Developing and Evaluating User-Designed Data Displays

Hospitalized children with respiratory disease are commonly monitored with continuous pulse oximetry and heart rate-respiratory rate monitors. These data streams generate >4,000 unique data points each patient-day, yet only a tiny fraction are used to inform care decisions. Failure to adequately summarize this large amount of data for clinicians may result in suboptimal care because clinicians may miss important data signals and may under- or over-react to individual data points. In children hospitalized with respiratory disease and in need of supplemental oxygen, there are a number of care decisions, currently made without adequate data, which could be informed by intelligent data visualization tools. This study has employed user-centered design to develop data displays that inform nurses' and respiratory therapists' decision-making in supplemental oxygen delivery. The investigators are now evaluating the effectiveness of these displays in the clinical care of patients with two common respiratory conditions-infants with bronchiolitis admitted to the general pediatrics ward and preterm infants requiring supplemental oxygen who are cared for in the neonatal intensive care unit. By reducing patient's time on supplemental oxygen and improving time with optimal oxygen saturations, this work has the potential to lead to a breakthrough innovation that improves both outcomes and value.

The goal of the study is to evaluate the effectiveness of using user-centered designed data displays in the clinical care of patients admitted to Cincinnati Children's Hospital with respiratory illnesses in two clinical settings. In the general pediatric floor the investigators will evaluate the effectiveness of user-centered designed data displays in the safe and efficient weaning of oxygen in general pediatric patients hospitalized with bronchiolitis. In the NICU the investigators will evaluate the effectiveness of user-centered designed data displays in the accurate targeting of oxygen saturation levels among premature infants. The investigators will address these aims by conducting parallel studies on the Cincinnati Children's Hospital hospital medicine (HM) unit and in the NICU comparing the data visualization application to standard of care. The investigators will examine intervention effectiveness using an adaptation of a quasi-experimental, equivalent time series design with randomization of intervention and control periods. The data visualization will be applied at the unit level and directly affect all the nurses and patients cared for in those units.

During intervention periods staff in both units will have access to the data visualization tool in the electronic health record.

During control periods, the staff will not have access to the data visualization and will continue to use standard of care electronic health record and monitor data to understand oxygen status and trends.

The data visualization contains current and recent (over last 4, 8 or 12 hours) pulse oximetry readings and trends as well as the monitor alarm limits and most recently recorded supplemental oxygen content and flow.

Inclusion Criteria: patients with bronchiolitis (HM) patients with supplemental oxygen need during hospitalization (HM) patients <12 months of age (HM) patients <32 weeks gestational age (NICU) patients with consistent supplemental oxygen need in 48 hours prior to study entry (NICU) Exclusion Criteria: patients with supplemental oxygen need prior to admission (HM) hospitalization occurs during the change from intervention to control periods (HM) patients with congenital anomalies of the lung (NICU) patients who are not in a weaning mode (e.g. on pre-discharge low flow oxygen) (NICU)