Our qualitative exploration shed new light on women’s experiences of perinatal psychological distress during the COVID-19 pandemic. We now consider our synthesised findings in relation to previous research to identify some key psychological indicators of perinatal distress and highlight opportunities to support perinatal women within and beyond the COVID-19 pandemic.
Identifying perinatal distress
In keeping with previous research , tearfulness was the most expressed symptom associated with distress, experienced by almost half of participants. Anxious-type symptoms of ‘worry and overthinking’, ‘fear’, ‘anxiety and nervousness’, and ‘panic’, were, when combined, also apparent in more than half of respondents. This was not surprising given PASS scores suggested a similar proportion of the sample had clinically significant anxiety symptoms. ‘Worry and overthinking’ was the second most reported symptom theme, in contrast with previous evidence of ‘worry and fear’ being among the least prevalent symptoms of perinatal distress . This may reflect the context of the pandemic, as several COVID-specific factors may have increased the salience of worry in the present sample (e.g., [27, 49]), or could equally be the result of methodological differences. For example, Coates et al.  analysed a small number of in-depth interviews, providing opportunity for an array of symptoms to be expressed; in comparison, survey questions (as used in the present study), typically prompted a brief response. Furthermore, Coates et al. treated ‘overthinking’ as an independent theme and clustered ‘worried’ with ‘scared’ which produced their least reported symptom theme. Had they combined worry and overthinking as we did, they may have found a similar prevalence. Finally, it is notable that prenatal women were not included in Coates et al.’s research. Fear was indeed less commonly reported amongst postnatal women than prenatal women in our analysis.
In our prenatal subsample, fear was common and typically associated with pregnancy-specific experiences, such as the fear of miscarriage. It has been suggested that pregnancy anxiety should be recognised as a unique construct [50, 51], and the present findings suggest there may be some benefit to this. Whilst the prenatal and postnatal sub-samples contained a similar proportion of participants scoring above the threshold for perinatal anxiety and depression, their qualitative experiences of distress differed. For example, guilt, frustration and feeling overwhelmed were all more commonly reported by postnatal women. Such differences have important clinical implications. Not only do they highlight the need for screening tools which assess the distinctive experiences of distress for women in the prenatal versus postnatal periods, they also suggest that pregnant and postnatal women may benefit from individual therapeutic intervention programmes specifically targeting the symptoms most distressing to them.
Despite the known comorbidity of perinatal anxious and depressive symptoms , and the prevalence of clinically relevant depressive symptoms in the present sample (Table 2), women’s descriptions of distress gravitated towards feelings typically associated with anxiety, as well as transdiagnostic symptoms of distress, rather than those traditionally related to depression. This reinforces the proposal that it may be more appropriate to focus on broader experiences of perinatal distress than address the two disorders separately in perinatal populations [16, 18]. Furthermore, although not part of the main analysis in this study, a post-hoc comparison revealed all of our themes were prevalent amongst women with both clinically relevant levels of anxiety and/or depression. It is also vital that screening tools capture the full range of distress, including anxiety-related symptoms, as even subclinical symptoms have been reported to impact mother and infant wellbeing (e.g., [21, 52]), we therefore recommend that all symptom themes be considered when screening for clinically concerning psychological symptoms.
Sources of perinatal distress
Six themes and sixteen sub-themes captured the psychosocial stressors that perinatal women associated with the abovementioned feelings of distress. Three COVID-specific themes broadly echoed findings from elsewhere in the world during the pandemic (e.g., [30–33]), whilst a further three themes described more generic perinatal triggers. Meaney et al.  reported an array of factors to have increased stress amongst pregnant women across multiple geographical locations during the pandemic. Just over ten percent of their sample were from the UK, and most of their participants were in the USA or Ireland. Despite this, many similarities were apparent between their findings and the present analysis, highlighting some of the shared experiences faced by pregnant mothers throughout Western populations.
Thematic analysis revealed shared psychological phenomena underlay many of the psychosocial stressors attributed to distress. For example, the sense of conflict between expectations of pregnancy or motherhood and reality, and an associated guilt for not achieving expected mothering standards, was observed across multiple themes and in relation to both COVID-specific stressors, as well as general perinatal experiences. These findings are consistent with the results of previous qualitative research which have attributed unrealistic expectations of motherhood, and guilt and self-blame, to psychological difficulties in the postnatal period [13, 18, 53]. This also fits with quantitative evidence of maladaptive beliefs towards motherhood increasing the risk of perinatal anxiety and depressive symptoms [54–56], and the reported relationship between dysfunctional perfectionism and postnatal distress . As such, interventions to manage misconceptions around mothering ideals and better prepare women for the challenges of pregnancy and motherhood, such as infant crying and sleep deprivation, may be helpful in reducing perinatal distress.
