Maternity care during COVID-19: a qualitative evidence synthesis of women’s and maternity care providers’ views and experiences

The inclusion criteria are defined in our protocol [18]. In brief, using the SPIDER (Sample, Phenomenon of Interest, Design, Evaluation, and Research type) acronym [20], the Sample was primiparous and multiparous women who were pregnant or up to six months postpartum at the time of study, and maternity care providers (midwives, obstetric nurses, obstetricians, doctors, and allied maternity care professionals) who were directly involved in maternity care provision during the COVID-19 pandemic. The Phenomenon of Interest was women’s and maternity care providers experiences of maternity care during COVID-19. For this QES maternity care is broadly defined as the care provided, inclusive of health and wellbeing monitoring and assessments and the provision of perinatal health education and information to women, babies, and their families during pregnancy, labour and childbirth and in the postpartum period, up to six weeks following childbirth. Care settings may be the hospital, community, or home birth settings. Study Designs were published and unpublished qualitative studies and studies of mixed methods design where the qualitative data could be extracted separately. Survey designs with free-text response options were also considered for inclusion if the available qualitative data were of sufficient depth and had been analysed formally using a structured approach (e.g., thematic analysis, content analysis, etc.). The Evaluation of outcomes was centred on the narrative views, experiences, and perspectives of pregnant and postpartum women and maternity care providers. The Research type included primary research studies, in the English language, available from 01 January 2020 to the date of our search. Study abstracts were also considered if they reported sufficient data to contribute to the synthesis in a meaningful way.

To identify eligible studies, the electronic databases of MEDLINE, CINAHL, EMBASE, PsycINFO and the Cochrane COVID study register (https://​covid-19.​cochrane.​org) were systematically searched from 01 Jan 2020 to 22 Feb 2021. Given the pandemic context and the rapidity with which new studies were becoming available, we updated our searches on 13 June 2021 prior to commencing data synthesis. To avoid potential misrepresentations arising from language and contextual nuances in translating text, non-English full-text publications were excluded from the QES, however we included all languages in our search strategy. This allowed us to identify the extent of potentially eligible non-English publications and whether this presented as a source of possible language bias. The search terms, and their combinations, which were guided by our SPIDER inclusion criteria and adapted as relevant for database specific subject terms, were detailed in our protocol [18] and independently peer reviewed prior to implementation. These search terms were:

To further enhance sensitivity, we extended our search to include searches of the reference lists of included studies, the grey literature websites of http://​www.​opengrey.​eu/​ and https://​greylit.​org and the proceedings of the international Normal Labour and Birth Research Conference (Dec 2020) and the Maternity Expo Conference: Maternity Services after COVID-19 (Sept 2020).

Once all searches were complete and the citations were exported to EndNote reference manager, duplicate citations were removed. The remaining records were uploaded to Covidence, a software package designed to assist with preparing systematic reviews. Three members of the review team (SJF, KMS and VS) screened the records on title and abstract, with each record independently screened by at least two reviewers. Records forwarded for full text review were independently screened by two reviewers (SJH and VS). Disagreements at each stage of the selection process were resolved through discussion and consensus.

An adapted version of a quality appraisal tool developed by the Evidence for Policy and Practice Information and Co-ordinating (EPPI) Centre for use in a systematic review of healthy eating in children [21] and used in previous QES by review authors [22, 23], was used to assess the methodological quality of the included studies. Using the tool, each included study was assessed independently by pairs of reviewers (SJF and VS; KMS and HD) on the extent to which the study met the tool’s 12 quality appraisal criteria. Minimum standards for a Yes, No or Partially met judgement were agreed in advance (Supplementary File 1). The 12 assessment criteria spanned three domains: i) the quality of the study reporting, ii) the reliability and validity of data collection and analysis, and iii) the quality of the study methods. A decision to include all studies following quality assessments was agreed because qualitative data providing perspectives on views and experiences, irrespective of methodological quality “…could have led to important new angles of consideration” [24] p.1718].

Data were extracted from the included studies using a pre-designed data extraction form (see https://​osf.​io/​bzt38/​ for the template form). The form was initially piloted on two studies and would have been refined, if necessary, but this was not required. Relevant data were extracted from each study independently in pairs (SJF and VS; KMS and HD) and cross-checked for accuracy. The following information was extracted as available: study reference (including publication type and year published), study aim, description of the participants and the study setting, dates when the study was conducted, data collection and analysis methods, funding details, and all findings related to pregnant and postpartum women’s and maternity care providers’ views of maternity care during COVID-19. For studies that reported on both women’s and maternity care providers’ views and experiences, the data were extracted and tagged to the relevant population category for synthesis.

