outcomes of pregnant women and their infants.8,11,12 In fact, the implemented pandemic measures are believed to have exacerbated the inequalities already present because of various social determinants of health across many countries.13,14 Pre-pandemic research suggests that pregnant women with lower socioeconomic status (SES) are at higher risk for inadequate prenatal care, leading to an increased risk of adverse pregnancy outcomes, including PTB and stillbirth.15–17
Provincial and local governments across Canada have cooperated to impose measures to reduce the COVID-19 burden.18 Studies in two Canadian provinces, Ontario and Alberta, examined the influence of the implementation of COVID-19 first wave lockdown measures on perinatal health outcomes and found non-significant changes in both PTB and stillbirth rates compared to pre-pandemic period.19–22 In Manitoba, different pandemic strategies were implemented affecting antenatal care services which could have influenced the health of pregnant women and their infants.23–25 During the first wave, the daily number of COVID-19 cases in Manitoba was relatively low (ranging from 0 to 33 cases in the general population and significantly lower among pregnant women)26. Unprecedented changes in the prenatal practice included limiting in-person prenatal, shifting to some virtual care, and applying preventive measures for hospital deliveries and newborn care. The policy responses adopted in Manitoba and the low rates of COVID-19 infection provide a unique opportunity to serve as a natural experiment to determine their effects on perinatal outcomes.27 We, therefore, aimed to use a quasi-experimental approach to assess the short- and long-term impacts of the pandemic restrictions implemented during the first and second waves in Manitoba on the incidence of PTB and stillbirth. In addition, we aimed to evaluate if the effect of the COVID-19 restrictions was modified by the maternal SES.
Methods
Data source
We used data from the Manitoba Population Research Data Repository at the Manitoba Centre of Health Policy (MCHP) to conduct the study. The Repository is a secure data-rich environment containing person-level health information on virtually the entire population of Manitoba28. All records are de-identified and linkable at individual and family levels using a scrambled health number. The validity and reliability of the MCHP Repository for epidemiological studies has been previously reported.29–31We used the following databases: 1) the Manitoba Health Insurance Registry (date of birth, sex); 2) hospital abstracts and physician visits (to identify the perinatal outcomes of interest); 3) Drug Program Information Network (prescription drug data to identify comorbidities); 4) Manitoba laboratories (for COVID-19 testing data); 5) postal codes from the Canada Census (to distinguish urban from rural regions), and 6) census data for income quintiles based on ranges of mean household income, and grouped into five categories with each quintile assigned to approximately 20% of the population (Quintile 1 [mean income= CAD 17,910] to Quintile 5 [mean income=CAD 46,230]).
Study design and population
Using the Manitoba linked administrative health data, we conducted a quasi-experimental study. The linked data from Manitoba was used to create a large, pregnancy cohort pre-COVID-19 and during-COVID-19. All pregnancies (live birth or stillbirth) in Manitoba were included in the final cohort during the pre-pandemic period (Q4-2016 to Q1-2020) and during the pandemic period (Q2-2020 to Q1-2021). Subjects required a 1-year residency in Manitoba before study entry to enable sufficient covariate creation. The delivery date was defined as the date that a procedure code of delivery was recorded in the database. Maternal age at delivery, comorbid diseases (MCHP validated definitions of asthma32, diabetes33, and hypertension34), income (quintiles), parity (primipara and multipara), and residence (urban and rural) (urban: Winnipeg and Brandon and rural: all other areas) were extracted from the database. As a proxy for the maternal SES, we used the MCHP validated classification based on the median neighbourhood income quintile: lower income (individuals in the lowest and second lowest median neighbourhood income quintile), higher income (individuals residing in the neighbourhoods with the three highest median neighbourhood income quintiles), and income unknown (individuals who cannot be assigned a neighbourhood income from the census data).35
Exposure and Outcomes
Pregnant women exposed to the pandemic restrictions during the first and second waves between March 1, 2020, and March 31, 2021, were compared to those who were pregnant before the pandemic period. A strict mitigation strategy was implemented in Manitoba from March 13, 2020 (first wave) followed by a subsequent ease of some restrictions between June and July 2020. More restrictions were applied in August 2020, with Manitoba meeting their peak lockdown stringency index of 81.8/100 during the second wave.26,36 Preterm birth was defined as a live birth where an infant is born at less than 37 weeks of gestational age37, while stillbirth was identified as a fetal death with a gestation of 20 weeks or greater.38
Statistical Analysis
We examined pregnant women demographics age: (<19 years, 20-34 years and >35 years), area of residence (urban vs. rural), income (quintile), and comorbidities were described before and during the pandemic. We estimated the quarterly rates of preterm birth and stillbirth from October 2016 to March 2021. We chose Q2-2020 as the intervention point for the COVID-19 period since the restriction measures started to impact clinical practice in Manitoba by the end of Q1-2020 (March 2020).
