Women's experiences of maternity service reconfiguration during the COVID-19 pandemic: A qualitative investigation

Objective To explore women's experiences of maternity service reconfiguration during the first wave of the SARS-CoV-2 (COVID-19) pandemic. Design Qualitative interview study. Setting South London, United Kingdom. Participants Women (N=23) who gave birth between March and August 2020 in one of the ten South London maternity hospitals. Methods Semi-structured interviews were conducted (N=23), via video-conferencing software. Transcribed interviews were analysed ‘by hand’ using Microsoft Word. Template analysis was selected to code, analyse, and interpret data, according to the findings of a recently-published national survey of maternity service reconfiguration across the UK in response to COVID-19. Findings Three main themes emerged through analysis: (i) Disruption to In-Person Care and Increased Virtual Care Provision, (ii) Changes to Labour and Birth Preferences and Plans, (iii) Advice for Navigating Maternity Services During a Pandemic. Key Conclusions Women reported mixed views on the reduction in scheduled in-person appointments. The increase in remote care, especially via telephone, was not well endorsed by women. Furthermore, women reported an under-reliance on healthcare professionals for support, rather turning to family. Implications for Practice We provide insight into the experiences of women who received antenatal, intrapartum, and postnatal care during the first wave of the COVID-19 pandemic. Our findings should inform healthcare policy to build back better maternity care services after the pandemic.


Introduction
In December 2019, the SARS-CoV-2 or COVID-19 outbreak was first reported in Wuhan, China. Within weeks, the virus had spread throughout China, and cases were being reported elsewhere in the world. By 31 January 2020, the UK recorded its first case, and by 26 March 2020 ( Lillie et al., 2020 ), the government legislated national lockdown, through mandated and enforced stay-at-home orders ( UK Government, 2021a ). In the year since the first reported case, the UK has in the 11th version of the Royal College of Obstetricians and Gynaecologists ( RCOG; 2020 ) guidelines. Therefore, urgent modifications were made to maternity care services. The National Health Service [NHS] and RCOG published and rapidly updated their guidance on all aspects of maternity care, to mitigate the risk of SARS-CoV-2 infection, and in response to government-imposed social distancing and movement restrictions ( NHS, 2020a ;2020b ;RCOG, 2021 ).
A recent national survey (conducted between May-July 2020) documented substantial pandemic-responsive changes to UK maternity services ( Jardine et al., 2021 ). Most sites reported a reduction in scheduled antenatal (by 70%) and postnatal (by 56%) appointments, particularly for low-risk women. Almost all sites (86%, particularly in London) reported provision of at least some component of care using remote methods, usually telephone, particularly in early pregnancy and less so near or at term gestational age. Over half of units (59%) at least temporarily suspended support of birth at home or in a midwifery-led unit. Further, most health visitors 1 were re-deployed, meaning responsibility for ongoing postnatal care was delegated to other healthcare professionals or was delivered at reduced capacity ( Institute of Health Visiting [iHV], 2020 ).
Globally, maternity care services changed the way in which they delivered care to reduce risk to mothers, their newborns, and the staff who were providing care ( Grünebaum et al., 2020 ;Jardine et al., 2021 ;Montagnoli et al., 2021 ;Szabo et al., 2021 ;Wu et al., 2021 ). Whilst this differed the world over, the common factors included increased use of virtual care and reduced in-person or 'face-to-face' care, reduction of choice with regards to desired location of birth, and fewer people allowed at points of care or during birth, meaning women were often separated from their partners or chosen birth companions ( Fumagalli et al., 2021 ;Jardine et al., 2021 ;Sweet et al., 2021 ).
Using established qualitative research methodology, we sought to explore the experiences of women in South London, UK, receiving antenatal care both before and during the pandemic, and who subsequently gave birth and received postnatal care during the pandemic. Our findings contribute to ongoing development of healthcare policy and guidelines for the provision of maternity services during subsequent waves of COVID-19 and post-pandemic service re-build, as well as to prepare for future health system shocks.

