ABSTRACT
An assumption behind the designs of the previous chapters is that the response of each patient becomes completely available at a single time point. Almost all research on design construction is for such responses, whether they are instantaneous or delayed, univariate or multivariate. But the responses of many trials are longitudinal, consisting of repeated measurements over time on each patient. An example is the asthma prevention study described in §0.6.5. After treatment, the individual may be examined once a week over a period of four weeks for the assessment of ‘asthma’ status, giving a sequence of up to four binary responses. A similar situation occurs in the treatment of patients having rheumatoid arthritis described by Biswas and Dewanji (2004c), where patients are treated and followed up over an extensive period, again giving rise to longitudinal responses. In this chapter we describe the construction of longitudinal response-adaptive designs, mainly for these repeated binary responses, whilst keeping the possible covariate information in mind.
