OBJECTIVES: The aim of the present study was to develop a patient reported outcome (PRO) instrument which would be suitable for use in clinical practice to measure the intensity and impact of cancer-related fatigue (CRF), as well as patients' attitudes and beliefs regarding the condition. METHODS: Questionnaire content was generated from literature review, focus groups with oncology patients, and expert meetings with oncologists and specialists in the production of PRO instruments. Potential items were administered to oncology patients with CRF in a multi-center, cross-sectional, item reduction study. Patients answered all items twice to obtain data on both item frequency and importance using 5-point Likert-type scales. Item reduction was performed using a combination of clinimetric (calculation of impact score by multiplying frequency and importance scores for each item, expert opinion) and psychometric analysis (factor analysis, evaluation of scale internal consistency), and Item Response Theory (IRT) techniques. RESULTS: The initial pool of 75 items was administered to 238 cancer patients (mean age 57 years, 56% women, 30% breast cancer, 64% with metastasis, 46% with anemia). The 35 items with the lowest impact score were eliminated in clinimetric analysis; statistical analyses eliminated a further 15 items, and 13 items were eliminated on the basis of expert clinical opinion, supported by findings from the IRT analysis and item-scale correlations. The final measure includes 12 items. Factor analysis confirmed the presence of 3 dimensions: physical function (4 items), activities daily living (4 items) and beliefs/attitudes (4 items). Cronbach's alpha values for the overall score and individual dimensions were 0.92, 0.78, 0.85, and 0.81, respectively. CONCLUSION: The combination of methods for item reduction has led to the production of a new instrument with 12 items and 3 dimensions, which it is hoped will be suitable to measure aspects of CRF which are important in clinical practice.