OBJECTIVES: Cancer-related fatigue (CRF) is a frequently reported complaint in cancer patients and survivors. The Perform Questionnaire (PQ) is a recently validated scale to assess perceptions and beliefs about CRF. This study aims to determine how sensitive to change is the PQ as well as to compare it with the sensitivity of the FACT-F. METHODS: An observational and longitudinal multi-centre study was carried out on a sample of cancer patients with a moderate level of CRF. PQ and FACT-F were administered at inclusion and 3 months later, as well as sociodemographics and key clinical indicators. Patient improvement or worsening related to CRF was assessed by means of a health status item (HSI) self-administered at the second visit on a Likert-type ordinal scale with 13 response options. RESULTS: Baseline patient characteristics (n=437) were: 60.5% women, mean age 59.1 years, an average of 2.21 years since diagnosis, 33.6% breast cancer, 54.7% with metastasis, Karnofsky mean score 80.9, and 29.1% with anaemia. Of the 350 patients who assessed their change with HSI: 208 (59.4%) reported improvement (‘slightly' to ‘greatly'), 84 (24%) reported worsening (‘slightly' to ‘greatly'), and 58 (16.6%) reported no significant change. The overall PQ score showed a better sensitivity to clinical deterioration (effect 1.04) than to the improvement (effect size=0.57), similar to the magnitude of the effect sizes obtained with FACT-F (0.91 for deterioration; 0.53 for improvement). The effect sizes of PQ dimensions were also higher for the patients reporting worsening (ranging from 0.92 to 1.06) than for those reporting improvements (0.51 to 0.59). CONCLUSION: The score of PQ has demonstrated a good level of sensitivity both in patients reporting improvement and in patients reporting deterioration of health status, in a similar magnitude than FACT-F. PQ could be helpful in monitoring cancer patients from clinical research or practice.