1002
EMT-P JUDGMENT
Qazi et al.
EM"-P JUDGMENT
Paramedic Judgment of the Need for Trauma Team Activation for Pediatric Patients KHAJISTA QAZI, MD, MS, JEFFREY A. KEMPF, DO, NORMANC. CHRISTOPHER,MD, LOWELLW. GERSON,PHD Abstract. Objective: To determine the value of paramedic judgment in determining the need for trauma team activation (TTA) for pediatric blunt trauma patients. Methods: A prospective, observational study was conducted a t the ED of Children's Hospital Medical Center of Akron between July 12, 1996, and February 28,1997, in cooperation with Akron Fire Department emergency medical technician-paramedics (EMT-Ps). The ED provides on-line and off-line medical control for pediatric transports. Patients with minor or no identifiable injuries are released a t the scene with the instructions to see a physician. The remainder are transported to the ED. The decision for TTA is based on ED trauma protocols a s well a s emergency physician judgment of injury severity in combination with the judgment of the treating paramedic. During the study, EMT-Ps were asked (before physician input) whether, based solely on their judgment, a patient needed TTA. Patients 0-14 years old who were involved in motor vehicle crashes, bike crashes, or falls from a height of 210 feet were included in the study. TTA was defined as necessary if the patient was admitted to the intensive care unit (ICU) or operating room (OR) for nonorthopedic surgical procedures. Out-of-hospital, ED, and hospital records were reviewed. Coroners' records as well a s medical records of all trauma admissions during the study period were reviewed to ensure that the patients released a t the scene were not mistriaged. Results: One hundred ninety-two patients met study criteria. Eighty-five patients (44%) were transported to the ED, of whom 12 had TTA. EMT-Ps requested TTA for 10 of these patients, and 2 patients had TTA per ED trauma protocol. Two of the patients who were judged by EMT-Ps to need TTA were admitted
T
RAUMA triage protocols that are used to determine the need for trauma team activation
From the Division of Emergency and Trauma Services, Children's Hospital Medical Center of Akron (KQ, JAK, NCC), Akron, OH, and Division of Community Health Sciences (LWG), Northeastern Ohio Universities College of Medicine, Rootstown, OH. Received J a n u a r y 1, 1998; revision received May 27, 1998; accepted J u n e 3, 1998. Presented a t the American Academy of Pediatrics annual meeting, New Orleans, LA,October 1997. Supported in p a r t by a grant from Akron Children's Hospital Foundation. Address for correspondence and reprints: Khajista Qazi, MD, Children's Hospital Medical Center of Akron, One Perkins Square, Akron, O H 44308-1062. Fax: 330-258-3761; e-mail:
[email protected]
to the ICU/OR, and neither of the patients identified by ED trauma protocol to require TTA were admitted to the ICU/OR. Two initially stable patients who did not have ITA deteriorated after arrival to the ED. Both were admitted to the ICU. The sensitivity and specificity of paramedic judgment of the need for "A for pediatric blunt trauma patients were 50% (95% CI 9.2-90.8) and 87.7% (95% CI 78.0-93.6), respectively. The positive and negative predictive values were 16.7% (95% CI 2.9-49.1) and 97.3% (95% CI 89.6-99.5). None of the patients released a t the scene was mistriaged based on the review of the coroners' and trauma admission records. Conclusion: Results of this investigation indicate that a small percentage of pediatric blunt trauma patients require TTA. EMTP judgment alone of the need for TTA for pediatric blunt trauma patients is not sufficiently sensitive to be of clinical use. The low sensitivity is explained by the deterioration in the clinical condition of 2 initially stable patients. The paramedic disposition decisions from the scene were always accurate. Nontransport by emergency medical services (EMS) may be acceptable in some uninjured pediatric trauma patients. Injured pediatric trauma patients who appear to be stable may deteriorate shortly after injury. However, if a pediatric patient appears injured, transport from the scene and examination by a trauma specialist are needed. Finally, the role of paramedic judgment must be further defined by larger studies with urban, rural, and suburban EMS systems before it can be used as a sole predictor of "A. Key words: emergency medical technicians; trauma; judgment; assessment; pediatrics. ACADEMIC EMERGENCY MEDICINE 1998; 5:1002-1007
"M'A) consider physiologic, anatomic, and mechanism of injury This protocol-based approach to trauma care may lead to over- or undertriage of trauma victims. An ideal trauma triage system with no undertriage and minimum overtriage does not exist for adult or pediatric trauma patients. Physiologic and anatomic trauma triage criteria may not be obvious or readily available on the scene, resulting in transport of patients to facilities not properly equipped for their management. Such undertriage may delay appropriate patient care and adversely affect patient outcome. Blunt injury mechanisms are poor predictors of injury severity in adult and pediatric trauma
