Abstract
Background.
This study aims to (i) describe frailty in the subgroup of independent community-dwelling seniors consulting emergency departments (EDs) for minor injuries, (ii) examine the association between frailty and functional decline 3 months postinjury, (iii) ascertain the predictive accuracy of frailty measures and emergency physicians’ for functional decline.
Method.
Prospective cohort in 2011–2013 among 1,072 seniors aged 65 years or older, independent in basic daily activities, evaluated in Canadian EDs for minor injuries.
Frailty was assessed at EDs using the Canadian Study of Health and Aging-Clinical Frailty scale (CSHA-CFS) and the Study of Osteoporotic Fracture frailty index (SOF). Functional decline was defined as a loss ≥2/28 on the Older American Resources Services scale 3 months postinjury. Generalized mixed models were used to explore differences in functional decline across frailty levels. Areas under the receiver operating characteristic curve were used to ascertain the predictive accuracy of frailty measures and emergency physicians’ clinical judgment.
Results.
The SOF and CSHA-CFS were available in 342 and 1,058 participants, respectively. The SOF identified 55.6%, 32.7%, 11.7% patients as robust, prefrail, and frail. These CSHA-CFS (n = 1,058) proportions were 51.9%, 38.3%, and 9.9%. The 3-month incidence of functional decline was 12.1% (10.0%−14.6%). The Areas under the receiver operating characteristic curves of the CSHA-CFS and the emergency physicians’ were similar (0.548–0.777), while the SOF was somewhat higher (0.704–0.859).
Conclusion.
Measuring frailty in community-dwelling seniors with minor injuries in EDs may enhance current risk screening for functional decline. However, before implementation in usual care, feasibility issues such as inter-rater reliability and acceptability of frailty tools in the EDs have to be addressed.
Key Words: Frailty, Functional decline, Injuries, Seniors, Emergency department
Every 4 minutes, an injured senior visit a Canadian Emergency Department (ED) (1). While 65% of injured seniors who live in the community seek treatment in EDs, the majority (60%–70%) are discharged home with injuries that may limit mobility and normal activities (2,3). Recent evidence shows that minor injuries are associated with declining function (inability to perform Activities of Daily Living [ADLs]) in approximately 15% of otherwise independent seniors up to at least 6 months postinjury (4,5). Injuries to upper and/or lower limbs (4), fractures (6), requiring help with ADLs prior to injury (4), slower mobility, and depression (6) have been associated with declining function up to 4 weeks after injury-related visits to EDs by seniors. More recently, in a multicenter Canadian ED study, predictors of functional decline 3 to 6 months after minor injuries in previously independent seniors were found to be: age ≥85 years, presenting numerous comorbidities, occasional use of walking aid, slow walking speed, not going outside of the home daily, fear of falling, falls, and visits to a general practitioner in the previous 3 months (5).
Strikingly, most of these risk factors have been described as correlates or consequences of frailty (7–11), which is recognized as a distinct clinical syndrome resulting in a reduced physiological capacity. It affects multiple systems translating into loss of reserve (energy, physical ability, cognition, emotion, health) causing increased vulnerability to stressors and putting seniors at increased risk of disability even following a minor external disturbance such as minor injury (12).
Despite their critical role in treating injured seniors, EDs have no standard approach to identify and manage those 15% independent seniors at risk of declining function. Moreover, ED screening tools for seniors and the emergency physicians (EPs) clinical judgment on the risk of functional decline have not proven good prediction in this population (5). Improving risk screening is needed as postinjury decline among these seniors is likely to continue in the current context (13). Understanding frailty and its relation to postinjury outcome may help ED clinicians better manage seniors. However, given the scarce ED resources and a growing elderly population (1,14), measuring all frailty attributes in all seniors presenting to ED is impossible. Frailty is rarely assessed in EDs but measurement tools are available (15) that may help screen injured seniors at risk of declining function. Recently, Stiffler and colleagues (16) used a five-item modified version of the Fried frailty index (17) to assess 90 seniors discharged home from ED. Hasting and colleagues (18) derived a 44-items deficit accumulation index from a Medicare database to describe frailty in 1,851 seniors discharged from EDs. As no Canadian ED study has reported on frailty measurement, we wanted to test the potential of two frailty tools to help screen seniors with minor injuries for risk of functional decline after the ED visit. The Study of Osteoporotic Fracture frailty index (SOF) (9) and the Canadian Study of Health and Aging-Clinical Frailty scale (CSHA-CFS) (7) were chosen as considered quicker than the fried frailty index or the deficit accumulation index for bed-side use in our EDs.