Thomas and Harden’s thematic synthesis framework, which involves line by line coding of extracted text, developing descriptive themes and generating analytical themes, was used to guide the synthesis of the findings data [25]. This synthesis method was chosen over other methods (e.g., framework synthesis, meta-ethnography), as it provides a process that can be used for synthesising the findings from most, if not all qualitative enquiries [25], and allows for the inductive identification and development of themes that reflect the included studies data, overall. To enhance rigour, two members of the review team (SJF and VS) independently coded data from three included studies, initially, and met to compare the codes for consistency and congruity. Following this, the extracted data were categorised into women’s data and maternity care providers’ data. These category data were coded separately by two reviewers (VS for women’s data and SJF for providers’ data), using the comment function in Microsoft Word to add codes to the text, and the descriptive themes were developed. A meeting involving all four review authors was then held where the descriptive themes and associated codes were reviewed, refined (if required) and agreed based on discussion, reflection, and iteration. A similar process was used in determining the analytical themes; that is, one reviewer (VS) developed the analytical themes relating to women’s views and experiences, one reviewer (SJF) developed the analytical themes relating to maternity care providers’ views and experiences, and all members of the review team met to discuss and agree the final analytical themes.

The database searches yielded 7461 records, with a further 32 records identified from searching additional sources. Of these 7493 records, 239 were duplicates and removed. The resulting 7254 records were screened on title and abstract against the QES inclusion criteria; 7007 of these were clearly ineligible and excluded. Full texts of the remaining 247 were retrieved and assessed for eligibility, of which 196 were excluded. During data extraction, one further study was subsequently excluded as it became clear that the sample were not maternity care providers and the data related to women’s views were not specific to COVID-19 [32]. Details of these 197 excluded records with reasons is available at https://​osf.​io/​bzt38/​. This screening process resulted in the inclusion of 50 records reporting on 48 studies, of which 32 were included based on our initial search (Feb 2021) and 16 included following our updated search [2, 15, 17, 33‐79]. For three records arising from one study, each of the records reported on a discrete population, that is, doctors [73], midwives [59] and women [56], with nuanced methods for recruitment and data collection as applicable to each population. For quality assessment, data extraction and synthesis purposes, we thus considered these records as single ‘studies’ contributing to the QES. Twenty-seven of the included 50 records provided data from pregnant and postpartum women, 17 provided data from maternity care providers and six provided data from both populations (Table 1). The screening and selection process, including results, is presented in Fig. 1 using the PRISMA flowchart [80].

 

The summary descriptive characteristics of the included studies, organised by population groups and alphabetically, are presented in Table 1. The summary methodological characteristics of the studies, similarly organised, are presented in Supplementary File 2. A total of 9,348 women and 2,538 maternity care providers were included in the 50 records and contributed data to the QES. The studies spanned the globe with 16 conducted in the USA, seven in the UK, six in Australia, three in Ireland, two in Italy, Turkey and India, and one in each of Iran, Spain, Brazil, Kenya, Ethiopia, South Korea and Puerto Rico. A further three studies were international studies, involving up to 81 countries, and for two, the country of origin is unknown (Table 1). The studies were conducted between December 2019 and December 2020, with the majority (n = 35) carried out between March and August 2020. The data collection methods were semi-structured interviews (21 studies) mainly conducted remotely via telephone, Zoom or other such platforms, questionnaires with open-text response options (16 studies), personal narratives of lived experiences (six studies), focus group discussions and individual semi-structured interviews (two studies), unstructured interviews (two studies), and in the remaining three studies, a structured interview, a sharing circle and questionnaires combined with in-depth conversations were respectively used (Supplementary File 2). Data analysis involved thematic analysis in 22 studies, qualitative content analysis in nine studies, personal story telling in six studies, constant comparative method in two studies, Colaizzi’s seven-step content analysis in two studies, framework with thematic analysis in two studies, inductive process consistent with Grounded Theory in one study, the Attride-Sterling Framework in one study, immersion and crystallisation in one study, Giorgi’s four-step phenomenological approach in one study, and for the remaining three studies, although the data were thematically organised, the data analysis method was not explicitly described.