Design
Individual interviews were utilised to facilitate understanding of women's experiences of pregnancy and childbirth during the COVID-19 pandemic, between March and August 2020, thus we interviewed all participants (N = 23) who showed interest in the study, and who met eligibility criteria in this set period of time. Interviews were semistructured to ensure common questions were asked of all participants and responses could be analysed across the dataset, whilst still allowing sufficient flexibility in the interview schedule to follow interesting lines of inquiry pertinent to individual participants ( McIntosh and Morse, 2015 ). Appendix 1. Interviews were conducted chronologically, covering women's experiences of antenatal, intrapartum, and postnatal maternity care services, in addition to women's psychosocial experiences of pregnancy, labour, childbirth, and motherhood, and how the COVID-19 pandemic may have affected those experiences.

Ethical approval
Ethical approvals were sought and granted from the King's College London Biomedical & Health Sciences, Dentistry, Medicine and Natural 1 In the UK, Health Visitors are Nurses or Midwives who undergo additional training in Public Health Nursing. It is therefore common in the UK for Health Visitors to provide postnatal care from soon after birth. & Mathematical Sciences Research Ethics Sub-Committee, in June 2020 (project reference:-HR-19/20-19486).

Patient and public involvement and engagement
This study was presented at a National Institute for Health Research [NIHR] Applied Research Collaboration [ARC] South London, Patient and Public Involvement and Engagement [PPIE] meeting for maternity and perinatal mental health research (July 2020), which has a focus on co-morbidities, inequalities, and maternal ethnicity; an NIHR ARC South London Work in Progress Meeting (October 2020) focusing on maternity and perinatal mental health research; and at an NIHR ARC South London Public Seminar (February 2021) which focused on COVID-19 rapid response research. We received advice on study design, recruitment strategy, and interpretation of our findings as well as support to share the study details in the local community from both lay and expert stakeholders, including members of the public, those with lived experience, health and social care professionals, researchers, and policy makers.

Setting
The participants in this study received their antenatal care and gave birth in one of ten South London maternity hospitals. South London is recognised as the region of London under the River Thames spanning approximately 250 sq mi, with a population of approximately three million people ( Office for National Statistics [ONS], 2019 ). The area is home to high levels of ethnic diversity ( ONS, 2019 ), and indices suggest high levels of social complexity, including multiple deprivation ( Greater London Authority [GLA], 2019 ). Types of deprivation include barriers to housing and poorer living environments, higher levels of crime, and increased deprivation affecting children and older people; although employment rates remain high ( GLA, 2019 ).

Participants and recruitment
Women were eligible if they had given birth in South London during the study period and had received at least part of their maternity care prior to both the COVID-19 pandemic and the UK 'lockdown' restrictions (from 23 March 2020 until easing began on 13 May 2020).
Recruitment took place on-line, with posts on social media platforms, and by word-of-mouth. We also tried to specifically recruit women from Black, Asian, and Minority Ethnic backgrounds through our established Patient and Public Engagement and Involvement [PPIE] network, so as to achieve a more demographically representative sample of South London itself.
A critical case purposeful sampling technique ( Farrugia, 2019 ) was used to identify newly parous women who had recently given birth in South London, UK. This meant we recruited from one area, aiming to extrapolate our findings to the wider general population (e.g. to other cities where ethnic diversity is high or where there are high levels of social complexity and multiple deprivation). In addition, this approach aids comparison between participants and the exploration of commonalities and differences within a single given context ( Farrugia, 2019 ) in this case, geographically-bounded to South London.

Data collection
Interested participants e-mailed the research team and were sent a participant information sheet and consent form, to be completed electronically prior to their interview or verbally at the beginning of the interview. Due to UK Government-imposed lockdown and physical distancing restrictions associated with the COVID-19 pandemic, all interviews were conducted by one researcher [SAS], remotely using videoconferencing software. Interviews were recorded and ranged between 30 and 90 minutes ( M Time = 52 minutes). The audio was transcribed by a professional transcription company using intelligent (or 'standard')  Brooks et al., 2015 andKing &Brooks, 2017 ). transcription, i.e. not following verbatim which would have included every single incidence of "contextual matter such as false sentence starts; filler-words, or those which are emphasised or repeated; grammatically incorrect phrases, and those spoken in a different way (including foreign words and the demarcation of whispers, mumbles, raised-or acted-voices e.g. falsetto), whilst also noting coughing, laughing, crying etc., as well as any interruptions to the interview. " ( Silverio et al., 2019 ;p.44). This decision was taken to match the rapid research approach we had adopted.

