Cancer Patient Safety Learning Laboratory (CaPSLL): Preventing Clinical Deterioration in Outpatients

Principal Investigator: Matthew Weinger, M.D., Vanderbilt University Medical Center, Nashville, TN 
Co-PI: Daniel France, M.P.H., Ph.D., Vanderbilt University School of Engineering, Nashville, TN
AHRQ Grant No.: HS26616
Project Period: 09/30/18-09/29/23
Description: The overall goal of CaPSLL was to improve detection and response to clinical deterioration in cancer outpatients, who often suffer the effects of adverse events and disease progression.¹

The specific aims were to:

1. 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. Create and refine processes and training that engage patients and their caregivers as active and reliable participants in detecting and reporting potential clinical deterioration.
3. Implement in the operational environment and formally evaluate the integrated detection and response tools and processes.

CaPSLL used a human-centered integrated approach to successfully predict cancer outpatients’ 7-day risk of experiencing unplanned treatment events (UTEs), such as readmissions, emergency department visits, and cancer therapy changes. To start, lab researchers observed 100 patient encounters over 80 hours and conducted 17 interviews with oncologists to understand ambulatory cancer care and the processes of detecting deterioration. 

This approach led to using the following technological mechanisms, which make up the CaPSLL surveillance-and-risk prediction system (one of only a few such systems for cancer patients)^1-3: 

• Fitbit wristwatches to collect vitals and activity data from patients (e.g., calories, sleep, steps walked, resting heart rate).
• Smartphone app for patients/caregivers to answer questions weekly (i.e., patient-reported outcome measures [PROMs]).
• Google Maps data from smartphones to monitor patients’ location.
• Clinical data from the electronic health record (EHR).

Of all the data methods, the PROMs and clinical EHR data (used together) were the most reliable in predicting the 7-day risk of UTEs among 50 patients undergoing ambulatory cancer treatments over 3 to 6 months. The Fitbit and Google Maps methods were less successful due to several barriers, such as the COVID-19 pandemic, the lengthy process of educating patients and setting up devices, and the burden of daily device management (e.g., wearing, charging, and synching smartphones). Despite the barriers, the accessibility and affordability of these technological tools make them, according to the researchers, worthwhile in monitoring cancer outpatients; however, more research needs to be conducted.^1,3

An EHR-integrated risk communication system prototype was also developed to communicate predicted risk scores to oncologists and nurses. The prototype’s usability evaluation showed healthcare professionals liked its high-level dashboard, which allowed them to quickly scan their patient panel to identify patients needing immediate care. Healthcare providers also found the visibility of patients’ weight change over time and symptoms (e.g., nausea, swelling) useful in predicting UTEs. 

After postintervention interviews of 25 physicians and patients, common barriers found were workload and work balancing, lack of a care management system, limited family caregiver support, and suboptimal systems. Facilitators included patient engagement and development of trust, formal and informational huddles, multiple communication channels for patients and physicians, quick response to deterioration, and ancillary staff support.¹

To date, this PSLL’s work has resulted in at least three peer-reviewed journal publications that have been cited 139 times in other publications, along with 20 scientific abstracts or posters at local and national conferences during the project period.

Publications

2024

• France DJ, et al. Development and validation of a surveillance and risk prediction system for clinical deterioration in ambulatory cancer. ACI Open 2024 Sep.

2023

• Anders S, et al. Look before you leap: insights on the implementation of AI across healthcare settings. Proc Hum Factors Ergon Soc Annu Meet 2023;67(1):589-592.
• Salwei ME, et al. Barriers and Facilitators to Resilient Cancer Care. AHRQ Patient Safety Learning Laboratories Annual Meeting. Rockville, MD, September 2023.
• Slater H, et al. Implementation of Passive Surveillance Techniques for Cancer Outpatients Using Activity Monitors and Smartphones: Challenges and Mitigation Strategies. AHRQ Patient Safety Learning Laboratory Networking Conference (virtual). Rockville, MD, February 2023.
• Slater H, et al. Implementation of Passive Surveillance Techniques for Cancer Outpatients Using Activity Monitors and Smartphones: Challenges and Mitigation Strategies. Vanderbilt University Medical Center, Anesthesiology 18th Annual Research Symposium. Nashville, TN, May 2023.