Of the 50 records, seven were not subjected to an assessment of methodological quality due to their study design (i.e., case report of lived experiences or study abstract) [33, 38, 54, 58, 60, 61]. The quality of the remaining 43 records ranged from four studies meeting all 12 quality assessment criteria to two studies meeting one quality criterion only. Twenty-five of the studies either fully (Y) or partially (P) met 11 of the 12 criteria because they did not meet criterion L, actively involving the participants in the design and conduct of the study. Table 2 presents the results of the quality assessment process.

The data were synthesised and presented separately by participant category, that is, pregnant and postpartum women and maternity care providers. The totality of the synthesis is represented by eight analytical themes of which five themes and six associated sub-themes represent women’s views and experiences, and three themes and four associated sub-themes represent maternity care providers’ views and experiences. Although there is overlap in some of the themes identified in the participant categories, the themes are presented separately so that the reader can interpret and consider the respective findings explicitly in the context of the participants from which these themes emerged. Table 3 and Table 4 respectively present these themes and the audit trail from the codes (condensed for illustrative purposive) to the descriptive themes and finally to the analytical themes. Codes denoted in bold represent ‘new’ codes from studies that were identified during our updated search. These codes were very few (n = 5 for women’s data and n = 3 for maternity care providers data), providing reassurance that the inclusion of additional studies is unlikely to alter the overall findings. Illustrative participant quotes from the included studies are presented to support the synthesised findings.

The theme of Altered maternity care (women) reflects how care changed for pregnant and postpartum women during the pandemic and the impact these changes had on their experiences of care. Thirty-one of the included studies contributed data to this theme. The two sub-themes of Altered care structures, processes, provision, and access, and Telehealth represent this analytical theme.

The pandemic resulted in considerable changes in how maternity care was structured, which affected care provision and access to care. Examples of these changes included: women attending clinics for care on their own, care transferred from in-person to virtual (sub-theme Telehealth), reduced choice for childbirth, inconsistent care, altered continuity of care (reduced and enhanced), schedules of fewer, postponed or cancelled antenatal or postnatal appointments, and altered maternity pathways based on COVID-19 test results [2, 35, 37‐41, 44‐47, 49‐51, 55, 57, 58, 79]. For some women, the changes to maternity care provision were viewed positively. For example, antenatal care for some women involved a ‘…less busy waiting room’ [40, p.2019] and was reportedly more ‘streamlined, with reduced waiting times’ [2, p.6]. Reduced ‘inefficiencies’ were also noted, such as eliminating low-value visits, although this change was recognised as better serving women who had low-risk pregnancy [79]. In one study, where alterations to midwifery care were reportedly abrupt and continuous, most women felt they had received information on these changes that was ‘clear and timely’, which helped them adapt to and cope with the changes [44]. Overall, however, the alterations to maternity care were unsettling for women, causing increased stress, anxiety, worry, uncertainty, or dissatisfaction [38, 40, 41, 46, 49, 50, 52, 54‐57]. Care provision for many women felt rushed, with limited time available to talk to maternity care providers or for normal checks, such as assessing blood pressure and performing scans, which were now not being done. As a result, women were left feeling anxious, overwhelmed, unsupported, or concerned [33, 37, 39, 46, 47, 52, 54, 55].

“Everything felt very rushed…. Nobody spent more than 10 minutes with me…. The entire time (in the hospital) I just felt rushed and alone.” [37, p.8]

Cancelled or postponed maternity care appointments, arising as a direct result of the pandemic, were commonly experienced by women [2, 34, 35, 43, 45‐47, 49, 52]. This resulted in women feeling confused, worried, fearful, and abandoned [2, 39, 49‐51]. Uncertainty and the ‘not knowing’ surrounding maternity care was also a considerable source of stress and anxiety for many women [2, 34, 39, 42, 44, 49, 56], resulting in some women entering the hospital for labour and childbirth already ‘tired’, ‘stressed’, ‘fearful’, and feeling ‘disillusioned’ [79]. Increased medicalisation of childbirth because of COVID-19 was also a concern for women. For example, having limited or no access to birthing pools as a birth option [39, 41, 47], ‘having alleged procedures (e.g., epidurals) forced on them’ or not being able to ‘have a normal delivery’ [39, p.25], and the ‘risk of a C-Section’ if diagnosed positive for COVID-19 [47] featured in women’s narratives. Pressures to be induced, and early discharge from the hospital after childbirth due to the pandemic were also cited as concerns for some women [37].

