This study was a secondary analysis of an existing data set. The purpose was to explore the problem-solving cognitive structure that nurse practitioners (NPs) used in their assessment and management of a patient with acute pneumonia superimposed over chronic bronchitis, as measured by a patient management problem with 296 items. Results of a factor analysis suggested that the activity of patient education was different from assessment and medical management functions. A discriminant analysis was used to explore which variables correlated with high and low performance on the simulation. Laboratory proficiency scores were the first, patient education management the second best discriminating variables. Routes through the simulation were labeled typical, suggesting the usual sequence of history, physical, laboratory tests, and management, and atypical, suggesting variations on this route. The typical group performed significantly better than the atypical group. At the end of each section within the simulations, subjects were asked to list potential problems and possible medical diagnosts. Subjects were able, with a minimal amount of data, to list correct problems and diagnosis and appeared to proceed through the simulation attempting to confirm or deny those hunches. These findings support the concept that there appears to be a nursing focus withing NPs' performance on a clinical simulation.