ACADEMIC EMERGENCY MEDICINE
October 1998, Volume 5. Number 10
TTA based on mechanism of injury alone leads to overtriage and inefficient utilization of trauma center resources, and may prove to be cost-ineffective.B-l1 Several studies of adult trauma patients attempted to assess the value of emergency medical technician-paramedic (EMT-P) judgment of trauma patients as a n alternative t o objective trauma triage instruments. EMT-P judgment of adult trauma patients is comparable to that of out-of-hospital trauma triage instrum e n t ~ . ’ * - Inclusion ’~ of EMT-P perception of injury severity in the triage decision process increases the sensitivity of select out-of-hospital trauma triage criteria.16J6 No published studies have evaluated the use of EMT-P judgment of injury severity in pediatric blunt trauma patients. Involvement of paramedics in the triage decision process may potentially improve the quality of triage decisions by minimizing both undertriage and overtriage. The objective of this preliminary study was to estimate this potential by determining the accuracy of paramedic judgment in predicting the need for TTA for pediatric blunt trauma patients.
METHODS Study Design. This prospective study was conducted at the ED of Children’s Hospital Medical Center of Akron between July 12, 1996, and February 28, 1997, in cooperation with the Akron Fire Department and Emergency Medical Services (EMS) system. The study was approved by the institutional review board. Setting and Population. Children’s Hospital Medical Center of Akron serves as the sole tertiary pediatric care center for 17 counties. The ED has an annual census of approximately 50,000 visits with 500 trauma admissions. Pediatric emergency physicians supervise patient care in the ED 24 hours a day. The Akron Fire Department has 13 EMS squads that serve a n area of 62.2 square miles with a relatively stable population of 223,000 people. The system responds to approximately 4,200 pediatric calls annually. All squads are staffed exclusively by EMT-Ps. The EMT-Ps are taught the Akron Fire Department trauma protocol prior to starting their duties as paramedics. The EMT-Ps are required to take Pediatric Advanced Life Support (PALS) and Basic Trauma Life Support (BTLS) courses, as well as continuing education lectures on trauma. The PALS course includes a lecture on trauma resuscitation; the BTLS course was designed for out-of-hospital care providers and emphasizes recognition and assessment of trauma in the out-of-hospital setting. The quality assurance officer at Akron Fire Department reviews all
1003
TABLE1. ED Protocol for Trauma Team Activation Blunt or vehicular trauma + burn or inhalation injury Death in the same automobile/vehicle Fall from >20 feet Penetrating injury to the head, neck, or torso Paralysis Extremity amputation proximal to the knedelbow Pelvic fracture T w o or more long bone fractures Pediatric trauma score 5 8
trauma run reports regularly to provide feedback on performance as needed. The trauma run reports also are reviewed by the EMS coordinator at the ED. Assessment and out-of-hospital management of trauma patients are discussed in the monthly run reviews. In addition, lectures are given on trauma assessment and management in the run reviews. Patients 0-14 years of age injured in motor vehicle crashes (MVCs), bike crashes or falls from a height of >10 feet who were evaluated by City of Akron EMT-Ps were eligible for study. Medical direction for pediatric transports is proved on-line by ED attending physicians and by written (off-line) protocols. Study Protocol. The approved procedure for disposition of trauma patients is that the EMT-Fs analyze the information available at the scene and establish contact with medical control. Trauma patients who are determined to have minor or no injuries may be released at the scene, based on paramedic judgment in conjunction with on-line medical control. All scene-released patients are given instructions to follow-up with their physicians. All other injured patients are transported to the ED. The planned level of response in the ED is determined by a preexisting protocol (Table 1) or on-line medical control, taking into consideration the treating paramedic’s assessment of injury severity. Those patients who qualify by either criteria are evaluated by a trauma team. The leadership and composition of trauma team are shown in Table 2. When TTA is considered necessary, the members of the trauma team are summoned to the ED prior to patient arrival. Patients who are transported to the ED, but are judged not to need TTA, are initially evaluated by the ED medical staff. Surgical consultation is later obtained as appropriate. A senior surgery resident is available for consultation 24 hours a day. The on-call surgery attending is available immediately and within 15 minutes of the patient’s arrival during the regular business hours and after hours, respectively. During the study, at the time the report was called by the EMT-P to the ED, the treating paramedics
1004
EMT-P JUDGMENT
Qazi et al.
