Background

Having a pre-existing maternal medical condition is a key risk factor for adverse pregnancy outcomes for mother and baby. Indeed the review of maternal deaths in the UK during 2009–2013 [1] found that indirect causes (exacerbation or new onset of medical or psychiatric disease) accounted for two thirds of maternal deaths during or after pregnancy. Two medical conditions that are increasingly common in pregnancy are diabetes and cardiac disease. Between 0.2–2% of pregnant women in the UK have pre-existing diabetes [2] and 1% are affected by heart disease [3]. These pregnancies are associated with increased risks of adverse outcomes for both mother and baby [1, 2, 4,5,6,8].

In the UK, National Institute for Health and Care Excellence (NICE) guidelines recommend women with pre-existing diabetes are referred immediately once pregnant to a ‘joint diabetes and antenatal clinic’ [9] and a National Enquiry into diabetes in pregnancy recommended the minimum team composition (obstetrician, diabetes physician, diabetes specialist nurse, diabetes midwife and dietician) [2]. Similarly, numerous publications recommend MDT management for women with pre-existing cardiac conditions [5, 7, 10]. The European Society of Cardiology published consensus guidelines recommending that ‘high-risk patients should be treated by an MDT in specialised centres’ [7], and the Royal College of Obstetricians and Gynaecologists (RCOG) recommend all women are at least initially referred for risk assessment by a core MDT including an obstetrician, cardiologist and anaesthetist [5] (with midwives, neonatologists and intensivists involved when appropriate) [11]. Similar recommendations for multidisciplinary management appear in guidelines globally ([6]). However, implementation of guidance has not been audited, nor does the guidance specify how these MDTs should be operationalised (e.g. leadership, mode/frequency of meeting with each other and with women and their partners, pathways into and out of the MDT). Furthermore, a systematic review by the authors found no critical evaluation of MDT models or impact on maternal or infant outcomes [12].

Consequently, the objectives of this audit were to evaluate the implementation of UK recommendations for managing pregnancy in women with pre-existing diabetes or cardiac conditions, and to describe and compare current service provision.

Methods

Sample and setting

An online UK survey aimed to achieve geographical representation by targeting senior specialists involved in referring or managing pregnant women with either pre-existing diabetes and/or cardiac conditions. There is no single data source to ascertain these senior specialists, so national organisations were approached who agreed to distribute the survey link to their members. The organisations included: British Maternal and Fetal Medicine Society (BMFMS); Royal College of Obstetricians and Gynaecologists (RCOG) Clinical Directors’ members; National Institute for Health Research (NIHR) Diabetes in Pregnancy Network (subgroup of the Diabetes Research Network); McDonald UK Obstetric Medicine Society (MOMS); NIHR Reproductive Health and Childbirth Research Network; and NIHR Cardiovascular Research Network. In addition, authors circulated the invitation to their networks of colleagues.

In the UK, maternity care is mostly provided in NHS hospitals that either serve their local population only (secondary care) or also receive referrals from other hospitals (tertiary care). Hospitals are managed by NHS trusts in England (N = 139 trusts provide maternity care in England, within 10 health regions), and by unified Health Boards in Scotland (N = 14), Wales (N = 7) and Northern Ireland (N = 5). Health trusts/boards vary in size and may include one or more hospitals with one or more maternity units.

Survey

Respondents were screened to confirm they either referred or managed pregnant women with pre-existing cardiac conditions and/or diabetes. The survey comprised: background information (professional discipline; geographic location and type of unit - secondary/tertiary provider); and details of MDT management for women with pre-existing a) cardiac conditions; b) Type 1 or 2 diabetes. Piloting with five volunteer obstetricians highlighted the range of care defined as multidisciplinary teamwork (from specialists working in parallel with limited or ad-hoc direct communication, through to joint clinics where specialists met together with the women). The stem question and response options about MDT management were therefore designed to capture this variation, and based upon a framework distinguishing degree of integration between specialists [13] (Table 4).

