telemetry utilization
Inpatient Cardiac Telemetry Monitoring: Are We Overdoing It? Manreet Kanwar, MD, Riyad Fares, MD, Steven Minnick, MD, Howard S. Rosman, MD, and Louis Saravolatz, MD Abstract • Background: To evaluate the effect of educational tools in improving compliance with telemetry monitoring guidelines. • Methods: Retrospective data collection was performed for all inpatients on telemetry in May 2006. Educational interventions to improve physician awareness were performed over the next 4 months and follow-up data were collected. • Results: 416 of 972 patients (43%) admitted in May 2006 were placed on telemetry, of whom only 238 (57%) had a class I/II (appropriate) indication. After interventions, data collection in September 2006 showed a reduction in utilization to 37% (316 of 856 patients; p = 0.03) and an increase in appropriate class I/II indication to 71% (p < 0.001). Average duration of telemetry trended down from 4.3 ± 4 days in May to 3.8 ± 3 days in September (p = 0.18). Patients did not differ with regard to average age, sex, and average length of stay. • Conclusion: Improving physician awareness for standard telemetry guidelines promotes compliance, eases the burden of limited telemetry bed availability, and may lead to health care cost savings.
C
ontinuous electrocardiographic (ECG) telemetry monitoring of patients is used extensively in hospitals throughout the United States. Since its introduction more than 40 years ago, the goals of monitoring have expanded from basic rhythm monitoring to the diagnosis of complex arrhythmias, the detection of myocardial ischemia, and identification of prolonged QT interval [1,2]. In recent years, the demand by physicians that their patients be monitored by telemetry has surged, thereby creating the need for studying the trends of telemetry utilization and introducing ways to reduce inappropriate utilization [1–3]. The American Heart Association (AHA) published a comprehensive position paper in 2004 providing recommendations for best practices in hospital ECG monitoring [2]. Indications for telemetry for arrhythmia, ST segment, and
16 JCOM January 2008 Vol. 15, No. 1
QT interval monitoring were classified into 3 groups: class I (indicated in all patients), class II (indicated in most patients) and class III (not indicated) (Table 1). We conducted a study to evaluate compliance with telemetry guidelines in our hospital and the effect of educational tools on increasing physician awareness of appropriate utilization of cardiac monitoring. It was our hypothesis that there would be an improvement in the appropriate use of telemetry as a result of our interventions. Methods The study was performed in a 608-bed urban teaching hospital with 32,000 annual admissions and 185 nonintensive telemetry beds. Data were collected for all adult patients admitted to nonintensive telemetry beds from the emergency department (ED). Patients admitted to or transferred from intensive care units, pediatric, obstetric, or surgical floors were not included. We initially obtained data for the month of May 2006 as a baseline for comparison before initiating various educational measures from June 2006 onwards to assess the effects of our interventions. Baseline Measures Retrospective data for admitting diagnosis and indication of telemetry use were obtained for the month of May 2006. Data collected included patient demographics (age, sex), admission diagnosis (primary, secondary, or tertiary per ED documentation), duration of hospital stay (calculated from date of admission and discharge), and duration of telemetry (per hospital electronic records). Telemetry indications were divided into AHA telemetry class based on the admitting diagnoses. If the admitting diagnosis per ED documentation included the conditions listed under classes I or II, telemetry was considered appropriate. However, if the condition fell under class III or in none of the categories, it was considered inappropriate.
From the Department of Medicine, St. John Hospital and Medical Center, Detroit, MI.
www.turner-white.com
Original research Table 1. American Heart Association Indications for Telemetry Monitoring Class I: Cardiac monitoring is indicated in most if not all patients in this group 1. Patients who have been resuscitated from cardiac arrest 2. Patients in the early phase of acute coronary syndromes (ST-elevation or non–ST-elevation myocardial infarction [MI], unstable angina/“rule-out” MI) 3. Patients with unstable coronary syndromes and newly diagnosed high-risk coronary lesions 4. Adults who have undergone cardiac surgery 5. Children who have undergone cardiac surgery 6. Patients who have undergone nonurgent percutaneous coronary intervention with complications 7. Patients who have undergone implantation of an automatic defibrillator lead or a pacemaker lead and are considered dependent 8. Patients with a temporary pacemaker or transcutaneous pacing pads 9. Patients with atrioventricular block 10. Patients with arrhythmias complicating Wolff-Parkinson-White 11. Patients with long-QT syndrome and associated ventricular arrhythmias 12. Patients receiving intra-aortic balloon counterpulsation 13. Patients with acute heart failure/pulmonary edema 14. Patients with indications for intensive care 15. Patients undergoing diagnostic/therapeutic procedures requiring conscious sedation or anesthesia 16. Patients with any other hemodynamically unstable arrhythmia Class II: Cardiac monitoring may be of benefit in some patients but is not considered essential for all patients 1. Patients with postacute MI 2. Patients with chest pain syndromes 3. Patients who have undergone uncomplicated nonurgent percutaneous coronary intervention 4. Patients who are administered an antiarrhythmic drug or who require adjustment of drugs for rate control with chronic atrial tachyarrhythmias 5. Patients who have undergone implantation of a pacemaker lead and are not pacemaker-dependent 6. Patients who have undergone uncomplicated ablation of an arrhythmia 7. Patients who have undergone routine coronary angiography 8. Patients with subacute heart failure 9. Patients who are being evaluated for syncope 10. Patients with do-not-resuscitate orders with arrhythmias that cause discomfort Class III: Cardiac monitoring is not indicated because a patient’s risk of a serious event is so low that monitoring has no therapeutic benefit 1. Postoperative patients who are at low risk for cardiac arrhythmias (eg, young patients without heart disease who undergo uncomplicated surgical procedures) 2. Obstetric patients, unless heart disease is present 3. Patients with permanent, rate-controlled atrial fibrillation 4. Patients undergoing hemodialysis (when patients have a class I or II indication and undergo dialysis in the hospital, electrocardiographic monitoring is recommended) 5. Stable patients with chronic ventricular premature beats Reprinted with permission from reference 2.
