Case study Trust selection and data collection
The two NHS organizations are part of seven NHS maternity providers within England included in the wider ASPIRE COVID-19 project. Unit A is located in the north of England. It manages around 4,500 births each year across three sites. It services a mixed socio-economic population, with some areas among the top 25 most deprived in the country, while others are much more socially advantaged. Less than 1% of the population are from ethnic minorities. Trust B is in the Midlands and manages around 2,500 -3,000 births a year. It is an area of low deprivation with around 7% of the population from ethnic minorities. Both organizations were chosen for the case study because they were able to provide at least some timely and high-quality clinical and service-related quantitative data (there were difficulties accessing quantitative data for a number of the other Trusts). These two trusts represent geographic, and to some extent socio-economic diversity, but we acknowledge these are not representative of UK maternity providers as a whole.
The case studies draw on both quantitative and qualitative data from a range of sources. Trusts were requested to provide data on key quantitative indicators from routinely collected data. As there is no agreed list of variables to measure safe and personalised care, a list was identified through an initial analysis of key documents, in particular the NHS England Better Births Report which introduced these concepts into NHS policy (8). These were then revised through discussion with the wider ASPIRE research team and stakeholder group (see Acknowledgements for further details). They were then reviewed with the research staff taking part in ASPIRE COVID-19 at Trust level to establish which indicators were accessible and available (see appendix 1). Other quantitative data were taken from the Family and Friends test (9), Safe Staffing data (10)UK Government data on COVID-19 incidence (11) and NHS data on hospital admissions (12). The data were requested from the beginning of 2018 to allow an understanding of pre-pandemic trends.
For the qualitative component, 30 interview transcripts (half from each Trust) from service users and staff were randomly sampled from 96 in-depth interviews undertaken in the included Trusts as part of the wider ASPIRE COVID-19 study between March 2021 and October 2021. These were not sampled based on data saturation, as we were not undertaking an inductive analysis, but looking for processes and experiences that could explain components in the ASPIRE model or direct the quantitative analysis, or that might contextualise the quantitative findings.
Full details of the methodology for the qualitative data collection can be found in Appendix 2. Of these transcripts, 17 were from service users and 13 were from staff. Transcripts were analysed to gain insight into experiences and perspectives relating to the changes and impacts of the pandemic, for both staff and service users. Additionally, we gathered documentary evidence from Board Reports and Trust communications outlining changes in policies and practice.
Use of a case study methodology
Case studies offer the opportunity to explore a phenomenon, policy, or situation from a range of perspectives. They aim to offer in-depth insights on complex situations. These characteristics make case studies popular for health systems research, which requires the analysis of multiple factors from the viewpoints of different stakeholders which can be placed over time within a changing context (13,14).
We used a mixed-methods approach drawing on qualitative and quantitative data to triangulate our findings and provide greater depth and scope; total reliance on qualitative material would mean there was no possibility to include objective measures of change in service delivery and outcomes, whereas quantitative data alone would lead to areas being missed where no indicators were available, and an inability to explore the drivers and pathways of change and impacts on stakeholders. Our approach is essentially descriptive, with the purpose to describe a phenomenon in its real-world context (15). Adequate integration is key to an effective mixed-methods case study, and we adopted a convergent design where the qualitative and quantitative data were extracted and analysed during a similar timeframe (although the qualitative data had been collected earlier). Based on Stake’s definitions of case study purpose we suggest that our studies are defined as instrumental in that we use them to gain a broader appreciation of the impact of COVID-19 on the delivery of maternity care by Trusts in England, but there is also a collective element as we compare data from both Trusts to provide wider insights (16).
Applying a conceptual framework for data analysis
The analysis was structured around the ASPIRE Trust Level Conceptual Framework (see Figure 1). This framework was developed from existing literature on the impact of emergency health care delivery and with reference to ongoing qualitative work from interviews and documentary reviews in earlier phases of the ASPIRE project. It suggests a starting point of the pre-pandemic baseline, and then a flow from inputs through to processes and health and wellbeing outcomes for both staff and service users. Directional arrows suggest paths of influence, and the possible influence of policy changes on demand, processes and outcomes are included. It is designed to be adapted to other future crises, or for analysis and development of routine maternity and neonatal care. Use of a framework provides a clear structure for comparing the two Trusts. Each component of the framework was expanded and illustrated with Trust-level data from the two case studies.
