Background
Perinatal mental health (PMH) problems, encompassing mental disorders occurring during pregnancy or up to 2 years after childbirth, are a significant public health concern. These conditions, marked by distressing symptoms and impaired functioning are a leading cause of maternal deaths.1,2 The most common disorders are depression and anxiety disorders; they can predate conception, persist through the perinatal period, or emerge during pregnancy or the postnatal period. Common risk factors include psychosocial stressors related to childbirth, such as unplanned and unwanted pregnancy, lack of social support, domestic abuse, childhood trauma, financial hardship and young age. A history of prior mental health issues is the strongest predictor of mental health problems.1 Moderate to severe depression affects around 4% of the maternity population,1,3 with severe depression being the most common diagnosis associated with maternal suicide. Obsessive–compulsive disorder (OCD) is highly prevalent during the perinatal period, particularly postnatally, with an average prevalence of 7%.4 The most common anxiety disorders are generalised anxiety disorder and social phobia (4–5% prevalence); while post-traumatic stress disorder (PTSD) occurs at 4%5,6 and PTSD themes may focus on childbirth and infant loss trauma. Less common disorders include eating disorders, psychotic disorders, tokophobia (fear of childbirth) and personality disorders, each occurring in around 1–3%.1,2 Psychotic disorders during pregnancy have a prevalence similar to that outside the perinatal period (around 1%), but the first few weeks after childbirth sees increased risk of psychosis, which may be particularly rapid in onset and severe, and is especially among those with a personal or family history of bipolar disorder.2,7 Relapse rates are reported in a systematic review and meta-analysis to be around 20% and 35% for severe (psychotic) and episodes overall, respectively.8 However, pregnancy-related relapse can also be problematic for women with severe disorders – for example, for those who may stop prophylactic medication due to their concerns about its impact on the baby.2
Perinatal mental health disorders not only affect the mother but are also linked to pregnancy complications like low birthweight and preterm birth and carry risks for long-term adverse effects on child’s cognitive, social and emotional development.9 These effects are thought to be mediated by prenatal cortisol exposure on the developing fetus leading to alterations in infant stress reactivity, and postnatal issues related to the quality of the mother–infant interaction.6 Although some studies offer contradictory findings, recent research suggests that the children of mothers who have had effective treatment for maternal depression will not have developmental problems at age 210 (few studies have examined the impact of treatment of other disorders on child outcome). Child development issues are more likely with persistent and severe symptoms11 or when the mother has additional personality difficulties.12,13 Effective PMH services could mitigate the substantial economic burden of PMH disorders, which are estimated to cost the UK economy £8.1B per annual birth cohort in the UK (28% due to direct costs, 72% to the effect on the infant).14
The perinatal period is a critical period for maternal and infant mental health and long-term well-being. Women have frequent interactions with healthcare providers during pregnancy and postnatally, including midwives, health visitors and general practitioners (GPs), offering numerous opportunities to detect mental health issues and provide appropriate intervention. However, without integrated care across the care pathway that effectively identifies women with PMH needs and rapidly and effectively engages them with evidence-based care, few women (around 15%)15 receive interventions, due to stigma, lack of training and a small workforce.16 Before the National Health Service England perinatal programme only around 15% of trusts had comprehensive specialist community perinatal mental health teams (CPMHTs).
The 5 Year Forward perinatal transformation investment (£365M) by National Health Service England aimed to address this imbalance by expanding CPMHTs and delivering evidence-based perinatal care pathways.17 These pathways included: (1) preconception advice to women with severe mental disorders; (2) identification and assessment by universal services, with referral for assessment and care by CPMHTs for women with moderate to severe and complex mental disorders and their partners who may also suffer from mental disorders; (3) emergency assessment for acute crises, carried out by psychiatric liaison or crisis teams, supported or led by the CPMHTs; (4) timely access to evidence-based psychological interventions, and co-ordination with primary care, maternity and secondary mental health services on identifying, preventing and caring for mild to severe PMH problems; and (5) arranging admission/post-admission support. The long-term plan further extended PMH care to 24 months post delivery, expanded access to psychological therapies within CPMHTs, and offered fathers psychological assessment and signposting. The guidance emphasised the broader impact of well-functioning CPMHTs on the wider health system, ensuring comprehensive care for parents with PMH problems across universal services, NHS Talking Therapies (primary care psychology services), generic secondary mental health care, children and young people’s care, and social care.
