The status of quality of life research in oncology is assessed, and priorities for future research with regard to conceptual and theoretical developments, focus and content of research, research designs and practical strategies for research implementation, and transferring information to clinical practice and medical policy decision-making are identified. There is general agreement that quality of life is a subjective and multidimensional construct, yet comprehensive theoretical models have not been developed and applied fully. We recommend that future research be based on conceptual models that explicate the interrelationships among quality of life domains throughout the stages of cancer care. These models, and the longitudinal research that follows from them, should attend specifically to cross-class and cross-cultural issues to avoid overgeneralization from theory and research that are based largely on the views of the majority culture. We encourage the inclusion of this theory-based quality of life assessment as a standard component of clinical trials. Success in this endeavor will require additional standardization of quality of life measures for use across a range of cancer patient populations, including the development of age-specific norms and instruments designed to assess the entire family system.

Poor drug compliance may cause a decreased survival of children with ma-lignancies. Children who fail to take their medications are not receiving optimum amounts of chemotherapy and suboptimal therapy causes a shortened survival in children with cancer. This is a study of prednisone compliance in 52 children with cancer during three distinct phases of therapy. The patients were either known to be taking prednisone (on-therapy group), off prednisone (off-therapy group), or their compliance was unknown (unknown group). Evaluation of prednisone compliance was attempted by measuring hemoglobin level changes, weight changes, and random urinary 17-ketogenic steroids. The results obtained show that while hemoglobin and weight changes are not helpful, a random urine 17-ketogenic steroid assay is able to differentiate clearly those patients who are taking their prednisone. By the use of this assay it was found that 33% of patients who by protocol and instruction were supposed to be receiving prednisone were not complying. Separate analysis of adolescents revealed an even more alarming 59% noncompliance rate. This striking level of noncompliance strongly suggests that the survival of patients may be threatened by noncompliance. Cancer 43:169-173, 1979. OPULATIONS OF CHILDREN with acute leu-P kemia that have been treated with the same drug regimens have shown different induction rates and remission lengths.2,6 One possible explanation for these varying results may be the number of patients who fail to take all of their prescribed medications. Studies of drug compliance in various other disease states have documented that therapeutic regi-mens are not always followed completely. Compliance is known to be related to the severity of the illness, the complexity of the treatment, the duration of therapy, and the severity of the patient's symptom^.',^ Children with cancer, when in remission, are indeed on prolonged, complex therapy while being relatively asymptomatic and poor drug compliance might be expected. This study has evaluated prednisone com

The financial burden of cancer treatment is a major source of anxiety for the families of pediatric cancer patients. Parents of these patients report that nonmedical, out-of-pocket expenditures are the most troublesome because, unlike medical bills, nonmedical costs must be paid immediately and are rarely reimbursed. Data on nonmedical expenditures (transportation, food, lodging, clothing, family care, and miscellaneous) were collected from 70 patients' families for one-week periods at three-month intervals. Multiple regression analysis was carried out to ascertain the factors influencing costs. Factors contributing to nonmedical costs included level of care, the patient's Performance Status, distance from the treatment center, and family size. Thirty-two families reported loss of pay when one or both parents accompanied the child to the hospital. When this loss was combined with the nonmedical costs, the results indicated a serious strain on the family budget. For half the families surveyed, the total expenses plus loss of pay amounted to more than 25% of the weekly family income.