Quantitative variables and findings from the qualitative interviews and documents were mapped against the conceptual framework. Quantitative data were analysed over time using appropriate graphic methods, and notable trends were recorded. Qualitative and documentary data sources were used to triangulate trends and identify information gaps, as well as identify discord and concurrence. The conceptual framework was adapted as a result of our findings to reflect the process.
Findings
Our findings are structured under sub-headings that relate to components of the framework in figure 1.
Inputs: demand and need for maternity services in case study Trusts
A crisis can directly change the level of need by changing the number of service users, intensity or type of services required. In the case of COVID-19, this may be either through the need to care for COVID-positive women within the maternity health care system or through the need for protective measures to prevent the spread of infection.
Direct changes in need: COVID-19 incidence and admissions
An examination of monthly bookings and births, as well as the booking profiles by parity and decile, found there was no evidence of any change in demand for services occurring in either provider. We were unable to acquire information on the number of women testing positive for COVID-19 within the maternity health system, but analysis of regional incidence data and Trust admission data enabled us to map the waves of the pandemic for both Trusts (see figure 2). At the beginning of the pandemic, cases and admissions were fairly low in the locality of both case studies (although cases are subject to under-reporting because of a lack of testing services). After the first wave, there were then two spikes in cases, between August 2020 and February 2021 with a concurrent increase in COVID-19 admissions that peaked in December – January 2021. Following a dip in cases in March and April 2021, cases then increased again and remained high up to the end of September 2021, but in this wave hospital admissions did not rise so markedly. These data give some indication of when stresses are likely to be greatest on the health system, both in terms of cases, which will affect both the need for measures to reduce transmission and possible staff sickness and admissions that will increase demand on the hospital infrastructure.
Indirect need: mental health concerns
A crisis can also produce indirect changes to the need for services. For instance, during the COVID-19 pandemic, there was a documented increase in intimate partner violence (17) and a risk of increased mental health concerns (18). Both Trusts acknowledged these increased risks in communications and guidelines provided to staff, but there were no data to track the incidence of domestic violence. Data were, however, available for both Trusts related to mental health concerns at booking, and referral at any time during pregnancy to mental health services (see figures 3 and 4). Trust A shows a marked and consistent increase in reported mental health concerns (from a very low baseline) to a height of around 50% at the end of 2020. Referrals to mental health services at any time during the pandemic also increased from mid-2019. Trust B however, demonstrated no change in reported mental health concerns, and a decline in referrals.
As it is unlikely there will be such marked differences between regions in mental health and wellbeing, it is likely that to some extent patterns reflect the way midwives identify women with mental health concerns and the availability of specialist referral facilities. The increased identification and referral in Trust A could be due to strong perinatal mental health services in Trust A, which appears to have prioritized this area. Maternal anxiety was described in the qualitative data for both Trusts. Trust B did acknowledge the issue since they introduced a system through which inpatients could indicate if they wished to talk about anxieties or concerns.
Inputs: Resources
Staffing
COVID saw rises in staff sickness and self-isolation and redeployment. Data from the safe staffing datasets (which provides information on actual staff hours worked versus those estimated as required) for both Trusts showed that before the pandemic in Trust A there was already a staff shortfall for registered midwives, while for Trust B this was much less marked. Both Trusts saw an increase in staffing for registered midwives in the early phase of the pandemic (see Figure 5), which may reflect the national drive to increase the NHS workforce at that point. In Trust A, the increased staffing levels continued until around the end of 2020 (although the increase was inconsistent), but then reverted to pre-pandemic levels before summer 2021, subsequently falling further.
In Trust B there was a shorter period of increased staffing (March-June 2020) before returning to pre-pandemic levels, and then figures showed a marked decrease from June 2021 onwards. Data for non-registered staff (e.g., support workers) was more inconsistent, particularly for Trust B (possibly due to small numbers not shown), but Trust A again showed a marked decrease in staffing from around June 2021. No data were available on trends in other health professionals.
Although the quantitative data suggest a more concerning situation in Trust A, the qualitative data from both Trusts highlighted the negative impact of staff shortages. Several responses from Trust A acknowledged that staffing was a long-term problem pre-pandemic and not specific to the COVID-19 situation:
I think, to be quite honest, the pandemic's come at a time when we were probably at a national shortage of midwives. We had stress before. Eighteen months ago, we were short of staff. So, you know, so even probably before the pandemic hit, I think we were short of staff. So, it just cemented it even more, really, I think the unit was run on goodwill, and I think to a degree, it still is. Health care provider, Trust A.
