cost-effectiveness of defibrillation by emergency medical technicians
TRANSCRIPT
Cost-Effectiveness of Defibrillation by Emergency Medical Technicians
JOSEPH P. ORNATO, MD,* EDWARD J. CRAREN, MA,t EDGAR R. GONZALEZ, PharmD,* A. RANDOLPH GARNETT, MD,* BARBARA K. McCLUNG, RN, BSN,* MARY M. NEWMAN, BS*
Effective emergency systems using emergency medical techni- cians (EMTs) trained to defibrillate or paramedics can save more lives fmm out-of-hospital cardiac arrest due to ventricular fibrii- iation than can emergency systems staffed with basic EMTs who cannot defibrillate. This article focuses on the cost-effectiveness of systems staffed with each type of EMT. Data were collected from ail 50 states and from the District of Columbia to determine the number of hours and estimated cost of initial training for the three types of EMTs in the United States in 1986. The median initial training hours for basic EMTs, EMTs trained in defibriiia- tion, and paramedics were 110, 129, and 700, respectively. Me- dian costs for initial training at each EMT level were $123, $150, and $1580/student. According to published survival data for emergency medical systems staffed with EMTs at each level, the total initial training personnel and equipment cost per life saved from ventricular fibrillation was $7687, $2126, and $2289 for systems staffed by the respective EMTs. The initial cost per life saved from ventricular fibrillation is more than three times greater in systems staffed by basic EMTs than in systems staffed by EMTs trained in defibrillation or paramedics. From a medical and a cost-effective standpoint, ail communities served by basic EMTs should consider upgrading them to at least the defibriiia- tion-trained EMT level. (Am J Emerg Med 1988;6:108-112)
Rapid electrical countershock of patients in ventric- ular fibrillation by emergency medical technicians (EMTs) trained to defibrillate can save lives in urban,’ suburban,* and rural settings.3 The cost-effectiveness of systems served by basic EMTs who cannot deti- brillate has not been compared with that of systems served by EMTs trained to defibrillate or paramedics.
From the ‘Section of Emergency Medical Services, Department
of Internal Medicine, Medical College of Virginia, Richmond, Virginia, and the tDivision of Emergency Medical Services, Ne- braska State Department of Health, Lincoln, Nebraska.
Manuscript received 11 February 1987; revision accepted 10
June 1987.
Address reprint requests to Dr. Ornato: Medical College of Vir- ginia, Box 525, Richmond, VA 23298.
Key Words: Cost-effectiveness analysis, defibrillation, emer- gency medical system, emergency medical technician, para- medics, ventricular fibrillation.
0 1988 W.B. Saunders Company
0735-8757188 $0.00 + .25
108
This article compares the average time and cost needed initially to train a basic EMT, an EMT trained to defibrillate, and a paramedic and compares the cost- effectiveness of each of these providers to the com- munity in terms of dollars spent per life saved from out-of-hospital cardiac arrest due to ventricular fibril- lation.
METHODS
The number of training hours and the cost of training EMTs were determined by a mail survey sent to the director of the state division of emergency medical services in all 50 states and the District of Columbia. Telephone follow-up was done when we received no response to the initial questionnaire or when we had questions about the answers provided on the question- naire .
Median values for EMT initial training hours and cost for the United States were determined. Using data from published studies,‘” the average survival to
hospital discharge in systems staffed with each of the three types of EMTs was used to predict how many lives (out of 100 out-of-hospital cardiac arrests due to ventricular fibrillation) would be saved by rescue teams at each of the three levels of training. Survival from out-of-hospital cardiac arrest due to ventricular fibrillation treated by the three types of EMTs was estimated to be 6%,3*536 25%,le3 and 28%,’ respective-
ly. The relative training cost for each prehospital pro-
vider per life saved was defined as the training cost (in dollars) for each provider divided by the number of lives saved out of 100 out-of-hospital cardiac arrests. In states that had different hours of training or differ- ent costs within each category of service, an average number of training hours and an average cost were determined by weighting the sum for the number of EMTs trained in each category in that state. If the training costs could not be accurately estimated by the Division of Emergency Medical Services in a state be- cause no figures were available, training costs for that state were listed as “not available” and that state was not included in the cost analysis.
Emergency medical technicians trained to “intermediate” levels other than to defibrillate were
ORNATO ET AL n COST-EFFECTIVENESS OF EMT-DEFIBRILLATION
not included in the analysis, because there was no con- sistent pattern to the training from state to state and because many states have multiple “intermediate” EMT categories.
