for a patient in isolation. In response to the social isolation and boredom experienced by patients with SARS, an attempt was made to provide access to the Internet, telephone, newspapers, television and books. Identifying families’ needs, offering an opportunity to express feelings, and supporting effective coping strategies helped to enhance the families’ sense of competence and control. Pharmacological and behavioural interventions to treat insomnia were used extensively. Support of staff On units that received patients with SARS, the initial reactions of uncertainty and fear of the unknown among the staff were met with immediate clear information in repeated, succinct messages, staff meetings, and provision of appropriate equipment and supplies. Occupational therapists developed a pamphlet identifying signs of anxiety and stress and information about support resources, which was distributed to every nursing unit and program area. Psychiatric staff who were on the units to see patients lingered to chat with staff. Informal individual contacts occurred between psychiatric staff and colleagues in medicine, surgery and administration in which simple gestures of support and advice, for example, about sleep, were offered. When it became apparent that some staff were reluctant to talk about personal concerns with psychiatrists with whom they had working relationships, another psychiatrist offered time to any staff at the request of the nursing unit administrator. This resource was used briefly during a period in which nurses’ anxieties were high after several staff had become ill over a short period. A drop-in support centre in the now-vacant medical ambulatory care clinic was provided immediately and then modified because it was not being used. It was replaced with a drop-in lounge in an open setting with soothing music, comfortable chairs and snacks. Senior staff acted as role models by making use of this support service and bringing others with them. A confidential telephone support line staffed by inpatient psychiatric nurses was set up for all hospital staff and was used particularly effectively by those in quarantine. An informal network of mutual telephone contact and support was arranged by quarantined staff of the intensive care unit. Staff on home isolation who had email access were able to receive all communications from the hospital. Interpretation The psychosocial response to an infectious event of this magnitude is complex. In the SARS outbreak, as in previous outbreaks of disease,5 the hospital’s response emphasized clear communication of directives and disease information and a high degree of collaboration between disciplines. In addition, our experience highlights the importance of leadership during times of crisis, consistent with group psychology theory, which emphasizes the effect of leadership on maintenance of team cohesion.11 Although systematic methods of study and longer follow-up are required to determine the burden of the SARS outbreak on nursing staff, our immediate experience suggests that SARS-unit nurses may have experienced less distress than nurses on other medical wards caring for patients with SARS. This may be because, in contrast to the burden reported previously upon nurses acting as VRE gatekeepers,6 the “gate” was at the hospital entrance, education was institutewide, the SARS-unit nurses had a greater sense of competency and multiple support measures were quickly put into place. The most prominent emotional effects upon patients with SARS were feelings of fear, loneliness, boredom and anger. Patients with SARS worried about the effects of quarantine and contagion on loved ones. They also experienced the psychological effects of physical symptoms, especially anxiety about fever, dysphoria due to nausea and the effects of insomnia on mood and coping. Staff were adversely affected by fear of contagion and of infecting loved ones. Caring for health care workers as patients increased discomfort for many. Uncertainty and stigma were prominent themes for both staff and patients. The validity of these observations is limited by the nonsystematic methods of data gathering and interpretation of data by expert opinion and consensus.