Altered systems of maternity care were identified in all studies contributing to this sub-theme, irrespective of country and birth setting, however, the extent of the changes appeared varied. For example, in Meaney’s international study [49], with participants mainly from the USA, UK, and Ireland (84%), restrictions on access to care varied across regions, and between hospitals within the same region, with some hospitals imposing less restrictions than others. These variations and inconsistencies in maternity care left women feeling frustrated and dissatisfied [37, 49, 79]. Continuity of care was also affected by the pandemic resulting in care continuity that was reduced, disjointed or non-existent [46]. Contrastingly, continuity of care in some regions was enhanced, especially between community and hospital settings [58]. Innovative and person-focused services which enabled home birth and continuity of services throughout the pandemic highlighted that supporting choice during COVID-19 was a possibility [41].

The sub-theme of Telehealth reflects women’s views and experiences of virtual rather than in-person maternity care during COVID-19. Telehealth was noted by women to confer some benefits; for example, avoiding travel time to the hospital or clinic for antenatal or postnatal appointments, overcoming long waiting times in clinics, being in the comfort of their homes and minimising exposure to risk of COVID-19 infection [52, 55, 74, 79]. The overarching narrative, however, was that telehealth was problematic for women and was favoured less than in-person care. Women were concerned that important information would be missed during telehealth consultations, especially regarding pregnancy or postpartum complications [33, 42, 45, 47, 52, 54, 55, 74]. This was especially evident in women who had pregnancy health or childbirth associated issues, such as hypertension, gestational diabetes, or perineal or caesarean wounds, and in women who had a previous stillbirth or miscarriage.

“And this telehealth situation, this monitoring from home, that’s a joke. It’s not going to work. How can you tell me that my C-section isn’t hurting when I’m telling you that it is hurting but you can’t see it” [74, p.4]

Online consultations for breastfeeding were described as ‘awkward’ and ineffective [2, 55, 75], and communication with maternity care providers was hampered as women found it difficult to develop a rapport during telehealth conversations [2, 45, 47, 74].

“And over the phone just doesn’t do it like. You don’t get the same, to look into somebody’s eyes and to trust them and for them to say, you’re okay” [2, p.14]

Women described feeling ‘embarrassed to talk about mental health concerns over the phone’ [47, p.3] such that telehealth was deemed inappropriate by women for discussing sensitive health issues. It also left some women feeling unprepared for birth, as in-person support mechanisms such as childbirth and parenting education classes were moved to the virtual space or were cancelled altogether [34, 74]. Telehealth also reduced the connection with maternity care providers that women considered important in pregnancy and postpartum care. The lack of in-person assessments resulted in many women feeling isolated, frightened, and anxious, and it led to mistrust amongst women; what women needed were in-person reassurances from their maternity care providers rather than virtual care that was largely perceived as inadequate or unsatisfactory [2, 34, 39, 46‐48, 52, 74].

Eighteen studies contributed data on women’s views and experiences of pandemic imposed restrictions. The synthesis of these data is represented in two sub-themes; ‘It felt cruel’—restricting partners attendance and Restrictions in general: pros and cons.

Sub-theme 2.1: ‘It felt cruel’ – restricting partners attendance

The degrees to which COVID-related restrictions were imposed on women’s partners or primary support person varied. For example, for some women, partners could remain throughout the birth and postpartum periods [52, 55], or throughout the birth and for a short period afterwards [38, 58]. For others, partner presence was not permitted at all during labour and childbirth [36, 44], or permitted only when women were deemed to be in ‘active’ labour [2, 76]. In almost all studies, however, participants indicated that partners were prohibited from attending antenatal appointments or routine postnatal follow-up visits [34, 40, 42, 43, 46, 47, 58]. Restrictions on partner attendance throughout the maternity care continuum evoked a wide array of emotions for women. These emotions ranged from feelings of guilt [38, 44, 55], anger [38, 49], emptiness [44], sadness [2, 44, 46, 49, 58], bitterness [44], anxiety or stress [39, 42, 46, 49, 55], fear [39, 52], worry or concern [38, 39, 43, 47, 52, 58], and disappointment [52]. Significantly, women expressed intense feelings of being alone, isolated, and lonely because of the restrictions [2, 37‐39, 56, 76]. It was clear that women had a strong desire or ‘needed’ to have their partner present throughout the maternity care experience, even for those simple supportive and reassuring gestures “…. like, to hold your hand, or to tell you that it would be ok” [2, p. 15]. Imposed partner separation at the time of birth was extremely distressing for women and resulted in a labour and birth that felt ‘unfulfilled’ [44]. Concerningly, restrictions on partner attendance evoked intense emotions which could potentially have lasting effects:

