Page 8 - Kansas Journal of Medicine, Volume 10 Issue 3
P. 8

KANSAS JOURNAL of M E D I C I N E                                            for MS3s and 100% incorrect for MS4s). Neither were statistically
                                                                             significant (95% CI: 7.2 to 1.1% and -9.7 to 0.49%, respectively).
TRAUMA                                                                          Percentage of incorrect responses ranges from 18.5% to 100%
continued.                                                                   for MS3s and from 0.0% to 100% for MS4s. The image with the
                                                                             least incorrect responses by both groups was epidural hemor-
    Table 1. Life-threatening computed tomography injury and                 rhage, 33.9% and 18.5% incorrect by MS3s for diagnosis and
    treatment images.                                                        treatment, respectively, and 11.1% and 0% for MS4s. Excluding
                                                                             epidural hemorrhage, the range of incorrect for MS3s was 75.4%
          Injury Type                         Treatment                      to 100% and 44.4% to 100% for MS4s. The median percentage
                                                                             of incorrect responses for MS3s was 93.1% and 77.8% for MS4s.
    Head                                                                     No MS3s were able to identify or propose a correct diagnoses or
                                                                             treatment for one of two grade III liver lacerations. No MS4s cor-
          Epidural hematoma (EDH) Operative intervention or drainage         rectly diagnosed the grade IV renal injury and no MS4s correctly
                                                                             identified a treatment for one of two grade III liver lacerations.
          Subdural hematoma (SDH) Medical management
                                                                             Figure 1. Percentage of incorrect responses for diagnosis by group.
    Intraparenchymal hemorrhage Medical management

    Chest, Abdomen and Pelvis

          Pulmonary contusion and Chest tube

          Grade III liver laceration Observation in the intensive care

          Grade III liver laceration Observation in the intensive care

          Grade IV splenic injury Embolization or operative

          Grade IV renal injury Embolization or operative

          Pelvic fracture Operative intervention or
                              sacroiliac screw

          Small bowel thickening Serial exam or operative

          Right colon mesenteric injury Serial exam or operative

       Survey forms were de-identified when scored and were re-              Figure 2. Percentage of incorrect responses by treatment by group.
    viewed separately and scored by two trauma surgeons. One
    point was assigned for a correct response, half a point for a               Five of the twenty-two questions reflected a statistically signifi-
    partially correct response, or zero points for an incorrect re-          cant reduction in incorrect responses between MS3s and MS4s.
    sponse. Each student’s scores were averaged for each of the              These included: diagnosis of subdural hemorrhage (85.2% incor-
    22 questions. Both groups were compared by total points for              rect MS3 versus 44.4% MS4, 95% CI: 8.2% to 75%); diagnosis of
    each image, and proportion of incorrect responses was cal-               grade III liver laceration (100% MS3 versus 66.7% MS4, 95% CI:
    culated by group. Differences in proportions were calculated.            2.5% to 64%); treatment of epidural hematoma (18.5% incorrect
                                                                             by MS3 versus all correct by MS4, 95% CI: 9.0% to 28%); treat-
       Due to sample size, focus on the error reduction comparison was       ment of small bowel thickening (98.5% MS3 versus 66.7% MS4,
    viewed through proportional differences, assessing statistical sig-      95% CI: 0.85% to 63%); and treatment of right colon mesenteric
    nificance with a 95% confidence interval, corrected for continuity.      injury (90.8% MS3 versus 55.6% MS4, 95% CI: 2.0% to 68%).
    Confidence intervals of 95% were calculated, of which those inter-
    vals, including zero, were considered statistically insignificant. Bon-
    ferroni correction indicated one out of 18 comparisons, given this
    confidence interval. Analyses were conducted using SPSS release
    19.0 (IBM Corp, Somers, New York). The study was approved by the
    Institutional Review Board of Via Christi Hospitals, Wichita, Inc.

       Survey results were collected from 65 of 65 (100%) MS3s and

    9 of 60 (15%) MS4s (N=74). Overall, MS4s performed better than
    MS3s, with fewer incorrect responses on 20 of 22 questions (Fig-
    ures 1 and 2). Two exceptions were noted: diagnosis of a grade IV
    renal injury (96.9% incorrect for MS3s and 100% incorrect for
    MS4s) and treatment of a grade III liver laceration (95.4% incorrect

   3   4   5   6   7   8   9   10   11   12   13