Initial equipment cost was defined as the equipment that a prehospital provider needs to care for and trans- port a patient in cardiac arrest. Items included in the analysis were an ambulance, radio equipment, and miscellaneous medical equipment (stretcher, oxygen, first aid supplies) for all categories of EMTs. For EMTs trained to defibrillate and paramedics, a de- fibrillator monitor was included. A UHF radio with telemetry and a cardiac drug box were added to the equipment cost for paramedics only. Initial equipment cost was estimated from 1986 prices of standard am- bulance equipment items obtained from nationwide distributors. The total initial training and equipment cost for each prehospital provider per life saved was defined as the sum of the training cost (in dollars) for each provider plus the equipment cost for that provid- er to function divided by the number of lives saved out of 100 out-of-hospital cardiac arrests.
Because the distributions of training hours and train- ing cost were skewed, calculations were made with the median rather than the mean values. The nonparamet- ric Kruskal-Wallis test was used for statistical analysis.’ A p value of ~0.05 was considered to be statistically significant.
RESULTS
Initial training and cost data for basic EMTs were obtained for all 50 states and the District of Columbia (Table 1). Comparable data for EMTs trained to de% brillate were obtained for the 29 states that have at least one such program.
Median initial training time for EMTs trained to de- fibrillate was 19 hours more than the median initial training time for basic EMTs (Table 2). The median initial training time for paramedics was five to six times greater than the median initial training time for the other two categories of EMTs. The number of ini- tial training hours for paramedics was significantly higher (p < 0.0001) than the number of training hours for the other EMTs.
The median cost of initial training was only $27 higher for EMTs trained to defibrillate than for basic EMTs. The median initial training cost for paramedics was almost 10 to 13 times greater than the initial train- ing cost for basic or defibrillation-trained EMTs. It costs significantly more @ < 0.0001) to train paramed- ics than to train the other two groups of EMTs.
Because complete initial training hour and initial cost data for all three categories of EMT service could be obtained from only 12 states, the analysis was re- peated for only these states. The results were not sig-
niticantly different from those of the aggregate analy- sis for all states. In the 12 states, median initial training hours for basic EMTs, EMTs trained to defibrillate, and paramedics were 110, 127, and 563, respectively; median initial training costs were $110, $145, and $1585, respectively.
With either form of analysis, the conclusion was the same: the initial cost per life saved from out- of-hospital cardiac arrest due to ventricular fibrillation is highest for a system using basic EMTs. Such sys- tems have more than triple the initial setup cost per life saved than systems staffed by EMTs trained to defi- brillate or paramedics.
DISCUSSION
An analysis of the influence of the EMT provider level on cost-effectiveness for treating out-of-hospital ventricular fibrillation is appropriate because survival from this condition is most influenced by provi- sion of advanced life support in the field.’ For initial startup costs, systems using EMTs trained to defibril- late and paramedics are of comparable economic value to the community in treating patients with out- of-hospital cardiac arrest due to ventricular fibrilla- tion. Basic life support systems staffed with basic EMTs save few lives from out-of-hospital cardiac ar- rest. A basic EMT system requires a major investment in equipment and training without providing the res- cuer with the single most effective life-saving treat- ment for out-of-hospital cardiac arrest: electrical coun- tershock.
Wages and fringe benefits for EMTs, dispatch costs, and vehicle maintenance costs were not included in this analysis because of the extreme diversity in each of these variables between volunteer systems and sys- tems staffed by full-time paid providers. It is likely that for each system, wages and fringe benefits will in- crease minimally from basic EMTs to EMTs trained to defibrillate, with a larger increase to the paramedic level. Dispatch and vehicle maintenance costs are rel- atively independent of the level of training of the EMT staffing the vehicles. If these assumptions are correct, the system using EMTs trained to defibrillate may be the most cost-effective of the three in paid provider systems, strictly in terms of cost per lives saved from ventricular fibrillation.