While Trust B’s staffing figures may not have looked too concerning until mid-2021, the qualitative evidence suggests that staffing levels were maintained at some individual cost to health care providers who felt morally obliged to work more hours than they wanted, with detrimental effects on their mental wellbeing:
“I can remember walking into work literally in tears thinking, what am I doing going in here? You know, I'm retired. I shouldn't be coming in. You know, I'm bank [temporary staff]. I shouldn't really be working here. But I also knew from a, from the point of view of staffing, having been the coordinator, how you reliant on your bank, everybody to turn up…. So, I felt a moral obligation to come in and I do enjoy going to work but of recent I do feel very anxious because again, because of the staffing…” Health care provider, Trust B.
When asked what is needed to ensure safe and personalised maternity care in the future, most respondents said that the answer lies in safe staffing. Interview data indicate that women also noted the low staffing levels, particularly on the postnatal ward.
Access to resources
Beyond staffing levels, concern with excessive bureaucracy was reported as a way to free up time to care for front-line care providers and to create conditions for system agility e.g. making resources available without spending excessive time negotiating opaque organisational systems or being denied requested resources altogether. A reduction in bureaucracy early on in the pandemic was one of the few positive factors raised across several sites in the wider ASPIRE study, which meant they could rapidly access resources that had previously been tied up in ‘red tape’. Streamlined processes meant they could access equipment needed, quickly and easily:
“We bought, we used every resource we had. So, and actually, what was quite liberating at the beginning was everything was, you can have the resources there, you can have this, you can have it, whatever you need to preserve life, you can have…. And actually we have the permission to do that, how we always want to run the service, you know, and that was quite liberating. And, you know, you weren't questioned because it was like, if that's what you need.” Manager, Trust B.
Processes and outputs for safe and personalised care:
In terms of the framework, changes in demand/need and resources are likely to directly impact on the processes and outputs for safe and personalised care. However changes in policy and guidelines either at national or provider level may also, in turn, bring about more structured changes in delivery of care. Changes in processes and output are hypothesised to affect access, acceptability, quality and delivery of care across antenatal, intrapartum and PN care. Our case studies produce a wealth of information on a range of components of care in line with the framework. Here, we focus on summarizing key findings that illustrate the pathways to change.
Processes of ANC
In the case of ANC, both Trusts implemented policies to carry out some contacts remotely. Neither seemed to set a particular target for the proportion that should be remote, although Trust B produced advice about which visits could be remote and Trust A published public-facing guidance saying appointments would be remote where possible. Available data suggest that the percentage of appointments delivered remotely was low in both Trusts: the highest proportion of remote contacts in any month was 12% in Trust A and around 10% for Trust B.
Possibly more importantly than the way services were accessed, the percentage of women who received fewer than six ANC contacts (the minimum recommended by the Royal College of Midwives) increased in both Trusts, with a revised and reduced schedule of ANC introduced in Trust B (although still with a recommended minimum of over six visits).
As the pandemic progressed, in Trust A the patterns show an increase in the percentage of women experiencing less than six ANC contacts in the months after the start of the first wave (see figure 6). This percentage then decreased, whereas in Trust B (which had a much higher percentage of women receiving less than six ANC contacts than Trust A) there is a steady increase in the percentage of women receiving less than 6 ANC contacts before the start of the pandemic, followed by a sharp increase in the percentage of women receiving a lower number of ANC contacts from September 2021. This suggests that the number of AN contacts may have been reducing in Trust B before the pandemic. Changes in Trust B may be less related to changes made in service delivery as initial responses to the pandemic, but the ongoing crisis may have exacerbated a situation already in existence. There was no clear pattern of change for gestational age at first booking for either Trust.
It should be highlighted that here the routinely collected quantitative data contrasts with the experiences and perspectives of service users and staff. The interview data suggest that many women experienced remote appointments, particularly at the beginning of the pandemic, and that some staff were significantly concerned about the impact of this on both the psychological and physical health of service users. Interviews with both staff and service users in the Trusts indicated a feeling that this potentially reduced the quality of service and raised levels of anxiety, and, along with fewer AN contacts, raised concerns about problems not being adequately detected.