“I’m so angry that neither I, nor [name], will ever get that day back. I will never be able to correct it or make it a better experience … it felt cruel [38, p.1]

“… denying my husband, the right to be there, or me the support he provides is a disgusting standard of care which will have lifelong effects” [39, p.6]

This sub-theme reflects women’s experiences and views of restrictions beyond those related to restrictions on partner attendance. Restrictions, based on hospital policy, because of being positive for COVID-19 or restrictions imposed when babies were admitted to the neonatal intensive care unit (NICU) appeared especially harrowing and distressing for women [36, 37, 44, 45, 58, 75]. Being separated from their baby left women feeling like they had done something wrong [75] or that they had ‘abandoned’ their baby [45]. New visiting rules meant parents were only allowed to visit their baby for a very short period during the day [58] or not at all [45]. This separation resulted in women being unable to touch, feel, cuddle, or smell their babies, and, in recounting their experience, women’s voices ‘trembled with tears alternating to silences’ [44, p.8]. Isolation and separation from friends and the wider family also affected women, albeit in various ways. Women expressed disappointment that they were not able to engage in traditional pregnancy rituals, share their pregnancy journey with family and friends or celebrate their pregnancy and birth with others, resulting in feelings of missing out or of loneliness [34, 42, 49‐51, 55].

“It’s made it definitely a more somber experience and it has been difficult to be excited because you can’t share it with people.” [42, p.5]

Attending or being in hospital alone without visits from friends and family was difficult for many women [37, 38, 40, 41, 43, 44, 56]. The separation and isolation from wider support networks resulted in women experiencing fear [36, 47] or feeling ‘cut-off’ [41], without freedom to move around [49] as ‘you are in a room, you can’t go out and you’ve got to stay between those four walls’ [44, p.8]. This isolation and loneliness continued for many women into the postnatal period, especially as women were denied opportunities to introduce their new baby to loved ones. Lack of interactions with family and friends following childbirth affected women’s mood negatively. Women recounted feeling ‘overwhelmed’ or anxious without having the help from family and friends in caring for their baby that they would otherwise have had [51, 55, 56, 75]. Contrastingly, the wider visiting restrictions in hospital beyond partner visiting, and when women returned home, were a positive experience for some women. Women attributed reduced visiting in the postnatal ward as providing extra space and time to bond with their babies [2, 40]. Quieter postnatal wards facilitated a private space for women to establish breastfeeding more comfortably [2, 40] and women drew comfort from the ‘peace and quietness’ offered by less crowded postnatal wards.

“It is a lot quieter, more time to adjust and try to get a hang of breastfeeding without an audience” [40, p.220]

Visiting restrictions and being isolated encouraged women to form close relationships with other women in a similar situation [2, 44]. On returning home from the hospital, because they could ‘politely decline people coming over’ [56, p.3], this enabled some women to enjoy quiet, precious time as a family unit.

Navigating information and psychosocial support during the pandemic appeared especially challenging for women. Women experienced navigation difficulties for various reasons and due to different causes. Information support, for example, was affected by a lack of consistent messaging, conflicting information or a lack of clear guidance surrounding the virus and how this affected women’s care, both during pregnancy and postnatally [37, 39, 43, 46, 49, 51, 53, 56].

“One doctor would say one thing and then the next would say another.” [37, p.6]

This resulted in women feeling lost, confused, or helpless [39, 46, 55]. Having trustworthy information became a key concern for women and many women recounted a need for further information [36, 42, 47, 49, 52]. In navigating their information needs and support, many women resorted to alternative sources, mainly social media, television, and online sources, as well as friends and, in some cases, government sources [34, 38‐40, 44, 47, 50, 52, 53, 55, 56]. The use of these alternative sources was, however, problematic as women recognised that such sources can be unreliable or were a causal source of stress and fear [44, 47, 50, 53, 56]. Although some women were sent information by their maternity care providers, many women recounted a desire for greater levels of official communication from the hospital or their care providers:

“…and I think probably one thing that maybe could be improved is just that extra information of what you are doing with the COVID stuff in terms of precautions, what it's going to look like when I come in to have bubs, just what to expect.” [34, p.6]

