Creating New Opportunities to Educate Families on the Impact of Frailty and Cognitive Impairment in a Trauma Intensive Care Unit: Results of a Quality Improvement Project

Abstract

Background:

Older adults comprise a rapidly growing proportion of admissions to acute care hospitals and trauma centers. Older adults admitted to a trauma intensive care unit (TICU) often have a more complicated inpatient and posthospital course. This is the most pronounced in frail elders with cognitive dysfunction. We aimed at integrating validated screening instruments for physical frailty and cognitive impairment into the standard nursing assessment of all older trauma patients admitted to our TICU and stepdown unit.

Objectives:

Our goal, for positive screens, was to trigger earlier referrals to palliative care for patient and family education on the range of likely clinical outcomes.

Methods:

In February 2015, our study team trained bedside trauma nurses to implement a validated frailty screening process on all patients at least 65 years of age or older who were admitted to the TICU and stepdown unit. Between March and May 2015, the number of older adults admitted, mechanism of injury, numbers of patients screened, and positive screens, along with volume of palliative care referrals, were tracked.

Results:

During the three-month period, the mean age of all older admissions (N = 131) was 75.5, of which 49% were screened. Among the patients screened, 38% screened positive for frailty, 45% screened positive for possible dementia, and 23% screened positive for both conditions. Palliative care consultations for older adults increased from 13% (before study) to 33% during the study period.

Conclusion:

A screening process designed for older adults to assess both physical frailty and cognitive impairment can be standardized into the routine care of older adults admitted to a busy trauma service. Positive screens can serve as a trigger for earlier palliative care assessments, with opportunities for educating patients and their families on the range of clinical trajectories that these vulnerable patients face.

Introduction

The U.S. population is aging, with a significant number of older adults (OAs) living longer with an increasing risk for injury.¹ U.S. trauma center trends demonstrate that a large and growing percentage of all trauma admissions comprised geriatric patients.² Injured OAs admitted to a trauma intensive care unit (TICU) have a high prevalence of functional and cognitive impairments,³ and increased length of stay and mortality as compared with younger patients.^4,5 Maxwell et al. reported the influence of preinjury cognitive impairment and physical frailty on one-year functional decline and mortality.⁶

Growing evidence shows that palliative care (PC) consultations that are aimed at engaging patients and families in earlier discussions about goals of care can potentially reduce 30 day readmissions, resulting in significant cost savings^7,8 Few studies have identified methods to trigger earlier discussions. The purpose of our study was twofold. First, we aimed at determining the feasibility of incorporating a validated screening tool into the daily workflow of bedside clinicians to assess for both cognitive impairment and physical frailty among OAs admitted to a trauma unit. Second, we aimed at tracking and reporting our screening results and at determining whether the screening process could serve as an objective trigger for PC consultations and lead to an increase in overall referrals to PC earlier in patients' clinical trajectories.

Methods

The project received IRB approval and took place at the Vanderbilt University Medical Center Trauma Unit from March through May 2015.

Procedures

Training

Before the project, bedside nurses received group and individual training on administration of the 5-item FRAIL questionnaire,⁹ as well as the 8-item AD8 Dementia Screen.¹⁰ To our knowledge, these instruments have not been utilized in the geriatric trauma population. The FRAIL Scale is a validated instrument⁹ that is predictive of disability and mortality.¹¹ Five items (Fatigue, Resistance, Ambulation, Illnesses, Loss of weight) create an acronym to facilitate administration. A score of ≥3 indicates frail, and a score of 1–2 indicates prefrail. The AD8 Dementia Screen is a validated screening instrument that differentiates individuals with and without dementia,^10,12 and it has been utilized with both patients and proxy respondents.¹³ The AD8 consists of eight yes/no questions related to memory and cognition that identify early signs of dementia. A score of ≥2 indicates possible dementia.

Project coordinators attended trauma and PC provider meetings to explain the project. Trauma care providers were asked to evaluate screening results on a daily basis and to determine whether an early PC consultation was appropriate, based on the screening results. In addition to traditional PC triggers, trauma care providers were encouraged to consider a consultation for patients who screened positive for both physical frailty and possible dementia. PC providers were encouraged to utilize the screening results as a discussion point.