Our specific aims were to (i) describe frailty status in community-dwelling seniors presenting to EDs with minor injuries, according to SOF and CSHA-CFS, (ii) examine the association between frailty status and functional decline in these seniors and, (iii) compare the capacity of the SOF, CHSA-CFS, and EPs to predict declining function in this population.
Methods
Design and Participants
This prospective multicenter study was conducted within the Canadian Emergency Team Initiative (CETI) program on mobility and aging, between March 2011 and January 2013 in eight EDs (Quebec, Montreal, Ottawa, Hamilton, Toronto, Calgary). Inclusion criteria were: age ≥65 years, ED consultation within 2 weeks of a minor injury (significant lesions such as lacerations, contusions, sprains, extremity fractures, minor thoracic injuries, concussion), independence in basic ADLs (described below) prior to injury and home discharge. Patients hospitalized, living in nursing homes, unable to provide consent (eg, patients with known Alzheimer’s, dementia, or those with incident confusional state in the ED as based on clinical judgment of research assistant [RA] and/or treating physicians) or to attend follow-ups were excluded.
Procedure
Recruitment occurred 7 days/week. Participants were identified by EPs, nurses, and RAs. Except for diagnoses, injury mechanism and the EPs’ assessment, the RAs evaluated patients in the ED or by telephone within 7 days. In-person and telephone follow-ups were conducted 3 months postinjury. Ethics approval was obtained from each center.
Measures
Predictors of Functional Decline
Frailty measures: The SOF (9) was measured at the initial ED visit. It has three components: unintentional weight loss ≥10 pounds, leg strength, and low energy (19). The SOF defines robust, prefrail and frail (0, 1, ≥2 components) seniors. The “CSHA-CFS” (7) was used by the EPs. This scale classifies patients based on clinical judgment as (i) very fit, (ii) well, (iii) well with treated comorbidities, (iv) apparently vulnerable and (v) mildly, (vi) moderately, or (vii) severely frail.
The EPs’ assessment of the risk of patients’ declining function was measured at the time of the ED consultation asking the EP to draw a line on a visual analog scale in response to the following question: “What is the risk that the patient will have a decline in function in mobility in 3 months?” (0cm/10cm = none, 10cm/10cm = very likely).
Main Outcome Measure
Functional status was measured with the Older American Adult Resources and Service (OARS) (20) scale which includes seven basic ADLs (eating, grooming, dressing, transferring, walking, bathing, and continence) and seven Instrumental Activities of Daily Living (IADLs; meal preparation, homemaking, shopping, using transportation, using the phone, managing medication, and money). Scores range from 0 (dependent) to 28 (independent). A loss ≥ 2/28 OARS points (7%) from baseline to 3 months was considered clinically significant (5).
Other Variables
Demographic variables included age, gender, education, and living situation. Clinical variables included types and mechanisms of injuries and comorbidities. Cognitive status was evaluated with the Montreal Cognitive Assessment for in-person interviews using a normal cutoff ≥23/30 (21). For phone interviews, the modified “Telephone Interview for Cognitive Status” (TICS-m) was used with a normal cutoff TICS-m >32/50 (22).
The Short Falls Efficacy Scale (Short FES-1) (23) was used to evaluate fear of falling. Mobility measures included: The “Timed Up & Go” (TUG) (24); Occasional use of a walking aid which signals performance issues in lower extremities (6,25); number of falls in the previous 3 months; and the number of times/week the individual leaves home (26).
Analyses
Univariate and bivariate analyses were performed to summarize the patient’s characteristics. Linear trend analyses using multilevel modeling were performed to explore functional decline at 3 months and patients’ characteristics across the frailty measure categories. Polychoric correlation coefficients and kappa statistics were used to assess the agreement between the SOF and the CSHA-CFS as well as two aggregated versions of it: CHSA-CFS-V.1 collapsing levels 1 and 2, levels 3 and 4, levels 5 and above; and CHSA-CFS-V.2 leaving level 1 as is, collapsing levels 2 and 3, levels 4 and above. Prevalence ratios (27) were used to assess the associations of functional decline at 3 months across CSHA-CFS and SOF categories as well as with the EP’s assessments that were categorized into three levels. Generalized linear mixed models with a binomial distribution and a log link function were used to examine prevalence ratios of outcomes across frailty measures (27) that maximized the area under the receiver operating characteristic curves (AUCs) for this evaluation. The ability of frailty measures and EPs to predict functional decline were compared with AUCs (28). The cut-points for the three-level EPs’ assessment were chosen to maximize the AUC for this evaluation. All analyses were conducted using the SAS software V 9.02. (SAS Institute, Cary, NC).
