Many women who experience perinatal mental health problems do not receive the care they require (Reference Hearn, Illif and JonesHearn et al, 1998; Reference OatesOates, 2004). Even those with serious conditions often do not have access to specialist psychiatric services or to a mother and baby unit (Reference OatesOates, 2000; Scottish Intercollegiate Guidelines Network, 2002; National Institute for Health and Clinical Excellence, 2007). Recent policy initiatives have emphasised the need for every maternity locality to have in place a perinatal mental health strategy to ensure that women who experience maternal mental health problems of any type are treated effectively at all levels of healthcare (Oates, Reference Oates2000, Reference Oates2004; Department of Health, 2004; National Institute for Health and Clinical Excellence, 2007) The purpose of this qualitative investigation was: to identify the core elements of care and service delivery considered essential to the management of perinatal mental illness; to establish problems with and shortfalls in the provision and organisation of services; and to identify specific areas where improvements were deemed necessary.
Method
Our area of interest was the East Midlands. This involved the participation of two strategic health authorities, six health communities, four mental health trusts, twelve maternity hospitals and twenty-four primary care trusts. Two groups of informants were recruited – health services professionals and recovered women.
Semi-structured individual interviews were conducted with 39 health professionals who were recruited through a combination of non-probabilistic purposive sampling with pre-defined criteria (professional group and locality) and snowball sampling, in which new informants were identified through existing contacts. Informants from involved localities and professions, including mainstream and specialist perinatal services, were recruited to ensure a representative range of views (Table 1). These comprised general adult and perinatal psychiatrists, obstetricians, health visitors, midwives, general practitioners, primary care mental health practitioners and health services managers.
Composition of interview samples
| Trent SHA | Leicestershire, Northampton and Rutland SHA | Total | |
|---|---|---|---|
| Health visitor | 2 | 2 | 4 |
| Midwife | 3 | 1 | 4 |
| General practitioner | 3 | - | 3 |
| Mental health practitioner | 2 | - | 2 |
| Community psychiatric nurse | 5 | 3 | 8 |
| Obstetrician | 4 | 1 | 5 |
| Psychiatrist | 7 | 2 | 9 |
| Health services manager | 3 | 1 | 4 |
| Total | 29 | 10 | 39 |
Two focus groups of five and seven recovered women were conducted. Group size was kept relatively small in order to facilitate participant interaction and maximise group focus (Reference MorganMorgan, 1997). Participants were all women who had been admitted to a specialist mother and baby unit and we hoped to capitalise on their shared experiences (Reference KitzingerKitzinger, 1995). Admissions to all three mother and baby units within the study area were sampled, so as to capture a wide range of views. Former in-patients were recruited through their consultant perinatal psychiatrists, who ensured that all identified informants had recovered sufficiently in order to participate. A further focus group was conducted with staff from two of the three mother and baby units, so as to broaden the range of experiences.
Using the principle of analytic induction (in which data collection and data analysis occurred simultaneously), participants were recruited for individual interviews and focus groups were continued until no new issues emerged (‘data saturation’).
Based on the broad area of enquiry and previous research, an interview schedule was developed to explore participants’ beliefs and understandings of the nature and management of perinatal mental illness, their views about shortfalls in care and service provision (and the implications of these), and their suggestions for improving healthcare and outcomes for women and their families.
Data were analysed using data reduction and data complication techniques (Reference Seidel, Kelle and KelleSeidel & Kelle, 1995; Reference Coffey and AtkinsonCoffey & Atkinson, 1996) and simplified into initial themes, or codes, which reflected participants’ meanings and experiences of services. These were used to identify similarities, patterns and inconsistencies in the data; the coding framework was then refined and developed and the process repeated until no new themes emerged.
Results
Five categories were identified. These represent the views and beliefs that were common among informant groups. Individual themes highlighted by stakeholders (health professionals) are presented in an online data supplement to this paper.
Knowledge and skills
Many non-specialist healthcare practitioners felt they lacked the knowledge and skills required to effectively manage the whole range of perinatal mental health problems. They were unaware of the significance of previous history of mental illness and the risk of recurrence following delivery, which resulted in a lack of forward planning. A lack of awareness over how to access specialist services for referral and advice and under which circumstances this would be appropriate was also evident. At all levels of healthcare and service provision an absence of systematic training was noted. In order to acquire the necessary skills and knowledge, non-specialist professionals agreed that they should have access to specialist education, training, enhanced guidelines and care pathways, and standardised referral criteria.
