An integrative review and meta-analysis of oncology nursing research: 1981-1990.

The systematic accumulation of widely scattered research results is essential to developing a scientific basis for clinical practice. The purposes of this study were to describe 10 years of patient-related oncology nursing research in the United States, assess effectiveness of interventions, and test a prototype for research synthesis projects. Using a nursing framework and strategies of integrative research review and meta-analysis, 428 studies were selected for the study. The studies were found to be predominantly descriptive (90%), with convenience samples (79%), and a median sample size of 48. Subjects were predominantly in treatment or rehabilitation (52%), with demographic and stage of disease characteristics largely unreported. Of the 428 studies, 42 met the criteria for meta-analytic treatment. Interventions were found to be significantly effective, with membership in the intervention group improving success rate by 22%. The prototype was judged highly for adherence to standards and usefulness for adaptation to other research syntheses. Because of wide diversity in topics studied, missing information on subjects, and lack of replication, implications for research-based practice are premature at this time. Recommendations for future research include new studies, replications, and new research syntheses with an international scope for nurses studying and providing cancer care.