Although current salary information is unavailable for a nationwide comparison, one system has com- pared operational costs between basic EMT and para- medic service. Urban and coworkers’ estimated that it cost $42,358/1ife saved from out-of-hospital cardiac ar- rest in King County, Washington, in 1981 to upgrade its basic life support system to an advanced life sup- port system staffed by paramedics. The analysis in- cluded direct and indirect costs of operating an ad-
109
AMERICAN JOURNAL OF EMERGENCY MEDICINE n Volume 6, Number 2 I March 1988
TABLE 1. Training Hours and Cost by State for Emergency Medical Technicians (EMT)
Training Hours Training Costs ($)
EMT-A EMT-D
Alabama 140
Alaska 110
Arizona 110
Arkansas 120 California 104
Colorado 110
Connecticut 120
Delaware 110
Dist. Columbia 110
Florida 110
Georgia 150 Hawaii 315 Idaho 105
Illinois 126
Indiana 110
Iowa 120
Kansas 120
Kentucky 103
Louisiana 110
Maine 110
Maryland 110
Massachusetts 110
Michigan 115
Minnesota 81
Mississippi 90
Montana 110
Missouri 138 N. Carolina 110
N. Dakota 110
Nebraska 81 Nevada 110
New Hampshire 110
New Jersey 120
New Mexico 105
New York 109
Ohio 90
Oklahoma 124
Oregon 110
Pennsylvania 126
Rhode Island 110
S. Carolina 109
S. Dakota 110 Tennessee 130
Texas 130
Utah 118 Vermont 120
Virginia 210 W. Virginia 109
Washington 81 Wisconsin 110 Wyoming 120
NA 570 122 1110 114 784 NA 720 NA 916
126 910 132 560 126 910 NA 400 NA 710 NA 700 NA 1530 121 825 141 456 NA 1210
136 480 146 1000 NA 603 NA 1070 146 510
118 410 130 1360 125 515
105 1281 NA 890 NA 610 NA 763 NA 528 128 1110
97 496 140 1010
134 1910 NA 920 NA 925 124 494
130 590
140 364 126 536 NA 926 122 340 129 533 NA 610 NA 630 NA 520 134 703 137 1280 NA 663 NA 441 96 881
130 860 130 520
EMT-P -
EMT-A EMT-D EMT-P
NA 223
2585 100 NA 175 NA
45
400 NA
NA 1704
NA NA
115 NA NA
75 140 160 NA
315 NA 200 NA NA
140 NA
90 65
123 105 NA NA 115 NA
90 180
1260 200 NA
80 100 150 NA NA NA NA 120 NA
80
NA
253
NA NA
NA
NA
NA
60
NA
NA
NA
NA
NA NA
NA
NA
NA
NA
NA
184
NA
NA
NA
NA
NA
NA
NA
NA 105 140 135 195 NA NA
135
NA
NA
212
NA
212
NA
NA
NA NA NA
NA
NA NA
150 NA
115
NA NA
26585 1200 NA
1925 NA
4210
7000 NA NA 8294 NA NA
1315 NA NA
800 1190 1360 NA
4215 NA
1950 NA NA
820 NA 1590 2565 1923
10105 NA NA
690 NA
240
795 9260
650 NA 1580
800 1150 NA NA NA NA 2120 NA 1580
ABBREVIATIONS: EMT-A, basic EMT; EMT-D, EMT trained to defibrillate; EMT-P, paramedic
vanced life support system, including paramedic sala- ries, benefits, vehicle dispatch, and vehicle maintenance.
Survival statistics from out-of-hospital cardiac ar- rest are specific to each system. The published surviv-
al figures used for our analysis may be higher than comparable figures noted in other communities, be- cause the published figures come from systems partic- ularly noted for their excellence. The relative cost per life saved among the different levels of service is likely
110
ORNATO ET AL n COST-EFFECTIVENESS OF EMT-DEFIBRILLATION
TABLE 2. Cost-Effectiveness of Emergency Medical Technician (EMT) Services in Treating Cardiac Arrest Due to Ventricular
Fibrillation (VF)
Median initial training hours EMT per
Median initial training cost EMT ($) per
Equipment costs ($)
Ambulance
Medical equipment
VHF radio with HEAR capability UHF radio with telemetry, no repeater
Defibrillator/monitor Drug box and supplies
Total equipment costs Total initial costs
Lives saved 100 VF arrests per
Total initial costs life saved (9) per
EMT-A EMT-D EMT-P
110 129 700 123 150 1580
40,000 40,000 40,000
3000 3000 3000
3000 3000 3000 NA NA 9000
NA 7000 7000 NA NA 500
48,000 53,000 62,500 46,123 53,150 64,080
6 25 28
7687 2126 2289
to be insensitive to local variation in the survival rate, because there is generally a severalfold relative im- provement in survival when basic life support systems are compared with systems that can at least provide rapid defibrillation.
