The Circle of Security Intervention (COSI) Study

Perinatal mental health problems are a significant public health concern, affecting up to 20% of new mothers and other birthing parents and, when untreated, costing £8.1 billion per birth cohort. Seventy-two per cent of this cost is attributed to adverse impacts on the child, as a result of impaired mother-infant relationship. Perinatal mental health services (PMHS) provide multi-disciplinary support for women with mental health difficulties in pregnancy and the first post-natal year. However, these services are relatively new, and gaps in provision remain in group therapeutic approaches and mother-infant bonding therapies. COS-P is an intervention that can meet these needs, and which shows preliminary efficacious findings and user acceptability.

To conduct a multi-site national RCT comparing COS-P to treatment as usual in improving perinatal mental health and the quality of parent-infant relationships. Our secondary objectives include exploring whether the intervention improves maternal sensitivity, emotion regulation skills, attachment security and has an impact on infant development. Additionally, cost-effectiveness and service-user and practitioner acceptability will be assessed.

We will randomly allocate 369 women across nine PMHS in England to treatment as usual or COS-P. We will assess perinatal mental health conditions using the Clinical Outcomes in Routine Evaluation – Outcome Measure (CORE-OM) at baseline, 3 months later (once the intervention has finished), and at 7- and 12- months after baseline. We will use a mixed effects linear regression model to estimate the mean difference in CORE-OM between treatment group at 3-, 7- and 12-months. We will also carry out analyses of secondary outcome measures, as well as process evaluations. We will use thematic analysis to analyse qualitative data. An economic analysis will also take place that compares the costs and outcomes of COS-P to treatment as usual.

This research will substantially add to the evidence base for psychological interventions that target maternal mental health problems and mother-infant relationship quality, areas which are highlighted in government policy. If effective, the research will lead to improved short- and long- term outcomes for mothers and children across a range of domains. Furthermore, COS-P has the potential to be used widely across all PMHS, given the established gaps in evidence-based provision that it has the potential to fill and as such, this research can make a substantial contribution towards addressing this area of concern for the NHS. COS-P has the potential to reduce the costs of perinatal mental health problems to the NHS and wider society, through reduced cost to health, social, educational and criminal justice systems. 

• Dr Zoe Darwin (Reader, University of Huddersfield) 

• Dr Victoria Cornelius (Reader in Medical Statistics, Imperial College of Science, Technology and Medicine)

• Professor Steve Pilling (Professor of Clinical Psychology and Clinical Effectiveness, UCL)

• Professor Pasco Fearon (Chair in Developmental Psychopathology, UCL)

• Dr Elena Pizzo (Senior Research Fellow, UCL)

• Lani Richards (Patient and Public Involvement, Anna Freud)

• North Yorkshire & York

• South West Yorkshire

• Northumberland

• Northamptonshire

• Sussex

• Cheshire and Wirral

• Mid-Mersey: Halton, Knowsley, St Helens and Warrington

• Liverpool and Sefton

• Devon

• Hampshire, Southampton city, Portsmouth city and the Isle of Wight

The COSI Study includes a panel of experts by experience (EBE) led by Lani Richards, who has completed a COS-P group. The EBE panel members have all accessed PMHS in the recruitment sites and surrounding areas. They are consulted regularly to ensure the study remains inclusive and accessible to the participants who are recruited. We have also formed two subgroups: the equity, diversity and inclusion panel, which focuses on the recruitment of diverse participants into the study; and the qualitative team, which meets at four points during the study to assist in analysing the qualitative data.