Data analysis
Template analysis ( King, 2012 ) was selected to evaluate women's experiences of pandemic-related maternity care changes according to the findings of a recently-published national survey of maternity service configuration across the UK in response to COVID-19 ( Jardine et al., 2021 ), and conduct our evaluation rapidly enough to inform care during subsequent waves of COVID-19 and post-pandemic. As a qualitative methodology it is philosophically flexible, but has been used in this study with a critical perspective ( King and Brooks, 2017 ;p.15), whereby we are interested in the "realities which exist independent of human activity " meaning contextual factors outwith human control (in this case, the COVID-19 pandemic) "may not directly determine behaviour, they are nonetheless recognized as having important influences in understanding experience ". Template analysis follows a stepped procedure (see Fig. 1 ), broadly categorised as: (re)familiarisation with the data; preliminary coding; organisation of themes; defining an initial coding template; application of the initial template; finalisation of the template and application to the full dataset ( Brooks et al., 2015 ). Importantly, the initial template can be modified to ensure completeness of analysis, as described below ( King, 2012 ).
An analyst who had not conducted the interviews [KDB] checked every transcript for accuracy against its corresponding audio, which also allowed her to familiarise herself with the data. The analyst who had conducted the interviews [SAS] was able to re-familiarise himself with each transcript in the context of the dataset. Transcripts were analysed 'by hand' [SAS] using annotation tools on Microsoft Word, rather than qualitative data analysis [QDA] software, to allow multiple researchers to access the dataset at the same time using shared files, and facilitate more rapid analysis. Rigour is maintained in Template Analysis and in our study by reflexively engaging with data and analytical processes throughout all six analytical steps, as detailed above. Furthermore, as per Brooks et al., (2015) , iterative coding from step three onwards allowed for thorough and methodical analyses, whilst accuracy checking employed in the final two stages allowed for certainty and confidence in thematic saturation (i.e. final themes were well supported by data contained within the dataset).
The coding template initially included the following key changes in maternity care services in the UK (as described by Jardine et al., 2021 ): (i) Disruption to Routine, In-Person Care; (ii) Increased Provision of Virtual Appointments; and (iii) Changes to Labour and Birth Preferences and Plans. An iterative approach ( King and Brooks, 2017 ) was taken to coding data and extracting key quotations [SAS], involving reading and re-reading transcripts, and ensuring comparisons were being made between all participants' data to that point. The coding template was tested with the initial transcripts [SAS], modified accordingly, and then re-applied to all data within the dataset. The modified template incorporated a fourth theme of (iv) Advice for Navigating Maternity Services During a Pandemic. Later iterative analytic work noted similarity between the first two themes (i & ii), and so these were merged and renamed into (i) Disruption to In-Person Care and Increased Virtual Care Provision; thus, results are presented as three themes. The cohesion, completeness, and meaningfulness of all data, coding, and analysis was checked and confirmed with regular consultations with the wider study team [KDB, JS, LAM] throughout the data collection and analysis stages.
Qualitative analysis of the data resulting in three themes is presented below, with the most illustrative quotations presented in text and supplementary quotations found in Table 2 . Where appropriate, representative figures have been used to further illustrate participant data (see Figs. 2 & 3 ). Each quotation is presented with its corresponding participant identifier.