Results
The recruitment flow chart can be found in Supplementary Figure 1. There were 1,531 patients meeting inclusion criteria, among whom 409 (26.7%) were excluded because they refused assessment or could not be located within the allowable time frame. A total of 1,122 patients were evaluated at ED, among whom 1,072 (95.5%) had at least one frailty measure available. There were three subsamples for this study: with available CSHA-CFS (n = 1,058); with available SOF (n = 342, 30.5%), with both assessments (n = 268). A reflected by these numbers, 30% of the patients had a complete in-person evaluation in the ED. Of the 1,072 patients with frailty assessment, 855 (79.8%) were reassessed at 3 months and 20.2% were lost to follow-up. Little baseline differences existed between participants and lost to follow-up, except that participants had more comorbidities (44.7% with ≥5 comorbidities vs 31.8%, p < .001) were more educated (26.9% university degrees vs 18.3%, p < .001) and had less cognitive deficits (27.0% vs 34.7%, p < .03). Distributions of lost to follow-ups did not differ according to SOF or CSHA-CFS.
Table 1 shows the baseline characteristics of the CSHA-CFS subsample and the SOF one. The mean age of participants was 77.1 (±7.5) years. Most frequent injuries were contusions (39.8%), lacerations (26.8%), and fractures (24.0%) that were mostly the result of falls (≥70.0%). Participants were independent in basic ADLs prior to injury, although 22% occasionally used a cane (eg, during period of exacerbating pain or for safety on icy side-walks). Overall, 60.4% of them were independent in IADLs.
Table 1.
Characteristics of the Subsamples of Independent Seniors Who Consulted EDs for Minor Injuries
| Characteristics | Total (n = 1,072) |
CSHA-CFS (n = 1,058) | SOF (n = 342) |
|---|---|---|---|
| N * (%) | N * (%) | N * (%) | |
| Age | |||
| 65–74 | 438 (40.9) | 431 (40.7) | 150 (43.9) |
| 75–84 | 443 (41.3) | 438 (41.4) | 134 (39.2) |
| 85+ | 191 (17.8) | 189 (17.9) | 58 (17.0) |
| Men | 339 (31.7) | 337 (31.9) | 139 (40.9) |
| Number of comorbidities ≥5 | 450 (42.1) | 447 (42.3) | 140 (41.1) |
| Emergency personnel assessment of the risk of functional decline ≥3/10 | 341(33.4) | 341(33.5) | 103 (32.1) |
| Lives alone | 392 (36.7) | 384 (36.4) | 122 (35.8) |
| General practitioner visits in the last 3 months ≥3 | 103 (9.7) | 102 (9.7) | 36 (10.6) |
| ED visits in the last 3 months | 123 (11.6) | 121 (11.6) | 45 (13.2) |
| Falls in the last 3 months | 231 (21.7) | 227 (21.6) | 65 (19.1) |
| Occasional use of a walking aid | 235 (22.0) | 233 (22.1) | 67 (19.6) |
| Instrumental Activities of Daily Living score = 14/14 | 647 (60.4) | 637 (60.2) | 211 (61.7) |
| Mechanism of injury | |||
| Falls | 771 (74.3) | 768 (74.6) | 222 (68.3) |
| Motor vehicle accident | 43 (4.1) | 42 (4.1) | 15 (4.6) |
| Other | 224 (21.6) | 220 (21.4) | 88 (27.1) |
| Type of injury | |||
| Mild traumatic brain injury | 173 (16.1) | 173 (16.4) | 59 (17.3) |
| Contusions | 420 (39.2) | 419 (39.6) | 135 (39.5) |
| Lacerations | 287 (26.8) | 285 (26.9) | 104 (30.4) |
| Sprains | 106 (9.9) | 106 (10.0) | 28 (8.2) |
| Fractures | 257 (24.0) | 255 (24.1) | 57 (16.7) |
| Slow walkers (Timed-Up-Go ≥20 seconds)† | 58 (16.7) | 56 (16.7) | 44 (13.8) |
| Normal Short Fall Efficacy scale (≥9.8/10) | 545 (51.0) | 538 (51.0) | 185 (54.1) |
| Cognitive deficit (MoCA <23/30 or TICS-m ≤31/50) |
297 (28.6) | 296 (28.8) | 102 (30.9) |
Notes: ED = emergency department; CSHA-CFS = Canadian Study of Health and Aging-Clinical Frailty scale; MoCA = Montreal Cognitive Assessment; SOF = Study of Osteoporotic Fracture frailty index; TICS-m = modified Telephone Interview for Cognitive Status.
*Because of missing data, numbers do not always add to the total.
†In-person interview.