All of it is over telephone conversation and it isn't unheard of, even at 16-weeks they still haven't seen their midwives and they're coming through at twenty-four, twenty-eight weeks, having not seen an actual midwife, only had a virtual contact. When you speak to the patient for them, for some of them who've had babies before they go, oh, it's very different, but for those new mums, first-time mums find it really difficult. They are quite anxious, the not knowing and because obviously it's all new, it makes it very difficult for them. Health care provider, Trust B.
I had less face-to-face appointments with midwives. It was telephone appointments or just a couple of face to face. I wasn't able to go and look around the hospital to see the labour ward or anything like that, all I could do was look on the website. But I would say it was more like, I couldn't really go and hear the baby's heartbeat. I think it was that, it was that lack of face-to-face contact with the midwife team that I found the saddest. Service user, Trust B.
Home births
In terms of home birth, the policy responses of the two Trusts played out differently. In both Trusts overall numbers of home births were low, but in Trust A the initial response was to suspend home births for several months. However, they were reinstated after pressure from service users:
I think if I remember correctly, we stopped the home birth. We stopped the homebirth. And then there was a serious objection came from the patients. And then our senior management was questioning that decision because from a PPE, you know, thinking from a patient point of view, what they said is that home is safer than your hospital, so we want to deliver at home. So, if I remember correctly, I think we went back and said, it is OK, we will do it. But the initial reaction was to stop it. Manager, Trust A.
Trust B, however, actively promoted home births as safer in terms of infection control. As a result, the Trust provided extra training and support to enable the increased and more seamless provision of their home birth service.
And we were saying, right, let's suspend the home birth service…. suddenly our head of service said, hang on a minute, we need to turn this on its head. We need to, where's the safest place for people to be at home, so we need to keep them at home…... And so then we decided actually let's shift the priority. So let's... let's … because what we actually did was expand our home birth service at that point. Manager, Trust A
Quantitative trends are based on small numbers, but the data do demonstrate a cessation of home births for several months in Trust A followed by a modest increase, in contrast to a sharp increase from the outset of the pandemic in Trust B followed by a gradual decline (see figure 7). These differences occurred under the same national guidance, which recommended continuing all options for place of birth unless specific criteria of low staff or lack of availability of ambulance services were met (19).
Induction of labour
Another example of diversity between the two Trusts was the process of induction of labour. Early in the pandemic the Royal College of Obstetricians and Gynaecologists issued guidance to avoid induction of labour “for indications that are not strictly necessary” (20). Trust A showed a marked fall in inductions that may have been in response to this guidance, whereas Trust B shows a slight increase (see figure 8), which is supported by qualitative evidence from one interview suggesting some inclination to move births forward to protect vulnerable pregnant women from the increased risk they face from COVID-19.
Obviously, the third trimester was the point at which they were vulnerable. I wouldn't say I was inducing earlier, but I would say and I don't think I'm alone in this, but I think a lot of us had a mind-set that let's get them delivered and then they're no longer pregnant and their risk profile reduces, you know, because obviously the outcome, pregnant women on ICU were the worst. So, that was something I was very aware of, I wanted those women delivered. Health care provider, Trust B.
Postnatal care
Quantitative data for both Trusts showed a reduction in the length of time women spent in hospital after the birth, with a decrease in stays over 48 hours (see figure 9). This does not seem to be the result of a specific policy, but a combination of staff concerns over the risk of COVID-19 transmission and women discharging early, largely due to restrictions to partners being present.
The staff obviously were very anxious about visitors on the ward. And, you know, in a bay of six once you have got another six partners and six babies your air exchange has completely gone down, and it was very difficult. We were often stuck between a rock and a hard place, you know, and the women didn't want to stay in very long. So, we were worried about the impact of that you know, would the success of feeding be OK? You know, we kept thinking we were getting more wound infections, but we haven't actually, you know, seen that evidence. It felt like it felt like we were getting a lot of people coming back in postnatally with problems but, you know, on audit, it doesn't look like, you know, that was a correlation, although it felt like it. Manager, Trust B.
Women reported challenging experiences on the PN ward including reduced or restricted partner visits, a lack of support by staff and general confusion about what was going on in the ward.
