Web-based forecast tool helps predict hospital admissions from the cardiac cath lab

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon
 - Physician_computer

Cardiology providers at an urban tertiary care hospital used an Internet-based cardiac catheterization laboratory inpatient forecast tool to help predict daily demands for beds at the hospital. The automated tool extracts data from the catheterization scheduling system and forecasts daily cardiology bed needs.

During the study period, the daily aggregate forecasts were accurate to within one bed for 70.3 percent of days and within three beds for 97.5 percent of days.

Lead researcher Matthew F. Toerper, of the Johns Hopkins department of emergency medicine in Baltimore, and colleagues published their results online in the  Journal of the American Medical Informatics Association on Sept. 5.

They inputted variables such as demographics, scheduled procedures and clinical indicators into a logistic regression model and embedded the model into an Internet-based application connected to the local EMR system.

“This study represents the integration of predictive analytics into a local EMR system that aims to improve the management of healthcare resources (inpatient beds),” they wrote. “Although increasing emphasis has been placed on using data-driven analytics to increase organizational efficiency, a gap still largely exists in integrating these methods into health information technology, especially for real-time decision-making.”

In 2007, an estimated 85 percent of U.S. hospitals provided cardiac catheterization services. Since then, the volume of cardiac catheterization laboratories has decreased, but the variety of procedures has increased, according to the researchers. They noted that different procedures and complex patients make it difficult to predict what patients need after their procedures.

Recently, the U.S. Institute of Medicine, the National Academy of Engineering and the President’s Council of Advisors on Science and Technology recommended that engineers and informaticists develop ideas and tools to improve patient flow. The researchers mentioned that overestimating admissions from the catheterization laboratory may impact patients who need beds, while underestimating admissions may lead to patients recovering in sub-optimal areas or cancellation of the procedure.

Researchers performed this study at a 1,059-bed hospital. The cardiology department manages 52 inpatients beds, including 12 for intensive care patients. In addition, the adjacent catheterization laboratory has nine procedural rooms, a 12-bed preparatory area and a 12-bed recovery area. Of the cardiology admissions, 37 percent come from the catheterization laboratory.

To develop the tool, the researchers analyzed 6,384 catheterization patients between June 1, 2012, and July 1, 2013. They then developed and evaluated the model on an out-of-cohort sample of 7,029 patients between July 2013 and August 2014.

During the 26-month period, 13,292 patients received catheterization procedures: 52.4 percent were admitted after the procedure and 47.6 percent were discharged to home. Of the patients, 86.9 percent were outpatients. A logistic regression model predicted 46.8 percent of outpatients would be admitted.

Patients who were older and male had an increased probability of admission, according to the model.

Further, patients with invasive procedures, congestive heart failure, positive cardiac enzymes or coronary artery bypass grafts were more likely to be admitted, while patients with diagnostic procedures or less severe indicators such as hypertension, hyperlipidemia, angina and smoking were less likely to be admitted.

The researchers cited a few study limitations, including that it occurred at one hospital and that some information was manually inputted and subject to human error. They also mentioned that the forecast tool’s accuracy was tested prospectively.

“In the future, we plan to evaluate the forecast tool’s operational impact on bed management decision-making and patient flow,” they wrote.