2022

• Salwei ME, et al. Preventing clinical deterioration in cancer outpatients: Human centered design of a predictive model and response system [abstract]. JCO 2022;40(16):e13567.
• Salwei ME, et al. Understanding Barriers and Facilitators to Resilient Cancer Care. American Medical Informatics Association Annual Meeting. Washington, DC, November 2022.
• Scheer E, et al. User-Centered Design and Implementation of a Clinical Deterioration Risk Prediction Tool for Providers of Cancer Outpatients. Human Factors and Ergonomics Society (HFES) 66th International Annual Meeting. Atlanta, GA, October 2022.
• Slater H, et al. Implementation of Passive Surveillance Techniques for Cancer Outpatients Using Activity Monitors and Smartphones: Challenges and Mitigation Strategies. International Society for Research on Internet Intervention’s (ISRII) 11th Scientific Meeting. Pittsburgh, PA, September 2022.
• Vasquez E, et al. Role of Network Analysis To Inform the Design of a Clinical Deterioration Response System for Cancer Patients. International Symposium on Human Factors and Ergonomics in Health Care. New Orleans, LA, March 2022.

2021

• Anders S, et al. Reducing Unplanned Treatment Events in Cancer Outpatients. International Forum on Quality and Safety in Healthcare. Europe (virtual), June 2021. 
• France DJ, et al. Using Fitbit data to predict clinical deterioration and unplanned treatment events in cancer outpatients. JCO 2021;39(15):e13560.
• Salwei ME, et al. Barriers and Facilitators to Resilient Cancer Care. Vanderbilt Translational Research Forum. Nashville, TN, November 2021.
• Salwei ME, et al. Engaging Patients With Health IT for Resilient Cancer Care. International Symposium on Human Factors and Ergonomics in Health Care. April 2021. 
• Salwei ME, et al. User Centered Design of a Clinical Deterioration Response System for Outpatient Cancer Patients. AMIA 2021 Annual Symposium. San Diego, CA, October-November 2021. 
• Salwei ME. Cancer Patient Safety Learning Laboratory (CaPSLL): Preventing Clinical Deterioration in Outpatients. Vanderbilt Department of Anesthesiology 2021 BH Robbins Research Symposium (virtual). Nashville, TN, May 2021. 
• Salwei ME. Design of a Clinical Deterioration Response System for At Risk Cancer Patients. National Library of Medicine Informatics Training Conference (virtual). Seattle, WA, June 2021. 
• Salwei ME. Designing Resilient Cancer Care. Vanderbilt Department of Biomedical Informatics Seminar Series (virtual). Nashville, TN, October 2021. 
• Salwei ME. Human Centered Design of a Clinical Deterioration Response System for Cancer Outpatients. Vanderbilt Department of Biomedical Informatics Research Forum (virtual). Nashville, TN, May 2021. 

2020

• Beebe AC, et al. Developing a Decision Support System To Detect and Enhance the Response to Clinical Deterioration in Patients Receiving Outpatient Care for Cancer. Society for Technology in Anesthesia Annual Meeting. Austin, TX, 2020.
• Cinar P, et al. Safety at the time of the COVID-19 pandemic: how to keep our oncology patients and healthcare workers safe. J Natl Compr Canc Netw 2020;18(5):1-6.
• Patterson ES, et al. Predicting the future: diverse perspectives on the transformation of healthcare delivery over the next 15 years. Proc Hum Factors Ergon Soc Annu Meet 2020;64(1):671-674.

References

1. Weinger MB. Final Report: Cancer Patient Safety Learning Laboratory (CAPSLL): Preventing Clinical Deterioration in Outpatients. Nashville, TN: Vanderbilt University Medical Center; 2023, pp. 1-20.
2. Anders S, et al. Patient Safety Learning Labs: what are we actually learning. Proc Hum Factors Ergon Soc Annu Meet 2020;64(1):593-597.
3. Salwei ME, et al. Preventing clinical deterioration in cancer outpatients: human centered design of a predictive model and response system [abstract]. JCO 2022;40(16):e13567.