The use of predictive modeling to test nursing practice outcomes.

A middle range, multivariate practice model was constructed to test the impact of a perioperative teaching program on multiple patient outcomes: coping, anxiety, recovery pattern, pain, and patient satisfaction. In addition, the influence of such a program on nursing staff was examined. A quasi-experimental, causal modeling approach was used. The operative trajectory of 88 patients and their operating room nursing staff was traced from two pre-operative points through the intra-operative stage to two points post-operatively. The experimental patients (n = 54) evidenced higher satisfaction with their care, a tendency to experience less post-operative vomiting, but also tended to recover more slowly and cope less well than the control patients (n = 34). Multiple regression analysis of the practice model suggested the perioperative teaching program affected patient outcomes, indirectly, through the functioning of the staff nurses intra-operatively.