This report describes three services supported by the Working Together in London Programme, a partnership of the King’s Fund, the Sainsbury Centre and the Department of Health. A chapter discussing ‘assertive outreach’ suggests all three services use assertive outreach principles, but only one, the Camden and Islington Team, is an assertive outreach team as conventionally understood. The other two, the Antenna Outreach Service in Haringey, aimed at young black people, and the Lambeth Early Onset (LEO) team, exclude those with long term illnesses and interpret ‘assertive outreach’ rather flexibly in practice. Stretching the concept may be useful at a time when new teams are needed to meet targets, but risks loss of meaning. A chapter discusses the role of the services in combating social exclusion, though in a later section on regeneration, the report acknowledges the way current ‘punitive policies’ may work counter to this intention by worsening the stigma of involvement with services. This was clearly a concern to the services themselves.

The strength of the report lies in the description of the development of the actual services, their contexts and their activities. The accounts convince the reader that much of value was achieved, while being frank about difficulties. Many of these (staff recruitment and retention, inability to solve problems with housing, employment and benefits, and difficulty forming partnerships with other stake-holders, including primary care) are common to many London services. In the face of these, even successful service developments may be insufficient to get us ‘out of the maze’.

An evaluation of the services is given, based on stakeholder interviews. The conclusions tend to be cautiously positive, though the real difficulties of developing integrated services in the London context also emerge. More rigorous evaluations of these service models are needed, and two of the services are carrying out randomised trials to be reported separately. A final chapter makes common sense recommendations and this should ensure that the report finds a place as a useful primer for service developers. However, its clear description of a fragmented and complex service context should warn policy makers that there is unlikely to be a single ‘London Model’ to rival Birmingham’s.