Access
Difficulties in accessing psychiatric services, particularly in emergency situations, was a common factor. This was attributed to rigid protocols or practices used by non-specialist psychiatric services, slow response times or insufficient awareness or understanding of the referral criteria of specialist services by referring professionals. Practitioners also experienced difficulties in obtaining specialist advice and information from psychiatric colleagues, particularly when following formal communication routes such as telephone support. Informants felt that more effective systems are needed to ensure that non-specialists are able to access specialist care and treatment and/or advice and information, when required (see below).
Pathways and protocols
The development of care pathways and protocols was felt to be important in the management of serious perinatal illness but this was seen to have been adhoc and unsystematic. Such tools were considered particularly important in helping non-specialist professionals decide when and how to make referrals for specialist care and treatment, although it was emphasised that they would need to be flexible both to the skills and judgment of the practitioner (in order to maintain clinical autonomy and decision-making) and tailored to the patient's individual needs. It was agreed that systematic pathways, protocols and guidelines should be developed at all levels of perinatal healthcare provision and for all localities.
Comprehensiveness and integration
Informants felt that the lack of any systematic development of specialist services in the region had resulted in a fragmented and disjointed approach to the management of perinatal mental illness. This, coupled with the fact that health professionals often failed to communicate important information to one another, meant that services were unable to provide the most effective response to meet the needs of their service users.
The need to develop better links and improve liaison between maternity, psychiatric and primary care services, together with systems that encourage integrated or joint working, such as care pathways and protocols was emphasised. It was felt that improving liaison and interagency working could also be achieved through designated link workers, who would provide a single point of contact for specialist advice and information.
Roles and responsibilities
Health professionals across all levels of service provision were unclear about their roles and responsibilities. They felt this often led to misunderstandings and uncertainty and could result in delayed care. These concerns were more pronounced in non-specialist services and where there were many interfaces between services and levels of provision, such as in primary care. Informants agreed that all services involved in the management of perinatal mental illness should be clear about their own responsibilities. It was felt that the use of systematic procedures, including guidelines, improved access to specialist advice and information, referral criteria, supervision and training, could help achieve this.
Discussion
The aim of this qualitative study was to identify – from a practitioner's and patient's perspective – the core elements of service provision and delivery considered vital to the proper management of perinatal mental illness. It was intended that the data collected should help inform both the development of an innovative approach to the management of perinatal mental illness – a ‘managed clinical network’ (Reference Baker and LorimerBaker & Lorimer, 2000; NHS Scotland, 2002) and the development of national minimum core standards for perinatal mental health. Of most importance, these developments needed to be informed by the very people who would be directly affected by the proposed changes (patients and professionals). The other purpose of our study was to show whether the core elements of care and service provision identified by informants were consistent with previous developments in this area and the objectives of the proposed network.
Key findings
The key themes identified demonstrated a high degree of consistency across informant groups, both professional and patient. There was strong agreement that non-specialist practitioners need to have the right skills and knowledge to properly manage perinatal mental illness. Tied to this, practitioners need to have access to specialist care and treatment for women who require these services and to specialist advice and information to assist in management of complex and serious conditions. Issues around knowledge and access were strongly related to the use of care pathways and clinical guidelines to assist professionals in managing perinatal illness and referring women more appropriately. Concerns over the lack of clarity and definition of professional roles and responsibilities at all levels of healthcare provision were linked to the need for systematic procedures for managing perinatal mental illness. Finally, the need for improved liaison and communication between services at all levels of healthcare was emphasised, with systems in place to increase and support joint working.
Implications for service provision
The major themes highlighted by the health professionals and service users who participated in this project demonstrate the importance of the interrelationships between different elements of service provision and delivery essential to the proper care and management of perinatal mental illness. These findings are consistent with the provision and organisation of healthcare around ‘whole:systems working’ (Reference Senge, Ross and SmithSenge et al, 1994; Reference Plsek and GreenhalghPlsek & Greenhalgh, 2001), rather than focusing on discrete aspects of service development and improvement. For perinatal mental health, such an approach emphasises the need for integrated and joint working at all levels of healthcare, particularly at the interfaces between maternity, psychiatry and primary care. Policy at both the national and local level supports the implementation of a perinatal mental health strategy in every health community (Oates, Reference Oates2000, Reference Oates2004; Department of Health, 2004; National Institute for Health and Clinical Excellence, 2007). By promoting a hierarchical, tiered system of healthcare, such a strategy supports a comprehensive and integrated model of service provision, which should enable women to receive care and treatment at a level most appropriate to their needs.
Declaration of interest
None.
Declaration of interests
None.
Open access
eLetters
No eLetters have been published for this article.