Remote clinical assessment for acute trauma: an initial experience

Remote Clinical Assessment for Acute Trauma: An Initial Experience John A. Aucar, M.D., Robert Eastlack, MS III, Matthew J Wall, Jr, M.D., Kathleen R. Liscum, M.D., Thomas S. Granchi, M.D., Kenneth L. Mattox, M.D. Department of Surgery Baylor College of Medicine Houston, Texas Optimal care of the injured patient requires an organized approach, dedicated resources and clinical expertise. Victims of major trauma, however,frequentlypresentto ruralandsuburban hospitals regardless of whether a dedicated trauma system in place. Immediate consultation by a trauma expert could potentially expedite effective evaluation and management of trauma victims, reducing the occurrence of unnecessary transport and leading to efficient stabilization and transport when needed. Remote assessment of trauma patients must be assesses for feasibility, safety and efficacy before widespread implementation. This project represents the initial steps towards the development of a functional telemedicine system for trauma care.

approach was used following the Advanced Trauma Life Support (ATLS) guidelines set forth by the American College of Surgeons Committee on Trauma ' . We have previously described the accuracy of remote clinical data assessment by video observation alone 2. In that report, an ATLS based data tool with 44 defined clinical variables was used to compare observations derived from 24 video taped acute trauma resuscitation to those documented in the clinical records. Greater than 90% accuracy was demonstrated for 19 of the 44 clinical variables, while 10 variables showed an accuracy from 70% to 90% and 15 variables showed an accuracy of less than 70 %. Prior to that, video observation had only been reported as a quality assessment and educational tool in the context of trauma care 3-5 . Our prior study served as the pilot project upon which the current interactive protocol is based.

INTRODUCTION Trauma care has evolved into a highly specialized field of surgery where experience and judgment have a major effect on clinical outcome. Unfortunately, the modem trend towards regionalization oftrauma care to designated trauma centers does not eliminate the need to deliver efficient trauma care at rural and suburban hospitals. Telemedicine holds significant potential for resolving this disparity between the ubiquitous nature of traumatic injuries and our desire to centralize specialty care. However, the feasibility, safety and efficacy ofremote clinical evaluation of acute multi system trauma has yet to be determined.

METHODS The Ben Taub General Hospital, a 600 bed, urban, Level I trauma center is equipped with a closed circuit analog television system, linking five emergency center resuscitation rooms on the first floor to the Trauma Coordinators office on a distant floor. The system was installed primarily for quality assurance purposes. Video taping was not used during any of the remote evaluations to avoid violation of quality assurance protocols. Seventeen trauma victims were selected at random, who presented to Ben Taub via ambulance after sustaining either penetrating (n= 3) or blunt (n= 14) trauma. The mechanisms of injury are presented in Table I.

This report describes an initial experience with the successful, comprehensive remote evaluation of 17 trauma victims, using real time video observation combined with verbal interaction. A structured

1091-8280/98/$5.00 C) 1998 AMIA, Inc.

396

Table I - Mechanism of In ury Stab Wound Gun Shot Wound Auto-Pedestrian Motor Vehicle Accident AggraatedAssault Fall

2 1 1 7 2 4

,

Immediate evaluation was conducted remotely by a surgical attending, observing the resuscitation from the trauma coordinator's office, in communication with a third year surgical resident, positioned within the resuscitation suite. A 3 meter high wall mounted video camera and microphone provided a room view perspective.

Based on our previous report, we have found it useful to catagorize the 44 variables contained on the clinical assessment form into 8 groups: demographics (Demo), primary survey (ABCDE), initial vital signs (Initial VS), ancillary studies (Ancillary) secondary vital signs (20 VS), secondary patient history (2° Hx), secondary physical exam (20 PE ), diagnosis and disposition (Dx/Dispos). Table II shows the 44 variables documented by the evaluation form.

Direct vocal communication between the

attending surgeon and the surgical resident was conducted over a dedicated telephone line using a cordless, hands free telephone set, worn by the resident. An ATLS based clinical assessment form with 44 specific variables was used by the attending to document findings as they occurred. The remote evaluations began immediately upon the patients arrival into the resuscitation suite and continued until the primary and secondary survey were complete, and a disposition from the resuscitation suite had been determined. The time of specific clinical observations, diagnoses and decisions were noted. The evaluation documented remotely was then compared, by an independent party, to the clinical chart and discrepancies were noted. All evaluations were performed in real time. Resident consent was implied by their participation in the two way teleconversation, in compliance with an institutional review board approved protocol. Patients were treated by standard existing protocols and their treatment did not vary from non-study patients by inclusion into the study. Implied consent for urgent treatment extended to these and all other emergency patients. Patients were excluded from the study if they met specific trauma center criteria for immediate staff presence. These criteria resulted in the exclusion of pediatric, geriatric and pregnant patients, those with hypotension (systolic less than 90), and severely impaired mental status (Glasgow Coma Score less than 9).