TABLE2. Leadership and Composition of t h e Trauma Team Leadership Surgery attending* Composition ED attending* Surgery resident Two ED nurses Laboratory technician Respiratory therapist Radiology technician Nursing supervisor Anesthesia resident? OR personnelt
E m - P JUDGMENT
ED. The EMT-P responses were based solely on their judgments of severity of injury. The paramedics did not calculate a formal trauma score. No extra training was given to the EMT-Ps or the ED dispatch officers. Standing patient care policies and procedures were not altered during the study.
Measurements. Out-of-hospital, ED, and hospital records were abstracted for patient demographics, disposition from the scene and ED, and final diagnoses. Patients with incomplete records were excluded (n = 1).We reviewed medical records of all trauma admissions during the study period to *The ED attending is the t r a u m a team leader until the surgery ensure that no patient released at the scene was attending arrives to the ED within 20 minutes after trauma mistriaged. In addition, coroners of the 17-county team activation ('ITA). referral base as well as Cuyahoga County, located t T h e operating room (OR) supervisor is notified of t h e patient's directly north of Akron, were surveyed. arrival. The anesthesia attending arrives to the ED within 20 minutes after "A.
Data Analysis. TI'A was defined as necessary if the patient was admitted to the intensive care unit (ICU) or operating room (OR) for nonorthopedic operative procedures. Admission to the ICU/OR was 7.6 2 4.2y r used as a criterion standard for calculation of sensitivity, specificity, and predictive values. Epi Info 100 (52.0%) (USD Inc., Stone Mountain, GA) was used for data 92 (48.0%) analysis. Mean values a r e reported as t standard deviation.
TABLE3. Demographics, Injury Mechanisms, and Scene Dispositions of the Study Subjects Age-mean
-t
SD
Sex Male Female Injury mechanism Motor vehicle crash Bicycle crash Fall from a height of >10 ft
137 (71.3%) 42 (21.9%) 13 (6.8%)
RESULTS
One hundred ninety-two patients met the study criteria. Fifty-two percent of the patients were 85 (44.3%) male, and the average age for all the patients was 107 (55.7%) 7.6 5 4.2 years. MVCs were the leading mechanism of injury, followed by bike crashes and falls were asked to determine the level of response re- (Table 3). Eighty-five patients (44%) were transquired in the ED. Their responses were recorded ported to the ED. Of these, 12 (14.1%) had TTA. on the dispatch form by the dispatch officer in the Two patients who were judged not to need TTA
Scene disposition ED Home
TABLE4.Characteristics of Patients with Trauma Team Activation (TTA) and/or Intensive Care Unitloperating Room (ICUI OR) Admission
TT Activation Yes Yest No No Yes Yes Yes Yes Yes Yes Yes$ Yes Yes$ Yes
Mechanism of Injury
MVC* Bike crash Fall MVC MVC MVC MVC
MVC MVC MVC MVC Fall Fall Fall
Disposition ICU/OR Yes Yes Yes Yes No No No No No No No No No No
Length of ICU Stay (Days)
Final Diagnosis Liver laceration Intracranial hematoma a n d closed head injury Liver laceration Subdural hematoma and basilar skull fracture Open fracture of tibidfibula Superficial soft-tissue injury Open fracture of tibia Open fracture of tibidfibula and open fracture of ulna Closed head injury Superficial soft-tissue injury Superficial soft-tissue injury Superficial soft-tissue injury Superficial soft-tissue injury Closed head injury
*MVC = motor vehicle crash. tTTA per emergency medical technician (EMT) judgment and per ED trauma protocol. ST" per ED trauma protocol; t h e remainder of the patients had TTA per EMT judgment.
