Postpartum psychosis and management: a case study

Role of the midwife

The midwife is key in detecting deviations from the norm and making appropriate timely referrals to relevant specialists (Cantwell et al, 2017). This leads to effective multidisciplinary team input by way of co-operative interprofessional communication (Murray-Davis et al, 2011). The Nursing and Midwifery Council (NMC) Code (NMC, 2018) concurs, stipulating the importance of prioritising the patient and protecting safety while providing sensitive, compassionate care. Midwives have a pivotal role in advocating for the woman's wishes and being a point of contact for safety, support and continuity, particularly in cases of complex social and mental health (Bayrampour et al, 2018).

An example of ineffective recognition of mental decline is when Sarah's community midwife visited her at home on day 11. The midwife encountered Sarah agitated and restless, while expressing distorted thoughts and anxiety about ‘overheating’ her baby due to a fever she believed she had, particularly as she was breastfeeding. The midwife documented the strange behaviour and recorded her temperature as ‘no abnormalities detected’. Sarah was advised to go to hospital, which she declined, but the midwife failed to escalate the situation by communicating to a specialised professional, such as the specialist perinatal mental health midwife, as soon as possible. This was not safe or dutiful practice, as monitoring for signs and symptoms of mental illness is an important midwifery role (National Institute for Health and Care Excellence (NICE), 2014), particularly as early detection allows for effective postpartum psychosis management and maximises the safety of mother and infant (Nahar et al, 2017).

However, given the rarity of postpartum psychosis (VanderKruik et al, 2017), midwives may lack sufficient experience to confidently recognise its signs and symptoms (Noonan et al, 2017). Consistent training (Cantwell et al, 2017) should therefore be provided by healthcare institutions and specialist perinatal mental health midwives to raise and maintain competence (Maternal Mental Health Alliance (MMHA) et al, 2013). However, midwives are expected to update and maintain their own knowledge (NMC, 2018). Furthermore, human factors may have played a role (Derickson et al, 2015), if the midwife was scared or apprehensive to refer Sarah's disordered behaviour to a specialist, for fear of raising a false alarm and appearing inadequate. Healthcare institutes can cultivate ‘psychological safety’ by being aware of human factors, which enables courage to communicate across multidisciplinary team hierarchies (Aranzamendez et al, 2015).

The community midwife's lack of action did not fulfil the midwife's role to protect women's dignity (NMC, 2018), as evidenced by Sarah's husband bringing her to A&E the next day in a state of pyrexia and confusion. Effective timely referral to the specialist perinatal mental health midwife, leading to robust risk assessments, could have avoided this (Cantwell et al, 2017). Moreover, continuity of care throughout Sarah's perinatal experience was not provided, likely due to staff shortages; however, this probably impeded the development of a trusting relationship between Sarah and midwifery staff (NMC, 2018). This may have discouraged Sarah from communicating concerns, and inhibited knowledge of Sarah's usual behaviour, hence delaying the recognition of her decline (Cantwell et al, 2017).

A review of postnatal notes from day 0 and at discharge on day 1 did not describe Sarah's emotional wellbeing, her adaptation to motherhood or her competence at recognising her infant's needs, and the notes were inadequate in providing an immediate postnatal baseline of her mental health. The Whooley questions (Table 1) were asked and recorded as ‘no abnormalities detected’; however, their design favours false positives (Bosanquet et al, 2015). Alternative questions (Table 1) were not considered. Instead documentation encompassed Sarah's physical health and although this is important, the absence of mental wellbeing documentation (NICE, 2014) implies its exclusion from midwifery care.