Disruption to in-person care and increased virtual care provision
Lockdown restrictions and subsequent maternity service reconfigurations affected routine appointment schedules and the frequency of those appointments. Antenatal care visits were found to have been either altered frequently or cancelled altogether, with women discussing their perceptions of virtual care appointments as having less value or importance: "…they showed us in the book [maternity record] at the first appointment, 'These are all the weeks you should be having these appointment and this is how it should all work', then it was the polar opposite to that because pretty much every single one of those was cancelled or turned into a telephone call. " (Participant-021) In addition, women discussed how reduced frequency of antenatal care appointments made them feel, with provision of virtual care not equated to the in-person care they had either expected or wanted: "I would say: don't have any expectations for the midwife care up until the point of labour. At the time you feel cheated out of these appointments. You keep looking at this book that tells you, you should have been seen in all these weeks and you are thinking, I have been seen face-to-face twice. You feel a bit cheated and anxious that they maybe are missing out on something…" Home SVD 2nights † Participant 015 is missing, as they withdrew from the study after consenting (and being assigned a participant number), but before taking part in the interview. ‡ Ethnicity was defined by participants in response to the question: "Could you tell me the ethnicity with which you identify? " (Participant-022) Whilst many women understood why appointments were changing so rapidly, concern often arose about the lack of in-person care, in re-lation to the growth and wellbeing of the baby during pregnancy or of their newborn infant (see Fig. 2  In relation to the reduction of antenatal and postnatal care, women often mentioned virtual care was sometimes used as an alternative, however, was recognised as not being accessible to all: "I think I would question the accessibility of that. Not everyone does have a smartphone and so expecting people to be able to receive a video call is not necessarily the most inclusive thing. Although where it is available, it would be nice. I think the weirdest call was the one I had with the consultant at < Hospital > just before giving birth, which felt very impersonal because it was someone I had not met in person and never to hear from again. " (Participant-003) However, where accessibility was not an issue, virtual care was often discussed as an inappropriate medium through which to conduct health checks, especially postnatally, during which time both newborns and new mothers required attention: "I was also referred by my health visitor for a breastfeeding Zoom call. That was ridiculous. I needed to see someone face-to-face because they have to check your position, your latch and whether your baby has tongue tie. Feeding support has to be there face-to-face and it needs to be available. " (Participant-005) Whilst not favoured in the postnatal period as much as it had been accepted for certain aspects of antenatal care, virtual care was tolerated as a better alternative to no care at all, which many postnatal women faced: Sometimes, this extended to the communication between maternity care providers and women during their intrapartum care: "So anyway, I went into this room and I basically said to the midwife, 'Look, I'm getting really panicky; I need to go and get some fresh air outside and I need to speak to my husband. I need to go outside.' And I think she could see that I was getting a bit stressed and getting a bit panicky. And it was a bit odd. She had to go out and ask more of a senior midwife is it all right if she goes out for 20 minutes to the car park. And I was a bit like 'hang on, I'm not a prisoner'. " (Participant-004) Occasionally, the restriction on birth partners meant that nonbirthing parents missed the birth of their baby (see Fig. 3 ):

"…she was born within half an hour of getting proper contractions [laughs]. I was still in the induction bay, basically, and I was like, ' Oh excuse me, I think I'm having my baby'. [Laughs] Then they just wheeled me round incredibly quickly to a room and she was born within 10 min-utes……… so, yes, all that morning he wasn't there and because they only realised it was an established labour half an hour before she was born, he didn't get there……… It was a shame really. "
(Participant-021)

Fig. 3. Artistic depiction of findings -birth partners
However, restrictions were often viewed as arbitrary and, therefore, changes to birth plans, including presence of birth partners, were frequently reported as unnecessarily frustrating:

"I'm not clear that the policy on fathers not being able to be present until you're four centimetres dilated is a necessary decision. "
(Participant-001)

