Indication Based Ordering of Ecg Monitoring Reduces Inappropriate Utilization

1University of California San Diego Health System, San Diego, CA
2UC San Diego Medical Center, San Diego, CA
3University of California San Diego Health System, San Diego, CA
4UC San Diego, San Diego, CA

Meeting: Hospital Medicine 2014, March 24-27, Las Vegas, Nev.

Abstract number: 132


Overutilization of ECG monitoring (EM) can negatively impact patient flow and increase healthcare costs. In 2004, an American Heart Association (AHA) consensus statement detailed indications for EM in hospital settings. Monitoring was suggested for patients with Class I and II indications, but discouraged for Class III. We hypothesized that “indication‐based ordering”, which requires clinicians to document a reason for EM, would reduce EM orders for Class III indications.


At our university teaching hospital, a multidisciplinary group of key stakeholders adapted the AHA consensus statement into a list of locally accepted indications for EM, with an associated duration of EM for each indication. The computerized order entry system was then revised to require that orders for EM indicate a specific indication. Providers selected from a prepopulated list that included all locally adopted indications; an “Other” indication was also available that required free‐text entry of the indication. An automatic expiration time for EM was included in each order.Chart audits were conducted on 100 patients on the telemetry unit at baseline and 3 months following implementation. Indications for EM were abstracted, and chart notes were used to validate the selected indications in the post‐intervention period. Other outcomes were measured using administrative data from the identical 3 month period in the year before and after the intervention.


At baseline, 44% of patients on EM had a Class III indication, which decreased to 22% following the intervention (p = 0.005). Frequently listed Class III indications included sinus tachycardia related to acute illness that did not warrant EM, alcohol withdrawal, anemia, and stable atrial fibrillation. These indications were similar in both the pre‐and post‐intervention groups. No change in hospital‐wide frequency of non‐ICU cardiac arrests was observed. A reduction of this magnitude equates to an estimated savings of 195 hours/month in nursing time, and $6809/month in staffing costs for a single 24 ‐bed telemetry unit.Mean time on EM increased from 2.22 to 3.80 days after the intervention (p =.0005); median times were lower (2.0 to 2.55), but the increase remained significant even after exclusion of outliers (time > 3SD above the mean). Mean time from admit order to arrival on the unit decreased from 4.93 to 3.73 hours (p < 0.0005).


Indication‐based ordering for EM improved appropriate utilization in this single‐center study, with no apparent increase in adverse events. An associated improvement in ED throughput was noted after implementation, though the intervention was one of several factors that influenced that metric. Auto expiration of EM by indication seems to have paradoxically increased the telemetry length of stay; however this likely reflects a reduction in utilization by short‐stay patients with Class III indications, leading to a cohort of patients more appropriate for prolonged cardiac monitoring.

To cite this abstract:

Popa R, Clay B, Burtson P, Garza A, Seymann G. Indication Based Ordering of Ecg Monitoring Reduces Inappropriate Utilization. Abstract published at Hospital Medicine 2014, March 24-27, Las Vegas, Nev. Abstract 132. Journal of Hospital Medicine. 2014; 9 (suppl 2). Accessed March 28, 2020.

« Back to Hospital Medicine 2014, March 24-27, Las Vegas, Nev.