Care by mental health staff

In A&E, a mental health nurse specialist extracted a highly detailed psychosocial history, in order to individualise risk and provide care that contextualised Sarah's culture and individuality (Di Florio et al, 2014). A family history of learning disability and a challenging housing situation were discovered. The couple rented a room in a five bedroom house along with four other couples, two of whom were recently postnatal. This may signal poverty and stress, which could lead to mental health vulnerability. Sarah's housing situation was also a mental health risk factor as maintaining functional relationships with other families within a confined, shared space can be stressful and difficult, particularly as the home is meant to be a space of relaxation (Franks et al, 2017). However, living with other families can, in some cases, be a display of effective community, where families in lower-resource settings assist each other to transition into motherhood and collectively raise children (Al-Maliki et al, 2012). Sarah may not have confessed her housing situation at booking due to negative stigma (Davis et al, 2018); therefore midwives and specialist perinatal mental health midwifes (MMHA et al, 2013) need to challenge this stigma with open communication and education to prevent secrecy and misplaced shame (VanderKruik et al, 2017).

The mental health nurse was effective in creating a comprehensive biopsychosocial picture, which aided direction of management by highlighting risk factors (Table 2) (NICE, 2014). However, requiring Sarah to interpret, rather than a professional, neglected her partner's needs. It proved futile (as she repeated her own disjointed thoughts) and was highly improper due to her compromised cognition (Mannion and Slade, 2014). Although it was necessary to prioritise Sarah, targeting the family holistically is needed for full recovery (Glangeaud-Freudenthal et al, 2014) and for prevention of long-term sequelae to the infant (Thippeswamy et al, 2017). Partners of women with postpartum psychosis have been shown to experience fear, confusion and anger; and are reported to be reluctant help-seekers, which can contribute to stress and marital collapse (Wyatt et al, 2015). Midwives should therefore provide affirmational support while informing and educating the family to increase understanding of the illness. This enables all to cope with the associated stress and guides the partner to help the patient, as their support is crucial to recovery (Doucet et al, 2012).

The mental health nurse specialist also discovered Sarah's antenatal anxiety over the small symphysiso-fundal height and reduced fetal movements, despite normal growth scans, and a non-suspicious CTG. Although the midwives discharged Sarah with warnings to return if reduced fetal movements repeated, they did not document any reassurance given, or note her anxiety. Considering the growing evidence that maternal stress leads to negative neonatal outcomes (Zijlmans et al, 2015), midwives here were therefore not effective in monitoring perinatal mental health risk (NICE, 2014). This disregard continued in the intrapartum notes, where Sarah grew anxious at sighting thin meconium in her amniotic fluid, and midwives did not document that they informed her that this was normal in postdates pregnancies (England, 2014), or that they gave reassurance. Midwives may have assumed that due to Sarah's background, she would have had health-based knowledge, and thereby not provided as much information or psychological support. This violates a tenant of practice (NMC, 2018) to not discriminate between patients, and to provide quality care, including information, to all.

Sleep deprivation may have contributed to Sarah's development of postpartum psychosis as she was working night shifts before commencing maternity leave, which was compounded by gestational sleep disruption and postnatal insomnia. Although no reports associate night shift work with postpartum psychosis (Aiken et al, 2016), evidence of the effect of sleep deprivation on postpartum mental illness and postpartum psychosis aetiology exists (Lawson et al, 2015). Midwives did not take note of this aspect of Sarah's employment, and therefore inadequately screened Sarah's physical and mental wellbeing, both antenatally and postnatally. This is important as the information could have alerted staff to a potential perinatal mental health risk (NICE, 2014).

Interprofessional working

The mental health nurse specialist performed her role well in communicating to perinatal mental health, obstetric and midwifery staff, who then orchestrated a side room in the postnatal ward for Sarah's readmission and privacy. This was a good display of communication across multidisciplinary teams and interprofessional collaboration in the best interests of the woman (Murray-Davis et al, 2011). Sarah's history was then communicated to maternity staff and a mental health nurse was stationed with her to monitor her condition and protect her safety (NMC, 2018). Sarah's sister-in-law stayed throughout the readmission, caring for the infant and ensuring that Sarah continued to breastfeed (Wyatt et al, 2015). Although these actions are a midwife's duty (Posmontier, 2010), staff shortages and time restraints make it difficult to provide the complete attention required. Therefore, recruiting more midwives and retaining them, with a work culture that promotes staff wellbeing and happiness, is needed for effective and safe midwifery care (Aquino et al, 2016).