Interrupted times series analysis using autoregressive integrated moving average (ARIMA) models were used to examine the association of the pandemic with PTB and stillbirth rates, while accounting for autocorrelation between consecutive quarterly observations (to avoid bias due to seasonal variability that is known to affect pregnancy outcomes).39,40 We used the step and ramp intervention functions to evaluate the immediate change in level (sudden change) and change in slope (lagged effect) in the quarterly outcome rates within 3 models: 1) overall incidence, 2) among lower income group and 3) among higher income group.41 We generated ARIMA forecast models to compare the observed quarterly PTB and stillbirth rates during the pandemic with the counterfactual expected rates in the absence of the pandemic. In addition, we used generalized linear models to test for the potential effect modification by maternal SES using multiplicative interactions.42 Analyses were conducted using SAS, version 9.4 (SAS Institute, Inc). P-value <0.05 was used as the threshold for statistical significance.
Results
During the study period, we included 70,931 pregnant women: 54,306 in the pre-pandemic period and 16,625 during the pandemic period. Of the pregnant women included, the mean age was 29 years and over half lived in urban areas. Twenty five percent of the pregnancies were among women in the lowest income quintile, while 15% of pregnancies were among women in the highest quintile. Asthma (24%), hypertension (8%) and diabetes (6%) were the most common comorbidities (Table 1).
Maternal characteristics before and during COVID-19 periods in the Canadian province of Manitoba.
Preterm Birth
The rates of PTB were 8.77% and 9.45% before and during the pandemic period, respectively, with a relative increase of 9%. In the lower income group, 10.86% of women had preterm delivery compared to 8.13% in the higher income group during the pandemic (Table 2). Figure 1 illustrates the trends of the observed versus expected PTB rates during the study period. During the first quarter of 2021, the absolute differences in the observed and expected PTB was 2.05%. The implementation of mitigation measures was associated with a statistically non-significant 4.5% change in PTB rates (p=0.094). However, we observed a significant rebound decrease by 0.63% (p=0.005) among the quarterly PTB rates (Table 2). Among pregnant women with lower income, the onset of pandemic measures was associated with an immediate and significant 20.2% increase in PTB rates (p=0.013), while a statistically non-significant 12.2% reduction was observed among higher income women (p=0.332). Moreover, the rebound decrease among the lower income group was statistically significant (1.34% change, p=0.001).
Preterm birth and still birth outcomes before and during the pandemic periods in the Canadian province of Manitoba.
Observed and expected preterm birth (PTB) rates among lower income, higher income, and all pregnant women in Manitoba from Q4-2016 until Q1-2021.
Stillbirth
The rates of stillbirth were 0.57% and 0.76% before and during the pandemic period, respectively, with a relative increase of 33%. Among pregnant women with lower income, the rate of stillbirth was 0.95% compared to 0.6% in the higher income group during the pandemic (Table 2). Figure 2 demonstrates the trends of the observed and expected stillbirth rates pre-pandemic and during the pandemic. During the first quarter of 2021, the absolute differences in the observed and expected stillbirth was 0.04%. Immediately after the onset of the pandemic, there was a statistically non-significant 18.6% change (p=0.959). The pandemic period was associated with a non-significant rebound decrease among the quarterly stillbirth rates (β=-0.0122p=0.878) (Table 2). Within the income subgroups, the rates of stillbirth deliveries did not differ significantly between the pandemic and pre-pandemic periods.