397

Table II Clinical Variables Assessed Variables Group Demo Age, Sex, Mechanism of injury ABCDE Airway, Breathing, Circulation, Disability, Exposure BP, HR, RR, Temp. Oxygen Initial VS Saturation Ancillary CXR, Pelvis, C-spine, Foley, NG tube, IV fluids, EKG, Type and Cross match, Labs 20 VS BP, HR, RR, Temp, Oxygen Saturation HPI, PMH, PSH, 2° Hx Medications, Allergies 20 PE Head & Neck, Chest, Abdomen, DPL, Pelvis & Perineum, Back & extremities Dx / Dispos Diagnosis, Disposition, Therapy, Time in EC, Mortality, TRTS DPL= Diagnostic peritoneal lavage; TRTS = Triage revised trauma score 6

The information documented by remote evaluation of each patient was compared to the clinical chart as documented by a trauma team member present within the resuscitation suite. Each variable was considered to agree if there was concordance between the two sources of documentation; to disagree if there was discordance between the two documents; and considered insufficient if a finding for that variable was documented on only one source or the other. Items not documented on either source are considered to agree.

RESULTS A total of 17 remote evaluation and management sessions have been successfully conducted thus far. One patient was excluded from analysis because the clinical chart was not available for comparison. All patients had a Triage Revised Trauma Score of 12 and were subsequently discharged alive. The average session lasted 21 minutes (range 10 - 40). On two occasions, two simultaneous resuscitations were performed concurrently in two rooms by one senior resident and twojunior residents. There were 12 male and 5 female trauma victims studied with an average age of 31 (range 13 - 52) years. The 17 remote evaluations data forms documented on the average 3 1(range 24-35) of 44 possible variables. The 16 available clinical charts documented on average 28 (range 3-35) of 44 clinical variables. Two charts were found to have minimal physician documentation. An addendum was subsequently entered into those records based on the Telemedicine documentation. Four of the 16 charts showed a greater number of variables documented than their corresponding remote evaluation form, while 9 remote evaluation forms contained more documentation than the corresponding clinical chart. Three cases showed equivalent documentation.

The results of this comparison for 16 patients, summaring the 44 variables into 8 groups, is displayed in Table III. Table III - Agreement Rates Group Dis- Insufficient Agree (Variables) (/0) (%) agree Demo 44 (92) 0 (0) 4 (8) ABCDE 62 (76) 0 (0) 18 (23) Initial VS 55 (69) 1 (1) 24 30

Ancillary 2 0 VS 2 0 Hx 2 0 P.E.

Dx / Dispos

81 (56) 45 (56) 62 (78) 79 (82) 73 (76)

0 (0) 1 (1) 0 (0) 1 (1) 0 (0)

63 (44) 34 (43) 18 (29) 16 (17) 23 (24)

Limited documentation in the clinical record accounts for the majority of the data points listed as insufficient. Some groups of variables are prone to

398

higher agreement rates, presumably because of greater attention devoted to these during the evaluation. For instance, the temperature, which is a component of both vital signs groups, is seldom recorded on routine trauma evaluations. Documentation of ancillary studies and labs also showed persistent limitations. Quality of care, efficiency of clinical performance and quality of documentation are three distinct parameters that affect the assessment of any clinical process. Because of the relatively stable nature of this patient group, the project served as an exercise to examine the documentation process as much as to examine the resuscitation process. There were only 3 instances of disagreement between specific findings documented remotely and the corresponding clinical chart. Two of these represent disparity in pulse oximetry readings and may be due to recording of values at different points in time. The disparity which occurred during the secondary physical exam was due to mis documentation of the head and neck exam on the remote evaluation form of a patient with multiple facial lacerations. There were no untoward clinical events or complications as a result of this remote evaluation exercise. In several cases, specific recommendations for further evaluation or treatment were made by the attending surgeon, which expedited the clinical course. The vast majority of information documented remotely was based on verbal interaction between the resident and the attending surgeon. None of the remote evaluation, once initiated, were terminated prior to completion. Technical difficulties with the communication link occurred in only one case. This was the result ofloose a telephone wire connection and was easily corrected. Each of the 5 residents involved in this project, when interviewed, considered the process helpful from both a practical and educational standpoint. None of the residents felt that it was cumbersome or impractical. All of the remote evaluations were conducted by a single attending surgeon with extensive experience in trauma evaluations, in person and by telephone consultation. Based on this experience, combined video/voice interaction is considered far superior to telephone communication alone.