Good interprofessional collaboration was also seen when the obstetric and perinatal mental health teams communicated about Sarah's management (Murray-Davis et al, 2011), as there was indecision over whether to choose a sepsis or a mental illness pathway. Initially, a sepsis pathway was chosen, but later changed after further discourse between the two clinical teams, who elected the consultant perinatal mental health psychiatrist as the lead clinician. Upon the change, the perinatal mental health midwife and other midwifery staff were informed, exemplifying model communication across a multidisciplinary team (Murray-Davis et al, 2011). The specialist perinatal mental health midwife held an important office of acting as a co-ordinator of care and advocate for the women (Posmontier, 2010), which was vital considering Sarah's vulnerable mental capacity.

Kalanithi (2016) has reported dangerous power dynamics and tugs of control between clinical teams, along with egocentric decision-making that impedes interprofessional communication and collaboration and has detrimental effects on patients' health and safety. The good communication seen in Sarah's case meant that this was unlikely; however, it remains a barrier to effective interprofessional working in general, as does lack of trust, respect, patient-centred work ethic and understanding of others' professional roles (Aquino et al, 2016).

Sarah's management started with antibiotics to treat her infection, followed by oral paroxetine for mood stabilisation. This was in line with NICE guidelines, which recommend stabilising physical health before treating mental health (NICE, 2014). Following the Department of Constitutional affairs' code of practice (2007), midwives first assumed that Sarah had capacity to consent, which was gained before medicine administration, in concordance with local and NMC standards (2007) (Table 3).

As immunology can play a role in postpartum psychosis pathogenesis (de Witte et al, 2018), it was important to use a sepsis bundle screen to ascertain the presence of an infection. This was performed and proved positive. Infection may have been missed earlier when Sarah complained of fever to her community midwife, despite a normal temperature reading. Raised monocyte and microglia inflammatory activation patterns are thought to be biomarkers of postpartum psychosis (Bergink et al, 2014); screening for these may be performed in the future, as could scanning for structural cortical changes in women with postpartum psychosis (Fusté et al, 2017). However, screening women with MRI scans, even women at risk (Table 2), is unlikely to be cost effective due to the rarity of postpartum psychosis (Dias and Jones, 2016).

Throughout her 12-day readmission, Sarah's mental health fluctuated between recovery and decline. This follows postpartum psychosis's natural history, which, without proper clinical knowledge, can result in repeated discharge and readmission of the patient (Cantwell et al, 2017). This did not happen with Sarah due to co-ordinated multidisciplinary team efforts; however, her recovery became dangerous as Sarah mentioned thoughts of killing her child and was declining medication. Despite Sarah's right to refuse medication (NMC, 2007), the risk to safety induced the midwives and the mental health nurse to document and communicate this promptly; resulting in the presence of perinatal mental health and obstetric teams.

Sarah's mental capacity was assessed by two perinatal mental health psychiatrists, in line with the code of practice (Department for Constitutional Affairs, 2007). She was then sectioned under the Mental Capacity Act 2005, but only after a joint multidisciplinary team meeting passed to incorporate the input of all members, communicate understanding of the situation and create a plan of care. This was good clinical practice, as communication was made in a timely fashion and action was taken promptly in order to protect the safety of mother and infant (NMC, 2018). Unfortunately, in this instance, no documentation related to the partner or family members was made, implying their exclusion. This may further propagate their isolation, pain and distress (Wyatt et al, 2015).

Transfer and care planning

Discussions began over transfer to a psychiatric unit. An MBU was preferred; however, due to the severity of Sarah's postpartum psychosis, it was not known if it would be suitable. Sarah's husband agreed to care for the infant, should separation from the mother occur; however, there were no discussions over extending his leave from work, organising childcare, infant feeding, visits to his wife, financial support or protecting his own mental health (Doucet et al, 2012). A translator was again not used, impeding effective communication and understanding (NMC, 2018), which is necessary for partners' confidence in staff (Davis et al, 2018). The perinatal mental health staff therefore failed to respect the degree of complexity that accompanies caring for a child, which is another example of staff not targeting the family as a whole (Glangeaud-Freudenthal et al, 2014) or providing sensitive care (NMC, 2018).