ACADEMIC EMERGENCY MEDICINE
1005
October 1998,Volume 5, Number 10
eventually required it. TTA was a t the request of the treating paramedics in 9 cases, and according to ED trauma protocol in 2 (Table 1). One patient was judged to need TTA both by paramedics and ED trauma protocol (pediatric trauma score 18). Two of the 10 patients judged by paramedics to require TTA were admitted to the ICU or OR, and neither of the 2 patients identified by ED protocol as requiring TTA required admission. Two patients who were not identified prospectively as requiring TTA were admitted to the ICU. Both patients were asymptomatic with stable vital signs upon arrival to the ED, but deteriorated within the first 30 minutes of their evaluation and treatment. One had sustained a liver laceration as a result of a fall from a height of 12 feet, the other a basilar skull fracture in an MVC. The characteristics of the 14 patients who had TTA and/or ICU/OR admission are shown in Table 4. The sensitivity and specificity of paramedic judgment of the need for TTA for pediatric blunt trauma patients were 50% (95% CI 9.2-90.8) and 87.7% (95% CI 78.0-93.61, respectively, and the positive and negative predictive values were 16.7% (95% CI 2.9-49.1) and 97.3% (95% CI 89.6-99.5) (Table 5). No patient released at the scene was later hospitalized for sustained injuries. The survey of coroners' records revealed no trauma deaths among patients released by paramedics at the scene of their injuries.
TABLE5. Accuracy of Paramedic Judgment
for Dauma
Team Activation Intensive Care UniUOperating Room Admission
Paramedic recommendation Yes No TOTAL
Yes
No
Total
2 2
10 71
12 73
4
81
85
attempted to establish the sensitivity of paramedic judgment for assessment of injury severity in order to determine alternatives to traditional trauma scores. In a prospective study,15 the inclusion of EMT-P perception of injury to select anatomic, physiologic, and mechanism-of-injury criteria increased the sensitivity of the criteria. A prospective study by Emerman et a1.I2found that the EMT predictions of mortality and the need for emergent operation were as accurate as 3 trauma scores. In a prospective study, Fries et al.13 reported that the paramedic judgment of the need for trauma center resources was more sensitive than the trauma triage rule proposed by Baxt e t a1.22A retrospective study by Ornato et al.14 showed that paramedic judgment was better at identifying patients in need of major surgery than the patients' trauma scores. Esposito e t a1.16 found that EMT judgment alone was a low-yield out-of-hospital triage criterion to identify major trauma patients. Adding DISCUSSION mechanism-of-injury criteria to EMT-P judgment An accurate assessment of injury severity in the improved accuracy in identifying major trauma. out-of-hospital setting is an essential component of These findings were corroborated by a study by the decision-making process for TTA. The appro- Hedges et al.23Most of the studies cited had either priate use of a trauma team is important for safe, no or inadequate numbers of pediatric patients. Our study suggests t h a t neither preestablished efficient, and cost-effective utilization of ED and hospital r e s o ~ r c e s . Several ~ ~ J ~ trauma scoring sys- criteria nor paramedic judgment may be sufficient tems have been evaluated for use in the out-of-hos- in the determination of the need €or TTA in pedipita1 setting. Their relatively low sensitivity pre- atric blunt trauma patients. Both over- and undercludes their use as out-of-hospital trauma triage triage occurred. The overtriage occurred in 83% t o ~ l s . Application ~~J~ of the pediatric trauma score (10/12) of the patients with TTA who did not re(PTS)has been advocated for assessing severity of quire ICU/OR. Of greater concern, there was a n injury of pediatric trauma victims in the out-of- undertriage rate of 14.2% (2114) for the patients hospital setting. Proponents suggest