Just under half the study sample (48.9%) reported at least one consultation or pregnancy check-up being postponed or cancelled during the first lockdown, whether on their own initiative (23.4%) and/or the initiative of the hospital and/or that of a health professional (36.3%).
Women who i) were inactive, ii) perceived received little or no social support, iii) experienced violence, and whose iv) healthcare professional changed during the first lockdown, were all significantly more likely to voluntarily change their pregnancy monitoring. Conversely, worry about the pandemic was inversely associated with changing pregnancy monitoring.
Covimater’s results show an association between violence and a voluntary change in pregnancy monitoring. The percentage of pregnant women who reported experiencing violence or serious arguments during the first lockdown was high (28.1%), but significantly lower than that obtained for women of childbearing age (18–49 years) in CoviPrev, a French general population-based repeated cross-sectional study which used the same methodology as Covimater and conducted data-collection waves at the same time (CoviPrev study, 28.1% vs. 32.9%, p = 0.03) (33). This result is in line with several studies suggesting that the prevalence of violence on women during pregnancy is no higher than in other situations. However, there is no international consensus about whether the risk of violence is higher in pregnant women than in women who are not pregnant (34–36). Violence during pregnancy not only negatively impacts mothers’ health, but also that of their unborn children. It also impacts success of antenatal care (40,43–45). Furthermore, violence is significantly associated with an increased risk of obstetrical complications (40,43–45). In terms of antenatal care, a survey held by the World Health Organisation in Tanzania studying domestic violence on women showed that it was significantly associated with fewer consultations for antenatal care because partners prevented or discouraged women from having them (36). Efforts to detect violence against pregnant women at an early stage must be continued in order to prevent its harmful impact on health (40).
In Covimater, perceiving little or no support during the lockdown was associated with voluntary change in pregnancy monitoring. These results reflect findings from the 2010 French National Perinatal Survey (NPS), where women who declared having no social support were significantly more likely to forego care (41). The perception of receiving little support may have been accentuated by the fact that during the first lockdown, in many maternity hospitals and private practices in France, neither partners nor people providing support to pregnant women were allowed to be present at consultations, obstetrical examinations, and hospitalisation for childbirth, except under certain conditions (15,18)). Only partners were allowed to visit after childbirth (15).
In our analyses, a change in health professional during the lockdown was associated with a higher likelihood of voluntarily postponing or foregoing pregnancy monitoring. As reported in several studies showing the importance of the patient/caregiver relationship in medical follow-up (in terms of treatment adherence, health examinations, etc.), it seems fundamental to ensure that the monitoring of pregnant women is as personalized as possible in the context of an ongoing pandemic (42,43).
In our study, women who had a higher worry score about the pandemic were less likely to change their pregnancy monitoring. This result suggests the need to communicate with pregnant women with a double objective: i) to avoid any increase in existing worry about the pandemic, and ii) to foster their adherence to health authorities’ recommendations concerning uninterrupted pregnancy monitoring. To ensure the quality and regular updating of information received by pregnant women, it is important to involve health care providers so that they can inform or direct their patients to reliable and responsive sources of information (44). French laws for patients’ rights and the Public Health Code stipulate that patients have the right to have access to information (45) and that doctors must inform them of advances in science according to their needs (46). Access to reliable information is therefore an essential element in effective patient follow-up.
Finally, in line with Ancelot et al.’s findings in the NPS study in France in 2010 (41), having a chronic illnesses or a pregnancy-related illness was not significantly associated with a voluntary change in pregnancy monitoring in Covimater. Furthermore, participants in Covimater with a deteriorated psychological state during the first lockdown were not more likely to change their pregnancy monitoring than those with no such condition (p-value = 0.89).
In addition to characterising women with a higher prevalence of modifying their pregnancy monitoring during the first COVID-19-related lockdown in France, our study also aimed to stress the reasons for these voluntary postponements or waiving of care. In addition to those related to the pandemic (i.e., fear of being infected by SARS-CoV-2, compliance with restrictions on movement), some of these reasons were organisational in nature, whether related to healthcare provision, or personal organisation problems linked to the pandemic. In Covimater, 28.6%, 17.7% and 15.3% declared, respectively, that they had not managed to make an appointment, that they had not been able to contact the health professional who usually followed them, or that it had been impossible for them to take days off work to attend their pregnancy appointments. Despite French authorities’ recommendations to promote video and telephone-based consultations when possible outside of the three compulsory ultrasounds requiring physical presence, a relatively large proportion of women were unable to contact healthcare structures, and as a consequence, modified their pregnancy monitoring (15,53–55). In terms of personal organisation of healthcare schedules, discussions are currently underway at the national level to provided current and future parents with greater flexibility to better reconcile their professional and parenthood (50).
Strengths and limitations
To the best of our knowledge, Covimater is the first national study in France to explore the experiences and behaviours of pregnant women during the SARS-CoV-2 pandemic. It used the same methodology as another study conducted in France at the same time on the general population entitled Coviprev. This choice was made to ensure comparison with women of childbearing age. Unlike studies from other countries which mostly focused on the third trimester of pregnancy during the current pandemic, Covimater included women with different gestational ages.
Covimater has some limitations. First, the probability of inclusion of individuals from the whole population of pregnant women in France was unknown. Consequently, greater caution is required when interpreting the statistical inference than would be needed for random sample studies. Second, sampling bias could explain the overestimation of the percentage of pregnant women with pre-existing chronic diseases or obesity. Third, as the study questionnaire was self-administered, there is always the risk that respondents misunderstood or misinterpreted questions and a risk of potential social desirability and recall biases. However, there is no reason to suppose that either of these biases should be limited to the particular sub-group of pregnant women who had postponed/forgone their pregnancy monitoring.