Large-scale controlled trials such as the CPPT offer an excellent opportunity to test the effectiveness of adherence procedures in small controlled substudies. Such studies should be of the additive variety (i.e., adding a new procedure to the ongoing program so that adherence will not be diminished in the experimental group), thus protecting the power of the main study. Procedures that enhance adherence can then be applied trial wide. In multicenter trials the generalizability of successful strategies across centers can be examined, exploiting a relatively unique situation. Thus, research and clinical application can proceed hand in hand.
The thesis of this introduction to the chapters on compliance intervention is that factors at multiple levels of the medical system and the patients life affect compliance with the medical regimen. This is further complicated in the pediatric domain by the likelihood of differential interactions among the infant, child, or adolescent, and their families as a consequence of different stages of maturation. Very different sets of behaviors will be associated with compliance problems at different ages. A comprehensive approach to compliance maintenance and remediation demands a consideration of the effect of each of these levels on the patient behaviors associated with compliance. As exemplified by the CPPT, these included the national level in terms of the NIH program office involved with the trial, because this office set policy and maintained funding priorities that could affect adherence by, for example, altering staffing levels in a particular way. At the next level were the physicians involved in the trial, particularly the LRC and trial directors. These key investigators needed to be aware of the technical problems involved in compliance and the potential solutions to such problems, so that they could adapt their clinic staffing to meet the demands involved in maintaining compliance. At the next level came the organization of the LRC, whether this involved a particular staffing pattern, or the provision of an adequate environment for the participant, or of access to computers for the adherence staff. The latter staff were also key individuals over the duration of the trial. The critical decision here was to adopt an adherence counselor model, to provide national and later local training for such counselors, and eventually to provide certification. The development of a data-oriented decision-making system and the use of specific counseling strategies were important developments at this level. All this provided the framework for the important continuing relationship between the counselor and participant, and the participant's family.
This chapter was completed while I was a Fellow at the Center for Advanced Study in the Behavioral Sciences. I am grateful for the support provided by the John D. & Catherine T. MacArthur Foundation.