Navigating psychosocial support also featured considerably in women’s narratives. Although some women recounted receiving good support from maternity care providers [40, 55, 58], for many women this type of support was significantly diminished, leaving women finding their own way ‘as there was nobody else to help’ [38, p.1]. Furthermore, because of reduced psychosocial support, many women were left feeling concerned that mental health problems would go unnoticed or left not knowing where to seek support should problems arise for them [2, 47, 52]. Women viewed dedicated formal support from maternity care providers as essential for their psychosocial wellbeing. When this was lacking or diminished, many women turned to and relied heavily on informal supports such as family and friends as a substitute [2, 34, 36, 44, 49]. Concurrently, however, informal supports such as peer, family, and other social supports (e.g., mother and baby groups), were no longer available for many women [2, 34, 37, 41, 42, 44, 45, 49, 51, 55, 56, 74] leaving women with limited or no support at all. Lack of breastfeeding support, especially in-person support, also featured heavily in women’s narratives [2, 47, 52, 54, 55]. This left many women having to work through issues on their own which was a source of disappointment for them:

“I was struggling breastfeeding. I would have gone to breastfeeding group, but that’s been cancelled…. I was in pain and I felt let down” [52, p. 20]

Women, in navigating their maternity experience during the pandemic, self-implemented solutions as a means of coping, including adjusting their plans or exploring other options for care [39, 47, 56]. Many women demonstrated resilience by trying to ‘forget’ the past and build a new normal for themselves [44], or by trusting their own judgement and instincts [51]. Women were proactive in preparing themselves for birth, with some describing how they actively sought out online classes in the absence of formal professional supports [34, 41]. Some women went to extreme measures to ensure they were prepared for all eventualities such as buying their “own IV fluids and cannulas and respiratory equipment, so that if the baby wasn’t breathing then we could do something about it” [56]. Women also compared themselves to other women, and drew comfort from each other:

“And you know if you were kind of just worried, but you were able to talk to each other. And just comfort each other.” [2, p.18]

Women spoke of advocating for themselves to achieve the maternity care they desired or needed, amidst the constant changes occurring within maternity systems [2, 37, 39, 46, 56]. For example, some women took the decision to book for an induction of labour or elective caesarean to minimise uncertainty and gain ‘some control over a situation that was uncontrollable’ [2, p.17]. While women described not wishing to ‘fight’ for what they required, many felt that they had no choice but to do so [39, 46, 56].

“I am forced to continually fight to be seen and have to reiterate my situation and reasoning over and over …. and now I have no choice but to advocate for myself but it has been very difficult.” [46, p.5]

Other women, alternatively, found themselves simply accepting of and being adaptive to the present situation which helped them feel prepared [2, 44, 51, 76].

The theme of Altered maternity care represents narratives from maternity care providers on how the provision of care substantially changed during the pandemic and the impact that this had on their capacity to provide appropriate and effective care. Twenty-two studies contributed data to this theme. The two sub-themes of Altered care structures and provision and Capacity to provide care were identified within this analytical theme.

Sub-theme 6.1: Altered care structures and provision.

Across all settings, the provision of care had substantially changed for maternity care providers. There was a focus on reducing in-person appointments with a move to virtual or telephone appointments, where possible, which required maternity care providers to be innovative and adaptive in identifying alternative ways to provide care [17, 64, 66, 71, 72]. Maternity care providers also described a reduction in the numbers of women accessing certain types of maternity services (e.g., inpatient antenatal care, postnatal clinics, and infant immunisation appointments) due to concerns about attending in-person [72, 78], while, concurrently, the demand for midwifery care at home and homebirth in some settings had increased [59, 69, 71, 72]. A feeling of uncertainty was dominant in maternity care providers’ narratives, largely influenced by the rapid speed with which care protocols were changing [17, 60‐63, 65, 68, 70, 72, 74]. Constant change and inconsistencies across settings often led to confusion and differences in interpretation [66, 70, 73]. Some expressed that this may have negatively influenced the care that was provided as it was unclear if the new care protocols were sufficiently evidence-based [69, 70, 76].

“Departmental protocols…were changing rapidly, leading to confusion and unclear interpretation by staff members… variation in practice and misinterpretation of guidance were expressed…especially where limited evidence is available” [70, p.1125].

Many maternity care providers, however, acknowledged that such uncertainty lessened over time as clear national guidelines became established and implemented, and the communication around care protocols improved [17, 62, 65, 66, 73].

Many maternity care providers also believed that the pandemic exacerbated existing inequalities in maternity care. The closing of some services or moving maternity appointments to virtual or tele settings were viewed as having a greater impact on at-risk communities as this group were less likely to be able to access these types of alternative services [61, 64, 67, 69, 71, 72]. Furthermore, some maternity care providers held the view that other providers were instigating racist or sexist practices based on inappropriate or misconstrued beliefs around the risk of COVID-19 in certain population groups, subsequently exacerbating the existing challenges that pregnant or postpartum women may already be facing [61, 68, 74].