If I could change anything, it would be once you'd have the baby. And I totally understand that with COVID, visitors can't stay too long. But forget the visitors. I just think the care, I would make it so that I felt more looked after, more reassured, helped more with feeding. And just have that whole, it's such a … nothing can prepare you for it… I thought "Oh, I know what to expect." You just don't and that there with this newborn life and you're on your own, pandemic or not, you want people that are the professionals. And not make you feel like you're an inconvenience and they haven't got time for you, which is, I don't think it's any fault of their own. I think maybe they're understaffed. But it wasn't a positive experience afterwards, and it could have been so easily better. Service user, Trust B.
The number of PN contacts reduced in both Trusts from a mean average of around six before the pandemic in Trust A to an average of 5.4 during the pandemic, whereas the median number of visits in Trust B fell from between six and seven pre-pandemic to five during the pandemic (although again there was evidence of reducing numbers pre-pandemic: see figure 10). In Trust A they rose somewhat after an initial sharp drop at the start of the pandemic but fell again in early 2021. It is worth mentioning that there was some uncertainty in what was meant by a “PN contact”: these numbers are greater than might be normally expected from community post-discharge contacts and suggest that contacts on the ward before discharge were also included which means these data should be interpreted with caution.
Readmission to hospital
Fears about safety were raised by staff along with concerns about readmission: while an audit did not suggest any increase in readmission in Trust B, an analysis of the incident reports for Trust A showed a potential increase. Neonatal readmissions increased from approximately five reported incidents per quarter just before the start of the pandemic to 13 incidents in quarter two of 2020, remaining at an average of nine throughout 2020 (see Appendix 3). This coincides with the first wave of the pandemic, as well as a fall in PN care during the pandemic, reaching a maximum of eight reported incidents, up from an average of three just before the pandemic. However, as there were also obvious sharp increases in readmissions in 2018 (particularly for babies), this situation cannot be attributed with any certainty to the pandemic per se.
Companionship and neonatal visiting
For many women and healthcare professionals, one of the most impactful changes to maternity services were the restrictions on partners and other visitors throughout the maternity care journey. This was as a result of national and local policies(21), but these tended to be interpreted somewhat variably, particularly during birth. No quantitative data were available on this, but qualitative data graphically illustrated how women had to attend antenatal appointments, including scans, alone, which reportedly left their partners anxious and distanced from the experience.
Things like the scans and as well having a partner just missing from he wasn't part of the birth, apart from showing up at the end, for a couple of hours. You know, he wasn't part of it, of the whole thing. But yeah, I think it was had a huge effect. So things like my 12 week scan, that's one thing that sticks heavily in my mind. I previously had miscarriages, so I've had two miscarriages in the past. My last time I've had a 12-week scan, I found out I was going to miscarry. So when I went to the scan and he wasn't allowed in and I just sat in tears with these complete strangers who were lovely and did their best. But they… they knew that they weren't what was actually needed in support terms. My partner was sat in the car park waiting for me. Service user, Trust A
For consultant appointments in particular, women felt that they would have benefited from having their partners' input in decision-making, and to help them take in and process information. Although women reported concerns about not having their partners with them for labour, they did have companionship during the birth itself (although it wasn’t clear at what stage of labour this became possible).
Views on the impact of PN restrictions on visitors showed a division between users and staff. Some service users felt vulnerable and isolated without their partner there to support them emotionally and physically with their new baby. However, some staff members felt that reduced visiting on the PN ward had been beneficial to women, in terms of breastfeeding and bonding, and for staff, because they had fewer additional matters to attend to.
One participant described the difficulties her family experienced when their baby had to be admitted to the Special Care Baby Unit (SCBU) immediately following birth, whereby partner restrictions compounded the trauma of being separated from her baby, not knowing what condition she was in.
They said, well, you've got to go home now, and I was like, hang on a second, I'm like, you know, I can't move, so I couldn't go be with her. She was all alone, just born, in SCBU, awful. You know, oxygen, everything all over the place and all tubes and everything. And then [name] had to leave immediately…….and he had to go home and we were just, I couldn't move. She was in SCBU like who knows what, we didn't know yet what had happened to her. So we didn't know she had [condition]. We didn't know if she was going to recover, if she would die, like who knows, like we didn't know. Service user, Trust B.
