emergency triage centerâ into the community's EMS ... their members to call the gate- keeper prior to 9-1-1. ... vices; EMS; ambulance; 911; man- aged care ...
564
COMMENTARIES
Neely
RETHINKING GATEKEEPING AND EMS
emergency triage center” into the Health Sciences University, Portcommunity’s EMS system. This land, OR. center would receive requests for nonemergent assistance from the Keg words. emergency medical ser9-1-1 center based on protocol- vices; EMS; ambulance; 911; mandriven criteria. The nonemer- aged care organization;health maingency communication center tenance organization; HMO; MCO; would assign medical personnel public health. for an on-scene assessment and to further discern patient need. References I n a prior commentary,12I described a similar multiple-option 1. Schneider SM, Cobaugh DJ, Leahey decision point (MODP) model NF. Gatekeepers: a missed opportunity for safe transport. Acad Emerg Med. t h a t suggests nontraditional 1998;5587-92. EMS responses at each decision 2. Kerr HD. Prehospital emergency serpoint along the EMS response vices and health maintenance organizations. Ann Emerg Med. 1986;15727-9. continuum: at dispatch, on 3. Young GP, Lowe RA. Adverse outscene, and upon transport when comes of managed care gatekeeping. destination decisions are made. Acad Emerg Med. 1997;4:1137-41. 4. Franco SM, Mitchell CK, Buzon RM. This model also relies on proto- Primary care physician access and gatecol-driven criteria to determine keeping: a key to reducing emergency department use. Clin Pediatr. 1997; Feb: resource assignment. 63-8. Such models may reduce the . Gadomski AM, Perkis V, Horton L, CRITERIA FOR EMERGENT need for gatekeeping because the 6Cross S, Stanton B. Diverting managed NEEDS EMS system attempts to achieve care patients from pediatric emergency use. Pediatrics. 1995; 95: one of the purposes of managed department 170-8. O’Brien et a1.l0 provide a very care gatekeeping; that is, appro- 6. Glotzer D, Sager A, Socolar D, Weitzuseful comparison of 3 method- priate resource allocation, which man M. Prior approval in the pediatric room. Pediatrics. 1991; 88: ologies for determining need for in turn may lead to decreased emergency 674-80. ED services. These methodolo- cost. Such a n alignment may al- 7. Warren BH, Bell PS, Isikoff S, Hale gies were based on patient com- low health plans to feel more PL. Cost containment and quality of life: an experiment in compassion for physiplaints, ED procedures required, comfortable with EMS resource cians. Arch Intern Med. 1991;151:741and physician judgment. The 3 utilization. This may reduce the 4. 8. Congressional Budget Office. The pomethods produced somewhat dif- emphasis that plans place on tential impact of certain forms of manferent rates of “inappropriate” their members to call the gate- aged care on health care expenditures. Washington, DC: CBO, 1992. visits, suggesting caution when keeper prior to 9-1-1. However promising, these 9. Congressional Budget Office. CBO considering any single protocol Memorandum: The effects of managed that restricts access to ED care. models remain largely theoreti- care and managed competition. WashGatekeeping ED or EMS care cal. For them to succeed, their ington, DC: CBO, February 1995. 10. O’Brien GM, Shapiro MJ, Woolard is complex and not without risks, purposes must be very clear, al- RW, O’Sullivan PS, Stein MD. “Inapproyet this practice occurs every day. ternative EMS resources as en- priate” emergency department use: a comparison of three methodologies for As suggested above, an optimal visioned by these models must be identification. Acad Emerg Med. 1996;3: further developed, and, most gatekeeping system may require important, the hard work of de- 252-7. 11. Koenig KL. Unscheduled access to services and practices difficult to veloping well-thought-out and health care: reengineering the 911 sysimplement, such as clear, nonconwell-validated decision criteria tem [commentary]. Acad Emerg Med. flicting purposes, primary care must be completed.-KEITH W. 1996;3:989-91. physicians to provide the service, NEELY, MPA, Department of 12. Neely KW. Ambulance transports: what are the alternatives? [commeneasily accessible alternatives, and Emergency Medicine, Oregon tary]. Acad Emerg Med. 1997;4:1103-6. clear, reliable, valid criteria for determining caller need. I also advocate redesigning EMS systems to reduce the need Clinical Wound Evaluation Scales for this practice. Previous authors have suggested such models n this issue of Academic ference (MCID) on the visual in this journal. Koenigll reported Emergency Medicine, Quinn analog cosmesis scale.’ The work done in Alameda County, and Wells determine the min- importance of these results and CA, that would integrate a “non- imal clinically important dif- their relevance to clinical wound tance, because of clinical, ethical, or legal concerns, to deny care to a patient already brought to the ED. Other work also supports the view that ED gatekeeping may not be effective as a long-term strategy to decrease ED use, if this is a desirable goal at alls Associated with this discussion is whether there are readily accessible alternatives to ED (or traditional EMS) care if such is denied by the gatekeeper. Presumably a patient has an easier time seeking care if given sound guidance prior to presenting to a n ED. Several reports suggest the importance of having readily available alternatives, especially after hours, if gatekeeping is to be a safe p r a ~ t i c e . ~ - ~ J