“Restrictions and regulations in the time of COVID-19 have allowed for a resurgence of the racist and sexist policies…Black women’s bodies have continued to be seen as risky…leading to a lack of care and touch that continues to put Black birthing people in danger” [68, p7).

Some providers also held a belief that other maternity care providers were using pandemic associated changes as an excuse for inadequate care [69, 75].

“COVID has also given practitioners justifications for many unnecessary and excessive practices; when negligence is not the issue, increased intervention is” [69, p3].

Some practice changes, however, were viewed as having a positive impact on care due to fewer women attending services which increased the time available to spend with each woman and which facilitated a more efficient service [73]. Despite the pandemic and the associated changes to care structures and provision, maternity care providers were emphatic in describing that the need for maternity care did not stop as “women’s health care needs …could not be put on pause” [15, p.3]; something which was considered unique to maternity care provision [15, 60, 62, 63, 69].

Most maternity care providers felt that the changes due to COVID-19 had resulted in a reduced capacity to provide safe and effective care. A lack of access to adequate resources, such as PPE and training on safe practices, left maternity care providers feeling that they were not adequately protected [17, 63, 65, 66, 70, 74, 77]. As a result, providers limited their interactions with women during pregnancy or during the birthing process as they feared being infected and/or acting as a vector of infection [65, 75, 77].

“To decrease the risk of transmission, we usually compromise the routine antenatal care service. For instance, we may not perform physical examination or draw blood, even if necessary” [77, p.4]

COVID-19 related restrictions, such as reduced in-person appointments, cancelled support groups, and reduced numbers of maternity care providers in the birthing suite, were viewed as barriers to care [17, 59, 64, 68, 69, 71, 74, 75]. Personal contact was viewed as an essential part of maternity care, especially by midwives. Having to limit this element of their care was perceived as having a negative impact on their capacity to provide care, with some viewing it as a “dehumanization of childbirth” [65, p4]; especially for certain activities such as lactation support, where personal contact was considered critically important.

“Virtually the screen is small, I’m at the mercy of the person holding the phone…I have to verbally direct the mom over the phone, and many interrupt[ion]s on both sides of the conversations” [71, p.265].

A move to telehealth was viewed positively by some as it enabled the continuation of care in a safe environment and allowed for a more responsive approach to care in some circumstances. However, telehealth was described as having limitations. Due to the lack of personal contact, it was viewed as hampering the ability to build a trusting relationship with women which impacted on providing effective care [64, 68, 69, 71].

“Over the telephone, it is harder to read all the non-verbal cues as you would in an in-person counselling session” [71, p.265]

Furthermore, language barriers and insufficient access to digital resources or a lack of digital literacy sometimes hindered the provision of care in this mode. This was an issue especially for maternity care providers in low- and middle-income countries, or for disadvantaged populations in high income countries [64, 67, 71].

“One of the biggest challenges reported was poor internet connection and/or regular interruptions in connectivity. This was a global problem reported by providers from both LMICs and HICs” [64, p.8]

Professional impact was predominantly negative for maternity care providers. Staff shortages [65, 73], additional tasks due to new care practices [62, 66, 68, 69], and longer and more frequent appointments as pregnant and postpartum women required more interaction due to anxiety and uncertainty, all contributed to an increased workload [17, 59, 64, 68, 73]. For maternity care providers working outside the traditional hospital system, such as independent community midwives, their workload substantially increased as the demand for a non-hospital birth experience increased [17]. There was one ‘lone voice’, however, that expressed a more positive experience in relation to workload whereby previous practices such as ‘overbooking of inductions’ ceased and ‘caesarean section lists reduced’ as part of pandemic efforts to strictly manage numbers in the hospital [70]. A further ‘silver lining’ described by maternity care providers from the experience of working on the frontline during the pandemic was ‘the bond’ it created amongst providers as they worked together [15, 61, 70]. The uncertainty and chaos introduced by the pandemic helped to build relationships as maternity care providers supported one another in providing care during these unprecedented times [61, 62, 66, 74]. In some ways, life in a maternity care setting acted as a bubble to the outside world where it was perceived that people were cut off from each other to a greater extent, and the hospital was associated with a greater sense of calm and camaraderie [59, 60].

