A phased-outcome clinical pathway for the management of moderate traumatic brain injury.
Subject: Brain research
Hospital patients (Care and treatment)
Hospital patients (Health aspects)
Hospital patients (Medical examination)
Intensive care nursing (Methods)
Authors: Mcilvoy, Laura
Spain, David A.
Vitaz, Todd
Meyer, Kimberly
Pub Date: 02/01/2002
Publication: Name: Journal of Neuroscience Nursing Publisher: American Association of Neuroscience Nurses Audience: Professional Format: Magazine/Journal Subject: Health care industry Copyright: COPYRIGHT 2002 American Association of Neuroscience Nurses ISSN: 0888-0395
Issue: Date: Feb, 2002 Source Volume: 34 Source Issue: 1
Geographic: Geographic Scope: United States Geographic Code: 1USA United States
Accession Number: 92801136
Full Text: Abstract: The literature contains numerous articles examining all aspects of severe traumatic brain injury (TBI). Guidelines for the management of severe TBI are available to assist practitioners in the care of these patients. In contrast, very little literature is available and no care guidelines have been developed for patients with moderate TBI. For 3 years a Level I trauma center successfully guided the care of patients with moderate TBI utilizing a phased-outcome clinical pathway that reduced both hospital and intensive care unit (ICU) lengths of stay as well as number of ventilator days. During this time, a subset within moderate TBI was identified that benefited from more intensive therapies. Patients in this subset, labeled moderate TBI with severe tendencies, were older and suffered from multiple traumatic injuries in addition to TBI. Their response to a structured plan of care was especially significant in terms of decreased length of stay and number of ventilator days.

**********

Numerous articles have been published detailing the successful use of clinical pathways in improving patient outcomes while containing costs, particularly in patients with neurological problems (Guin & Nalli, 1998; Hainsworth, Lockwoode-Cook, Pond, & Lagoe, 1997; Hydo, 1995; Mcilvoy, Meyer, & Vitaz, 2000; Mcilvoy et al., 2001; Spain et al., 1998; Summers & Sopher, 1998). However, there is nothing in the literature about the use of a clinical pathway with a moderate TBI population. In 1995, a Level I trauma center implemented clinical pathways for both severe and moderate TBI populations. The outcomes achieved for the severe TBI population have been published (Mcilvoy et al., 2001). This article discusses the lack of published information available on patients with moderate TBI as well as the successful outcomes achieved utilizing a phased-outcome clinical pathway.

Review of Literature

Injury Severity

Historically, TBI has been classified as severe, moderate, or mild in intensity, based on severity of injury. The Glasgow Coma Scale (GCS) has been widely utilized to grade severity of neurological injury. The National Traumatic Coma Data Bank describes injury severity by the patient's GCS score: a GCS score of 3-8 denotes severe injury; a GCS score of 9-12 denotes moderate injury; and a GCS score of 13-15 denotes mild injury (Marshall et al., 1983).

Information on severe brain injury has routinely appeared in the literature since the early 1970s. In the last few years, the spotlight has been turned toward mild TBI and the frequently overlooked, misdiagnosed sequelae that can occur. However, very little has been written specifically about patients with moderate TBI, a population that is clearly defined and differentiated from those with severe TBI.

Rimel, Giordani, Barth, & Jane published the earliest article specific to moderate TBI in 1982. They looked at 199 patients with GCS scores of 9-12. Only 38% had computed tomography (CT) scans (30% of these were negative), 10% required craniotomies or burr holes, and 18% had intracranial pressure (ICP) monitoring. Only 159 patients had a 3-month follow-up with 38% reporting good recovery, though 93% of these complained of headaches. Of the patients who were previously employed, 63% became unemployed post injury.

Mortality

Several studies have looked at mortality differences between severe and moderate TBI. In 1988, Luerssen, Klauber, & Marshall examined mortality rates in patients with similar neurological exams stratified by age. For patients who were younger than 15 years, the mortality rate was 1.4% for those with moderate TBI, compared with the expected higher rate of 28.4% for severe TBI (n = 1,906), while for patients who were 15 years or older, the mortality rate was 6.8% for those with moderate TBI, compared with 47.7% for severe TBI (n = 6,908). In 1992, Ross, Pitts, & Kobayashi found a 69% mortality rate in patients with moderate TBI who were 65 to 99 years and had a GCS score of 9-12 (n = 42). Of the patients with moderate TBI who survived, 24% suffered moderate disability or worse as determined by the Glasgow Outcome Score. Moderate TBI also has been associated with doubling the predicted mortality rate in patients with multi-system trauma (McMahon, Yates, Campbell, Hollis, & Woodford, 1999).

