Clinical management of trauma patients is resource-intensive, as these patients often present with complex injuries requiring collaboration of specialists from multiple disciplines (Dutton et al., 2003). This is increasingly evident in the management of elderly trauma patients. Diminished physiologic reserve, a higher frequency of preexisting conditions, and a higher incidence of chronic illness contribute to challenges experienced in the provision of care for these patients (Katrancha, Zipf, Abrahams, & Schroeder, 2017; Kirshenbom, Ben-Zaken, Albilya, Niyibizi, & Bala, 2017). Accordingly, elderly patients carry an elevated risk for mortality, functional impairment, and loss of independence compared with their younger counterparts following traumatic injury (Olufajo et al., 2016). With a growing elderly population, the challenges experienced by institutions to provide optimal geriatric care continue to rise, further taxing an already stressed infrastructure for patient care. Rising costs, declining reimbursements, and limitations on resident work hours have challenged health systems to develop new and streamlined processes through which the quality of patient care can be maintained while also preserving institutions' economic stability (Dutton et al., 2003; Haan et al., 2007; Morris et al., 2012). In an effort to maintain operational efficiency, many institutions utilize nonphysician providers to facilitate the provision of patient care (Morris et al.; Rejtar, Ranstrom, & Allcox, 2017; Sise et al., 2011). Similarly, the trauma service at the study institution has explored opportunities to enhance efficiency and continuity of care through the addition of trauma nurse practitioners (TNPs).
The trauma service began expanding its nurse practitioner (NP) model in September 2013. Beginning in December 2014, low-acuity patients presenting at the study institution could be admitted to trauma services under the care of TNPs. Through the implementation of this new process, the department sought to increase the proportion of low-acuity patients who were admitted to trauma services. A protocol and an order set were created to streamline the admission process. Additional TNP practices were established that targeted accomplishing the following tasks within the first 24-hours of admission: communication with patients' family to describe plans of care, assessment of predetermined discharge disposition criteria, engagement of social workers to facilitate discharge planning, assessment of home medications, patient mobility, facilitation of incentive spirometry, and expedition of physical therapy involvement in patients' care.
Following admission, low-acuity patients were managed by a weekly rotating team consisting of two TNPs providing bedside care in tandem with a trauma attending. The process was first piloted in orthopedic patients with low acuity. After a successful 6-month pilot, additional subsets of trauma patients who were previously admitted to hospitalist services with sub-specialty consults (neurosurgical, oromaxillofacial, etc.) were allocated to the TNP service. Because of the challenges presented by the clinical management of elderly trauma patients and the need to further substantiate the quality of care provided by the institution's TNP model, the current study endeavored to evaluate outcomes in elderly trauma patients admitted to the study institution since the inception of the current TNP model.
PURPOSE
To demonstrate the impact of the TNP model, trauma leadership sought to evaluate the quality of care provided to elderly trauma patients. Accordingly, the current study aimed to compare outcomes in elderly patients whose care was coordinated by the TNPs versus nontrauma NPs (NTNPs).
RESEARCH QUESTIONS/HYPOTHESIS
Elderly trauma patients managed under the TNP model would have superior outcomes when compared with patients admitted to other surgical and nonsurgical services during the same time period.
METHODS
This is a retrospective cohort study of trauma patients admitted to a Level 1 trauma center between December 2014 and June 2017. Patients included in this study were 65 years or older and had an Injury Severity Score (ISS) of 10 or less. Included patients were admitted to noncritical patient care areas within the trauma center following initial presentation to the study institution. Those meeting inclusion criteria were divided into two patient care groups: admission to TNP or NTNP services.
Following approval from the institution's review board, patients were identified and data were obtained through the institution's trauma registry. Data included the following patient characteristics and outcome variables: age, ISS, gender, pre-existing comorbidities, admitting injuries, hospital length of stay (LOS), incidence of unplanned intensive care unit (ICU) admissions, discharge location, rate of 30-day readmission, in-hospital complications, in-hospital mortality, and total charges for respective hospitalizations. Hospital LOS data were analyzed for patients whose stay was 24-hours or more in duration. Evaluated complications were categorized into hematological, cardiovascular, infections, pulmonary, genitourinary, musculoskeletal, integumentary, substance withdrawal, gastrointestinal, surgical, and neurological groups.
Data analysis was conducted with IBM-SPSS 22.0 (IBM Corp, Armonk, NY). Descriptive statistics were computed for each variable to describe the patient population. Means and standard deviations were conducted for continuous variables, whereas proportions and frequencies were reported for categorical variables. Continuous variables were compared using independent t or Mann-Whitney U tests, and categorical variables were compared using chi-square or Fisher's exact tests. Study outcome variables including hospital LOS, incidence of unplanned ICU admissions, discharge location, in-hospital mortality, rate of 30-day readmission, and complications during hospital stay were analyzed with analysis of covariance or binary logistic regression adjusting for differences in patient characteristics. Significance was reported at a level of p <= .05.
