Treatment preference, adherence and outcomes in patients with cancer: literature review and development of a theoretical model

Objective:A patient's preference may guide their behavior and influence their willingness to take medication or undergo treatment affecting outcomes, such as health-related quality of life, or survival. The importance of understanding patient preferences within oncology is unclear and few adherence studies exist compared with other therapeutic areas.Research design and methods:This study was designed to review the literature regarding patient preferences, adherence and their link to outcomes specifically in the oncology setting and to propose a theoretical model. An in-depth review was conducted, using Embase, MEDLINE and Cochrane Library databases to search for published data examining patient preference, adherence and oncology-specific outcomes, from 1982–2012. Articles were reviewed independently by two authors and rated for relevance and quality. Information from high-quality articles and discussion with oncology and patient preference experts were used to identify associations between important individual concepts as a basis for a theoretical model.Results:In total, 1362 abstracts were identified. After removal of duplicates and initial review, 1269 were excluded and 93 reviewed in detail. Of these publications, 18 were deemed 'high-quality' and used to develop the final model. Variables associated with patient preference, adherence and outcome were identified. External variables included communication, treatment and mode of administration; patient beliefs and values were identified as cognitive variables; and adherence was attributed as a behavioral variable. Relationships between patient preference, adherence and clinical outcomes were established. Adverse events had a strong relationship with adherence; patient beliefs and values were identified as having a moderating effect on adherence. Adherence behavior had a direct relationship to outcomes.Conclusions:Improving our understanding of patient preference may improve clinical outcomes in oncology patients. Although the proposed theoretical model is limited, it provides a basis to develop testable hypotheses for the relationships between patient preference, adherence and outcomes specific to oncology.