Plan of Care

The Guidelines for the Management of Severe Head Injuries (Bullock, Chestnut, & Clifton, 1995), developed by the Aitken Neuroscience Center in partnership with the American Association of Neurological Surgeons, were introduced in 1995 to provide practitioners with a standardized systematic treatment plan for the care of patients with severe TBI. In 2000, they were expanded to include management and prognosis of severe TBI (Bullock, Chestnut, & Clifton, 2000). The Eastern Association for the Surgery of Trauma has developed Practice Management Guidelines for the Management of Mild TBI, which provide Level II and Level III recommendations for the management of patients suffering a mild TBI (East Practice Management Guideline Work Group, 2000). However, there is no national consensus on the treatment of patients with moderate TBI.

Two studies have advocated a systematic plan of neurotrauma care for patients with moderate TBI (Fearnside & McDougall, 1998; Stein & Ross, 1992). Both define moderate TBI as a GCS score of 9-13 and recommend hospitalization for all patients with moderate TBI. Patients with abnormal CT scans should be admitted to intensive care units and a scan repeated 12-24 hours later. With comparable sample sizes (n = 106, n = 110), both authors reported that 18% of the moderate TBI population required intracranial pressure monitoring and 8-14% required craniotomies.

University of Louisville Experience

A phased-outcome clinical pathway can be an excellent tool for providing a structured collaborative plan of care to a trauma population. The phases of care are determined by levels of care (specific treatment modalities) that all patients with that specific diagnosis require. Each phase of care identifies patient and family outcomes that indicate significant improvement. Achievement of these outcomes signals that the patient is ready to move to the next phase. This individualizes the pathway for each patient allowing the patient to progress at his or her own rate. A post-injury day (PID) target is assigned to priority outcomes to identify outliers.

In October 1995, the University of Louisville Hospital, a Level I trauma center, implemented clinical pathways for patients with severe and moderate TBI (Mcilvoy et al., 2001). A multidisciplinary team under the direction of trauma and neurosurgical attending physicians developed these pathways. A neurosurgical clinical nurse specialist (CNS) followed all patients admitted with a diagnosis of TBI from admission to the intensive care unit (ICU) until discharge from the hospital. She coordinated the care they received as outlined by the clinical pathway, provided education to patients and families, and served as liaison between different medical specialties.

Each TBI patient entered the system with pre-printed admission orders. The severe TBI and moderate TBI admission orders were identical except for the addition of phenytoin to the severe pathway orders. Phase I was the first 24 hours post admission during which routine diagnostic procedures were completed. Phase 2, acute critical care and weaning, encompassed spinal clearance, initiation of feeding, and implementation of ventilator weaning protocols. A fast wean protocol was used initially for patients with moderate TBI. This protocol involved placing the patient on continuous positive airway pressure (CPAP) and performing weaning parameters. Patients who did not tolerate this type of weaning were switched to the slow wean protocol in which the ventilator rate was decreased by 2 breaths per minute every hour or as tolerated until the patient was breathing on a rate of 4 breaths per minute, at which time they were placed on CPAP again and weaning parameters were obtained. If the slow weaning process was unsuccessful, patients received a tracheostomy and slow weaning was once again attempted. Early tracheostomy has been shown to be beneficial in trauma patients and is routinely done early in the severe TBI population at this facility (Mcilvoy et al., 2001).

Phase 3, mobility and pre-rehabilitation, initiated progressive mobility and resolved ongoing medical issues. The discharge planning formulated on admission was finalized. Speech therapists performed cognitive functional assessments to determine appropriate therapy and discharge placement requirements. Figure 1 illustrates the actual clinical pathway used, including phases of care with targeted time frames and expected patient and family outcomes.

Data Analysis

All patients admitted with a GCS score of 8-13 and placed in an ICU were placed on the clinical pathway. However, for statistical analysis, a moderate TBI was defined as a trauma patient with a GCS score of 8-13 at 24 hours post admission. This allowed time for high alcohol or drug levels that might falsely lower consciousness to clear. Patients with gunshot wounds to the head were excluded from statistical analysis. A retrospective chart audit was performed to identify moderate TBI patients during the previous 12 months to provide comparative data for a historical control (HC) group.