Respondents who stated their current practice was referral to an MDT were asked about team membership, whether the team met in clinic with the pregnant women and/or separate to clinic (e.g. as a clinical team without the woman present), and typical timing of first referral to the MDT. Those who stated they referred to a ‘link’ clinician were asked the discipline and timing of referral. Those having ‘no formalised procedure’ or selecting ‘other’ were asked to describe the disciplines involved, the typical timing of referral, and any variation in practice. The full survey is available as an online link (see Additional file 1).

Analysis

Data were imported into IBM SPSS v.22 for analysis. Responses were checked for completeness and eligibility (e.g. removing those not responsible for referral or management of either cardiac conditions or diabetes; and non-UK respondents). Multiple responses (between two and six) were received for some English trusts (n = 37), and their concordance was examined in relation to the overall ‘model’ of care they reported, and subsequent responses (i.e. membership, timing or referral). For the management of diabetes, responses in relation to all 25 trusts for which multiple responses were received were discordant, either in relation to the type of MDT model or details of the MDT model. For cardiac conditions, responses for 26 of the 29 trusts that had multiple responses were discordant. As data were at Trust/Board level (not hospital/unit), all responses were included as independent. Data were filtered by condition (diabetes/cardiac disease) and organisational model (as per Table 4) and analysed descriptively. Team composition for women with diabetes was evaluated against the recommendation that “as a minimum the MDT should include an obstetrician, diabetes physician, diabetes specialist nurse, diabetes midwife and dietician” [2]. Responses were coded as meeting this recommendation if the team included: any obstetrician (including those with or without Advanced Training qualifications); a diabetes specialist nurse; a diabetes midwife; a dietician and either a diabetologist or endocrinologist. In the UK there are usually two types of specialist dealing with diabetes: a) diabetologists who are general physicians with specialist interests in Diabetes (usually located in secondary care, District General Hospitals); b) endocrinologists who are specialists in endocrinology and/or diabetes (with less general medicine input), usually located in regional (tertiary) centres and oversee management of more complex patients.

For women with cardiac conditions, recommendations for core (obstetrician, cardiologist, anaesthetist), and extended (midwife, intensivist, neonatologist) membership [5, 11] were similarly assessed.

Results

Characteristics of the sample

A total of 179 responses were received (Table 1), over half from obstetricians (table 2).

Table 1 Source of survey responses
Full size table
Table 2 Respondents to the questionnaire by professional grouping
Full size table

Two thirds of respondents (120, 67%) worked in secondary provider settings, and a third (59, 33%) in a tertiary setting. Responses from England covered 92 (67%) of the 139 NHS trusts providing maternity care, and included all health regions. There were two responses from Northern Ireland, three from Wales and 12 from Scotland (Table 3).

Table 3 Geographical spread of responses
Full size table

Management of pregnant women with congenital or acquired cardiac disease

123 (69%) respondents stated they either referred or managed pregnant women with congenital or acquired cardiac disease. Responses covered all UK regions in similar proportions to the overall pattern of responses (Table 3).

Two thirds of respondents stated that such women would be managed by an MDT, either in a tertiary (38%) or secondary (24%) setting (Table 4). A fifth of respondents stated they referred to a named link/specialist clinician (46% referring to a cardiologist; 29% to an obstetrician with advanced training; 17% to an obstetrician without such training), and 15 (12%) replied that they had no formalized procedures in place. Five (4%) selected ‘other organisational model’. Those with no formalized procedures or ‘other’ models described a range of models and membership including letter/email referrals to non-specific individuals on an ad-hoc basis; referral to a separate anaesthetic clinic; and “close liaison with the local cardiologist”. There was no regional pattern in responses; management by tertiary or secondary care MDTs was stated in all regions. Responses stating they had ‘no formalized procedure’ came from trusts within eight regions in England, and two health boards in Scotland and Northern Ireland.

