This quasi-experimental study utilizes two control and intervention groups to detemination the effect of stress management program on response to stress in nurses working in neonatal intensive care units. The study setting is neonatal intensive care units in teaching hospitals affiliated to Iran University of Medical Sciences, and the study population is 70 nurses who comprises eligible working in these units. intervention and control groups are randomly selected to avoid likelihood of information leakage. Nurses are selected by convenient sampling method, and written consent is obtained from them. Their demographic data are collected and recorded using a questionnaire. In the pretest, all control and intervention groups complete Stress Response Inventory (SRI), containing seven subscales: tension, aggression, anger, depression, fatigue, frustration, and somatization. McNamara stress management teaching method is used in six one-hour sessions during six weeks. In each session, new topics are presented after reviewing previous ones. In this study, education program is not provided for participants in control groups during intervention. However, after completion of the intervention, all educational materials given to intervention groups are also made available to control groups. Eight weeks later, the questionnaire (SRI) is given to the intervention group nurses and inter- and intra-group results are compared.