Early during data collection a subset of patients within the moderate TBI population was identified and labeled moderate TBI with severe tendencies (ST). These patients exhibited the GCS score of a moderate TBI but the functional impairment of a severe TBI requiring a tracheostomy or a feeding tube at time of discharge. For statistical analysis the moderate TBI with ST were analyzed separately.

Moderate TBI

The 67 patients in the HC group (October 1994 to September 1995) were compared with the 96 patients treated with the clinical pathway (CP) group over 3 years (October 1995 to September 1998). Demographic and patient outcome data were analyzed by using the students' t test. The Injury Severity Score (ISS) was used to grade overall system injuries (Baker, O'Neill, Haddon, & Long, 1974).

Table 1 shows the differences between the HC group and the CP group. The percentage of isolated TBI versus TBI with multiple trauma was similar in both HC and CP groups. There was no statistical difference in mean age, mean GCS score 24 hours post resuscitation, or mean ISS between the HC group and the CP group.

However, there were statistical differences between the two groups in length of stay and time spent on a ventilator. Hospital length of stay was a mean of 14.2 days in the HC group and 11.9 days in the CP group (p = .005), resulting in a mean savings of 2.3 hospital days per patient. The ICU length of stay decreased from a mean of 6.9 days in the HC group to mean of 6.4 days in the CP group but was not statistically significant. Patients in the CP group had their ventilators discontinued 1 day earlier than those in the HC group.

Because these patients were admitted to the ICU, timing of when they were fed and pneumonia incidence was monitored. The CP group was fed 1.03 days earlier than the HC group (M = 3.46 versus M = 4.43) (p = .001). For ease of data collection, pneumonia incidence was defined as any patient receiving antibiotic treatment for pneumonia. The clinical pathway reduced pneumonia incidence by 21%, from 39% to 18%.

Moderate TBI with ST

There were 15 patients in the HC group and 57 patients in the CP group who had a GCS score of 8-13 and were discharged with the functional impairment of a severe TBI requiring either tracheostomy and/or feeding tube. Table 2 shows the differences between the two groups. Demographically there were no statistical differences between the HC and CP groups. The percentage of isolated TBI versus multiple trauma was similar in both groups.

Hospital length of stay for the CP group decreased by 17.6 days (p = .006). ICU length of stay also was significantly affected: 21.9 mean days for the HC group versus 15.4 mean days for the CP group (p = .06). Although the decrease of 5.9 ventilator days for the HC group versus the CP group was not statistically significant, it is obviously clinically significant.

Pneumonia incidence in this subset paralleled what was found in the severe TBI population. Of the HC group, 80% experienced pneumonia, whereas 77% of the CP group developed pneumonia (Mcilvoy et al., 2001). As with the severe TBI patients, the patient outcomes of PID feeding and timing of tracheostomy placement were monitored. The CP group was fed 2.23 days earlier (HC group mean PID 6.3, CP group mean PID 4.07, p = .01). To evaluate the effect of the clinical pathway on tracheostomy placement in the moderate TBI with ST subset, the CP group was compared with the severe TBI CP group, who all received tracheostomies. A total of 188 patients in the severe TBI CP group received tracheostomies on mean PID 4.3, while 48 (84%) of patients in the moderate TBI with ST CP group received a tracheostomy on mean PID 5.3, one day later. This delay was attributed to the time spent attempting the slow wean protocol before reaching the decision to proceed with tracheostomy.

Differences

On examination there were several notable differences between patients with moderate TBI and patients with moderate TBI with ST. The moderate TBI with ST group was older (N = 71, mean age 45.8, 61% > 40 years old, 38% > 50 years old) than the moderate TBI groups (N = 163, mean age 35, 23% > 40 years old, 11% > 50 years old.

Another major difference was percentage of multiple trauma. Of patients with moderate TBI with ST (n = 72), 71% had multiple traumatic injuries in addition to their TBI, while only 54% (n = 163) of the patients with moderate TBI experienced multiple traumatic injuries. The craniotomy rate was very similar in both groups: 15% in the moderate TBI group and 17% in the moderate TBI with ST group. Only one patient in the moderate TBI with ST population was monitored for increased intracranial pressure.

Post-discharge disposition was clearly different between the HC and CP groups. For the moderate TBI CP group, a higher percentage of patients was discharged to rehabilitation than the moderate TBI HC group. Patients in the CP group who were sent home received more outpatient rehabilitation than those in the HC group (Table 3). A team comprising both community and hospital rehabilitation experts, a social worker, and a CNS followed the CP patients from admission to discharge. This team was very aware of the rehabilitation needs of moderate TBI patients and facilitated both inpatient and outpatient placement. In contrast, patients with moderate TBI with ST, regardless of clinical pathway usage, were primarily sent to a rehabilitation setting or to a nursing home (Table 4). The higher incidence of functional impairment in the patients with moderate TBI with ST necessitated a higher level of post-discharge care.