The recommended workforce composition of CPMHTs was outlined in the 2015 Royal College of Psychiatrists’ report CR19718 (updated in 2021 to CR232). However, limited evidence existed regarding optimal CPMHT service models and their impact on women and infants. A National Institute for Health and Care Excellence (NICE) scoping review and consultation did not identify any new studies on service models,19 and there was considerable variation in team configuration and services provided, critical to consider as team composition affects the range and types of interventions offered. National Health Service England highlighted the range of diverse CPMHT approaches, including perinatal psychiatry and nursing-driven liaison, education, preconceptual counselling, care planning and medication optimisation and support. Other services built on this model to incorporate the skills of nursery nurses, social workers, clinical psychologists and occupational therapists. These expanded services offered psychological interventions, mother–infant interaction treatments, facilitated peer support, and support for partners and family relationships. 2018 National Health Service England benchmarking data showed that although use of other professions was limited, in some areas innovative approaches were being implemented. For example, specialist pharmacists were embedded in some CPMHTs, and additional psychological time was provided. Service configurations varied by trust type (acute vs. mental health), whether they were colocated within maternity hospitals or community secondary care [pre-coronavirus disease (pre-COVID)], where they employed in-person versus remote/home-based approaches or relied on outpatient/office-based visits, coworked with generic secondary care services versus care provided solely by CPMHTs, and whether they provided care across the perinatal period or for shorter periods.
Theoretical framework: As noted in the commissioning call (see Report Supplementary Material 1), despite the growing number of CPMHTs, there was a lack of data regarding how service-level variations or specific components impacted women and infants outcomes.19 This was especially important to understand, because with the increased number of CPMHTs since the first wave of National Health Service England investment, NHS benchmarking data reported a 49% increase in the number of women seen by CPMHTs over 2015–7 (due to the availability of new teams), representing 0.9% of the maternity population in 2017.20 Further expansion was expected, aiming for 5% of the maternity population by 2021, and 10% across the pathway. However, it was unclear whether women most in need were accessing CPMHTs, the extent of inequity of access across primary and secondary mental health care, and whether CPMHT access improved maternal and infant outcomes. We hypothesised that variations in service models and components influenced communication between CPMHTs and other healthcare professionals, as well as women’s willingness and ability to access care. Therefore, it was critical to identify optimal models and components for identifying women, improving access, and improving outcomes for women and their infants. Following questions posed in the Health and Social Care Delivery Research (HSDR) commissioning brief, we aimed to provide evidence to establish:
1. Which robust methods of assessment could be used by health and social care services, and are acceptable to potential service users, to identify those in need of intervention [2.1(i) of HSDR brief].
Evidence on how best to assess important domains of mother–infant interaction quality (e.g. sensitivity, responsivity, withdrawn behaviours, intrusiveness, warmth or positive regard) within clinical services, known to impact adversely on child mental health or cognitive development, was acutely needed. Existing observation measures are training and time intensive (both to become a reliable observer/coder and to code a single mother–infant observation) and lacked evidence of predictive validity to child mental health outcomes or required a lengthy filmed observation period not commensurate with clinical practice. Assessment of mother–infant interaction quality required a validated, brief, clinically feasible observational measure of mother–infant interaction which could be used in routine clinical settings; we aimed to develop this in work package (WP) 2.
2. Which interventions are attractive and acceptable to women, and clinically effective within/in conjunction with mainstream secondary mental health care, maternity and primary care [2.1.(ii) HSDR brief].
There were major differences nationally in provision of interventions and it was not known which interventions were acceptable, appealing or what impact they had on women and their families when delivered by CPMHTs. For example, because services varied on the size of geographical area they cover (dense urban or expansive rural), the provision of therapies varied in their delivery (e.g. remote or group vs. home-based, 1 : 1 care). With the coronavirus disease discovered in 2019 (COVID-19) pandemic, many services moved to remote delivery only, but the acceptability and effectiveness of this model of delivery for perinatal women and their families were largely unknown. Care planning or psychological interventions might also not be as effective when delivered in the broader system (e.g. generic secondary care or NHS Talking Therapies) as they were generally not tailored for the perinatal period, and this may also vary by the mother’s particular mental health problem or her sociodemographic characteristics. A range of psychological interventions and the ways in which they are delivered were therefore examined in WP3 informed by WP1 taxonomy.
3. What service models are effective for mothers with PMH problems and their babies, and whether this varies according to mental health needs …, who delivers these interventions, with what skill mix and competencies, in what settings? [2.1.(iii) HSDR brief].
It was not clear whether CPMHTs improved access to care and improved maternal, neonate and infant health outcomes, for whom and under what circumstances. This was examined in WPs 3 and 4.
Why this research is needed now: 35 National Health Service England PMH programme wave 2 sites were announced in May 2018 (20 wave 1 pilot sites were launched in 2017).21 By the start of this proposal, most areas in England had recruited staff for their CPMHTs and were operational. It was therefore timely to evaluate these services to inform commissioning beyond the National Health Service England 5-year Forward View22 and the NHS Long Term Plan.23
Aim: We aim to answer several inter-related questions from the HSDR call, using pre-existing data where possible, and primary data collection over four WPs. We aimed to produce high-quality evidence, examining which types of CPMHTs are most effective for which mothers and babies, in what circumstances.
- Develop a taxonomy of CPMHTs across England (WP1) to inform sampling in subsequent WPs.
- Validate a brief clinically feasible observational measure of mother–infant interaction (WP2).
- Examine which CPMHT model worked, for whom, in what circumstances, how and why, identifying underlying mechanisms for effectiveness (WP3).
- Investigate the effectiveness of CPMHTs in improving access to mental health care, improving birth outcomes, preventing relapse, and thus reducing cost, using national NHS data sets (WP4).