Background: Fatigue is a common and often debilitating stroke sequela, and it is important to accurately define and detect post-stroke fatigue. Often questionnaires are used but a case definition has been developed and proposed as a better tool. Objectives: The aim of the study was to determine validity and inter-rater agreement of the case definition of post-stroke fatigue, and to determine optimal cutoff scores for marked fatigue on the Multidimensional Fatigue Inventory-20 and the Fatigue Severity Scale-7 questionnaires. Methods: Stroke patients were interviewed with the structured interview schedule for the case definition and asked to complete the two questionnaires. To examine the inter-rater agreement of the case definition a second interviewer did another interview blinded to the result of the first interview. Results: Seventy patients were enrolled, 44% women. The median age was 74 years (interquartile range: 67–80) and the median time from stroke to interview was 8 days. The median Fatigue Severity Scale-7 score and the median Multidimensional Fatigue Inventory-20 (General Fatigue subscale) score were higher in the case definition positive than in the negative group (p <.001). The kappa value for the inter-rater agreement was 0.63. A cutoff score of 4.9 for the Fatigue Severity Scale-7 and a cutoff score of 12 on the Multidimensional Fatigue Inventory-20 were optimal to identify marked fatigue according to the case definition. Conclusions: The case definition was valid and had a substantial inter-rater agreement. A score ≥ 5 using the Fatigue Severity Scale-7 or a score ≥ 12 using the Multidimensional Fatigue Inventory-20 (General Fatigue subscale) may be used to detect potentially debilitating post-stroke fatigue in stroke survivors.