Discussion

The needs of patients with moderate TBI are in danger of being overlooked among the numerous publications that focus on patients with severe TBI. Yet patients with moderate TBI require a specific focused plan of care to maximize their outcomes. The subset of patients with moderate TBI with ST requires almost the same level of support as patients with severe TBI. When hospital and ICU lengths of stay are compared among patients with moderate TBI, moderate TBI with ST, and severe TBI, there is little difference between patients with severe TBI and those with moderate TBI with ST (Table 5).

Because there is very little in the literature about patients with moderate TBI and no information on clinical pathway usage with this population, this information is important for professionals caring for patients with moderate TBI. Recognizing the effect of age and degree of injury on patients with moderate TBI will allow for earlier identification of patients who require therapies such as tracheostomy, feeding tube placement, or both to facilitate their care.

Summary

Clinical pathways have been proven to be excellent tools in providing a systematic plan of care that improves patient outcomes while decreasing utilization of resources and associated costs. The University of Louisville Hospital has successfully managed the neuroscience populations of spinal cord injuries (Mcilvoy et al., 2000), severe TBI (Mcilvoy et al., 2001), and moderate TBI utilizing these tools. With the discovery of a subset within the moderate TBI population that requires more intensive therapies to achieve appropriate outcomes, a systematic plan of care becomes even more essential for patients with moderate TBI.

Acknowledgment

The authors acknowledge the nursing staff of the University of Louisville Hospital whose willingness to accept change and dedication to improve patient outcomes made implementation of the clinical pathway program a success.

Questions or comments about this article may be directed to: Laura Mcilvoy, MSN RN CCRN CNRN, Floyd Memorial Hospital, New Albany, IN 47150, lmcilvoy@worldnet.att.net. She is a staff nurse at Floyd Memorial Hospital and a doctoral student at the Indiana University School of Nursing, Indianapolis.

References

Baker, S.P., O'Neill, B., Haddon, W., & Long, W.B. (1974). The Injury Severity Score: A method for describing patients with multiple injuries and evaluating emergency care. Journal of Trauma, 14, 187-196.

Bullock, R., Chestnut, R.M., & Clifton, G. (1995). Guidelines for the management of severe head injury, Retrieved October 2000 from The Brain Trauma Foundation, Inc. Web site: http://www.braintrauma.org.html

Bullock, R.M, Chestnut, R.M., & Clifton, G.L. (2000). Management and prognosis of severe traumatic brain injury. Journal of Neurotrauma, 17 (6 & 7), 451-627.

The EAST Practice Management Guideline Work Group. (2000). Practice management guidelines for the management of mild traumatic brain injury. Retrieved October 1, 2000, from Eastern Association for the Surgery of Trauma Web site: http://www.cast.org

Fearnside, M., & McDougall, P. (1998). Moderate head injury: A system of neurotrauma care. Australian New Zealand Journal of Surgery, 68, 58-64.

Guin, P., & Nalli, B. (1998). Decreasing roadblocks and improving outcomes: The spinal cord injury (SCI) clinical pathway. SCI Nursing, 15, 66-74.

Hainsworth, D.S., Lockwoode-Cook, E., Pond, M., & Lagoe, R.J. (1997). Development and implementation of clinical pathways for stroke on a multihospital basis. Journal of Neuroscience Nursing, 29, 156-162.

Hydo, B. (1995). Designing an effective clinical pathway for stroke. American Journal of Nursing, 3, 44-51.

Luerssen, T.G., Klauber, M.R., & Marshall, L.F. (1988). Outcome from head injury related to patients age. Journal of Neurosurgery, 68, 409-416.

Marshall, L.F., Becker, D.P., Bowers, S.A., Cavard, C., Eisenberg, H., Gross, C.R., Grossman, R.G., Jane, J.A., Kunitz, S.C., Rimel, R., Rabaddor, K., & Warren, J. (1983). The National Traumatic Coma Data Bank. Part I: Design, purpose, goals, and results. Journal of Neurosurgery, 59, 276-284.

Mcilvoy, L., Meyer, K., & Vitaz, T. (2000). Use of an acute spinal cord injury clinical pathway. Critical Care Nursing Clinics of North America, 12, 521-530.