Advi ce for navigating maternity services during a pandemic
The final theme of this analysis covered women's advice for other pregnant and birthing women. Largely, advice centred around replacing the face-to-face support with virtual forms of social support, and a noticeable under-reliance on healthcare professionals for this type of support: "Trying to talk to people. Nothing can beat face-to-face in terms of being able to go somewhere and talk to someone face-to-face, someone coming into your house and having a cup of tea or you going in someone's house and having a cup of tea, but if that is not possible trying to talk to people, talking to people on the phone. When I came off the phone from particularly my friends and my mum -anyone who had had a baby, who understood what some of these difficulties felt or looked like -I always felt reassured. " (Participant-013) In addition, there was a firm belief amongst women that whilst managing COVID-19 was currently the immediate priority of all healthcare services, pregnant women were never going to be forgotten, and, more so, were always privy to prioritisation amongst healthcare services and professionals: "The advice my mum kept giving me was: Try and stay relaxed about it. Maternity services are never going to stop. People are going to be giving birth. My mum kept saying to me, 'Every hospital prioritises pregnant women and maternity services.' They are never going to compromise on the welfare of you. If they are restricting your visitors, that is only to protect you……… Clearly in my experience maternity services were being prioritised just as much and there was no question of compromising on the level of care. Try and stay relaxed. " (Participant-018) And finally, many women recognised the importance of mental health as well as physical health during pregnancy, childbirth, the postnatal period, and during the COVID-19 pandemic:

"…try to keep on top of how you are feeling about things, because there's a lot of being in pregnancy, when you feel like everyone else knows what is best for you and you don't feel that your own feelings, your own views, are heard. And sometimes it is hard to even identify how you actually feel about something. And I guess the other thing I would say is don't be afraid to -and I think this is easier for some people than others -don't be afraid to ask for things: advocate for yourself. "
(Participant-010)

Main findings
The landscape of healthcare service delivery during the COVID-19 pandemic changed rapidly and healthcare Trusts within the NHS adapted their services, and continue to do so, to ensure both their staff and the populations they serve remain safe. Our interviews of women in South London confirmed disruptions in maternity care, in terms of reduced frequency, reduced face-to-face visits, relocation of care, and/or cancelled antenatal and postnatal care, and the impact that this had on women's experiences.
Virtual care -either by video-call or telephone -was deemed more acceptable antenatally than postnatally, and there was an emphasis on prioritisation of face-to-face care for foetal and newborn health surveillance. However, virtual care was preferred to no care whatsoever -reported frequently, postnatally and especially in relation to health visitors. There was an emphasis on in-hospital restrictions placed on birthing mothers, from healthcare providers requesting women delay hospital attendance when in labour, to separation of women from their birthing partners until they were in established labour, and frequently, the non-birthing parent missing the birth of their child.
Finally, most women focused on replacing in-person social support with virtual alternatives to ensure they were seeking help and practicing wellbeing throughout their pregnancies and after their baby arrived. However, there was an under-reliance on seeking support from healthcare professionals ( Birthrights, 2021 ). This under-reliance on healthcare professionals, showed a social shift of emotional burden and advice seeking from trained professionals to family members or friends who had experienced pregnancy and childbirth beforehand. Whilst it may have reduced the workload of healthcare professionals and providers, it largely meant that women sought advice from those with experience, rather than those who were appropriately qualified ( Birthrights, 2021 ;Fumagalli et al., 2021 ;Sweet et al., 2021 ). The potential outcome of this -both positive and potentially negative -is yet to be substantiated, evaluated, or calculated.

Strengths, limitations, and future directions
This study was conceived and conducted as a rapid research response to the COVID-19 pandemic. To our knowledge, this is the first qualitative study of women's experience of maternity service reconfiguration in the UK during the COVID-19 pandemic. Other strengths include the large sample size for a qualitative study, and use of template analysis to relate women's experiences to reported service configuration, and to do so iteratively, rigorously, and at a rapid rate.
Limitations include limited ethnic, relationship, and socio-economic diversity amongst our participants, despite our efforts to recruit a more representative population and targeted advertisement of the study. Our participants represent views of women in South London, an area of the country particularly hard-hit by COVID-19. However, their descriptions of maternity care service changes mirror those described nationally and so may have broader applicability ( Jardine et al., 2021 ). We interviewed women who began receiving antenatal care pre-pandemic and then completed antenatal care and gave birth during the pandemic; it is not known whether their views may reflect re-alignment of expectations formed pre-pandemic and, therefore, may differ from the views of those whose expectations were formed during the pandemic. Future research should compare findings from new data collection with our findings whilst also making efforts to recruit a more ethnically diverse sample ( Fernandez Turienzo et al., 2021 ).
Finally, we interviewed few women with hypertension or diabetes. Therefore, we are unable to comment on the increased use of home blood pressure and glucose monitoring during the pandemic. Such practices were increasing in prevalence pre-pandemic, even by women's independent choice ( Tucker et al., 2021 ). Future research should consider these specific populations of pregnant women, including those with other pregnancy or neonatal complications (e.g. hyperemesis gravidarum, pre-eclampsia, NICU/SCBU admission).