I
ACADEMIC EMERGENCY MEDICINE
June 1998, Volume 5, Number 6
research should not be underestimated. Although wounds are one of the most common problems encountered by emergency physicians, well-designed clinical research regarding the management of lacerations has been lacking. Researchers such as Edlich and colleagues have helped pave the way with their excellent and systematic animal wound researchz; however, these results may not always generalize to the clinical ~ e t t i n g . ~ One of the major factors limiting clinical wound research has been the lack of good outcome measures with which to compare treatments. To date, most of our practices regarding wound management are based on animal studies or small case series involving p a t i e n t ~ .With ~ the recent shift toward evidence-based medicine, it is more important than ever to identify valid and reliable scales with which to measure the clinical outcome of wound repair. Until recently, the primary outcome measured after laceration repair has been the incidence of wound infections. However, this measure has several major drawbacks limiting its usefulness. First, the definition of a wound infection itself is contr~versial.~ Second, the overall incidence of wound infection is low.6Thus, a very large sample size is required to demonstrate a significant difference between treatments. For example, for a study to have an 80% power to show a 33% relative reduction in infection rates from 3% to 2%, a sample size of 3,800 subjects in each treatment group would be required. Finally, a good outcome measure should be clinically relevant. Most infected wounds heal without major sequelae. As Quinn and Wells point out in their article, for most patients who have sustained lacerations, the final cosmetic appearance of the wound scar is the most important outcome. It is therefore
evident that cosmetic appearance is the most appropriate method by which the outcome of laceration repair should be measured. How, then, does one reliably measure such a subjective parameter as cosmetic appearance? Quinn et al. were the first group to evaluate the use of a visual analog scale WAS) for the measurement of the cosmetic appearance of This approach was logical given that similar methods had been shown to be both valid and reliable for the measurement of other highly subjective parameters such as pain.8 Furthermore, VASs have been found t o be highly sensitive and generally amenable to parametric statistical analyses. Studies using these scales require relatively small sample sizes t o demonstrate a significant difference between treatments. As a result, they are very powerful research tools. Using a VAS, Quinn et al. demonstrated that plastic surgeons have a reasonable degree of reliability when reviewing photographs of healed scars.' However, as Quinn and Wells note, a major drawback of the VAS is that it lacks any descriptive reference that would otherwise help interpret the actual meaning of the quantitative value assigned to a wound on the
VAS .
In an attempt t o complement the VAS for cosmesis, our group developed a 6-item categorical wound evaluation s ~ a l e . This ~.~ scale is composed of 5 specific potential imperfections (such as a wide scar) and a sixth subjective component (overall appearance). Like the VAS for cosmesis, this scale was found t o be reliable. Its major advantage, however, is its ability to provide practitioners with specific feedback regarding the technical aspects of laceration repair. The disadvantage of this scale is that it requires a large number of subjects to demonstrate a significant difference
between those patients who receive an optimal score (score = 6) and those who are given a suboptimal score (score 5 6 ) . Thus, both wound evaluation scales have merit and complement each other. In order for a new measurement tool, such as the wound cosmesis scale, t o be clinically useful, it must be reliable, valid, and clinically significant. Reliability refers t o the test's consistency with repeated measurements and is relatively easy to assess. In the current study, Quinn and Wells have clearly demonstrated the reliability of both the VAS and the categorical wound evaluation scale. Validity refers to how well the scale truly measures the characteristic it is intended to study as judged by some external criteria. Ideally, criterion-related validity should be established in which the test or scale under consideration is compared with an absolute criterion standard. Unfortunately, in the case of the cosmetic appearance of wounds, no such criterion standard exists. In such cases, construct validity may be determined by comparing the new scale or test with other tests that purport t o measure the same characteristics. Construct validity also is satisfied when test performance is logical. By comparing the VAS for cosmesis with the categorical wound evaluation scale, Quinn and Wells have also helped validate these 2 scales. Quinn and Wells have gone a step further by attempting to determine the clinical significance of the VAS for cosmesis. The importance of this stems from the fact that not all statistically significant differences are clinically significant. Thus, to interpret clinical trials appropriately, it is essential to distinguish between statistical significance and clinical significance. In this issue of Academic Emergency Medicine, Quinn and Wells have made an