communication with the surrogate team leader via headset prevents room noise from confounding the efficiency of the evaluation. The ultimate quality and effectiveness of the resuscitative effort will depend greatly on the ability of the consultant to draw out the pertinent information over the communication link. The room view video perspective allows the consultant to monitor the activities ofeach member ofthe resuscitation team and ancillary personnel while focusing on the process of gathering and processing clinically relevant information about the patient. Physical separation, while maintaining a robust communication link, actually provides an advantage by promoting emotional detachment, allowing the consultant to focus on structured data gathering, documentation and decisionmaking. The quality of clinical documentation practices have a major impact on our ability to assess telemedicine techniques. This experience suggest that telemedicine may actually facilitate and improve documentation practices

DISCUSSION Trauma is the leading cause of death through the first 4 decades of life and the 4th leading cause of death nationally7. A broad range of medical, surgical, and ancillary disciplines are called upon to provide initial evaluation of injured patients. Frequently, errors of commission, omission and misprioritization may produces catastrophic delays in treating serious injuries, or result in costly, useless and dangerous interventions where none are needed. The ability of structured protocols, promoting a coordinated team approach, to reduce morbidity and mortality through timely recognition and treatment of injuries is one of the tenants of modem trauma care. This approach, combined with telemedicine, would likely reduce the need for non-therapeutic transfers and help to expedite inter-facility transfers when needed. The potential for improved utilization of resources, reduction of cost, and increased satisfaction for both referring and referral centers is enormous. The safety and efficacy oftelemedicine techniques for trauma care must be established by rigorous evaluation due to its potential for harm. Remote evaluation is not, by itself, a protection from inefficient resuscitation. A high degree of sensitivity for a wide variety of injuries, under a broad range of circumstances, must be established before any large scale implementation can be recommended. Its efficacy must be judged by the ability to conduct a thorough evaluation and guide appropriate treatment in a critical patient population with multiple injuries, maintaining a low rate of missed injuries.

The current study calls into question the need for the trauma team leader to be physically present for an accurate evaluation. The technological demands are not necessarily extensive and our technological achievements are progressing fasterthan our ability to use them. The addition of telemetric monitoring and digital image transmission would be obvious supplements to the evaluation process. CONCLUSION The current report establishes the feasibility of comprehensive remote evaluation ofrelative stable trauma patients under controlled circumstances. This series contained no significant missed injuries or complications related to the communication technique. Further evaluation under a broader range of circumstances and larger patient series is needed before the technique can be considered for widespread implementation. Its potential as a practical and an educational tool was deemed favorable by the residents involved and the attending conducting the remote consultation. Trauma Resuscitation will likely prove to be one of the most demanding and potentially gratifying applications for Telemedicine.

This initial assessment was conducted in house, at a Level I trauma center, under optimally controlled circumstances. This allowed immediate contact intervention, providing a safety net until the process becomes refined and established. Feedback from the participating residents indicated that participation in the telemedicine process was not cumbersome nor a burden, but became easier with each case experienced. Refinement of communication skills progressed throughout the series consistent with a short learning curve. Success of the technique will depend mainly on the ability to provide a focused and organized approach to traumna resuscitation combined with meticulous communication skills. Focused

399

References 1. Advanced Trauma Life Support Instructor Manual, 5th Edition, 1993. 2. Aucar JA; Villavicencio RT; Wall MJ Jr; Liscum KR; Granchi TS; Mattox, KL. Evaluation of clinical data by remote observation in trauma. Proc AMIA Annu Fall Symp 1997;:408-12

3. Hoyt DB, Shackford SR, Fridland. PH, et al. Video recording trauma resuscitations: an effective teaching technique. J Trauma, 1988;28:435-440. 4. Mann FA, Walkup RK, Benyman CR, Bessey PQ, Wilson AJVanner MW. Computer-based videotape analysis oftrauma resuscitations for quality assurance and clinical research. J Trauma, 1994;36:226-230. 5. Townsend RN, Clark R., Ramenofsky ML, Diamond DL. ATLS-based videotape trauma resuscitation review: education and outcome. J Trauma 1993;34: 133-138. 6. Champion HR, Sacco WJ, Copes WS, et al. A revision ofthe trauma score. J Trauma. 1989;29:623629. 7. Committee on Trauma Research. Injury in America. Washington, DC: National Academy Press, 1985.

D004758.wpd

400