The three subsamples were similar except that the SOF one included more men (40.9% vs 30.9%) and more mobile individuals (74.4% with ≥5 outings/week vs 67.0%). The distributions of patients according to frailty measures are depicted in Supplementary Figure 2. Of note, because of inclusion criteria, no participant was severely frail (CSHA-CFS = 7). The table showing the cross-tabulated distributions of SOF with CSHA-CFS, CSHA-CFS-V.1, and CSHA-CFS-V.2 in the subset with both measures (n = 328) can be found in the Supplementary Table 1. Agreement was low (polychoric correlation: 0.41, 0.43, 0.40, respectively). Best kappa statistics (0.25) was between SOF and CHSA-CFS-V.1.
The 3-month incidence of functional decline was 12.1% (10.0%–14.6%). The distribution of declining patients according to CHSA-CFS was: Very Fit: 4.8% (2.4%–9.5%); Well: 10.2% (7.0%–14.7%); Well + treated comorbidities: 11.0% (7.5%–16.3%); Apparently Vulnerable: 21.7% (14.2%–33.3%); Mildly Frail: 13.9% (7.2%–26.8%); Moderately Frail: 39.2% (22.3%–68.8%). Distributions of declining function according to the SOF and CHSA-CFS-V.1 are shown in Table 2 along with other frailty attribute. Except for education and gender, a positive trend is observed between all variables and SOF or CSHA-CFS scores (p < .0001) in the three subsamples.
Table 2.
Distribution of Declining Function at 3 Months and Patients’ Baseline Characteristics According to CSHA-CFS.V1 in Subsets With CSHA-CFS (n = 1,058), With Both Evaluations (n = 328) and With SOF (n = 342)
| CSHA-CFS-V1 (n = 1,058) |
CSHA-CFS-V1 (n = 328) |
SOF (n = 342) |
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Very Fit/Well | Well-treated Comorb./ Apparently Vulnerable | Mildly– Moderately Frail | Trend Test | Very Fit/Well | Well-treated Comorb./ Apparently Vulnerable | Mildly– Moderately Frail | Trend Test | Robust (0/3) |
Prefrail (1/3) |
Frail (≥2/3) |
Trend Test | |
| Patients, n (%) | 549 (51.9) | 405 (38.3) | 104 (9.8) | — | 174 (53.1) | 121 (36.9) | 33 (10.1) | — | 190 (55.6) | 112 (32.8) | 40 (11.7) | — |
| Decline at 3 months*, % (95% CI) |
8.1 (5.7–11.3) |
13.8 (10.1–18.8) |
20.4 (12.9–32.2) |
0.0003 | 3.4 (1.4–8.4) |
12.1 (6.4–23.1) |
7.7 (2.0 – 29.5) |
0.06 | 1.3 (0.3–5.2) |
14.3 (8.8–23.3) |
20.7 (10.1–42.3) |
<0.0001 |
| Age, mean (95% CI) |
75.9 (74.5–77.2) |
78.6 (77.2–80.0) |
81.5 (79.7–83.3) |
<0.0001 | 75.2 (73.5–77.0) |
77.7 (75.8–79.6) |
81.2 (78.2–84.1) |
<0.0001 | 74.8 (73.1–76.4) |
78.1 (76.2–79.9) |
81.3 (78.7–83.9) |
<0.0001 |
| Primary or secondary school, % (95% CI) | 51.9 (42.4–63.4) |
58.2 (47.5–71.4) |
58.8 (46.1–75.0) |
0.02 | 49.9 (37.7–66.1) |
52.8 (39.7–70.4) |
59.5 (40.0–88.6) |
0.3 | 50.8 (37.9–68.2) |
60.4 (44.8–81.5) |
59.2 (40.7–86.3) |
0.15 |
| Male, % (95% CI) |
32.8 (27.8–38.6) |
28.7 (27.8–38.6) |
28.6 (20.8–39.5) |
0.2 | 42.4 (33.4–53.7) |
42.3 (32.5–55.1) |
36.5 (22.0 –60.4) |
0.7 | 47.3 (37.5–59.7) |
29.4 (21.0–41.3) |
44.8 (30.6–65.4) |
0.1 |
| Comorbidities ≥ 5, % (95% CI) | 33.8 (24.5–46.5) |
48.7 (35.6–66.7) |
46.0 (32.4–65.4) |
<0.0001 | 30.9 (20.9–45.8) |
49.5 (34.4–71.3) |
50.1 (33.1–76.0) |
0.0009 | 34.4 (22.9–51.9) |
44.1 (29.1–66.8) |
49.9 (32.7–76.2) |
0.005 |
| Cognitive deficit, % (95% CI) | 22.4 (17.9–28.0) |
29.1 (23.3–36.4) |
50.5 (39.2–65.2) |
<0.0001 | 24.2 (18.1–32.4) |
36.7 (28.4–47.5) |
49.9 (34.6–71.9) |
0.0007 | 20.9 (15.2–28.7) |
34.5 (25.7–46.2) |
62.8 (47.7–82.8) |
<0.0001 |
| Falls last 3 months, % (95% CI) | 14.1 (10.0–19.8) |
23.5 (16.9–32.7) |
29.7 (20.1–43.8) |
<0.0001 | 13.4 (8.3–21.5) |