“It was ‘on the frontlines’ that I felt the most the distant from the pandemic itself. I felt guilty responding to messages from family and old friends, those not in medicine but trapped in their homes by an invisible enemy ravaging their cities and towns” [60, p.2]

Contrastingly, in some settings, divisions emerged as particular professions pitted themselves against others, for example, midwives and doctors [65, 69, 74, 76].

“The pandemic has impacted the medical culture in Puerto Rico, emboldening doctors to ‘protect’ their ‘domain’… medical professionals have taken to social media…gone on the news to argue that the absolute safest place to give birth is the hospital” [69, p.5]

Different professional beliefs or perceptions about status in the ‘hospital hierarchy’ also contributed to these divides. Altman and colleagues described how nurses working in perinatal settings felt that they were expected to take additional risks to those of other staff members, including consultants [74]. This contributed to a feeling that nurses were expendable and not valued for their contribution to care. There was an evident divide between frontline maternity care providers and management. Inadequate communication about changing protocols and a perceived lack of consideration of staff needs resulted in many care providers feeling abandoned by management [59, 63, 69, 72, 74, 76, 77]. This added to the feeling of uncertainty and was an additional emotional burden to burdens that were already present.

“Nurse participants described ...wanting more compassion and respect from hospital administration…a need to be seen as an individual who is being placed at risk” [74, p.6]

The pandemic presented a personal burden for all maternity care providers “including issues that affected personal health and well-being; challenges with family and parenting; and mental health concerns, stress, finances and loss of income. This was universal regardless of personal circumstances, type of practice, years in practice or geographical area” [73, p.4]. COVID-19 was considered a significant threat, and due to a fear of acting as a vector, to either patients or family members, maternity care providers isolated themselves from or interacted less with their families [17, 59, 61, 65, 66, 68, 69, 72, 73, 77, 78]. This carried a significant emotional burden as they dealt with the uncertainty and anxiety of the pandemic in isolation [63, 66, 68, 73]. This self-imposed isolation led to concerns about the wellbeing of dependent family members where maternity care providers questioned their parenting capacity; this further added to their experience of emotional burden, a burden which was further compounded by the feeling of abandonment by management [69, 74]. Maternity care providers described making substantial sacrifices to ensure the continuation of care but felt that little recognition was being given for this [62, 74]. Some maternity care providers also spoke about the financial burden that they were experiencing because of the pandemic, which added to existing emotional burden [64, 68, 73]. As the demand for certain services reduced, this had a negative impact for those working independently, such as community midwives, resulting in reduced income which was worrisome. Others spoke about needing to adapt their mode of care, such as moving to telehealth, but not being adequately reimbursed.

“Respondents themselves faced financial burdens from the use of telemedicine on two levels: not being able to afford the equipment and lack of reimbursement…for costs they incurred while providing telemedicine (including the telehealth consultation itself and its associated internet/phone/data costs)” [64, p.9]

The strength of this QES lies in the broad approach of including both women’s and maternity care providers’ views. This has enabled perspectives from both populations to be considered and compared in gaining insight and in understanding maternity care experiences in these unusual times. The rapidity with which studies on this topic were being published, however, could introduce a potential limitation for our findings. Although we updated our search prior to our synthesis, there is the likelihood that further studies will have since become available that meet the inclusion criteria for this QES. Reassuringly though, our updated search revealed only eight additional codes in total, from those already developed. In this sense, further studies are unlikely to significantly alter the findings of the QES. We accept however, that how maternity care has been experienced as the pandemic continued may have temporally changed, especially in the wake of the vaccination programme which may have helped reduce fears, lessened restrictions or impacted on how care was structured or provided. To assess for temporal trends, an update or, ideally, a second review from the time of our last search would be beneficial. A follow-on review would allow findings based on the initial year of the pandemic and later years to be compared or contrasted. We also acknowledge that maternity care experienced by women can be influenced by their partner’s feelings and experiences or by those of their wider family or circle of friends. To advance the evidence overall on the experience of maternity care, we plan to conduct a QES on partners’ experiences of maternity care during COVID-19 and explore further complementarity. A further limitation to our QES was the inclusion of studies in the English language only. To assess the potential for language bias, however, we unrestricted our search based on language. Seven records were retrieved that potentially might have meet the review’s inclusion criteria but were excluded as they were non-English language publications (see https://​osf.​io/​bzt38/​ for the list of excluded studies with reasons). Given the extent of data contributing to the QES, however, we feel reassured that the potential for language bias based on the exclusion of these records is minimal.