The Scottish Office defines managed clinical networks as “linked groups of health professionals and organisations from primary, secondary, and tertiary care working in a co-ordinated manner, unconstrained by existing professional and [organisational] boundaries to ensure equitable provision of high quality effective services.” Their report contrasted these with loose networks and suggests that they differ from hub and spoke models in that the interests of the network would dominate those of individual hospitals.^1,2 These networks may be grouped by function (for example, pathology, emergency medicine, critical care) or client group (children), disease (cancer, renal) or specialty (vascular surgery).

Network organisations have several theoretical advantages in terms of their flexibility, robustness, and ability to respond quickly to a rapidly changing environment.³ Formal NHS networks have started to emerge as a way of sustaining vulnerable services and maintaining access where the requirements of training or subspecialisation would otherwise mean complete closure of local services. A network may avoid the need to withdraw clinicians who form important parts of other services—for example, overcoming the problem of how to maintain emergency general services when vascular surgery is centralised.

Networks offer a way of making the best use of scarce specialist expertise, standardising care, improving access, and reducing any “distance-decay” effects that can result from the concentration of specialist services in large centres. They can create systems that ensure patients receive a standard investigation and are referred on rather than being held in a local service that may not have the full range of expertise. As a result, networks should be able to exploit any relationships between quality and volume and enable a faster spread of innovation. This appears to have been an important reason for the growth of networks in cancer and coronary heart disease.

The cancer networks have started to report significant benefits as a result of being able to focus on the needs of their patients without the distraction of managing the less patient focused parts of the system such as non-clinical support services. In critical care, networks have been used to increase efficiency and responsiveness by combining scarce resources to iron out the effect of variations in demand.

Networks may stimulate creativity and innovation by providing increased opportunities for interaction of people from different disciplines and organisations. Because of their flexibility they can also create an environment that allows self organisation, development and learning—features which seem to be related to improved outcomes and staff retention. Large networks are more likely to cover the large populations needed to support the different disciplines and expertise required for research and training in an increasingly competitive global market.^4,5

A real attraction of networks is that they focus on clinical issues and create organic and flexible organisations that can respond well to a changing environment. This and their collaborative nature seem to appeal to many clinicians. These positive features, however, that can put them directly at odds with the organisations in which their members sit. Who decides about a consultant appointment, drug formulary, or operational policy—the trust or the network? Who should be accountable for the clinical governance of network members? In fact pragmatic answers can be found, but many networks will have some difficult encounters over these and other issues and will need strong clinical and managerial leadership to deal with them.

A more hazardous possibility is that networks will be seen as the next structural panacea and turned into new NHS organisations. For some services there may be significant advantages from shifting the managerial focus away from institutions and towards services for patients. But this could destroy the creativity and fleetness of foot that networks can develop. The idea of putting all services into networks is also starting to emerge. If this helps to build better relationships and improve the flow of information the results could be beneficial. If it is simply a structural change it could create a model very similar to that adopted for the privatised railways, with the attending problems of competing priorities, a lack of connection between the parts, and confusion about responsibilities.