styles grieved more (p < 0.003) and felt it was wrong to sell the old hospice (p < 0.05). Both anxious and avoidant staff with higher scores were less likely to want to move (p < 0.05). Conclusions and Applications to Hospice practice Our data suggest that attachment style is stable despite the stress of working in a hospice environment. As in other workplaces anxiously attached personnel have predictably more negative emotional responses to life events. This is useful information for staff care and suggests some groups of staff may benefit from targeted clinical supervision during periods of great change.