Mcilvoy, L., Spain, D.A., Raque, G., Vitaz, T., Boaz, P., & Meyer, K. (2001). Successful incorporation of the severe head injury guidelines into a phased-outcome clinical pathway. Journal of Neuroscience Nursing, 33, 72-78.

McMahon, C.G., Yates, D.W., Campbell, EM., Hollis, S., & Woodford, M. (1999). Unexpected contribution of moderate traumatic brain injury to death after major trauma. Journal of Trauma, 47, 891-895.

Rimel, R.W., Giordani, B., Barth, J.T., & Jane, J.A. (1982). Moderate head injury: Completing the clinical spectrum of brain trauma. Neurosurgery, 11, 344-351.

Ross, A.M., Pitts, L.H., & Kobayashi, S. (1992). Prognosticators of outcome after major head injury in the elderly. Journal of Neuroscience Nursing, 24, 88-93.

Spain, D.A., Mcilvoy, L.H., Fix, S.E., Carillo, E.H., Boaz, P.W., Harpring, J.E., Raque, G.H., & Miller, F.B. (1998). Effect of a clinical pathway for severe traumatic brain injury on resource utilization. Journal of Trauma, 45, 101-105.

Stein, S. C., & Ross, S. E. (1992). Moderate head injury: A guide to initial management. Journal of Neurosurgery, 77, 562-564.

Summers, D., & Sopher, P. (1998). Implementation and evaluation of stroke clinical pathways and the impact of cost of stroke care. Journal of Cardiovascular Nursing, 13, 69-87.

David A. Spain, MD, is a professor at Stanford University, Stanford, CA.

Todd Vitaz, MD, is a neuro-oncology fellow at Memorial Sloan-Kettering Cancer Center, New York City.

Kimberly Meyer, MSN RN ARNP CNRN, is a neurosurgical acute care nurse practitioner in the Department of Neurological Surgery at the University of Louisville School of Medicine, Louisville, KY.
Fig 1. Moderate TBI phased-outcome clinical pathway

UNIVERSITY OF LOUISVILLE HOSPITAL

MODERATE TRAUMATIC BRAIN INJURY
PATIENT CARE MAP
ADMISSION GCS 8-13

* CONSIDER SEVERE TBI MAP IF > 50 YEARS OF AGE
AND/OR HAS SEVERE MULTISYSTEM INJURIES *

Aspects of           Phase 1:
Care                 Admission to Critical Care

MULTI-               VS & Neuro checks per standard
DISCIPLINARY         I & O's q shift
INTERVENTIONS        NG to LWS, NG pH q4h
All Interventions    F/C to BSD
require a            C-Collar
physicians order     Medical immobilizers prn
to be initiated or   Multipodus splits per PT
discontinued         TED hose
                     SCD's or plexipulse
                     Maintain normal PC[O.sub.2]
Interventions that   Preoxygenate with 100% [O.sub.2] before suctioning
occur within 24      Vent settings per trauma
hours of
scheduled time
do not create a
variance

MEDICATIONS &        D51/2NS 20 meq KCL adjust rate to maintain
IV'S                 euvolemia
                     Pepcid 20mg IV q12 hr when extubated D/C
                     MSO4 2-4 mgs IV q1hr prn pain/agitation
                     Tylenol 650 mg PR or PO q 4h Temp > 101.5,
                     Dulcolax supp prn
                     Maalox 30 cc per NG for pH <5
                     Lacrilube OU prn
                     KCL 10 meq in 25cc D5W over 1 hour x 3 for K < 3.5
                     (mix in 50cc for peripheral IV's)
                     Ativan/Versed 1-2 mgs IV q4h prn agitation,

NUTRITION            NPO

ACTIVITY &           Bedrest
POSITIONING          Reverse T-berg 30 degrees until spines are cleared
                     Logroll until spines are cleared

LAB TESTS            Admission ABG, CBC, Chem 7, PT, PTT,

DIAGNOSTIC           Chest x-ray in AM
PROCEDURES           Cat Scan of head and C-spines without contrast as
                     ordered by NES
                     VTX of T & L spines with AP and lateral views

CONSULTS             Social Worker
                     PT/OT
                     Rehab Consult

D/C PLANNING &       Discuss injury and care
EDUCATION

Aspects of           Phase 2: Acute Critical        Phase 3: Mobility
Care                 Care/Weaning                   & Pre-Rehab
                     (Target: 4 days)               (Target: 6 days)