Table 4 Organisational models for antenatal management
Full size table

MDT cardiac models

Membership

Membership of tertiary cardiac MDTs ranged from 2 to 7 (average 4 members); Membership of secondary MDTs ranged from 2 to 6 (average 3 members). The most commonly reported members were cardiologists, anaesthetists and obstetricians, two thirds had all three members as per the core membership guidelines [11] (Table 5). Only one tertiary team (and no secondary teams) reported also having the three recommended ‘extended’ members: Midwife, Intensivist and Neonatologist. All three were absent in 21 teams (20% tertiary and 40% secondary MDTs). A number of other disciplines were listed as team members by a minority of respondents (Table 5).

Table 5 Membership of tertiary and secondary MDTs for cardiac conditions
Full size table

Mode of working

Most MDTs (tertiary and secondary care) met within the clinic setting only (64% and 82% respectively). Some tertiary MDTs (8, 22%) and secondary MDTs (10, 14%) met as a team both in the clinic and separately. However a minority of tertiary MDTs (5, 14%) and secondary MDTs (3, 4%) only met separately to the clinic setting.

Timing of referral

Most women were referred to MDTs either at first contact with health services when pregnant (e.g. when visiting GP/family doctor to confirm pregnancy) or at the first hospital-based antenatal booking visit. However, in some units referral did not occur until first contact with the medical lead, or following the 18–20 week routine anomaly scan (Table 6).

Table 6 Timing of referral to tertiary MDT, secondary MDT or named link clinician
Full size table

Management of pregnant women with type 1 or 2 diabetes mellitus

132 (74%) respondents stated they referred or managed pregnant women with type 1 or 2 diabetes (Table 3). Most (116, 88%) stated that such women were managed by an MDT, either in a tertiary (32%) or secondary (56%) setting (Table 4). A minority (12%) reported referral to a link specialist clinician instead of an MDT, including diabetologists (n = 4), obstetricians (with advanced training n = 4; no advanced training n = 3), specialist diabetes midwives (n = 3) and obstetric physician (n = 1). One respondent stated that women were referred to a uni-disciplinary “diabetes or obstetric team”. There was no discernible geographic pattern: all regions reported both tertiary and secondary MDT models, and “named link specialist” models were reported in seven health regions in England and one health board in Scotland.

MDT diabetes models

Membership

Tertiary MDTs reported between 4 and 9 members (average 6), and secondary MDTs between 3 and 8 members (average 5). Less than half of all MDTs (18/41, 44% tertiary MDTs; 36/73, 49% secondary MDTs) had all five “minimum membership” specialists represented [2]. All MDTs included a diabetologist or endocrinologist, but most lacked at least one other specialist, particularly dieticians, specialist nurses and specialist midwives (though four tertiary and three secondary teams also lacked obstetric input). Other specialists represented in a small number of MDTs included anaesthetists, GPs, obstetric physicians, intensivists and neonatologists (Table 7).

Table 7 Membership of tertiary and secondary MDTs for diabetes
Full size table

Mode of working

Most MDTs (tertiary and secondary care) met within the clinic setting only (74% and 80% respectively). Some tertiary MDTs (8, 19%) and secondary MDTs (4, 5%) met both in clinic and separately. A minority of tertiary MDTs (2, 5%) and secondary MDTs (4, 5%) only met separately to the clinic.

Timing of referral

All referrals to MDTs occurred either at first contact with health services when pregnant or at the first hospital-based antenatal booking visit.

Discussion

Recommendations for MDT care during pregnancy for women with pre-existing diabetes or cardiac conditions have been implemented inconsistently across the UK. Although some form of MDT referral was standard practice in many units, the survey revealed wide variation in relation to membership, timing of referral and working practices. These inconsistencies were evident both within and between different trusts and regions of the UK.

For women with pre-existing cardiac conditions, a third of respondents (covering 47 UK units) stated that referrals were not to an MDT and instead to an individual “link” clinician, or there was ‘no formalized procedure’ of referral in place. Furthermore, in units where referral was to an MDT, the membership was typically medically dominated and often without midwifery/nursing and other extended membership particularly neonatologists and intensivists. Referral timing also varied; in some units not occurring until the fetal anomaly scan at 18–20 weeks gestation. For women with pre-existing diabetes, where NICE guidance recommends immediate referral once pregnant to a joint diabetes and obstetric team [9], most sites had MDTs, and referral was early in pregnancy. However, less than half of the MDTs comprised the ‘minimum’ recommended membership [2], most frequently omitting a dietician, specialist nurse and/or specialist midwife. Furthermore, a minority of MDTs only met separately to the clinic setting (and therefore by inference were not providing a joint clinic), and in a few units referral was to an individual specialist.