Interpretation
Our analysis details the experiences of women who were pregnant and subsequently gave birth during the COVID-19 pandemic, thus adding experiential evidence to the outcomes of the national survey of service reconfigurations ( Jardine et al., 2021 ). Women were unhappy about the reduction of relational care. While they understood the pandemic mandated certain circumstances to keep everyone safe , they questioned whether some of the apparently arbitrary decisions about provision of care really improved safety, as they certainly diminished care quality. It should also be noted that, whilst not elucidated in this study, there is widespread concern (see Montagnoli et al., 2021 ) that virtual care can lead to less frequent reporting or detection of unsafe living or relationship conditions (i.e. domestic violence).
Women were reassured that maternity care would always remain a priority of healthcare services, regardless of the pandemic, as babies will continue to be born, and women will continue to require maternity care, regardless of how the service is reconfigured ( Grünebaum et al., 2020 ). Whilst re-aligning women's expectations may have resulted in different views compared with women who booked during the pandemic, there may be lessons to learn about preparing women for uncertainties related to pregnancy complications and external forces which may negatively affect pregnancy experiences ( Fumagalli et al., 2021 ;Sweet et al., 2021 ;Wu et al., 2021 ).

Conclusion
Our study provides new insights into the experiences of women who received part of their maternity care as substantially reconfigured during the first wave of the COVID-19 pandemic, as documented by the recent national survey ( Jardine et al., 2021 ). Women variably endorsed the reduction in scheduled antenatal and, in particular, postnatal appointments. While women recognised that modifications in service delivery were necessary to protect themselves and others from SARS-CoV-2 infection, they expressed disappointment at the loss of choice, autonomy, and relational care, particularly with regards to their birth plan and labour. The increase in remote care, especially over the telephone, was viewed with disappointment, particularly for newborn care and infant wellbeing. Women's reliance on personal, rather than professional, relationships, could be due to women not wanting to burden their care providers during the pandemic, or because most women did not receive continuity of carer and, consequently, lacked the relational care required to feel safe in asking for support. Initially, our findings suggest four key recommendations: 1) if virtual care is to become part of routine maternity care, video methods should be employed; 2) staff and women may require training to utilise virtual methods of care and Trusts should be aware of data poverty which exists amongst especially deprived areas across the UK, meaning women do not always have access to the internet in order to access virtual care; 3) due to the nature of maternity care (whereby care cannot be delayed as it can for elective surgeries, for example), policy makers and healthcare providers should carefully consider whether maternity staff should in fact be re-deployed at all during a health crisis, so as not to compromise the health and welfare of pregnant women and their babies; and 4) chosen birth partners should be deemed essential to all aspects of maternity care and should not be excluded. Whilst countries around the world, and the NHS within the UK are looking at post-pandemic service delivery and care, our findings complement ongoing studies of healthcare delivery during the pandemic and can inform plans to build back better maternity services as this pandemic continues, thereafter, and in preparation for future health system shocks.

Funding
This project was funded by the King's College London King's Together Rapid COVID-19 Call, successfully awarded to LAM, SAS, AE, PvD, and colleagues (grant reference:-204823/Z/16/Z) as part of a rapid response call for research proposals. The funder had no role in the recruitment, data collection and analysis, or write-up associated with this work. This study was also supported by an NIHR Senior Investigator Award, held by JS (ref:-NIHR200306).

Details of Ethical Approval
Ethical approvals were sought and granted from the King's College London Biomedical & Health Sciences, Dentistry, Medicine and Natural & Mathematical Sciences Research Ethics Sub-Committee, in June 2020 (project reference:- HR-19/20-19486).

Data Availability Statement
The data supporting the findings of this study are available upon reasonable request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.