On the basis of this analysis, a strong argument can be made that all rescue vehicles responsible for initial care of patients in cardiac arrest should be staffed with providers who can defibrillate. Economic consider- ations (the cost of training and a defibrillator) strongly favor providing the skill. More lives can be saved from out-of-hospital ventricular fibrillation in an effective system using EMTs trained to defibrillate than in an effective system using basic EMTs.~*~ There is negli- gible risk of harm, provided that there is adequate medical control.“’ Even in systems that have a low frequency of calls, in which a potential continuing ed- ucation problem exists in maintaining EMT defibrilla- tion skills, safety and effectiveness can be ensured by using an automatic defibrillator.“5
Cardiac arrest due to ventricular fibrillation and ma- jor trauma account for the majority of potentially pre- ventable deaths in most emergency medical systems. Adding the skill of defibrillation to a promptly dis- patched basic life support system that serves an edu- cated public will improve survival from ventricular fi- brillation. From a medical and a cost-effective stand- point, all communities served by basic EMTs should consider upgrading to at least the defibrillation-trained EMT level. As a first priority, communities should concentrate on adding the skill of defibrillation for 1) basic EMT units that function without advanced life support backup, and 2) primary-response basic EMT units serving in geographic areas of two-tiered services where the response times of advanced life support units are prolonged.
Communities served by units that provide only basic life support care should strongly consider upgrading to the defibrillation-trained EMT or paramedic levels.
Published data suggest that more lives may be saved with a paramedic system than with a system using EMTs trained to defibrillate, but paramedic systems require more initial training time and setup cost. On- going personnel costs and continuing education re- quirements are greater in paramedic systems than in systems using defibrillation-trained EMTs. From a practical standpoint, some volunteers may not be able to dedicate hundreds of hours or more for paramedic training. Although not considered in this analysis, the paramedic’s additional training and skills in areas other than emergency cardiac care are relevant to the community’s choice of upgraded service. The optimal system from the standpoint of both service and cost- effectiveness must be determined by the unique needs and characteristics of each community.
REFERENCES
1. Weaver WD, Copass MK, Bufi D, et al: Improved neurologic
recovery and survival after early defibrillation. Circulation 1984;69:943-948
2. Eisenberg MS, Copass MK, Hallstrom AP, et al: Treatment of
out-of-hospital cardiac arrests with rapid defibrillation by
emergency medical technicians. N Engl J Med 1980;302:1379-1383
3. Stults KR, Brown DD, Schug VL, et al: Prehospital defibril-
lation performed by emergency medical technicians in ru- ral communities. N Engl J Med 1984;310:219-223
4. Eisenberg M, Bergner L, Hallstrom A: Paramedic programs
and out-of-hospital cardiac arrest: I. Factors associated
with successful resuscitation. Am J Public Health 1979;69:3&38
5. Eisenberg MS, Copass MK, Hallstrom A, et al: Management of out-of-hospital cardiac arrest: failure of basic emer- gency medical technician services. JAMA 1980;243:1049- 1051
6. Wilson BH, Severance HW, Raney MP, et al: Management of
out-of-hospital cardiac arrest by basic emergency medi- cal technician services. Am J Cardiol 1984;53:68-70
7. SAS User’s Guide: Statistics, version 5. Cary, NC, SAS Insti-
tute, 1985, pp 433-506
111
AMERICAN JOURNAL OF EMERGENCY MEDICINE n Volume 6, Number 2 n March 1988
8. Urban N, Bergner L, Eisenberg MS: The costs of a suburban paramedic program in reducing deaths due to cardiac
arrest. Med Care 1981;19:374-392
9. Ornato JP, McNeil1 SE, Craren EJ, et al: Limitation on effec-
tiveness of rapid defibrillation by emergency medical
technicians in a rural setting. Ann Emerg Med
1984;13:1096-1099
10. Rozkovec A, Crossley J, Walesby R, et al: Safety and effec-
tiveness of a portable external automatic defibrillator-
pacemaker. Clin Cardiol 1983;6:527-533
11. Cummins RO, Eisenberg M, Bergner L, et al: Sensitivity, ac-
curacy, and safety of an automatic external defibrillator:
report of a field evaluation. Lancet 1984;2:31&320. 12. Cummins RO, Eisenberg MS, Bergner L, et al: Automatic
external defibrillation: evaluations of its role in the home
and in emergency medical services. Ann Emerg Med
1984;13:798-801 13. Cummins RO, Eisenberg MS, Hallstrom AP, et al: What is a
“save”?: outcome measures in clinical evaluations of au-
tomatic external defibrillators. Am Heart J 1985;110:1133-
1138 14. Cummins RO, Eisenberg MS, Moore JE, et al: Automatic
external defibrillators: clinical, training, psychological,
and public health issues. Ann Emerg Med 1985;14:75>
760
15. Stults KR, Brown DD, Kerber RE: Efficacy of an automated external defibrillator in the management of out-
of-hospital cardiac arrest: validation of the diagnostic al- gorithm and initial clinical experience in a rural environ-
ment. Circulation 1986:73:701-709