| Description |
Computerized provider order entry (CPOE) is a component of electronic health records (EHR) that has been touted as a crucial means to support healthcare quality and efficiency. The costs of EHR implementation can be staggeringly high, and little literature exists to verify the hypothesized benefits of CPOE and EHRs. The purpose of this study, based on Coyle and Battle's adaptation of the classic Donabedian quality improvement framework, was to evaluate system-wide outcomes after CPOE implementation in a large academic setting. The specific aims were to describe the association between CPOE implementation and (1) mortality rate and (2) length of stay (LOS), controlling statistically for antecedent, structure, and process variables. The study used hierarchical linear modeling to analyze clinical and administrative data from 2.5 years before and 2.5 years after CPOE implementation. Aim 1 analysis included 104,153 hospital visits and aim 2 analysis included 89,818 visits. Two models were created for each analysis, (a) a model with individual patient care units as the unit of analysis and (b) a model with units aggregated by type. LOS decreased 0.9 days per visit in all models. Mortality decreased 1 to 4 deaths per 1000 visits, depending on the model; or 54 to 216 patient lives saved in the postimplementation period. Significant antecedents were patient demographics, insurance type, and scheduled versus emergency admission; structure variables included patient care unit, private room, and palliative care; and process variables included nursing care iv hours and the number of orders placed. Mortality models were variable by patient care unit, and strongly influenced by confounders such as rapid response team or code activation, suggesting the importance for future studies to account for those influences. CPOE was statistically associated with clinically significant improvements in the system-wide outcomes. Controlling statistically for antecedent, structure, and process variables, the analysis found that after the implementation of CPOE, there was a decrease in mortality and LOS. Future studies need to determine how CPOE implementation impacts nursing performance and how CPOE influences the effect of new physician resident arrival on patient outcomes. |