Processes

On a daily basis, medical receptionists distributed screening forms to bedsides of older patients (≥age 65). Based on the screening results, trauma care providers ordered PC consultations as appropriate for individual patients. Patients who received a PC consultation (or family caregivers) received a phone call six months after discharge to determine patients' overall condition and/or mortality status. Project coordinators entered the results of completed forms into SPSS 23.0¹⁴ for subsequent analysis.

Data analysis

SPSS 23.0 was used for the quantitative analyses. Median and inter-quartile range (IQR) were used to summarize the continuous distributions of ages; counts and percentages were used to summarize all other study variables. Group comparisons (screened vs. not screened; received consults vs. did not) were conducted by using Mann-Whitney tests (age) and Chi-Square Tests of Independence (all other comparisons). An alpha of 0.05 (p < 0.05) was used for assessing statistical significance.

Results

Aim 1: Feasibility of frailty and cognitive screening

Patient characteristics

Between March and May of 2015, a total of 131 patients with an age of greater or equal to 65 years old were admitted to the TICU. The median age of this sample was 73 years (IQR 69–80). The most common mechanisms of injury were falls from standing (53, 41%), motor vehicle events (41, 31%), and other types of falls (19, 15%). Within this cohort, 64 patients (49%) were screened by the bedside nurses. These characteristics, as well as the proportion of PC consultations for those screened and not screened are summarized in Table 1. There were no statistically significant differences between the screened and nonscreened groups in terms of age or mechanism of injury. Thirty-five (54%) of the screened patients received a PC consultation compared with only 9 (14%) of the nonscreened patients (p < 0.001).

Table 1.

Summary of Geriatric Trauma Patients Admitted and Screened Versus Nonscreened (N = 131)

	Total (N = 131)	Screened (n = 64)	Nonscreened (n = 67)	p
Age, median (IQR)	73 (69–80)	73 (69–81)	73 (70–80)
Age group, n (%)
65–74	75 (57)	36 (55)	39 (59)	0.752
75–84	35 (27)	17 (26)	18 (27)
85+	21 (16)	12 (19)	9 (14)
Mechanism of injury, n (%)
Fall from standing	53 (41)	25 (39)	28 (42)	0.159
Fall from other	19 (15)	11 (17)	8 12)
MVC	41 (31)	24 (37)	17 (26)
MCC	6 (5)	2 (3)	4 (6)
Pedestrian	1 (1)	1 (2)	0 (0)
Other	11 (8)	2 (3)	9 (14)
PC consultation, n (%)	44 (34)	35 (54)	9 (14)	<0.001

IQR, inter-quartile range; MCC, motorcycle crash; MVC, motor vehicle crash; PC, palliative care.

Frailty and cognitive screening

Available responses and scores from the FRAIL and AD8 measures within the sample of 64 patients who were screened are summarized in Table 2. Approximately 39% of patients (25 of 64) had a FRAIL score of ≥3, and 33% (21 of 64) had a score of 1–2 (prefrail). Among individual items, ∼64% (41 of 64) could not walk one block, ∼63% (40 of 64) could not walk up a flight of stairs, and 58% (37 of 64) reported fatigue or exhaustion. Forty percent (24 of 60) of the patients had a positive score of ≥2 on the AD8 screen, indicating possible dementia. Approximately 32% (19 of 59) of the patients or surrogates reported daily problems with memory or thinking; 31% (18 of 59) reported trouble remembering appointments; and 27% (16 of 59) reported repeating the same things over and over, trouble handling financial affairs, and having problems with judgment (Table 2).

Table 2.

