Review of a Large Clinical Series

Mature Trauma Intensivist Model Improves Intensive Care Unit Efficiency But Not Mortality

Journal of Intensive Care Medicine 2015, Vol. 30(3) 151-155 ª The Author(s) 2013 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0885066613507691 jic.sagepub.com

John Lee, MD, FACS1, Frederick Rogers, MD, MS, FACS1, Amelia Rogers, BS1, Michael Horst, PhD1, Roxanne Chandler, RHIA, CTR, CSTR1, and Jo Ann Miller, RN, CCRN1

Abstract Background: Although the Leap Frog intensivist staffing model has been shown to improve outcomes in the intensive care unit (ICU), to date, no one has examined the effect of an intensivist model in a dedicated trauma ICU. With stricter adherence to evidence-based protocols and 24-hour availability, we hypothesized that a mature intensivist model in a trauma ICU would decrease mortality. Methods: Level II trauma center trauma ICU admissions 2006 to 2011. The ICU care provided by 6 trauma intensivists. Two periods were compared: early (2006-2008) and mature (2009-2011). Patients matched on age, Injury Severity Score (ISS), preexisting conditions, and so on in a univariate analysis, with significant variables placed in a logistic regression model, with mortality as the outcome. Results: A total of 3527 patients (2999 excluding do not resuscitate status) were reviewed. Age 65 (odds ratio [OR] 2.38, P < .001), ISS 17 (OR 3.3, P < .001), coagulopathy (OR 1.64, P ¼ .004), and anemia (OR 1.73, P ¼ .02) were independent predictors of mortality. Multivariate logistic model encompassing these factors found no statistically significant differences in mortality across the 6-year period. The ICU efficiency showed significant improvements in terms of ventilator days (30.1% EARLY vs 24.4% MATURE; P < .001), decreases in mean consultant use per patient (0.55 + 0.85 EARLY vs 0.40 + 0.74 MATURE; P < .001), and increase in number of bedside procedures per patient (0.09 + 0.48 EARLY vs 0.40 + 0.74 MATURE; P < .001 Conclusions: Our mature intensivists staffing model shows improvement in ICU throughput (ventilator days, ICU days, decreased consultant use, and increased bedside procedures) but no survival benefit. Further improvements in overall trauma mortality may lie in the resuscitative and operative phase of patient care. Keywords trauma, intensivist model, efficiency, mortality

Background Trauma care is complex, involving not just system-wide readiness to rapidly mobilize costly resources but also depends on the interplay of multiple participants who care for these critically injured patients. Outcome studies of trauma care have examined both institutional effects (availability of dedicated odds ratio [OR],1 trauma center level status1-3) and personnel effects (part-time vs full-time trauma surgeons,4 trauma surgeon experiences,5 or annual volume6). For established trauma facilities, these studies show a general trend of improved outcome through resource enhancement and trauma system maturation. In spite of the proven value of trauma centers, the fact still remains that a significant number of patients with trauma continue to receive their primary trauma care at nondesignated trauma centers.7,8 Of concern is that resource enhancement alone does not confer a survival benefit if patients are treated at nondesignated trauma center.9 It would seem logical that high-

volume, stringently verified trauma centers, augmented by wellqualified surgeons and ancillary staff, would make the ideal trauma center. At the least, this type of trauma center most likely represents the pinnacle care achievable. By this logic, a mature trauma system would be anchored by a dedicated trauma center and producing best outcome. Peitzman and colleagues10 have described this superior result as they assessed the maturation of University of Pittsburgh Medical Center as a Pennsylvania Trauma Systems Foundation (PTSF)-verified level I trauma center, from 1987 to 1995.

1

Trauma Services, Lancaster General Health, Lancaster, Pennsylvania

Received April 16, 2013, and in revised form June 17, 2013. Accepted for publication July 10, 2013. Corresponding Author: John Lee, 555N. Duke St, Lancaster, PA 17602, USA. Email: [email protected]

Downloaded from jic.sagepub.com at UNIV OF MASSACHUSETTS on April 2, 2015

152

Journal of Intensive Care Medicine 30(3)

In the literature, improved outcome is synonymous with patient care efficiency, decreased complications, and improved survival.10,11 Although various theories have been proposed to explain the benefits of a trauma center, the precise mechanisms that may lead to improved survival are not clear. In the larger field of medicine, the concept of an intensivist-led, closed intensive care unit (ICU) is gaining traction because of its touted outcome benefits. Since the original report by Pronovost et al,12 multiple subsequent studies in diverse medical fields have affirmed the morbidity, mortality, and length of stay improvements associated with such a staffing model. With the maturation of our trauma intensivist model, we sought to demonstrate that, in addition to its proven benefits on efficiency, a realization in trauma mortality benefit would also be found.