666
COMMENTARIES
admirable attempt t o determine the MCID on the VAS. The approach taken by Quinn and Wells is similar to that of Todd et al.1° and Redelmeier and Lorig,ll in which a difference in a score on the VAS is believed to be clinically important if it is also associated with a difference that a typical patient or physician would notice. The actual threshold value thus derived, while not an absolute value, serves as a rough guideline to help plan and interpret clinical outcome trials such as in wound research. In Todd et al.’s study, the minimum clinically significant change in VAS pain score was defined as the mean difference between current and preceding VAS scores when the subject noted “a little less” or “a little more” pain between evaluations. In the current study, Quinn and Wells calculated the MCID on the VAS for cosmesis by subtracting the mean score of optimal wounds from that of suboptimal wounds. Remarkably, they found that this difference was 15 mm for both traumatic lacerations and surgical incisions. In a previous study comparing patient and practitioner satisfactions with wound appearance after laceration repair, we found a 10-mm difference between optimal and suboptimal wounds.12 Interestingly, this range of 1015 mm is quite similar to the MCID in pain scores of 13 mm found by Todd et a1.l0 Quinn and Wells should be commended for their contribution to clinical wound research. However, their method of having
Singer
CLINICAL WOUND EVALUATION SCALES
plastic surgeons evaluate the cosmetic appearance of scars based on photographs has several potential drawbacks. First, despite the high quality of the photographs, they cannot be expected to replace the actual physical assessment of a patient’s scar. Live assessment allows the wound to be evaluated from multiple visual angles and under various lighting conditions. Furthermore, it allows palpation of the wound as well as its inspection, making many contour defects more obvious. Second, by choosing plastic surgeons to evaluate the wounds, Quinn and Wells imply that plastic surgeons are the best judges of cosmetic appearance. While plastic surgeons have the most experience managing poor scars, it seems more appropriate for the patients themselves to be the judges of the cosmetic outcome of their wounds since the goal of laceration repair is t o achieve a functional and cosmetically appealing scar for the patient. In the future, the value of new wound management techniques should be assessed using clinically relevant wound healing and cosmetic appearance scales within a prospective, randomized, blinded design before new therapies are implemented into clinical practice. In their current publication, Quinn and Wells have helped pave the way for this type of valuable research.ADAM J. SINGER,MD, Department of Emergency Medicine, State University of New York, University Medical Center, Stony Brook, NY
Keu words. wound closure; cosmesis; healing; methodology; visual analog scale.
RMerences 1. Quinn JV,Wells GA. An assessment of clinical wound evaluation scales. Acad Emerg Med. 1998;5583-6. 2. Edlich RF’, Custer J, Madden J, Dajani AS, Rogers W, Wangensteen OH. Studies in the management of the contaminated wound 111. Assessment of the effectiveness of irrigation with antiseptic agents. Am J Surg. 1969;118:21-30. 3. Hollander JE, Richman PB, Werblud M, Miller T, Huggler J , Singer AJ. Irrigation in facial and scalp lacerations: does it alter outcome? Ann Emerg Med. 1998;31:73-7. 4. Singer AJ, Hollander JE, Quinn JV. The evaluation and management of traumatic lacerations. N Engl J Med. 1997; 337:1142-8. 6. Thomson PD, Smith DJ. What is infection? Am J Surg. 1994;167(1A suppl): 7s- 10s. 6. Hollander JE,Singer AJ, Valentine S, Henry MC. The wound registry: development and validation. Ann Emerg Med. 1995;25:675-85. 7. Quinn JV, Drzewieki A, Stiell IG, Elmslie TJ. Appearance scales to measure cosmetic outcomes of healed lacerations. Am J Emerg Med. 1995;13:22931. 8. Huskisson EC. Visual analogue scales. In: Melzack R. Pain Measurement and Assessment. New York Raven Press, 1983. 9. Hollander JE, Blasko B, Singer AJ, et al. Poor correlation of short- and longterm cosmetic appearance of repaired lacerations. Acad Emerg Med. 1996; 2: 983-7. 10. Todd KH, Funk KG, Funk JF’, Bonacci R. The clinical significance of reported changes in pain severity. Acad Emerg Med. 1995;2:369-70. 11. Redelmeier DA, Lorig K. Assessing the clinical importance of symptomatic improvements: a n illustration in rheumatology. Arch Intern Med. 1993; 153: 1337-42. 12. Singer AJ, Church AL,Forrestal K, Werblud M, Valentine SM, Hollander JE. Comparison of patient satisfaction and practitioner satisfaction with wound appearance after traumatic wound repair. Acad Emerg Med. 1997;4:133-7.