Summary of Frailty and Cognitive Screening Among Geriatric Trauma Patients (N = 64)

Item	n (%)
FRAIL scale core (N = 64)
0 (nonfrail)	18 (28)
1–2 (prefrail)	21 (33)
≥3 (frail)	25 (39)
Does the patient fatigue easily?	37 (58)
Inability to walk up one flight of stairs?	40 (64)
Inability to walk one block (1/4 mile)?	41 (64)
Does the patient have 5 or more illnesses?	20 (31)
Has the patient lost >5%–10% of weight?	13 (20)
AD8 Dementia Screen (N = 60)
0–1 (no dementia)	36 (55)
≥2 (possible dementia)	24 (40)
Does the patient have problems with judgment?	16 (32)
Does the patient seem less interested in hobbies?	11 (18)
Does the patient repeat the same things over?	16 (32)
Does the patient have trouble learning to use a new tool?	11 (18)
Does the patient forget the correct month or year?	13 (22)
Does the patient have trouble handling financial affairs?	16 (32)
Does the patient have trouble remembering appointments?	18 (31)
Does the patient have problems with thinking or memory?	19 (32)

Aim 2: Frailty and cognitive screening as a trigger for PC

PC consultations

Forty-four of 131 (34%) patients received PC consultations over the project period (March–May 2015). Table 3 summarizes the patient cohort by those who did and did not receive a PC consultation. Although patients who received a consultation were older (p = 0.038), there were no statistically significant differences by mechanism of injury. Table 4 summarizes the triggers for PC consultations, including those triggered by frailty and cognitive screening and traditional PC triggers. Eighteen (41%) of the 44 patients had a FRAIL score of ≥3 and 11 (25%) had a score of 1–2 (prefrail). Eighteen (41%) had a score of ≥2 on the AD8 Dementia Screen. Among traditional PC triggers, 20 patients (46%) had life-limiting conditions (e.g., severe injury) and 11 (25%) had complex care requirements. Seven (16%) had a history of completing an advance care discussion, and six (14%) experienced multiple admissions during the preceding six months. Thirty-seven (84%, 16% lost to follow-up) of the 44 patients or family caregivers received follow-up phone calls at six months posthospitalization. Seventeen of the 37 (46%) died by six months.

Table 3.

Summary of Patients Who Did and Did Not Receive PC Consultations (N = 131)

	Total (N = 131)	PC consultation (n = 44)	No PC consultation (n = 87)	p
Age, mean (SD)	75.5 (7.6)	78.1 (9.1)	74.2 (6.7)	0.038
Age group, n (%)
65–74	75 (57)	22 (50)	53 (61)	0.002
75–84	35 (27)	8 (18)	27 (31)
85+	21 (16)	14 (32)	7 (8)
Mechanism of injury, n (%)
Fall from standing	53 (41)	21 (48)	32 (37)	0.178
Fall from other	19 (15)	7 (16)	12 (14)
MVC	41 (31)	15 (34)	26 (30)
MCC	6 (5)	0 (0)	6 (7)
Pedestrian	1 (1)	0 (0)	1 (1)
Other	11 (8)	1 (2)	10 (12)

Table 4.

Summary of PC Consultations Among Geriatric Trauma Patients (N = 44)

Triggers	n (%)
Frailty category
Nonfrail	5 (11)
Prefrail	11 (25)
Frail	18 (41)
AD8 category
No dementia (0–1)	14 (32)
Possible dementia (≥2)	18 (41)
Unknown	3 (7)
Traditional PC triggers
Life-limiting condition	20 (46)
Multiple admissions in past six months	6 (14)
Complex care requirements	11 (25)
Admission from long-term care	1 (2)
Metastatic or advanced cancer	2 (5)
Chronic home oxygen use	2 (5)
Current or past hospice enrollee	2 (5)
Limited social support	2 (5)
History of completing an advance care discussion	7 (16)
Mortality: six months after discharge (n = 37 [16% lost to follow-up])	17 (46)

Discussion

Our results demonstrate the potential feasibility of training bedside nurses to screen OAs and/or family caregivers for frailty and cognitive impairment and incorporate a brief screening tool into the daily workflow of bedside trauma nurses. Further, our findings reveal that a brief screening tool was generally accepted and that positive screens flagged by the bedside nurses triggered a significant increase in PC consultations. Finally, our findings demonstrate that an interdisciplinary approach that involves bedside nurses and frontline providers is an effective mechanism to trigger earlier geriatric PC. To our knowledge, there are no published reports of implementation of tools that focus on frailty and cognitive impairment as a trigger for referral to PC services in acute care settings. Future research should include testing of such tools in other types of acute care units to determine appropriateness for PC referrals.