Educational Interventions Over the next 4 months, we implemented the following educational interventions: In June and July, lectures on telemetry use were given to ED and internal medicine residents and faculty. At the same time, guidelines were made available by reminder e-mail, telemetry indication laminated cards, and by postings in the ED and medicine floor units. In August and September, the same interven-
www.turner-white.com
tions were repeated and extended to nurse practitioners, physician assistants, and unit clerks. In addition, admitting physicians were asked to identify indication for telemetry as part of ordering process. This was done by including a “telemetry order sheet” that stated the AHA guidelines as part of the hospital admission orders. If the admitting resident/physician felt that the patient required telemetry but did not meet the criteria per the AHA guidelines, provision
Vol. 15, No. 1 January 2008 JCOM 17
telemetry utilization Table 2. Comparison of Telemetry Use in May, July, and September 2006 May n = 972
July n = 714
September n = 856
P Value
No. of patients on telemetry
416 (42%)
293 (41%)
316 (37%)
0.03
No. of telemetry patients with class I/II indication
238 (57%)
188 (65%)
226 (71%)
< 0.001
64 ± 17
65 ± 16
63 ± 17
0.16
195 (46%)
142 (48%)
167 (52%)
0.27
Mean length hospital stay ± SD (95% CI), day
5.07 ± 5 (4.6–5.6)
4.53 ± 3 (4.2–4.9)
4.74 ± 4 (4.3–5.2)
0.26
Mean days telemetry use ± SD (95% CI)
4.33 ± 4 (3.9–4.8)
3.97 ± 3 (3.7–4.3)
3.85 ± 3 (3.5–4.2)
0.18
0.89 (0.87–0.92)
0.88 (0.86–0.91)
0.84 (0.81–0.88)
0.03
Mean patient age ± SD, yr Male sex, n
Telemetry days/hospital days (95% CI)
CI = confidence interval; SD = standard deviation.
was given for listing the indication as “other.” Unit clerks were encouraged to call the admitting resident/physician to confirm the telemetry orders prior to placing the patient on the monitor. Further, residents and housestaff were encouraged to discontinue telemetry once the indication had been addressed. For example, if a patient was admitted for a “rule-out acute coronary syndrome” admission diagnosis and placed on telemetry, it was discontinued after 24 to 48 hours once the diagnosis had been confirmed. Data collection was repeated in July and September to compare with baseline data. Statistical analysis was performed using SPSS, version 12.0 (SPSS Inc., Chicago, IL). Chi-square analysis was used to evaluate changes in the proportion of patients on telemetry and on telemetry for appropriate indications (Table 2). One-way analysis of variance was used to compare age, length of stay, and days on telemetry for May, July, and September. Scheff post-hoc analysis was used to evaluate months pairwise for differences. A P value < 0.05 for a 2-sided test was considered statistically significant. Results Of the 972 patients admitted to medical floors in May, 416 patients (43%) were admitted on telemetry, 238 (57%) of whom had a Class I or II indication for telemetry. In July, 293 of 714 (41%) patients admitted were started on telemetry (p = 0.47), of whom 64% had the appropriate indication (p = 0.06). In September, only 37% (316 of 856; p = 0.03) were placed on telemetry, of whom 71% had a class I/II indication. The average duration of telemetry use showed a nonsignificant decline (4.33 ± 4 days in May, 3.97± 3 days in July, and 3.85 ± 3 days in September; p = 0.18), with more patients having telemetry discontinued once the indication was addressed. Also, more patients had inappropriately
18 JCOM January 2008 Vol. 15, No. 1
ordered telemetry discontinued within 24 hours of admission in September (24 vs. 15 in May [p = 0.003] and 19 in July [p = 0.02]). The percentage of telemetry patients with class I/II indications increased from 57% to 71% between May and September (p < 0.001) (Figure). In post hoc analysis, the months of May and September differed for class I/II indications. The average age, gender, and length of stay for these patients across the 4 months was similar. Discussion Analyzing the hospital’s telemetry monitoring needs and making valid decisions about telemetry acquisition requires a basic understanding of current telemetry technology and utilization issues. Several factors have been attributed to the increasing demand for telemetry monitoring: an aging, sicker population of hospitalized patients; higher nurse-topatient ratios on telemetry floors, independent of the need for ECG monitoring; and finally, the often erroneous expectation that telemetry will lead to prompt recognition and timely intervention for life-threatening changes in patients [3]. The role of telemetry in guiding patient management was felt to be overestimated, with significant arrhythmias having lead to urgent intervention in a small fraction of patients only [4]. The Emergency Care Research Institute telemetry utilization survey [5] noted that in hospitals with patient flow problems due to unavailable telemetry beds, 29% of respondents attributed the problem to improper use of telemetry monitoring, 16% to an insufficient number of telemetry beds, and 47% to combination of both factors. Rather than addressing inefficiencies in telemetry utilization, hospitals often acquire more telemetry units. This approach fails to address physician practice patterns or improve awareness of the admission and discharge criteria for telemetry and therefore may not improve patient flow.