21.6 (13.9–33.7) |
32.3 (18.2–57.3) |
0.004 | 13.2 (9.0–19.5) |
19.7 (13.4–28.7) |
44.9 (25.6–78.7) |
0.0004 |
| Walking aid last 3 months, % (95% CI) | 11.8 (8.6–16.2) |
30.4 (23.3–39.6) |
48.1 (36.1–64.2) |
<0.0001 | 7.7 (4.1– 4.8) |
25.4 (15.1–42.7) |
28.9 (15.4–54.1) |
<0.0001 | 8.3 (4.5–15.6) |
21.1 (12.2–36.7) |
39.3 (22.4–69.1) |
<0.0001 |
| Normal Short Fall Efficacy scale, % (≥9.8/10) | 55.7 (46.5 –66.7) |
44.1 (36.3–53.7) |
29.0 (20.9–40.3) |
<0.0001 | 56.8 (45.4–71.1) |
45.4 (34.9–59.1) |
30.1 (18.1–50.2) |
0.002 | 61.6 (51.6–73.6) |
49.3 (38.1 –61.1) |
15.4 (7.3–32.3) |
<0.0001 |
| TUG (seconds),‡ mean (95% CI) | 12.3 (10.6–13.9) |
16.4 (14.4–18.4) |
20.9 (17.3–24.5) |
<0.0001 | 11.7 (10.1–13.2) |
15.9 (14.0–17.8) |
19.4 (15.6–23.1) |
<0.0001 | 11.5 (10.0–12.9) |
16.3 (14.4–18.2) |
19.6 (16.5–22.8) |
<0.0001 |
| IADL score (<14), % (95% CI) |
26.8 (23.3–30.8) |
49.9 (45.2–55.0) |
69.2 (60.9–78.7) |
<0.0001† | 25.9 (20.1–33.3) |
52.9 (44.7–62.6) |
54.5 (39.9–74.6) |
<0.0001† | 27.8 (21.4–36.1) |
46.5 (36.6–59.0) |
60.7 (45.5–81.1) |
<0.0001 |
| Social engagement scale, mean/5 (95% CI) | 3.9 (3.7–4.0) |
3.4 (3.2–3.6) |
3.0 (2.7–3.3) |
<0.0001 | 4.0 (3.7–4.2) |
3.4 (3.1–3.7) |
3.2 (2.7–3.8) |
0.0004 | 4.0 (3.7–4.2) |
3.5 (3.2–3.8) |
2.9 (2.5–3.4) |
<0.0001 |
Notes: CI = confidence interval; CSHA-CFS = Canadian Study of Health and Aging-Clinical Frailty scale; IADL = Instrumental Activities of Daily Living.
*Because of loss to follow-up, patients with decline and CSHA-CFS data = 855; patients with decline and SOF data = 281.
†Armitage Cochran trend test.
‡In-person interviews.
Table 3 shows prevalence ratios with 95% confidence interval of functional decline according to SOF, CSHA-CFS-V.1, and to the three-group categorization of the EPs’ assessment of patients’ risk in the data set with both (SOF and CHSA-CFS) frailty measures available and in the one with CHSA-CFS only. Prefrail and frail SOF-defined patients were 11.0 and 15.9 times more at risk of declining function than robust ones (p = .002). According to CHSA-CFS-V.1, mildly/severely frail patients were around 2.5 times more at risk of declining function. Regarding the EPs’ assessment, a cutoff of 3.1/10 was associated with a two to five times risk of function decline at 3 months postinjury.
Table 3.
Prevalence Ratios and AUC* of Functional Decline 3 Months Postinjury According to the SOF, CSHA-CFS V.1, and the EPs’ Assessments in the Subset With Both Evaluations Available (n = 281) and in the CSHA-CFS Subset
| Prevalence Ratio (95% CI) |
p Value | AUC* (95% CI) |
||
|---|---|---|---|---|
| SOF (n = 281) |
Robust | 1.00 | .002 | 0.781 (0.704–0.859) |
| Prefrail | 11.0 (2.5–47.7) | |||
| Frail | 15.9 (3.4–75.6) | |||
| CSHA-CFS V.1 (n = 281) |
Very fit–well | 1.00 | .05 | 0.663 (0.548–0.777) |
| Well-treated comorbidities— apparently vulnerable | 3.5 (1.3–9.7) | |||
| Mildly–severely frail | 2.2 (0.5–10.9) | |||
| EPs’ assessments (n = 281) |
0–1.09 | 1.00 | .07 | 0.667 (0.562–0.772) |
| 1.1–3 | 4.3 (0.9 – 20.3) | |||
| 3.1–10 | 5.8 (1.3 – 25.8) | |||
| CSHA-CFS V.1 (n = 841) |
Very fit–well | 1.00 | .002 | 0.637 (0.529–0.746) |
| Well-treated comorbidities— apparently vulnerable | 1.7 (1.1–2.6) | |||
| Mildly–severely frail | 2.5 (1.5–4.3) | |||
| EPs’ assessments (n = 841) |
0 – 1.09 1.1–3 3.1–10 |
1.00 1.4 (0.8–2.3) 2 (1.2–3.3) |
.02 | 0.667 (0.562–0.772) |
Notes: CSHA-CFS = Canadian Study of Health and Aging-Clinical Frailty scale; EP = emergency physician; SOF = Study of Osteoporotic Fracture index.