The importance of multi-professional working to safe and effective maternity care is further emphasised by the recent National Maternity Review [14]. The omission of midwives and nurses from MDTs is of concern, but perhaps unsurprising given the recognised shortage – and projections of further decline - of NHS staff including midwives and nurses [15]. Whilst a recent study of the maternity workforce in England [16] found that increasing the number of obstetricians had the greatest impact on outcomes in high-risk women, this should be balanced against the critical role of MDT management of such women. MDT input from midwives and specialist nurses in particular is necessary to promote recovery, support breastfeeding, and provide advice on healthy lifestyle behaviours. Such extended MDT support to inform life-course health could have considerable benefit [17].

To our knowledge this is the first UK (or indeed global) study examining the organisation of care for women with pre-existing medical conditions in pregnancy. The survey design was informed by a framework of ‘degree of integration’ of healthcare [13], and team membership was assessed against existing guidelines. Due to limited resources we could not send reminders, which may have increased the response rate and thereby the generalisability of findings. However the responses represented two thirds of trusts in England and included representation from health boards in Scotland, Northern Ireland and Wales. Our findings are limited to provider-level interpretation as this was the only identifier in the dataset. If repeated it would be beneficial to include a unit/hospital level identifier to explore more fully variation within healthcare organisations as well as between them. The audit relied on self-reported data from one respondent in each site (in most cases). The lack of a single database of UK clinical leads for these medical conditions meant that it was necessary to seek the assistance of a range of organisations to distribute the survey to relevant professional members. However all included respondents confirmed they were responsible for either referring or managing women with cardiac conditions and/or diabetes. Data were not validated or checked for accuracy against practice and it is possible that some survey responses contained inaccuracies.

The diversity in practice uncovered is perhaps not surprising given the lack of guidance about operationalising multidisciplinary care for these conditions, and may also reflect limited resources. This differs from UK cancer care where comprehensive guidelines exist regarding team structure (at local, regional and national levels) (https://www.nice.org.uk/guidancemenu/conditions-and-diseases/cancer), and a national peer review programme ensures links to NHS commissioning. MDTs in cancer have been associated with better patient care [18, 19] but evidence to support MDTs in maternal medicine is lacking [12]. While there may be a number of explanations, including economic reasons, for the diversity in the models of care these may have important short and long term clinical implications for both mother and baby.

Further research is needed to identify the key elements of clinically (and cost) effective models of care before, during and after pregnancy for women with pre-existing medical conditions. Effectiveness should be considered in relation to outcomes for the women (including clinical outcomes and experience of care), the infant, the team, and wider organisation, and should take account of the different contexts and geographical settings in which maternity care is provided. Recent findings from the UK National Diabetes in Pregnancy Audit [20] show there is still much to be done to improve outcomes. The impact of the diversity of MDT management on outcomes is unknown and should be a priority focus for future research.

Conclusions

Despite current guidance and consensus opinion for the use of MDTs when caring for pregnant women with pre-existing medical conditions, there continues to be a lack of primary research to support the clinical and cost effectiveness of this approach to care or to define how such care should be implemented or evaluated. Life course health for women with serious medical conditions and their infants are compromised if pregnancy and birth are not optimally managed. If indirect causes of maternal death and maternal and fetal morbidity from medical disease in pregnancy are to be reduced, research is urgently needed to promote appropriate service provision, led by optimal MDTs which include clinicians with appropriate skills to provide evidence based care across the entire pregnancy pathway, including pre and post pregnancy. Without further research into composition, location and referral pathways, MDT care is likely to persist as ad-hoc and fragmented.