A common cause of preventable harm is the failure to detect and appropriately respond to clinical deterioration. Timely intervention is needed, particularly in medically complex (e.g., cancer) patients, to mitigate the effects of adverse events, disease progression, and medical error. This challenging problem requires effective clinical surveillance, early recognition, timely notification of the appropriate clinician, and effective intervention. In the hospital setting, ?failure to rescue? (FTR) is a recognized safety failure. To address FTR, hospitals have introduced new tools and processes (e.g., continuous monitoring, early warning systems, and `Rapid Response' teams). Yet, `death in bed' remains common. The Vanderbilt-Ingram Cancer Center, in collaboration with human factors and systems engineering faculty in the Center for Research and Innovation in Systems Safety (CRISS), as well as faculty in our Schools of Engineering and Management, will create the Cancer Patient Safety Learning Laboratory (CaPSLL). We will partner with surgeons, oncologists, nurses, staff, and adult patients with lung and head or neck cancer recovering from and/or undergoing treatment as outpatients, and their lay caregivers, to more reliably detect and respond more effectively to unexpected clinical deterioration. The details that follow in this proposal are based on our current understandings but will be modified as we employ a systems engineering oriented user-centered design (UCD) process to analyze, design, develop, implement, and evaluate innovative tools and processes to address this complex patient safety problem. We will achieve this through three Specific Aims: 1) To create and refine software tools and a predictive model for a surveillance-and-response system to prevent harm from unexpected all-cause clinical deterioration in outpatients receiving cancer treatment; 2) To create and refine processes and training that engage patients and their caregivers as active and reliable participants in detecting and reporting potential clinical deterioration. We will apply high reliability organizational (HRO) principles and theories to develop processes and training for the relevant ?team? ? the cancer patients, their caregivers, and the clinicians who need to respond to signals from the surveillance system; and 3) To implement in the operational environment and formally evaluate the integrated detection and response tools and processes. We hypothesize (H1) that this system will decrease the likelihood and severity of unplanned treatment events (UTE; e.g. hospital admission). Further, with the incorporation of a patient/family focused HRO framework, we hypothesize that the system will increase non-routine event (NRE; deviations from optimal care) reporting (H2) and decrease clinician response time (H3). The resulting tools, methods and predictive model will be scalable to other cancer types as well as being generalizable to other institutions and to other high-risk outpatient populations (e.g., heart failure).
To enhance patient safety and care quality in the ambulatory setting, we will develop and evaluate an integrated system of tools and processes to detect and respond to unexpected all-cause clinical deterioration. Data from wearable sensors of patient activity (i.e., steps), heart rate, and location, as well as patient-reported events and outcome measures, will drive an artificial intelligence engine that notifies clinicians of ominous changes in the status of outpatients undergoing or recovering from cancer treatments.