Changes in outcomes through the pandemic
In the framework, outcomes have been divided into maternal, fetal and neonatal mortality, morbidity and wellbeing, service user experience and staff wellbeing and safety. These can be influenced either through changes in the delivery of safe and personalised care, or more directly as a result of changes in demand and need: for instance, COVID-19 infection in pregnant women may lead to a high risk of poor outcomes even when quality of care is maintained.
At Trust level, there was no evidence of any adverse changes in neonatal or maternal mortality, birth weight, and gestational age at birth. The only noticeable change was a decline in infants born below the 10th centile for weight in Trust A (not shown). While this could be a data anomaly, it is interesting to note that this trend has been found in national-level data in several countries including the UK (22,23). Due to small numbers and natural variability, it would not be expected that major changes would be seen in mortality and morbidity statistics at Trust level, so these are not likely to be responsive quantitative indicators of changes in service quality unless conditions deteriorate to a catastrophic degree. It is notable that, as mentioned earlier, there were increases in readmission rates, as well as increases in the previously undiagnosed incidence of small for gestational age (SGA) in Trust A, which could reflect shortcomings in PN support and antenatal care (see appendix 4).
User experience
There is little quantitative evidence on either user or staff experience. The Family and Friends test was suspended for most of 2020, and when reinstated uptake was low. Qualitative evidence suggests service user experiences were mixed. Positive experiences included instances where health professionals had time for women and where there was constant provision of information. The difference that the individual midwife, doctor or health visitor can make was notable, and there is clearly appreciation for staff who went “the extra mile” in the face of acknowledged constraints. Several service users, however, suggested that their care was more cursory and less personalised than with previous births.
So, you know, there was a lot of communication. I didn't feel that that was missing or that because of COVID, that they didn't have enough time to talk these things through with me. I felt, I do, I say this all the time, thecare that I got what I would have expected, even if COVID wasn't on. So with all those extra steps in place, worked really well. Service user, Trust A
I definitely wasn't seen as often as I was with [name], that's for sure. Yeah, it was very short and brief on the on the times you know, that we did. Yeah, I did see them basically, I remember when I had [name] there was a lot more in-depth like questions and going through things and measurements and all that kind of stuff, whereas this was like, are you OK? Yes. OK, let's hear where baby is, brilliant, your urine's fine, crack on off you go. So it was, yeah, it was much more kind of like in and out. Service user, Trust B
Staff wellbeing and safety
Data are not routinely collected about staff wellbeing and safety. Information from Board reports, while not reported separately for maternal health staff, reported high proportions of staff sickness caused by depression, anxiety, and other mental health issues for both Trusts as a whole. In April 2020 in Trust A it was the second most common cause of sickness, but by November 2020 it was the most common cause. It was again reported as the most common cause in May and November 2021.
In direct response to national concerns on the impact of the epidemic on health care workers, Trust A instituted several support mechanisms fairly early in the pandemic including individual wellbeing conversations, telephone helplines, opportunities for face-to-face counselling, an online portal with peer-to-peer and personal resilience support and bereavement counselling. However, the qualitative data suggest that staff in both Trusts were struggling at the time of the qualitative interviews due to low staffing and the changing nature of their job. What is clear from the qualitative data is that staff continued to strive to provide safe and personalised care when they had the resources to do so, and some went above and beyond even without sufficient resources. However, this is not sustainable and does not provide individualised, person-centred care for the majority of women and their families.
I think, you know, from my point of view, I think our maternity services need to look on maternity. Bosses need to look at the providing of the staffing ratios per patient because I think you know the term, what is it now, making gold out of sows ear, you know comes to mind because you can't give such a high standard of care with minimal staff, and I think, I can see it now more than before the staff are absolutely exhausted because working with all these PPE is tiring…And I think, you know, especially with young junior midwives that are not getting the support they need, they will fortunately burn out or leave the profession. And that's what bothers me. Health care provider, Trust B
We need bums on seats because we never get to sit, particularly long, to be able to provide them and spend the extra time, happy face to face appointments, go to women's houses, see who they are, be available to them rather than this need to get as many in, as many out as we can, because that's all the time we've been given and, have those facilities to be able to them to get the scans when they need them and to be able to see their own midwife and things like that. I'm probably being idealistic, but that's certainly the job I wanted when I applied those years ago to be a midwife was, I could be there and have a relationship with the women and the families and just make them feel that they were well looked after and that I care. Healthcare provider, Trust A