Observed and expected stillbirth rates among lower income, higher income, and all pregnant women in Manitoba from Q4-2016 until Q1-2021.
SES effect modification
We assessed the interaction between COVID-19 restrictions and income (as a proxy for SES) on the overall incidence of PTB and stillbirth. Our models suggested a modest statistically significant effect modification by income for the overall incidence of PTB (p=0.047). Although there was an increase in stillbirth rates among the lower income group, the income effect modification was not statistically significant (p=0.643, Supplementary Table S1).
Discussion
In this Canadian province-wide quasi-experimental study, we observed no statistically significant changes in the rates of PTB and stillbirth immediately after the onset of COVID-19 restrictions. However, the observed change in the PTB rate could be considered clinically significant, followed by a return to pre-pandemic averages within the following 3 quarters. Such trend was observed among stillbirths but was hardly detected due to the low sample size. We also observed a statistically significant 20% increase in PTB rates among the lower income group immediately after the measures were applied. Such an impact was not observed within the higher SES group. Our data suggests that the pandemic restrictions had a rebound effect on the overall PTB rates, mostly attributed to the impact within the lower income group.
While the exact reasons for the observed increase followed by reversal to pre-pandemic level have yet to be determined, various factors may have contributed to the change in the overall PTB rates. The lockdown measures are hypothesized to be associated with social or behavioural changes among pregnant women and thus prevents potential adverse perinatal outcomes.43 Our study shows that changes in PTB rates are not consistent among pregnant women with different SES. The lower PTB rates among higher income group could be correlated to better hygiene measures, increased social support, reduced work-related or social stress, decreased anxiety and reduced exposure to environmental pollutants.12,44–47 On the other hand, pandemic restrictions among the lower SES group resulted in financial insecurities, increased stress and mental health concerns, and changes in maternity care practice that affected maternal health leading to higher PTB rates.20,48–50 Evidence indicates that the pandemic and associated lockdown measures exacerbated existing social inequalities.50,51 Our study showed the impacts of the pandemic restrictions were modified through the socioeconomic status of the mother. Whereas, in a published study in the Netherlands44, no association between COVID-19 restrictions and adverse perinatal outcomes were observed while accounting for SES as an effect modifier. However, corroborating our findings, Caniglia et al. showed a significant reduction in PTB rates post-lockdown period compared to during lockdown, highlighting that lockdown restrictions may have a delayed effect on adverse pregnancy outcomes.52 In contrast, evidence from other Canadian provinces suggests that the pandemic had no significant impact on PTB rate.19–22 In the largest city in the Canadian province of Alberta (Calgary), Alshaikh et al. reported that during the first wave lockdown, stillbirth and PTB rates remained unchanged, however, very preterm birth rate (<32 weeks of gestation) dropped significantly.20 Three retrospective cohort studies, in the Canadian province of Ontario, examined the association between the pandemic first wave and perinatal outcomes, and reported non-significant changes in PTB and stillbirth rates.19,21,22 Differences between these studies and the current study may be attributable to differences in the analytical methods performed, the pandemic timeframe examined, and the pandemic measures implemented among the Canadian provinces. Our study assessed 12 months during the pandemic period (up to March 2021), whereas these studies examined the period until October and December 2020. Additionally, previous studies used the Better Outcomes Registry & Network (BORN) Ontario and Alberta Perinatal Health Program40 databases which covered only maternal and neonatal data until discharge, whereas the current study provides more generalizable results due to the use of a province-wide database that includes the entire Manitoban pregnant population without restrictions of insurance coverage. In addition, the effect modification by SES was not assessed in the aforementioned studies.19–22
