The cost-effectiveness of in-patient mother and baby units

I read with interest the report of Professor Howard and colleagues on the effectiveness of in-patient mother and baby units.^{Reference Howard, Trevillion, Potts, Heslin, Pickles and Byford1,Reference Trevillion, Shallcross, Ryan, Heslin, Pickles and Byford2} The authors should be congratulated for obtaining the funds for this investigation. In the 1990s, John Cox and I submitted a protocol (unfortunately not funded) comparing the Queen Elizabeth unit in Birmingham with the Hanley day hospital and two general psychiatric services in the West Midlands; we planned to interview the mothers (when ill) to establish diagnosis and severity, in order to match the samples as far as possible.

The investigation published in May this year has shown that mother and baby units, costing £707/day, were no more effective than generic in-patient care, costing £385/day. Efficacy was measured by the readmission rate (22% v. 32%) and mother–infant relationship one month after discharge. This result will reassure high-income nations that have not invested in these expensive units and will worry National Health Service planners who may be spending as much as £50 million/annum on the 19 units we have in Britain.

I had the good fortune to work on in-patient mother and baby units in Manchester, Birmingham and Christchurch (New Zealand) and consider that the focus on severe maternal disorders has helped to construct the knowledge base, which is the essence of mother–infant (perinatal) psychiatry. But I can readily accept that many disorders can be treated equally well, and with less disruption, in day hospitals, and even psychoses can be treated at home, with daily visiting.

This was a welcome preliminary investigation. I believe that some maternal disorders cannot safely be managed in any other setting; for example, severe bonding disorders require mother and infant to be treated together but are too dangerous for home or day-patient care. I hope Professor Howard's initiative will stimulate health planners in Australia, Britain, France and other nations investing in these units to conduct a detailed investigation, similar to the one we planned in the West Midlands, and determine which disorders require conjoint mother and infant hospital admission and which can be managed equally well in other settings, without such huge expense.