Objectives The Circle of Security-Parenting (COS-P) group intervention has demonstrated efficacy in reducing maternal perinatal mental health difficulty (PMHD) symptoms in some contexts. The Circle of Security Intervention (COSI) study, a multisite, individually randomised, single-blind, parallel-arm controlled trial, was conducted in England to assess the clinical effectiveness of COS-P in reducing perinatal psychopathology, parenting and infant development, as well as its acceptability among the National Health Service (NHS) participants and staff. The main aim of this work is to estimate the cost-utility of COS-P plus treatment as usual (TAU) relative to TAU among mothers and birthing parents receiving NHS perinatal mental health services (PMHS) in England. Design A within-trial economic evaluation was performed comparing COS-P plus TAU with TAU alone, using data from the COSI trial, which employed a 2:1 randomisation ratio. Analyses were conducted from both NHS and personal social services (PSS) and societal perspectives. A 12-month time horizon was used, consistent with the final trial follow-up. Setting Secondary care NHS perinatal health services across multiple centres in England. Participants A total of 371 mothers and birthing parents with PMHD were randomised and had complete economic outcome data; 248 received COS-P plus TAU and 123 received TAU alone. Participants were eligible if they were receiving NHS PMHS; exclusion criteria were defined in the trial protocol. Interventions Participants in the intervention arm received the COS-P group programme in addition to TAU. The control group received TAU alone. Primary and secondary outcome measures The primary economic outcome was quality-adjusted life years (QALYs) over 12 months, derived from the 5-level EuroQol five-dimensional (EQ-5D-5L) questionnaire - responses. Costs were estimated from NHS and PSS as well as societal perspectives, including healthcare utilisation and productivity losses due to work absence. Results Compared with TAU, COS-P was associated with higher costs from both NHS and PSS (£180.58; 95% CI −£1075 to £1436) and societal (£72.94; 95% CI −£1473 to £1619) perspectives. COS-P was marginally less effective in terms of QALYs (−0.01; 95% CI −0.06 to 0.05). Probabilistic sensitivity analyses indicated substantial uncertainty around cost and effectiveness estimates. Conclusions On average, COS-P was associated with higher costs and did not demonstrate improvements in health-related quality of life compared with TAU alone. Given the uncertainty surrounding the estimates, further research is warranted to explore potential longer term economic and clinical impacts of COS-P in perinatal mental health settings. Trial registration number SRCTN18308962.
Wang et al. (Sun,) studied this question.