Methods Lancaster General Hospital is a not-for-profit, level-II regional trauma center, verified by the PTSF since 1987. It does not have a free-standing surgery residency. Since 1997, continuous 24/7 in-house trauma coverage became the standard of care, with a core group of 8 community-based general surgeons comprising the trauma call roster. Trauma care during this period was notable for its fragmented nature as typified by multiple specialist involvements that would take place in response to the needs of critically injured ICU patients. During a single admission, it was quite common for ICU patients to have both surgical (general surgery, neurosurgery, orthopedic surgery, urology) and medical (pulmonology, cardiology, nephrology, infectious disease, hematology, neurology) specialists participate in their trauma care. In 2006, 3 hospital-employed, full-time general surgeons with added qualification in surgical critical care were hired to oversee the trauma program. An important point of distinction was that all admitted patients with trauma came under the comprehensive care of only the 3 trauma intensivists, regardless of the identity of their initial resuscitating surgeon. Trauma ICU care was provided continuously in a 1-week block rotation schedule, with a 30-minute bedside response time if the intensivist was not physically in-house. Our coverage model was further refined over the next 3 years, such that, by 2009, a full complement of 7 trauma intensivists (between 7 and 30 years experience) staffed the service; this allowed continuous, in-house, 24/7 coverage for all trauma-related care (both resuscitation and ICU care). This robust coverage was further augmented by the hiring of advanced practitioners (APs; physician assistant and nurse practitioners). By 2011, Lancaster General Hospital had, at any given time, a trauma intensivist and a trauma AP on duty simultaneously. The EARLY phase (2006-2008) of the intensivist coverage witnessed a gradual cultural change on the part of the ICU nurses. They learned to rely on the trauma intensivists as the Captain of Ship (COS) who would coordinate the complex care of the patients. Preexisting ICU protocols were examined and updated as needed to reflect most current best practice (open lung strategy/low-tidal volume; ventilator bundles; insulin infusion for

tight glucose control; appropriate gastrointestinal and deep venous thrombosis/venous thromboembolism [DVT/VTE] prophylaxis; early enteral nutrition; daily sedation vacation for ventilator liberation). Concurrently, a daily multidisciplinary ICU rounding checklist was introduced, which would ensure compliance with these guidelines. Bedside ICU procedures (percutaneous tracheostomy, percutaneous endoscopic gastrostomy, inferior vena cava filter) were also introduced as facility support became available. Significant changes in the mature phase (2009-2011) include continuous in-house ICU coverage by the trauma intensivists, with augmentation to this staffing model by advanced practitioners toward the end of 2010. Notable protocol changes were the comprehensive revision of the Traumatic Brain Injury Guideline to be in complete accordance with the recommendations of the Brain Trauma Foundation. A new massive transfusion protocol reflecting contemporary understanding was also implemented, utilizing a packed cell–plasma ratio of 1.5:1.

Patient Population and Analysis All patients admitted to the ICU setting following initial trauma evaluation and resuscitation, from the year 2000 to 2011, were eligible for review. Three comparison periods were chosen: pre-intensivist (PRE, 2000-2005), EARLY intensivist (EARLY, 2006-2008), and MATURE intensivist (MATURE, 2009-2011). Additional user-defined data points were also gathered and maintained in Collector (Digital Innovation, Forest Hill, Maryland), a PTSF-mandated trauma database. To be included in the study, patients had to have an ICU length of stay (LOS) of 1day. Patients that died, had an in-hospital mortality, and were dead on arrival or died in the trauma bay were excluded on the basis of not meeting the study’s inclusion criteria. Variables were analyzed by univariate logistic regression for their association with mortality. Age  65 and ISS  17 were significant predictors of mortality and were subsequently included in the multivariate logistic regression model with the time periods. The PRE period was the referent. Then Fischer exact test was used to evaluate the differences between the time periods for hospital length of stay (LOS) >4days, ventilator days >2days, and ICU LOS >2days. A 1-way analysis of variance test was used to find the mean + standard deviation of the number of consultants and number of bedside procedures per patient during the different time periods. A P  .05 was considered significant.

Results From 2000 to 2011, there were a total of 6449 patients with trauma admitted to the ICU. During the pre-intensivist period, 2806 patients were admitted to the ICU, of which 169 (5.7%) died. The EARLY period saw 113 (7.0%) deaths of the 1514 ICU patients. The MATURE period had 1761 ICU patients, with 139 (7.3%) dying. In each of the 3 time periods, age was trimodally distributed, and so we used the median as a measure of central tendency. The median ages for each time periods are

Downloaded from jic.sagepub.com at UNIV OF MASSACHUSETTS on April 2, 2015

Lee et al

153

Table 1. Mortality Rates. OR Adjusted for ISS 17 and Age 65

Mortality Ndeaths/Ntotal

Raw, %

OR (95% CI)

P Value

168/2974 113/1627 139/1899 420/6500

5.65 6.95 7.32 6.46

Ref 0.98 (0.76-1.27) 1.02 (0.80-1.30)

– .902 .876

PRE (‘00-’05) EARLY (‘06-’08) MATURE (‘09-’11) All

Abbreviations: CI, confidence interval; OR, odds ratio.

Table 2. ICU Efficiency.

ICU LOS >2 days, % Vent days >2, % Hospital LOS >4 days, % Consultant use mean/pt (+SD) Bedside procedures mean/pt (+SD)

PRE (2000-2005)

EARLY (2006-2008)

MATURE (2009-2011)

P Value

41.1 50.9 49.6 0.81 (1.1) 0.01 (0.11)

35.2 50.41 48.7 0.55 (0.85) 0.09 (0.48)

32.8 41.9 40.7 0.40 (0.74) 0.21 (0.75)