*AUC = area under the receiver operating characteristic curves obtained with a single model.
Table 3 also shows the area under the AUCs with 95% confidence interval for each tool (SOF, CHSA-CFS-V.1, and EPs) in the above-mentioned data sets. As reflected by their AUCs, the CHSA-CFS-V.1 and the EPs’ capacity to predict declining function were similar in both data sets (p ≥ .05). The AUC of the SOF (0.78, 95% confidence interval: 0.70–0.86) was significantly higher (p < .05).
Discussion
This study examined two frailty measurement tools on their ability to correlate with frailty attributes in Canadian community-dwelling seniors consulting EDs for minor injuries, and to screen them for risk of functional decline as compared to EPs’ clinical judgment.
According to the instrument used, there were 9.9% to 11.7% individuals qualifying as frail and 32.7% to 38.2% as prefrail. These results compare with the systematic review by Collard and colleagues (29) who found a weighted average prevalence of frailty and prefrailty of 13.6% and 33.5% in community-dwelling seniors. Moreover, our distributions of frailty attributes compare with other population-based frailty studies (8,30).
In the ED, Stiffler and colleagues (16) found a 20% prevalence of frailty among 90 seniors discharged home from ED. Besides their small sample, differences between their results and ours are explained by 28% of their patients being dependant in at least one ADL and chief complaints such as chest/abdominal pain, heart problems, weakness, and dizziness, while we excluded previous ADL disability and recruited seniors with minor injuries only. On their part, Hasting and colleagues (18) described four quartiles of frailty levels based on deficit accumulation index scores extracted from a Medicare database (n = 1,851) and found that deficit accumulation index strongly predicted serious adverse outcomes following the ED visit.
In this study, while proportions of robust, prefrail, and frail seniors were similar across the SOF and CHSA-CFS levels, agreement between the two was only moderate. Because of the different frailty models that varied measurement tools are derived from, ascertaining frailty may capture different but overlapping groups of seniors (8,31). Cigolle and colleagues (31) have summarized differences and overlaps of three main models: the Functional model (32) based on deficiencies in physical, nutritive, cognitive, and sensory domains; the Rockwood Burden model (33) based on accumulation of diseases and disabilities, and the Fried (17) Biological model based on cumulative decline across physiologic systems. In addition, three measurement approaches have been described (14): rules-based, deficits counts, and clinical scales based on clinical judgment. In this complex context where concepts and methods are still debated, EDs face the additional challenge of requiring valid and reliable frailty tools that can be performed rapidly. Moreover, it has also been recommended that frailty screening tools should be interpretable by nonspecialists (34). In this study, both measures were quick but based on different models and measurement approaches. The CSHA-CFS is clinical judgment-based and derived from the Burden model, while the SOF is derived from the Biological model and is a clinical performance rules-based measure.
Both measures showed a dose response with decline and frailty attributes but did not risk screen patients for postinjury decline equally. Interestingly, CSHA-CFS and EPs had similar predictive ability. As both were performed by treating EPs, our results indicate that providing EPs with the CSHA-CFS that includes clear descriptors did not enhance their ability to predict functional decline. This is not surprising as the CSHA-CFS is based on judgment and was validated for use by specialists (14) while there is little evidence that front line EPs are aware of the potential impact of minor injuries on seniors (35). In such context, a more objective clinical performance-based tool as the SOF might be preferred.