that a PTS of who required ICU/OR but did not have TTA. Our results also emphasize that physiologic de4 8 indicates the need for treatment at a pediatric trauma center.20 Detractors suggest PTS scoring, rangements may occur after arrival to ED. The 2 including the use of this cutoff point for transport patients who deteriorated after arrival to the ED to a pediatric trauma center, has its drawbacks. A were initially asymptomatic. The patients were retrospective study by Aprahamian et a1.21 re- transported to the ED because the mechanism of ported that 17% of pediatric trauma patients with injury was considered significant by the treating a PTS of 2 8 were admitted to the OR and 9%were EMT-Ps and the on-line medical control. One child admitted to the ICU. The accuracy of the PTS for had flipped over twice after being hit by a motor recognizing severely injured children was shown to vehicle moving a t 25 miles per hour. The second child had fallen from a height of 12 feet. If the enbe only 68% in another retrospective study.20 Several studies of adult trauma patients have tire study population is taken into account (n =
1006
EMT-P JUDGMENT
Qazi e t al.
EMT-P JUDGMENT
192), based on review of coroners' and trauma ad- and/or rural communities. In addition, level of mission records, there was no undertriage among training and ongoing education of EMTs may vary the patients who were not transported to the ED. from one EMS system to another. Future studies Our tiered out-of-hospital and ED trauma response should look at the judgments of paramedics who systems discourage overtriage. In our study, sever- serve in different practice settings. Moreover, imity of injury of 2 patients who later required ICU pact of the level of training and ongoing education admission could not be predicted in the out-of-hos- on EMT judgment of pediatric blunt trauma papita1 setting. The outcome of these patients was tients is another interesting area for future studies unaffected because of immediate availability of re- to explore. Determination of factors that help sources for specialized pediatric trauma care. The EM% make their decisions regarding injury severresources needed by pediatric trauma patients ity of pediatric blunt trauma patients is another differ significantly from those needed by adult interesting area for future studies. trauma patients. Therefore, pediatric trauma triage schemes should be designed to ensure timely CONCLUSIONS care for those pediatric trauma patients who may develop symptoms and signs consistent with major Results of this preliminary investigation indicate trauma after arrival to the ED. Improved pediatric that a small percentage of pediatric blunt trauma trauma triage will result from research on the con- patients require TTA. EMT-P judgment alone of tribution of individual physiologic, anatomic, and the need for "A for pediatric blunt trauma pamechanism-of-injury factors in predicting injury tients is not sufficiently sensitive to be of clinical severity in pediatric trauma patients. Emphasis on use. The low sensitivity is explained by the deteimproved education and training of general emer- rioration in the clinical conditions of 2 initially stagency physicians in pediatric trauma care may ble patients. The paramedic disposition decisions prove to be helpful in the appropriate initial man- from the scene were always accurate. Nontransagement of pediatric patients when undertriage of port by EMS may be acceptable in some uninjured initially stable patients does occur. Expanded pediatric trauma patients. Injured pediatric paramedic education and training in pediatric trauma patients who appear to be stable may detrauma assesssment also are important in making teriorate shortly after injury. However, if a pediaccurate triage decisions. atric patient appears injured, transport from the scene and examination by a trauma specialist are LIMITATIONS AND FUTURE QUESTIONS needed. Finally, the