Our project has limitations and caveats. First, although our PC team was appropriately staffed to meet the usual needs of our institution, we were unsure of our capacity to absorb extra referrals. We found that a recent expansion of the PC team provided adequate band width to spend counseling both patients and their families and communicating discussions to the trauma team. Second, our PC and trauma teams have a history of working together as a result of a longstanding trusting relationship. As a result of this strong working relationship, we may have obtained greater buy-in from surgeons than would normally occur with other PC teams. Finally, the sample size for our project was small and only 49% of admitted patients were screened. Our prior work demonstrates the difficulties of screening injured older patients.³ In addition, as noted in our results, we found no statistically significant differences between the screened and nonscreened patients. We will continue to explore ways to optimize care of frail elders on the trauma service through early identification of risk and innovative communication methods targeted to patients and family caregivers.

Conclusion

Trauma teams can operationalize a screening tool for frailty and cognitive impairment into their daily workflow. Positive screens can effectively trigger an increase in earlier referrals to PC in the OA population, thereby identifying new opportunities for PC teams to educate families on the various clinical trajectories that these patients face.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Kehoe

, Smith

, Edwards

, et al.: The changing face of major trauma in the UK. Emerg Med J, 2015; 32:911–915.

Schonenberger

, Billeter

, Seifert

, et al.: Opportunities for improved trauma care of the elderly—A single center analysis of 2090 severely injured patients. Arch Gerontol Geriatr, 2012; 55:660–666.

Maxwell

, Mion

, Mukherjee

, et al.: Feasibility of screening for preinjury frailty in hospitalized injured older adults. J Trauma Acute Care Surg, 2015; 78:844–851.

Joseph

, Pandit

, Rhee

, et al.: Predicting hospital discharge disposition in geriatric trauma patients: Is frailty the answer?. J Trauma Acute Care Surg, 2014; 76:196–200.

Gill

, Murphy

, Gahbauer

, Allore

: The course of disability before and after a serious fall injury. JAMA Intern Med, 2013; 173:1780–1786.

Maxwell

, Mion

, Mukherjee

, et al.: Preinjury physical frailty and cognitive impairment among geriatric trauma patients determine postinjury functional recovery and survival. J Trauma Acute Care Surg, 2016; 80:195–203.

Morrison

, Penrod

, Cassel

, et al.: Cost savings associated with US hospital palliative care consultation programs. Arch Intern Med (Chicago), 2008; 168:1783–1790.

O'Connor

, Moyer

, Behta

, Casarett

: The impact of inpatient palliative care consultations on 30-day hospital readmissions. J Palliat Med, 2015; 18:956–961.

Morley

, Malmstrom

, Miller

: A simple frailty questionnaire (FRAIL) predicts outcomes in middle aged African Americans. J Nutr Health Aging, 2012; 16:601–608.

10.

Galvin

, Roe

, Powlishta

, et al.: The AD8: A brief informant interview to detect dementia. Neurology, 2005; 65:559–564.

11.

Malmstrom

, Miller

, Morley

: A comparison of four frailty models. J Am Geriatr Soc, 2014; 62:721–726.

12.

Galvin

, Roe

, Xiong

, Morris

: Validity and reliability of the AD8 informant interview in dementia. Neurology, 2006; 67:1942–1948.

13.

Galvin

, Roe

, Coats

, Morris

: Patient's rating of cognitive ability: Using the AD8, a brief informant interview, as a self-rating tool to detect dementia. Arch Neurol, 2007; 64:725–730.

14.

IBM Corp: IBM SPSS Statistics for Windows. Version 19.0 ed. Armonk, NY: IBM Corp., 2010.