BACKGROUND: Fear of recurrence (FoR) following head and neck cancer (HNC) treatment is a major patient concern but is infrequently discussed in outpatient settings and may cause significant detrimental effect on patient psychological well-being. AIM: The aim of this paper is to determine longitudinal trends and predictors of significant FoR. METHOD: Prospective data collection was obtained in 189 post-treatment HNC patients with 456 clinic attendances from October 2008 to January 2011. All patients completed Patient Concerns Inventory (PCI), University of Washington Quality of Life Questionnaire version 4 and the FoR questionnaires using a combination of paper and touch-screen technology. RESULTS: The prevalence of significant FoR was 35%, which was stable longitudinally. Logistic regression-delineated predictors of significant FoR were young age (<65 years), anxiety/mood and patient-reported FoR. Conventional prognostic factors including standard clinicopathological factors, treatment modality and length of follow-up did not predict significant FoR. Over time, 20% of patients reported significant FoR intermittently, whereas 30% experienced it consistently. The remaining 50% did not experience significant FoR. DISCUSSION: Patient-related characteristics were more important than clinicopathological prognostic factors in predicting significant FoR, potentially creating a mismatch between the doctors' perception of patient's experience of FoR with what patients may experience. In the long term, FoR is expressed across a spectrum extending from 'normal' FoR to persistent significant FoR with an intermediate category of intermittent significant FoR. Those patients experiencing intermittent and consistent levels of significant FoR may benefit from psychological support. This study confirmed that self-reported screening FoR using the PCI is a valuable tool in identifying significant FoR.