RESULTS
A total of 1,363 patients were included in this analysis (Table 1). One hundred forty patients were admitted to the TNP service and 1,223 to the nontrauma service. The NTNP service cohort consisted of patients admitted to hospitalists (90.4%), orthopedics (6.5%), plastic surgery (1.6%), oromaxillofacial (0.5%), and neurosurgical services (1.1%). Baseline demographics were similar between the comparator groups with the exception of age (74.99 +/- 8.08 vs. 80.63 +/- 8.70, p = .000) and ISS (6.02 +/- 2.56 vs. 7.25 +/- 2.69, p = .000), which were significantly lower in the TNP cohort. Comorbidities were also similar between the comparator groups with the exception of pulmonary, cardiovascular, neurological, and immunological diseases. The TNP cohort had significantly less pulmonary (3.6% vs. 15.2%, p = .000), neurological (19.3% vs. 29.0%, p = .015), and cardiovascular (70.0% vs. 81.8%, p = .001) disease states. In contrast, immunological diseases were significantly higher in the TNP cohort (6.4% vs. 2.9%, p = .038). Admitting injuries were also compared between the study cohorts. The TNP cohort had a significantly higher percentage of traumatic brain injury (11.4% vs. 4.8%, p = .001), dislocation, (5.7% vs. 0.8%, p = .000), sprain and strain (7.9% vs. 1.2%, p = .000), upper extremity fractures (30.7% vs. 13.2%, p = .000), and vertebral injuries (17.1% vs. 6.2%, p = .000). Lower extremity fractures, however, were significantly more prominent in the NTNP service cohort (41.4% vs. 67.0%, p = .000) (Table 2). The significant differences observed in these baseline variables were accounted for in multivariate analyses conducted for all outcome measures.
Comparisons of outcome variables revealed a significant difference in hospital LOS between the two cohorts. Patients under the care of TNPs had a shorter duration of hospitalization (1.22 days) compared with that of the NTNP service cohort (4.38 +/- 3.54 vs. 5.60 +/- 3.98, p = .048). Differences in LOS likely contributed to a significant decrease in average hospital charges between the compared cohorts, with charges for hospitalizations on the TNP service averaging $13,000 less per admission ($38,053 +/- $29,640.76 vs. $51,317.79 +/- $34,756.83, p = .016).
The majority of the patients in both cohorts were either discharged to their home or a skilled nursing facility. Comparison of the arms, however, revealed that a significantly higher percentage of patients admitted to the TNP service were discharged home (67.1% vs. 36.0%, p = .002) and a significantly lower percentage of patients from the TNP team were discharged to skilled nursing facilities (25.7% vs. 51.9%, p = .040) (Table 2). Despite the differences in the incidence in which patients were discharged home versus skilled nursing facilities, 30-day readmission rates remained similar between the study cohorts (Table 3).
Unplanned ICU admissions were lower in the TNP cohort (1.4% vs. 2.1%, p = .946); however, this decrease did not achieve statistical significance. Although there were no deaths in the TNP cohort, mortality rates were found to be similar to that seen with the NTNP cohort (0.0% vs. 1.1%, p = .996) (Table 3). Complications organized by the organ system affected (e.g., acute lung injury allocated to the pulmonary complication group) or grouped broadly by describing the complication (e.g., urinary tract infection allocated to the infection complication group) were compared and found to be similar between the study cohorts (Table 4).
DISCUSSION
It has been proposed that by 2050, patients 65 years or older will comprise 40% of all trauma patients (Tillou et al., 2014). These increasing numbers of elderly traumas present a unique challenge in the provision of medical care, as the elderly respond to traumatic injury differently than younger patients do. The altered response of the elderly patient is thought to be multifactorial likely resulting from a lower physiological reserve, an increased number of preexisting conditions, and a higher incidence of chronic illness (Katrancha et al., 2017; Kirshenbom et al., 2017). This paradigm is supported by literature evaluating outcomes with geriatric trauma patients, suggesting that age, injury burden, and baseline functional status are predictive of mortality or discharge to a skilled nursing facility (Kirshenbom, et al., 2017; Richmond, Kauder, Strmpf, & Meredith, 2002). Furthermore, preexisting comorbidities, common to this patient population, result in a threefold increase in complication rates during hospital admission (Richmond et al., 2002). As a result, elderly patients are at a much higher risk for poor outcomes than their younger counterparts following traumatic injury. Accordingly, existing literature emphasizes the need for focused care to improve outcomes in the geriatric trauma patient population (Fallon et al. 2006).
In December 2014, the Level 1 trauma center at the study institution implemented a practice model through which patients with low acuity were admitted to a TNP service with attending physician oversight. Given the challenges presented by the elderly patient population and an ongoing need to evaluate the quality of care provided by the TNP service model, the current study was developed for this select patient population. Current literature has frequently examined the impact of TNPs on the morbidity and mortality of trauma patients; however, there is a paucity of data examining the impact of patient care driven by TNPs in improving outcomes in geriatric trauma patients (Christmas et al., 2005; Collins et al., 2014; Holliday, Samanta, Budinger, Hardway, & Bethea, 2017; Nyberg, Waswick, Wynn, & Keuter, 2007; Sise et al., 2011). In the current study, elderly patients managed by TNPs had a significantly lower hospital LOS. In addition, the NP service was able to discharge a significantly higher percentage of patients home than skilled nursing facilities. Similar endpoints have been evaluated in previous studies; however, as previously mentioned, these studies' foci have not been specific to elderly trauma patients. Despite substantiating the impact of NPs in the provision of care to the trauma patient population, extrapolation of these studies' findings to the elderly patient population is limited (Christmas et al., 2005; Collins et al., 2014; Holliday et al., 2017; Sise et al., 2011).