Sarah was transferred to an MBU on day 24 after assessment from MBU staff. This is best practice (Cantwell et al, 2017), as separation of the infant from the mother has detrimental effects on maternal recovery, mother-child bonding and maternal parenting confidence (Glangeaud-Freudenthal et al, 2014). However, earlier transfer is preferred, as the postnatal ward is not a suitable environment for management, treatment and full recovery from postpartum psychosis (Cantwell et al, 2017). This is because, unlike MBUs, postnatal wards lack the continuous mental health staff, cognitive behavioural therapy, and facilities in which to support maternal parenting and child bonding in the context of severe mental illness (Glangeaud-Freudenthal et al, 2014). MBUs can also organise electroconvulsive therapy, which is recommended for postpartum psychosis women with catatonia (Tinkelman et al, 2017).

The delay may have been due to time restraints that occur when collaborating with clinical staff in a different locality, or because of uncertainty over Sarah's suitability for a MBU, but it is more likely that bed shortages were to blame (as this was also documented). These bed shortages are linked to the high cost and long stay that MBU admissions accrue (Glangeaud-Freudenthal et al, 2014), but more heavily associated with the scarcity of MBUs (MMHA, 2018c). Despite four new MBUs opening in 2018/19 (MMHA, 2018c), more are needed for women to access local specialist perinatal mental healthcare (Royal College of Psychiatrists, 2015).

No more information is known after Sarah's transfer, however, upon full recovery, a joint meeting with Sarah, the community health visitor, specialist perinatal mental health midwife (for advocacy and continuity), psychiatric team, pharmacist, GP and obstetric team will be needed (Glangeaud-Freudenthalet al, 2014). Sarah's husband can be included if she consents (NICE, 2014). This will allow MBU-to-community discharge planning, and ensure that Sarah has social support while at home, and good links with the MBU (Glangeaud-Freudenthal et al, 2014). This is to maintain Sarah's mental health, help her in adjusting to autonomous parenthood and prevent psychiatric readmission (Doucet et al, 2012).

Subsequently, a joint meeting with the couple, a translator and a psychiatrist or specialist perinatal mental health midwife is needed for pre-conception counselling (NICE, 2014) before MBU discharge. This is to inform the couple of the teratogenic effects of psychotropic medication that she may need prophylactically in a second pregnancy, or that she may still be taking, which should be administered alongside long-acting reversible contraception (Cantwell et al, 2017). Breastfeeding should be encouraged throughout her care, but tact is required, as some psychotropic drugs, such as carbamazepine, clozapine and lithium, are not suitable for breastfeeding, and Sarah should be made aware of their effects (NICE, 2014). If they have been prescribed while Sarah is breastfeeding, the reason needs to be documented, and regular checks on the infant's renal function should be conducted (NICE, 2014). Sodium valproate should not be considered for Sarah due to significant teratogenicity (NICE, 2014). This way, tailored, individualised care can be planned for Sarah, considering her feeding choices and her newborn child (NICE, 2014).

This meeting can also inform Sarah and her husband of the high risk of postpartum psychosis recurrence after a subsequent pregnancy (Dias and Jones, 2016). Education on the signs and symptoms should follow, along with how to escalate the situation effectively (Doucet et al, 2012) and access care quickly (Nahar et al, 2017). Given that partners can be reluctant help-seekers, this information is especially important (Doucet et al, 2012). The couple would also be informed that the psychosis may be the first episode of a chronic mental disorder, such as bipolar disorder, that will require more specialist antenatal attention should they decide to conceive again (Glangeaud-Freudenthal et al, 2014). As a result, the couple are empowered by being included in their care and decision-making process (NMC, 2018).