Chapter 17: Methods for Evaluating Organisation- or Area-Based Health Interventions
Martin Gulliford & Obioha Ukoumunne & Susan Chinn & Jonathan Sterne & Peter Burney & Allan Donner
Methods for evaluating organisation- or area-based health interventions In clinical research, intervention is usually at the level of the individual patient, but in health services research it is often at the level of the health service organisational unit or geographical or administrative area. Areas and organisations represent discrete clusters of individual patients or healthy subjects and evaluation may be difficult. This is because: (i) often only a few clusters may be included in a study; (ii) evaluation may be at the level of the individual subject whilst intervention is at the cluster level; and (iii) individual responses are often correlated within clusters. This chapter discusses the ways that these problems may be addressed in designing and analysing area- or organisation-based evaluations. It is important to include an adequate number of organisational clusters in each intervention group so that the effect of intervention may be distinguished from underlying variability between clusters. ...