Good recovery is expected after minor injuries. However, the latter are associated with functional decline in some community-dwelling seniors who are still independent at the time of injury. Patients at increased risk of decline are in a prefrail/frail state that is not currently recognized in EDs. Existing ED screening tools such as ISAR (36), BRIGHT (37), or SILVER CODE (36,38) mostly predict severe outcomes (hospitalization, institutionalization, and mortality) and although not derived from frailty models, they target the most impaired and severely frail individuals. These tools have shown moderate ability to predict functional decline in ED patients (37) and were not derived specifically for the subgroup that is the focus of the current study. In addition, even if senior populations are increasing in EDs (1), little changes are made to organize care accordingly (14), which remain inadequate (39). This is particularly the case for seniors with new functional limitations for whom the ED stay was described as a “Modern-Day Purgatory” (40) where patients do not meet neither criteria for hospitalization, placement in observation, alternative setting nor for safe return home. As a result, these patients wait long hours for proper disposition (40). Improving identification of seniors with minor injuries at highest risk of declining function may help expedite subsequent treatment and more appropriate disposition and should be a high priority for EDs (40).
The chief strength of this study is that it is the largest multicenter Canadian ED research regarding mobility in seniors. It includes EDs from eastern, central, and western Canada, thus representative of a wide segment of Canadian community-dwelling seniors with minor injuries. The study also targeted a well-defined functional population of seniors, while most previous ED studies were conducted with more disabled ones. Moreover, the broad inclusion criteria involving a wide spectrum of minor injuries are also a strength to this study making it representative of an important part of the daily ED clinical practice with injured seniors (70% are discharged back home).
Yet, this heterogeneity of injuries experienced by seniors participating to this study may also constitute a limitation. For instances, while extremity fractures are often associated with short-term impairment and may cause longer-term ones, lacerations rarely do. Moreover, predictors of functional decline in individuals with different types of injuries will most likely differ. In this context, using a frailty measure in the ED is likely to be useful across various types of injury.
Another limitation to the study is that we measured the SOF in 30% of patients (in-person evaluations) because the study lacked the funds to add RA staff on evening and night shifts or to compensate patients to return to EDs for in-person evaluation on the following day of their consultation. The reduced number of SOF evaluations has two types of consequences. First, although the AUC was good for the SOF (0.79), it will almost certainly not perform as well in a validation study. Second, the association between the SOF and functional decline might be underestimated as in comparison with seniors with a baseline SOF, those without declined more at 3 months (13.3% vs 7.8%, p = .02) were slower walkers (TUG >20 seconds: 48.3% vs 13.8%, p < .0001) and less mobile outdoor (35.4% vs 25.6%, p = .002).
Another limitation is that severely frail patients were excluded. Therefore, the results do not apply to the 30% hospitalized injured seniors. Finally, only 30% of the cohort was followed up in-person vs 70% over the phone. Although we chose validated tools for both types of assessments, phone evaluation of seniors with mild cognitive or hearing deficits may have introduced variability in outcomes.
This study showed that measuring frailty among community-dwelling injured seniors who seek treatment at EDs may enhance current risk screening and help with early identification of seniors at risk of functional decline after minor injuries. However, before implementing either SOF or CSHA-CFS in usual ED care, feasibility issues such as inter-rater reliability across ED professionals has to be determined, which was not done in the current study. First and foremost, the acceptability by ED staff of any of these tools in their daily practice will have to be demonstrated before any successful implementation. Further ED-based frailty research should focus on the latter as well as evaluating the added value of frailty assessment to existing risk screening tools.
Supplementary Material
Please visit the article online at http://biomedgerontology.oxfordjournals.org/ to view supplementary material.
Funding
This research is part of the Canadian Emergency Team Initiative (CETI) funded by the Canadian Institutes of Health Research through their Emerging Team Grant Program on Mobility in Aging (CIHR-229031, CIHR-23145) and by Université Laval de Québec. Information about the CETI can be found at www.cha.quebec.qc.ca/ceti/.
Conflict of Interest
The authors have no conflict of interest to report.
Supplementary Material
Acknowledgments
The authors thank all the CETI members for their valuable assistance in the completion of this study, as well as the research assistants who collected the data and helped with manuscript editing.
