Often the decision of whom to interview within a study is crucial to its ultimate success or failure, but when the study is to assess a very specific subset of health professionals, this decision may be even more vital.

One avenue that has been proposed within the AiMS Study, is to make use of the knowledge that the group of interest inherently possesses, the names of others working with the specific patient group. Using the tools of network analysis, the frequency and linkage of the names given will be analysed to assess who are the key or central persons whom the study should recruit for interview. Techniques will include degree, betweenness and eigenvector centrality, as while as core-periphery structure analysis of the network graph. The benefits of this approach are that rather than interviewing a large or random sample of health professionals, the core individuals are approached, reducing the number of interviews to be conducted, and in theory, conducting interviews with those having greater knowledge of the specific patient group. The protocol, ethical considerations and success of such a stratum for recruitment will be discussed.