Existing geriatric literature has evaluated practice models directed toward the provision of optimal trauma care. Initial publications described processes that focused on timely identification and thorough management of comorbidities through the consultation of a geriatrician. The incorporation of this care model was shown to improve mortality in elderly trauma patients in a study by Fallon et al. in 2006. Examples of geriatrician interventions in this model included the identification of inappropriate medication use, assistance in pain management, facilitation of advanced care planning, discharge disposition, and the recognition and management of delirium. Additional data from Mangram et al. (2012) demonstrated improvement in outcomes in geriatric trauma patients receiving care from a multidisciplinary team providing care exclusively to trauma victims who were older than 60 years. This model also resulted in improvements in duration of emergency department stay, time to surgical intervention, intensive care, and hospital LOS, as well as select posttrauma complications (Mangram et al., 2012). Data evaluating these models as well as additional data reported by Lenartowicz et al. (2012) prompted the American College of Surgeons to use the Trauma Quality Improvement Program as a platform to disseminate guidelines for the management of geriatric trauma patients (American College of Surgeons, 2018; Lenartowicz et al., 2012).
Literature clearly supports the paradigm that outcomes in elderly trauma patients are improved with the integration of specialized, focused care through geriatric consultations or the utilization of practitioners who maintain a focus on the specialized needs of this patient population (Fallon et al., 2006; Katrancha et al., 2017; Lenartowicz et al., 2012; Mangram et al., 2012; Olufajo et al., 2016; Tillou et al., 2014). These studies support the concept that early risk identification during hospitalization and smooth transition to discharge disposition are critical. Risk identification involves evaluation of preexisting cognitive and functional impairments, review of medications, social history, nutritional status, and advanced directives (Katrancha et al., 2017; Olufajo et al., 2016; Tillou et al., 2014). Similarly, the TNP care model at the study institution fosters similar practices and therefore improvements in outcomes identified by the current study resonate with the current literature.
Factors influencing outcomes in this study likely include the routine evaluation of patients upon admission with preset discharge disposition criteria and collaboration with social workers to initiate discharge planning within the first 24-hours of admission. During the initial 24-hours of admission, TNPs also evaluate home treatment regimens and communicate with patients' family to describe plans of care. In addition, the TNP care model emphasizes early patient mobility. Nurse practitioners make frequent bedside visits, assist patients with walking, and coordinate prompt placement of orders for physical therapy and behavioral medicine consults when applicable. Trauma nurse practitioners also formally evaluate these patients' incentive spirometry at least every 24-hours to ensure that pulmonary function is being maintained.
Economic endpoints of hospital cost and charges were also lower in the patients admitted to the TNP group. Specifically, per patient savings of $13,000 were seen when hospital charges were compared between the cohorts. These findings projected the potential for $1.8 million in decreased charges in the event that all patients included in the study were admitted to the TNP service. Similar endpoints have been previously examined by Collins et al. and Sisse et al. These studies also reported significant reductions in cost and hospital charges associated with the care of trauma patients following the integration of NPs into their practices.
LIMITATIONS
A large number of patients were included in this evaluation; however, the proportion of patients allocated to each respective arm was not uniform. In addition, the retrospective design of the current study makes it reliant upon the identification, availability, and completeness of patients' medical records. Accordingly, all patients admitted to the trauma center during the study period may not have been identified and included, resulting in selection bias. In addition, the validity of the evaluated data depends on the accuracy with which they have been documented and subsequently collected. This serves to limit researchers' ability to practice quality control with the data. The study also does not seek to establish a causal relationship between the implementation of the NP service model and positive outcomes. It only describes the changes in outcomes reported between the study arms.
CONCLUSIONS
The findings of the current study support the impact of the TNP practice model implemented at the study institution. Significant improvements in hospital LOS, the frequency with which elderly patients were discharged home, and economic outcomes were identified. These data add to the growing body of literature suggesting that outcomes in elderly trauma patients are improved with the utilization of practitioners who maintain a focus on the specialized needs of this patient population. These findings, however, provide inaugural data highlighting a practice model in which NPs' direction of bedside patient management has been associated with a significant enhancement in the care of elderly trauma patients. Accordingly, similar models initiated at other trauma centers may facilitate comparable improvements in outcomes. Moving forward additional research may endeavor to further delineate the impact of specific components and activities routinely performed by the TNPs in the current practice model.
KEY POINTS
* The current study supports the value of patient care coordinated by NPs in the elderly trauma patient population.
* Data revealed improvements in hospital LOS, percentage of patients discharged home, and hospital charges associated with the care of elderly trauma patients in the NP cohort.
* Further investigation should endeavor to evaluate routine components of the NP practice model to further substantiate current practices.
REFERENCES