role of paramedic judgment must be further defined by larger studies with urOur ED receives patients from 17 counties. To en- ban, rural, and suburban EMS systems before it sure consistency, we limited our study to the Akron can be used as a sole predictor of TTA. EMS system. This contributed to a relatively low number of TTAs. The presence of only one tertiary The authors thank Akron Fire Department paramedics for pediatric care center in Akron, the size of the city, their cooperation in the study. and a relatively stable population influence our out-of-hospital triage system. For this reason, we have reservations about generalizing the results of References our study to EMS systems that function in larger, 1. Emerman CL. Trauma triage: where do we go from here? more urban communities with several hospitals [commentary]. h a d Emerg Med. 1995; 2:1025-6. committed to pediatric trauma care. Rural EMS 2. Eastman AB, Lewis FR, Champion HR, Mattox KL. Retrauma system design: critical concepts. Amer J Surg. paramedics operate in areas in which transport to gional 1987; 154179-84. a tertiary pediatric care center may be impractical 3. American College of Surgeons Committee on Trauma. Rebecause of prolonged transport time. In addition, source Document for Optimal Care of the Injured Patient. ChiACS, 1993. paramedics from rural settings may not have ade- cago: 4. Long WB, Bachulis BL, Hynes GD. Accuracy and relationquate experience in the assessment and triage of ship of mechanisms of injury, trauma score, and injury severpediatric trauma patients. These special situations ity score in identifying major trauma. Am J Surg. 1986; 151: warrant the need for different strategies for safe 581-4. 5. Knopp R, Yanagi A, Kallsen G, Geide A, Doehring L. Mechpediatric trauma care for different practice set- anism of injury and anatomic injury a s criteria for prehospital trauma triage. Ann Emerg Med. 1988; 17:895-902. tings. This initial report suggests the need for larger 6. Walker PJ, Cass DT. Paediatric trauma: urban epidemiology and an analysis of methods for assessing the severity of studies in a variety of EMS systems. Since severe trauma in 598 injured children. Aust N Z J Surg. 1987; 57: injury in pediatric blunt trauma patients is uncom- 715-22. mon, a larger sample size is likely to include more 7. Henry MC, Alicandro JM, Hollander JE, Moldashel JG, Cassara G, Thode HC Jr. Evaluation of American College of patients with a need for TTA. The EMS systems in Surgeons trauma triage criteria in a suburban and rural setthe United States operate in urban, suburban, ting. Am J Emerg Med. 1996; 14:124-9.
ACADEMIC EMERGENCY MEDICINE
October 1998, Volume 5, Number 10
6. Simon BJ, Legere P, Emhoff T, Fiallo VM,Garb J . Vehicular trauma triage by mechanism: avoidance of the unproductive evaluation. J Trauma. 1994; 37:645-9. 9. Shatney CH, Sensaki K. Trauma team activation for ‘mechanism of injury‘ blunt t r a u m a victims: time for a change? J Trauma. 1994; 37:275-82. 10. Phillips JA, Buchman TG. Optimizing prehospital triage criteria for trauma team alerts. J Trauma. 1993; 34:127-32. 11. Hoff WS, Tinkoff GH, Lucke JF, Lehr S. Impact of minimal injuries on a level I t r a u m a center. J Trauma. 1992; 33:40812. 12. Emerman CL, Shade B, Kubincanek J. A comparison of EMT judgment and prehospital trauma triage instruments. J Trauma. 1991; 31:1369-75. 13. Fries GR, McCalla G, Levitt MA, Cordova R. A prospective comparison of paramedic judgment and the trauma triage rule in the prehospital setting. Ann Emerg Med. 1994; 24:885-9. 14. Ornato J , Mlinek E J , Craren E J , Nelson N. Ineffectiveness of the trauma score and the CRAMS scale for accurately triaging patients to trauma centers. 1985; 14:1061-4. 16. Simmons E,Hedges JR, Irwin L, Maassberg W, Kirkwood HA. Paramedic injury severity perception can aid trauma triage. Ann Emerg Med. 1995; 26:461-8. 16. Esposito T J , Offner PJ, Jurkovich G J , Griffith J , Maier RV.