MULTI-               Change to philadelphia
DISCIPLINARY         collar
INTERVENTIONS
All Interventions    Begin fast wean protocol
require a            when LOC improves and
physicians order     extubate (Target: PID 2)
to be initiated or
discontinued         If fails fast wean/initiate
                     slow wean protocol and
                     extubate as appropriate
Interventions that   (Target: PID 3)
occur within 24
hours of             Tracheostomy for failure
scheduled time       to wean (Target: PID 4)
do not create a
variance

MEDICATIONS &
IV'S

NUTRITION            Start Tube Feeds               TPN for pts not at
                     (Target: PID 3)                50% of nutritional
                     If extubated consider          goal by PID 6
                     bedside swallow study
                     metabolic cart per
                     dietitian order

ACTIVITY &           D/C Iogroll                    OOB to chair BID
POSITIONING          (Target: PID 1 )

LAB TESTS

DIAGNOSTIC           Flexion-extension films in
PROCEDURES           vascular lab with subseouent
                     C-Collar removal (Target:
                     PID 4

CONSULTS             Dietary
                     Pharmacy for Dilantin
                     Speech consult

D/C PLANNING &       Give family head injury book   Family to tour
EDUCATION            and patient pathway            rehab

Bolded interventions are EVIDENCE BASED GUIDELINES

Underlined Interventions are pathway outcomes that are monitored and
must be completed before patient progresses to the next phase

Reprinted with permission of University of Louisville Hospital,
Louisville, KY

EXPECTED PATIENT/FAMILY OUTCOMES:

1. The patient's T&L spines will be cleared by PID 1
2. The patient will have feedings started by PID 3
3. The patient will be extubated by PID 3
4. The patient/family will verbalize
   understanding/acceptance of rehab plans and goals

Table 1. Differences Between Historical Control (HC) Group and Clinical
Pathway (CP) Group of Patients with Moderate TBI

                       HC Group             CP Group             p

Isolated TBI           45%                  52%
Multiple trauma        55%                  48%
Number of patients     67                   96
Mean age               30.7 [+ or -] 14.8   31.4 [+ or -] 15.4   NS
Mean GCS 24 hours
  post-injury          9.5 [+ or -] 1.5     9.7 [+ or -] 1.5     NS
Mean ISS               19.8 [+ or -] 8.8    19.9 [+ or -] 7      N
Mean ICU days          6.9 [+ or -] 3.1     6.4 [+ or-] 3.2      NS
Mean hospital days     14.2 [+ or -] 5.7    11.9 [+ or -] 5.3    .005
Mean ventilator days   4.5 [+ or -] 2.3     3.5 [+ or-] 2        .003

Table 2. Differences Between Historical Control (HC) Group and Clinical
Pathway (CP) Group of Patients with Moderate TBI with Severe Tendencies

                       HC Group             CP Group             P

Isolated TBI           33%                  28%
Multiple trauma        67%                  72%
Number of patients     15                   57
Mean age               51 [+ or -] 18.7     44.4 [+ or-] 18.6    NS
Mean GCS 24 hours
  post-injury          9.5 [+ or -] 1.6     9.5 [+ or -] 1.1     NS
Mean ISS               27.5 [+ or -] 10.8   23.5 [+ or -] 8.9    NS
Mean ICU days          21.9 [+ or -] 14.5   15.4 [+ or -] 10.5   .06
Mean hospital days     40.5 [+ or -] 23.5   22.9 [+ or -] 14.3   .006
Mean ventilator days   15.9 [+ or -] 16.8   10 [+ or -] 10.5     NS

Table 3. Post-discharge Disposition of Patients with
Moderate TBI

                       HC Group   CP Group

Number of Patients       67         96
Rehabilitation           43%        60%
Home                     48%        33%
Home with Outpatient     22%        50%
  Rehabilitation
Hospital Transfer        4.5%        2%
Nursing Home             4.5%        3%

Table 4. Post-discharge Disposition of Patients with
Moderate TBI with Severe Tendencies

                     HC Group   CP Group

Number of Patients     15         56 *
Rehabilitation         80%        80%
Home                    0          4%
Hospital Transfer       0          5%
Nursing Home           20%        11%

* Disposition information unavailable for one patient

Table 5. Length of Stay Comparisons

CP Group               Hospital Days   ICU Days

Severe TBI                 22.5          16.7
Moderate TBI with ST       22.9          15.4
Moderate TBI               11.9           6.4
Gale Copyright: Copyright 2002 Gale, Cengage Learning. All rights reserved.