When pregnancy fears were reported by the present sample, they were often associated with unresolved trauma of historic obstetric events, such as miscarriage and pregnancy complications. As such, interventions aimed at supporting individuals with a known history of pregnancy and birth complications are recommended. Research suggests the unique concept of Childbirth-Related Post-Traumatic Stress Disorder (CR-PTSD), and secondary Tokophobia [58, 59] should also be considered. Improved postnatal debriefing may help to identify problems before subsequent pregnancies, reducing the risk of future perinatal distress. The present findings contribute to a large body of evidence pertaining to pregnancy-specific anxiety (e.g., [50, 51, 60]), suggesting improved availability of support and information around specific fears for the pregnancy and infant wellbeing may alleviate some distress not necessarily related to previous experiences. Themes also revealed the importance of co-parents in supporting women through pregnancy, particularly where specific pregnancy fears were described, reinforcing the need to include co-parents in perinatal primary care.
This research enriches understanding of the role social relationships play in supporting PMH (e.g., [6, 7, 38, 39, 61–63]). At the time of data collection, lockdown restrictions in place to mitigate the spread of COVID-19 forced families into physical social isolation. Being unable to spend time with friends and extended family was commonly attributed to feelings of psychological distress, particularly within the postnatal subsample. Many postnatal women had expected to spend considerable amounts of time with their family and friends during maternity leave and expressed concerned regarding lost opportunities for their infant to build relationships with their wider family, again pointing to the impact of discrepancies between expectations and reality. This echoed concerns surrounding the loss of social support, bonding rituals, and traditional birth celebrations reported in research conducted in Australia  and the USA .
Although many women in the present study spoke positively about their pre-COVID experiences of social support, and social support was positively associated with maternal psychological wellbeing (e.g., [6, 7]), it is important to note that not all women described beneficial social relationships. Some attributed distress to interactions with family and friends, and more commonly, their partner. This accords with evidence of relationship dissatisfaction being a significant risk factor for perinatal distress [64–67]. Furthermore, although not disclosed in the present dataset, domestic violence is reported to have increased during the COVID-19 pandemic [68, 69] and should always be considered in practice when women disclose distress associated with close relationships.
Interventions to support women in developing healthy expectations of perinatal experiences may help to reduce perinatal psychological distress. This could take many forms; however, peer support may be particularly effective. The peer support model is substantiated by evidence that support from friends (not family or significant others) was associated with lower levels of postnatal anxiety and depression , and chatting with other mums was associated with reduced loneliness and anxiety in prenatal women . Specific attention should also be paid to pregnancy related fears, particularly in women who have experienced previous obstetric trauma.
Should further restrictions be required to reduce transmission of COVID-19, policy makers should consider the distress associated with certain decisions, such as the exclusion of partners from perinatal care, the cancellation of antenatal classes and mother-baby groups, and the closure of childcare settings, and schools when calculating the risks versus benefits. The absence of antenatal classes is also likely to have made it difficult for women to gain accurate information regarding their perinatal experience. Antenatal classes are known to benefit women in the preparation for motherhood and it may be that they could be adapted for online delivery should further social restrictions be necessary .
Finally, the present evidence points to areas requiring further research, most notably, the need to develop appropriate PMH screening tools capable of identifying the most commonly occurring symptoms of perinatal distress in perinatal primary care.
Strengths and limitations
When interpreting these findings, it is important to consider several limitations. Firstly, the self-selected sample lacked diversity and participants’ responses may have been influenced by social-desirability biases. However, Braun et al.  point to the benefits of anonymous questionnaires when researching sensitive subjects, and Moore et al.  proposed that online data collection methods may encourage disclosure of PMH difficulties. Second, the single open-ended survey question provided limited access to women’s experiences, although it allowed for inclusion of larger sample than many other qualitative methods which was a significant strength. Third, the framing of the question may not have prompted information desired to answer the specific research questions, particularly with regards to the expression of symptoms, but the indirect nature of the question prevented wording bias and allowed women to share the thoughts and experiences most salient to them. Fourth, researcher bias is always possible in qualitative studies, however prior awareness of these effects meant effort was made to contain biases. Finally, findings cannot be generalised beyond the context of the COVID-19 pandemic, nevertheless they are valuable when considering screening and interventions to address the ongoing distress experienced as the pandemic continues.