I
1007
Do prehospital t r a u m a center triage criteria identify major trauma victims? Arch Surg. 1995; 130:171-6. 17. Tinkoff GH, O’Connor RE, Fulda GJ. Impact of a twotiered trauma Response in t h e emergency department: promoting efficient resource utilization. J Trauma. 1996; 41: 735-40. 18. Ochsner MG, Schmidt JA, Rozycki GS, Champion HR. The evaluation of a two-tiered t r a u m a response system a t a major trauma center: is it cost effective and safe? J Trauma. 1995; 39:971-7. 19. Baxt WG, Berry CC, Epperson MD, Scalzitti V. The failure of prehospital t r a u m a prediction rules to classify trauma patients accurately. Ann Emerg Med. 1989; 18:21-8. 20. Kaufman CR, Maier RV, Rivara FP, Carrico J . Evaluation of the pediatric t r a u m a score. JAMA. 1990; 263:69-72. 21. Aprahamian C,Cattey RP, Walker AP,Gruchow HW, Seabrook G. Pediatric t r a u m a score. Arch Surg. 1990; 125:112831. 22. Baxt WG, Jones G, Forlage D. The trauma triage rule: a new, resource-based approach to t h e prehospital identification of major trauma victims. Ann Emerg Med. 1990; 19:1401-6. 23. Hedges J R , Feero S, Moore B, Haver DW, Shultz B. Comparison of prehospital t r a u m a triage instruments in a semirural population. J Emerg Med. 1987; 5197-208.
Determination of the Minimal Clinically Significant Difference on a Patient Visual Analog Satisfaction Scale
I
ADAM J. SINGER, MD, HENRYC.THODEJR.,PHD Abstract. Objective: To determine the minimal clinically significant difference (MCSD) on a visual analog patient satisfaction scale. Methods: The authors prospectively collected patient satisfaction evaluations during a clinical trial assessing the effect of introducing personal television sets on overall patient satisfaction from their ED encounters. Patient satisfaction was assessed with 2 scales: a 100-mm visual analog scale (VAS) ( 0 = least satisfied, 100 = most satisfied) and a 7-point categorical scale (“terrible,” “mostly dissatisfied,” “mixed,” “partially satisfied,” “mostly satisfied,” “pleased,” and “delighted”). The differences between the mean VAS scores of “delighted” and “pleased” patients, and between “pleased” and “mostly satisfied” patients were used to determine the MCSD on the VAS. Reliability of each of the scales was determined. Results: 181 patients
From the Department of Emergency Medicine, State University of New York at Stony Brook, Stony Brook, NY ( A J S ,HCT). Received February 9, 1998; revision received May 26, 1998; accepted J u n e 3, 1998. Presented at the SAEM annual meeting, Chicago, IL, May 1998. Address for correspondence and reprints: Adam J . Singer, MD, Department of Emergency Medicine, University Hospital and Medical Center, ~ 4 - 5 1 5 ,Stony Brook, ~y 11794-7400, Fax: 516-444-3919; e-mail:
[email protected]
were evaluated. Mean age was 41 years; 59% were female. On a subset of 19 patients, the VAS yielded an interobserver correlation of 0.93. The kappa measurement of agreement on the categorical scale was 0.77. The mean difference between “delighted” and “pleased” patient VAS satisfaction scores was 6.8 mm (95% CI, 1.3-12.3 mm). The mean difference between “pleased” and “mostly satisfied” patient VAS satisfaction scores was 10.7 mm (95% CI, 5.5-15.8 mm). Conclusion: The MCSD in patient satisfaction scores measured with a 100-mm VAS was approximately 7-11 mm. Future studies evaluating differences in patient satisfaction should be designed to detect this difference. Key words: patient satisfaction; visual analog scale; clinical significance; reliability; validity. ACADEMIC EMERGENCY MEDICINE 1998; 5: 1007-1011
1
N TODAY’S consumer-oriented health care
marketplace, it is more important than ever to develop reliable and valid measures of patient satisfaction. Currently, most measurements of patient satisfaction are based on verbal-rating Likert-type categorical scales.’-” One of the problems associated with these semantic differential scales is a response bias known as acquiescent bias. This is the
