Using computational modeling to transform nursing data into actionable information

Transforming organizational research data into actionable information nurses can use to improve patient outcomes remains a challenge. Available data are numerous, at multiple levels of analysis, and snapshots in time, which makes application difficult in a dynamically changing healthcare system. One potential solution is computational modeling. We describe our use of OrgAhead, a theoretically based computational modeling program developed at Carnegie Mellon University, to transform data into actionable nursing information. We calibrated the model by using data from 16 actual patient care units to adjust model parameters until performance of simulated units ordered in the same way as observed performance of the actual units 80% of the time. In future research, we will use OrgAhead to generate hypotheses about changes nurses might make to improve patient outcomes, help nurses use these hypotheses to identify and implement changes on their units, and then measure the impact of those changes on patient outcomes.