Purpose: This paper aims to report the results of a replication study of perceived managerial and leadership effectiveness within a Romanian public sector hospital, and to discuss the extent to which they are similar to and different from findings from equivalent studies carried out in two British NHS Trust hospitals. Design/methodology/approach: Concrete examples (critical incidents) of effective and ineffective managerial behaviour were collected using Flanagan's critical incident technique (CIT). The critical incidents were content analyzed to identify a smaller number of behavioural statements (BSs). These were then compared and contrasted against two British BS data sets using realist qualitative analytic methods, and deductively coded and sorted into extant behavioural categories. Findings: A total of 57 BSs were identified of which 30 were examples of effective and 27 of least effective/ineffective managerial behaviour. The multi-case/cross-nation comparative analysis revealed high degrees of commonality and relative generalization between the Romanian and British findings. Research limitations/implications: Data saturation may not have been achieved during the CIT collection phase of the study. The relevance and transferability of the findings to other public sector hospitals in Romania have yet to be demonstrated empirically. The results have potential as "best evidence" to inform and shape "evidence-based HRD" initiatives designed to train and develop effective managers and leaders within the health services sector of Romania and the United Kingdom. Originality/value: The study is a rare example of indigenous managerial behaviour research in a non-Anglo country. The results lend strong empirical support for universalistic explanations of the nature of perceived managerial and leadership effectiveness. (Contains 2 tables.)