Increased physician and physical therapist communication is associated with earlier mobility and decreased length of stay in the cerebrovascular and trauma neuroscience population

1 Introduction

Acquired brain injuries are a prevalent cause of hospitalization and death. An estimated 1.7 million people each year sustain a traumatic brain injury (TBI) of which 275,000 people are hospitalized [5]. Additionally every year approximately 795,000 people have a stroke [9]. With growing healthcare costs, there is increased pressure to decrease hospital LOS, and improve patient outcomes. Early mobility in the ICU has a positive impact on patient outcomes including hospital costs, decreased falls, ventilator-associated events, pressure ulcers, and catheter-associated urinary tract infections (CAUTIs) [6]. In patients with acute intracerebral and subarachnoid hemorrhagic stroke, mobility interventions have been strongly associated with increased distance walked and appeared to promote stroke recovery [8]. Some data regarding early mobilization in the stroke population have been controversial. In the primary intention-to-treat analysis for A Very Early Rehabilitation Trial (AVERT), patients who received “usual care” had greater odds of a favorable outcome at three months compared to those who received very early mobilization [2]. Further analysis of this data has suggested that shorter, more frequent mobilization is associated with improved outcome when controlled for age and stroke severity, while increased minutes per day of mobilization reduced the odds of a good outcome [1].

Physical therapists are often the practitioners to initiate mobility with hospitalized patients, and can deliver a variety of modalities including passive and active exercise, cycle ergometry, early mobilization, and neuromuscular electrical stimulation (NMES). These interventions can improve outcomes, increase muscle strength, improve quality of life, and decrease ICU LOS [4].

Effective team communication is vital for patient safety [3]. Often the scope of practice of a physical therapist is not well understood by other professionals. Therefore, it is important for the physical therapist to advocate for patients who would benefit from their services [7]. The aim of this study was to identify whether increased coordination between the physical therapist and an attending cerebrovascular neurosurgeon correlated with improvements in inpatient care. We hypothesized that a physical therapist participating in morning neuroscience rounds would decrease the time to initial physical therapy (PT) consult, and lead to decreased hospital LOS, decreased Intensive Care Unit (ICU) LOS, and decreased ventilator days.

2 Methods

This retrospective study reviewed all patients admitted under a single neurosurgeon (JFF) over a sixteen-month period from April 1, 2014 to July 31, 2015 who received a PT assessment (n = 235). The ‘pre-intervention’ phase consisted of patients admitted during an eight month period (April 1, 2014 to November 30, 2014), and was followed by an eight month ‘post-intervention’ phase (December 1, 2014 to July 31, 2015). For this project, we used the institutional dataset available through our Enterprise Data Trust, which is accessed through an Umbrella IRB for Use of Health Information (IRB 11-0750-F6A).

During the pre-intervention phase physical therapists participated in a morning interdisciplinary meeting to facilitate timely discharges, and included the medical intern, patient care facilitator, social worker, and occupational therapist. In December 2014, a physical therapist began participating in morning weekday rounding with the admitting neurosurgeon. As each patient’s medical care was discussed any therapy related concerns were communicated (Fig. 1). The four areas in Fig. 1 were discussed as subtopics. An example for ‘Barriers to PT’ would be discussion about undergoing PT safely with a ventriculostomy (EVD) in place. With nursing in support, a PT would be allowed with work with selected patients with EVDs with the EVD clamped. If indicated, a PT consult would be placed and completed the same day. Additional topics discussed included types of interventions, safety for mobility with ventriculostomies, and estimation of LOS. The neurocritical care team would communicate if a patient could tolerate decreased sedation for PT. Once evaluated by the physical therapist, patients were followed per the therapist’s clinical judgment and plan of care.

OPEN IN VIEWER

Fig.1

Physical Therapist and Physician Rounding Communication Topics. Diagram showing grouped categories covered in discussions between the attending neurosurgeon and the physical therapist during morning rounds.

Data were obtained through the University of Kentucky’s Center for Clinical and Translational Science (CCTS), and through an approved umbrella Institutional Review Board protocol for the UK Enterprise Data Trust (EDT)-UK HealthCare dataset. Data were extracted from the electronic medical record, assigned to the pre or post group based on date of admission, and de-identified. Information extracted included the day of hospital stay PT assessment occurred, ICU length of stay, ventilator dependent days, and incidence of pressure ulcers. The data for pressure ulcers was unable to be analyzed due to low incidence (3 per 235 patients).

