This article examines the relationship between health managers' self-assessed empathy, their leadership behaviours as rated by their staff, and staff's personal ratings on a range of work satisfaction and related outcome measures. Empathy was conceived of as four distinct but related individual dispositions, namely empathic concern (EC), perspective taking (PT), personal distress (PD) and empathic matching (EM). Results showed three empathy scales (EC, PT and EM) were, as postulated, positively related to transformational behaviour (inspiring followers to achieve more than expected). The same three measures, also as expected, showed no relationship to transactional behaviour (motivating followers to achieve expected results) and were negatively associated with laissez-faire leadership (an absence of leadership style). Relationships between empathy scales and outcome measures were selective and moderate in size. Strongest empathy association was evident between the PT scale and most outcome measures. Conversely, the extra effort outcome appeared most sensitive to the range of empathy scales. Where significant relationships did exist between empathy and outcome, leadership behaviour was in all cases a perfect mediator. Whilst not denying the smaller dispositional effects on leadership outcomes, leadership behaviour itself, rather than individual traits such as empathy, appear to be major influencing factors in leadership effectiveness.