ABSTRACT
BACKGROUND: This study assesses the relationship that the brand of trauma program registry (TPR) has on mortality rate (MR) in the reports prepared by the American College of Surgeons Committee on Trauma (ACSCOT) trauma center (TC) site surveyors. METHODS: Data from 242 ACSCOT adult TC survey reports (88 Level I, 115 Level II, and 39 Level III) were analyzed for annual trauma volume, injury severity score (ISS), MR, and TPR. Six TPR (A through F) were identified; group F was a composite of several infrequently used TPRs. This report focuses on the ISS range 16-24 because of the likelihood that the mean for each TC would be near 20 and MR is high enough so that a difference, if present, could be statistically documented. RESULTS: For the total group, MR showed no correlation with TC volume or TC level for ISS 16-24. MR was significantly different according to which TPR was used by the TCs. The MR is less (4.8%) for 14 high volume TCs (over 1200 admits) using TPR A compared with 33 low volume TCs (below 800 admits) using TPR A (6.34%). CONCLUSION: The MR for ISS 16-24 in ACSCOT-surveyed TCs differs within subgroups based on type of TPR utilized. This may reflect improper use of the software programs. Enhanced skill in the application of software programs designed to generate ISS scores is essential if meaningful studies on the effects of improved trauma care on MR are to be conducted. Hand scored ISS by trained personnel may circumvent this problem.
Subject(s)
Diagnosis-Related Groups/statistics & numerical data , Registries , Trauma Centers/organization & administration , Wounds and Injuries/mortality , Confounding Factors, Epidemiologic , Data Interpretation, Statistical , Humans , Injury Severity Score , United States/epidemiology , Wounds and Injuries/classificationABSTRACT
BACKGROUND: Level II trauma centers may be verified (1999, American College of Surgeons Committee on Trauma) with an on-call operating room team if the performance-improvement program shows no adverse outcomes. Using queuing and simulation methodology, this study attempted to add a volume guideline. STUDY DESIGN: Data from 72 previously verified trauma centers identified multiple demographic factors, including specific information about the first trauma-related operation that was done between 11:00 PM and 7:00 AM each month for 12 consecutive months. RESULTS: The annual admissions averaged 1,477 for 37 Level I trauma centers, 802 for 28 Level II trauma centers, 481 for 4 Level III trauma centers, and 731 for 3 pediatric trauma centers. The annual admissions correlated with the number of operations done between 11:00 PM and 7:00 AM (p < 0.001). These 946 operations were performed by general surgery (39%), neurosurgery (8%), orthopaedic surgery (33%), another specialty (9%), or multiple services (10%). Admission to operation time was within 30 minutes for 12.1% of patients (2.6% for blunt and 24.1% for penetrating injuries). The probability of operation within 30 minutes of arrival varied with the number of admissions and with the percentage of penetrating versus blunt injuries. The likely number of operations from 11:00 PM to 7:00 AM would be 19 for 500 annual admissions, 26 for 750 annual admissions, and 34 for 1,000 annual admissions, with 5.83, 7.98, and 10.13 patients, respectively, going to operation within 30 min. The probability that two rooms would be occupied simultaneously was 0.14 and 0.24 for centers admitting 500 and 1,000 patients, respectively. CONCLUSIONS: Trauma centers performing fewer than six operations between 11:00 PM and 7:00 AM per year could conserve resources by using an immediately available on-call team, with responses monitored by the performance-improvement program.
Subject(s)
Models, Statistical , Operating Rooms/statistics & numerical data , Personnel Staffing and Scheduling Information Systems , Total Quality Management/organization & administration , Trauma Centers , Guidelines as Topic , Health Services Research , Humans , Linear Models , Needs Assessment/organization & administration , Patient Admission/statistics & numerical data , Predictive Value of Tests , Surveys and Questionnaires , Systems Theory , Time Factors , United States/epidemiology , Workforce , Wounds, Nonpenetrating/epidemiology , Wounds, Nonpenetrating/surgery , Wounds, Penetrating/epidemiology , Wounds, Penetrating/surgeryABSTRACT
This report uses a mathematical modeling system to define optimal orthopedic coverage for trauma centers. Data from 2,325 patients treated with emergency orthopedic operations within 24 hours of admission at 78 randomly sampled and at four totally sampled verified centers were used to create a profile of (1) admission by month, day, and hour; (2) operation times; and (3) operation duration. The reason for operation included (1) open fracture or crush (809 patients); (2) irreducible dislocations (164 patients); (3) fracture with vascular injury (seven patients); (4) dislocation with vascular injury (17 patients); (5) compartment syndrome (11 patients); (6) femoral neck fracture in young patients (36 patients); (7) combination of categories 1 to 6 (70 patients); (8) fracture with multiple injuries (171 patients); and (9) urgent not emergent (1,040 patients). The program defined the frequency that an injured patient needing an orthopedic consult would wait beyond 30 minutes because the orthopedic surgeon was doing a trauma related operation at a center with one or two orthopedic surgeons on call. The probability that a patient cannot be seen promptly by one orthopedic surgeon in a center doing 25, 50, 75, 100, 200, and 300 emergency procedures per year is 0.17, 0.74, 1.6, 3.1, 12.5, and 28 patients per year. When two are on call, 1.3 patients, yearly, will wait more than 30 minutes in a center doing 300 emergency procedures. Thus, mandatory orthopedic backup call for a trauma center performing fewer than 100 emergent trauma procedures within 24 hours is unwarranted.