www.turner-white.com
Original research On telemetry 75 Class I or II on telemetry
Patients, %
60
45
30
15
0
May (n = 972)
September (n = 856)
July (n = 714) Month
Figure. Percentage of patients on telemetry (white bars) declined (p = 0.03), while the percentage of patients on telemetry who had a class I or II indication (gray bars) increased (p < 0.001).
Health care delivery organizations worldwide are defining and employing clinical quality indicators to identify improvement opportunities, measure the efficacy of interventions, and quantify the link between quality of care and cost-effectiveness [6]. Our study demonstrates that up to 43% of decisions for telemetry are without appropriate indication. Further, telemetry was continued beyond the recommended 48-hour period in most patients with class I/II indications for initiation. The economic and personnel costs associated with the overuse are likely to be significant. In our hospital, a semi-private bed without telemetry is charged at $1209/night whereas a telemetry bed would cost $1439/night. Simple and persistent measures such as educational interventions along with electronic reminders can help reduce the number of inappropriate telemetry admissions based on the AHA telemetry guidelines. Our interventions resulted in patients less often placed on telemetry, but more often when indicated. These interventions should facilitate the transfer of patients from the intensive care unit, operating room, or ED to telemetry floors. Our study has several limitations. We have only shown short-term benefits from intensive educational interventions, which may lose effectiveness over time. We assigned indication class based on the written admitting diagnosis by ED physicians without an independent assessment of
www.turner-white.com
the cardiac history for each patient. We did not aim to study results of change in management or number of adverse events noted as a result of decline in telemetry utilization. In conclusion, telemetry monitoring currently is overused as a health care tool. Increasing awareness about proper utilization of telemetry can increase its appropriate use, improve patient flow, and likely reduce cost. Corresponding author: Manreet Kanwar, MD, St. John Hospital and Medical Center, 22151 Moross Rd., Ste. 126, Detroit, MI 48236,
[email protected]. Financial disclosures: None. Author contributions: conception and design, MK, RF, SM, LS; analysis and interpretation of data, MK, RF, HSR, LS; drafting of the article, MK, HSR, LS; critical revision of the article, MK, HSR, LS; administrative or technical support, SM, HSR, LS; collection and assembly of data, MK, RF.
References 1. Drew BJ, Funk M. Practice standards for ECG monitoring in hospital settings: executive summary and guide for implementation. Crit Care Nurs Clin North Am 2006;18:157–68. 2. Drew BJ, Califf RM, Funk M, et al. Practice standards for electrocardiographic monitoring in hospital settings: an
Vol. 15, No. 1 January 2008 JCOM 19
telemetry utilization American Heart Association scientific statement from the Councils on Cardiovascular Nursing, Clinical Cardiology, and Cardiovascular Disease in the Young: endorsed by the International Society of Computerized Electrocardiology and the American Association of Critical-Care Nurses [published erratum appears in Circulation 2005;111:378]. Circulation 2004;110:2721–46. 3. Pelczarski KM, Barbell AS. Making tough decisions about telemetry monitoring. Hosp Mater Manage Q 1993;15: 1–12.
4. Estrada CA, Rosman HS, Prasad NK, et al. Role of telemetry monitoring in the non-intensive care unit. Am J Cardiol 1995;76:960–5. 5. ECRI and American Association of Critical Care Nurses. Telemetry utilization—time to take a closer look; 1992 Mar; Plymouth Meeting, PA. 6. Curry JP, Hanson CW 3rd, Russell MW, et al. The use and effectiveness of electrocardiographic telemetry monitoring in a community hospital general care setting. Anesth Analg 2003; 97:1483–7.
Copyright 2008 by Turner White Communications Inc., Wayne, PA. All rights reserved.
20 JCOM January 2008 Vol. 15, No. 1
www.turner-white.com