Descriptive statistics including mean and standard deviation were calculated, and used to compute unpaired t-tests. Pre and post rounding overall LOS, ICU LOS, ventilator days, and day of PT assessment were compared using 2-sample t-tests. Additionally, correlations between data were quantified using Pearson correlation. A correlation coefficent (r) >0.80 was considered very strong, 0.60–0.79 strong, 0.4–0.59 moderate, 0.20–0.39 weak, 0.00–0.19 very weak. P < 0.05 was considered significant.

235 patients were admitted to the neuroscience service and received PT assessment during the study period, resulting in 117 and 118 patients in the pre and post-intervention groups respectively. The average age of patients in the pre-intervention group was 51.7 years and 59.3 in the post-intervention group. The pre-intervention group included 45 females and 72 males, the post-intervention group included 57 females and 62 males (Table 1). The mean hospital LOS for the entire group was 12.59 days. The distribution of patients with a stroke diagnosis was 48.7% in the pre-intervention group and 39.8% in the post-intervention group; this difference was not significant. The case mix index for the pre-intervention group was 6.84 ± 6.28, and the post-intervention group was 6.10 ± 6.16 (p = 0.36).

The primary outcome of this study was the time to the initial PT assessment, which was 4.38 ± 3.78 ‘pre-intervention’, while the ‘post-intervention’ mean was 2.81 ± 1.95 (mean difference 1.57 days; p < 0.001). Average ICU LOS decreased by 1.83 days from 8.02 ± 8.66 days to 6.19 ± 7.84 days (p < 0.05). Average hospital length of stay decreased from 14.32 ± 18.46 days to 10.86 ± 9.99 days (p < 0.04). No significant difference was found in average ventilator days, which decreased by 0.55 days on average from 3.53 ± 6.86 days to 2.98 ± 6.24 days (p = 0.26) (Table 2). A moderate positive correlation was found between time to PT assessment and ICU LOS (R = 0.53, p < 0.001). PT assessment and ventilator days were moderately positively correlated (R = 0.55, p < 0.001). A moderate positive correlation was found between hospital length of stay and time to PT assessment (R = 0.45, p < 0.001).

Effective interdisciplinary communication is vital to expedite the process of determining which patients are appropriate to mobilize. More complex patients may not be appropriate for early PT assessments due to medical contraindications, and subsequently have longer hospital lengths of stay. Interdisciplinary rounds are a beneficial opportunity to bring together perspectives from medical professionals with diverse areas of expertise, and create an optimal plan for activity. Adding a physical therapist to physician rounds can be a key reminder to include mobility and exercise into each patient’s plan of care. The physician team can inform the physical therapist of any medical barriers to PT activities. Additionally, the physical therapist and physician can discuss discharge planning, and plan to accommodate any hindrances to a timely discharge and involve other professionals as indicated.

The retrospective design of this study limits the ability to show causation, and only association can be ascertained. This study was performed in a narrow environment with neurosurgical patients under a single admitting physician, and may not be generalizable to other diagnoses or settings. However, the involved neurosurgeon’s practice is limited to cerebrovascular disease and traumatic brain injury, which are the most common types of neurosurgical patients to have prolonged ICU stays. There was a trend toward less stroke diagnoses in the post-intervention group, though this was not significant. This may have had an impact on LOS. However, given that expected LOS is often calculated based on CMI, we used this to show relative equivalence between groups. Another limitation is that outlying data points were present, which could skew data in a relatively small sample size. Outliers were maintained in the analysis to capture all patients. Finally, there could be confounding factors that would positively affect the outcome variables over time. However, in this case, no major changes in care occurred during the study period. The institution achieved Comprehensive Stroke Center designation prior to the initiation of this study, and maintained it throughout the study, with similar neurocritical care resources devoted to the patients treated. No additional major changes in care protocols occurred during the study period. Therefore, while this was not a case-control matched cohort, the study populations were similar. While these are notable limitations, we believe this study provides vital pilot data to support a more controlled evaluation of the effect of PT rounding in on patients in a neurocritical care unit.

5 Conclusion

In conclusion, increasing communication between the physician and physical therapist may expedite initial PT assessments, help develop interdisciplinary plans for patients, and improve outcomes. Involving a physical therapist as part of the daily rounding team with the attending physician is associated with decreased ICU length of stay, decreased time to initial PT assessment, and decreased hospital LOS. Additionally, the study indicates that decreased time to initial PT assessment is correlated with decreased ICU length of stay and ventilator days. Therefore, we advocate for early and collaborative involvement of physical therapy in daily planning of care for critically ill neurosurgical patients.

Conflict of interest

The project described was supported by the National Center for Advancing Translational Sciences, UL1TR000117. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.