Within the published literature, single-center45,51,53–55and national56–60 studies have reported conflicting evidence in relation to perinatal outcomes during the COVID-19 pandemic. In a meta-analysis, Yang et al. examined the influence of COVID-19 restrictions on maternal and neonatal outcomes.61 Among 36 studies, the authors found a non-significant reduction in the odds of PTB during the pandemic period in national studies (unadjusted OR 0.99, 95% CI 0.94-1.03, I2=76%), however, a significant reduction was observed in single-center studies (unadjusted OR 0.99, 95% CI 0.86-0.94, I2=76%).61 Moreover, there was no difference in stillbirth rates between pre-pandemic and pandemic periods in both single-centre and national and studies.61 Another meta-analysis of 17 studies examined the influence of the implementation of lockdown measures during COVID-19 first wave on maternal and neonatal outcomes in 15 countries.62 Vaccaro et al. observed a significant association between lockdown restrictions and an increased risk of stillbirth (RR= 1.33, 95% CI 1.04 −1.69, I2=71%), but not preterm birth (RR=0.93, 95% CI 0.84-1.03, I2=82%).62
Our study has major strengths to mention. We used province-wide administrative health data which included the entire pregnant population in the province of Manitoba. Furthermore, we included 1 year of the pandemic period using a quasi-experimental approach, thus facilitating a causal interpretation of the association between COVID-19 restrictions and the examined outcomes. Additionally, we used different model specifications to account for seasonality that could influence the study outcomes. The main limitations of this study should be acknowledged. We did not examine the influence of COVID-19 infection on the study outcomes; however, the number of reported COVID-19 positive pregnant women was small in relation to the total pregnancies in our cohort, and it is unlikely that viral infection would change our results.43 Although, we did not investigate variations across the different regions of Manitoba, the changes to maternal healthcare were implemented across the provincial healthcare system minimizing any differential variability within the data. However, some rural regions may have been substantially impacted during the pandemic compared to larger cities. Moreover, we did not include out of hospital births, maternal smoking and alcohol/substance use, and vaccination rates among pregnant women in Manitoba
Conclusion
The initial implementation of COVID-19 restrictions was associated with an increase in the rates of preterm births and stillbirths, followed by reversion to the pre-COVID-19 level. The observed rate changes were driven by socioeconomic status, with significantly higher rates of PTB among lower income pregnant women. Further studies are needed to monitor trend changes during subsequent pandemic waves and assess potential underlying mechanisms.
Data Availability
Data used in this article was derived from administrative health and social data as a secondary source. The data was provided under specific data sharing agreements only for approved use at Manitoba Centre for Health Policy (MCHP). The original source data is not owned by the researchers or MCHP and as such cannot be provided to a public repository. The original data source and approval for use has been noted in the acknowledgments of the article. Where necessary, source data specific to this article or project may be reviewed at MCHP with the consent of the original data providers, along with the required privacy and ethical review bodies. Statistical and anonymous aggregate data associated with this paper, along with metadata describing the original source, has been submitted to MCHP.
Competing interests
All authors declare that they have nothing to disclose.
Funding
This work was supported by a research grant from Research Manitoba. The authors have no financial relationships relevant to this article to disclose.
Ethical Consideration and informed consent
This study has been approved by the University of Manitoba Health Research Ethics Board (HREB #: H2020:335) and access to data was approved by the Health Information Privacy Committee of Manitoba Health (HIPC #: 2020/2021-33). Consent was not required as data are de-identified.
Acknowledgement
The authors acknowledge the Manitoba Centre for Health Policy for use of data contained in the Manitoba Population Research Data Repository under project (HIPC#2020/2021 - 33). The results and conclusions are those of the authors and no official endorsement by the Manitoba Centre for Health Policy, Manitoba Health, or other data providers is intended or should be inferred. Data used in this study are from the Manitoba Population Research Data Repository housed at the Manitoba Centre for Health Policy, University of Manitoba and were derived from data provided by Manitoba Health.
References