References
- 1. Ontario Injury Prevention Resource Center. Injuries among Seniors in Ontario: A Descriptive Analysis of Emergency Department and Hospitalization Data. Toronto, Ontario: Ontario Injury Prevention Resource Centre; 2007. ISBN:978-1-894828-00-0 [Google Scholar]
- 2. Scott V, Pearce M, Pengelly C. Technical Report: Injury Resulting from Falls among Canadians Age 65 and over. On the Analysis of Data from the Canadian Community Health Survey, Cycle 2.1 as Presented in Report on Senior’ Falls in Canada. Toronto, Ontario: Public Health Agency of Canada; 2005:16. [Google Scholar]
- 3. Bawa H, Brussoni M, De Gagné D. Emergency Department Surveillance System: Seniors Injury Data Report 2001–2003. Vancouver, BC: BC Injury Research and Prevention Unit; 2004. [Google Scholar]
- 4. Shapiro MJ, Partridge RA, Jenouri I, Micalone M, Gifford D. Functional decline in independent elders after minor traumatic injury. Acad Emerg Med. 2001;8:78–81. [DOI] [PubMed] [Google Scholar]
- 5. Sirois MJ, Emond M, Ouellet MC, et al. Cumulative incidence of functional decline after minor injuries in previously independent older Canadian individuals in the emergency department. J Am Geriatr Soc. 2013;61:1661–1668. doi:10.1111/jgs.12482 [DOI] [PubMed] [Google Scholar]
- 6. Russell MA, Hill KD, Blackberry I, Day LL, Dharmage SC. Falls risk and functional decline in older fallers discharged directly from emergency departments. J Gerontol A Biol Sci Med Sci. 2006;61:1090–1095. [DOI] [PubMed] [Google Scholar]
- 7. Rockwood K, Song X, MacKnight C, et al. A global clinical measure of fitness and frailty in elderly people. CMAJ. 2005;173:489–495. doi:10.1503/cmaj.050051 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Kiely DK, Cupples LA, Lipsitz LA. Validation and comparison of two frailty indexes: The MOBILIZE Boston Study. J Am Geriatr Soc. 2009;57:1532–1539. doi:10.1111/j.1532-5415.2009.02394.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Ensrud KE, Ewing SK, Taylor BC, et al. Comparison of 2 frailty indexes for prediction of falls, disability, fractures, and death in older women. Arch Intern Med. 2008;168:382–389. doi:10.1001/archinternmed.2007.113 [DOI] [PubMed] [Google Scholar]
- 10. Cawton PM, Marshall LM, Michael Y, et al. Frailty in older men: prevalence, progression, and relationship with mortality. J Am Geriatr Soc. 2007;55:1216–1223. [DOI] [PubMed] [Google Scholar]
- 11. Savva GM, Donoghue OA, Horan M, O’Regan C, Cronin H, Kenny RA. Using timed up-and-go to identify frail members of the older people. J Gerontol A Biol Sci Med Sci. 2013;68:441–446. [DOI] [PubMed] [Google Scholar]
- 12. Clegg A, Toung J, Iliffe S, Rikkert MO, Rockwood K. Frailty in elderly people. The Lancet. 2013;381:752–762. doi:10.1016/S0140-6736(12)62167-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Salter AE, Khan KM, Donaldson MG, et al. Community-dwelling seniors who present to the emergency department with a fall do not receive Guideline care and their fall risk profile worsens significantly: a 6-month prospective study. Osteoporos Int. 2006;17:672–683. doi:10.1007/s00198-005-0032-7 [DOI] [PubMed] [Google Scholar]
- 14. Goldstein JP, Andrew MK, Travers A. Frailty in older adults using pre-hospital care and the emergency department: a narrative review. Can Geriatr J. 2012;15:16–22. doi:10.5770/cgj.15.27 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. de Vries NM, Staal JB, van Ravensberg JSM, et al. Outcome instruments to measure frailty: a systematic review. Ageing Res Rev. 2010;10:104–114. doi:10.1016/j.arr.2010.09.001 [DOI] [PubMed] [Google Scholar]
- 16. Stiffler KA, Finley A, Midha S, Wilber ST. Frailty assessment in the emergency department. J Emerg Med. 2013;45:291–298. doi:10.1016/j.jemermed.2012.11.047 [DOI] [PubMed] [Google Scholar]
- 17. Fried LP, Tangen CM, Walston J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci. 2001;56:M146–M156. doi:10.1093/gerona/56.3.M146 [DOI] [PubMed] [Google Scholar]
- 18. Hastings SN, Purser JL, Johnson KS, Sloane RJ, Whitson HE. Frailty predicts some but not all adverse outcomes in older adults discharged from the emergency department. J Am Geriatr Soc. 2008;56:1651–1657. doi:10.1111/j.1532-5415.2008.01840.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Radloff L. The CES-scale: a self report depression scale for research in the general population. Appl Psychol Mea. 1977;1:387–401. [Google Scholar]
- 20. Fillenbaum G. Multidimensional Functional Assessment: The OARS Methodology-A Manual. Durham, NC: Duke University; 1978. [Google Scholar]
- 21. Luis CA, Keegan AP, Mullan M. Cross validation of the Montreal Cognitive Assessment in community dwelling older adults residing in the Southeastern US. Int J Geriatr Psychiatry. 2009;24:197–201. doi:10.1002/gps.2101 [DOI] [PubMed] [Google Scholar]
- 22. Knopman DS, Roberts RO, Geda YE, et al. Validation of the telephone interview for cognitive status-modified in subjects with normal cognition, mild cognitive impairment, or dementia. Neuroepidemiology. 2010;34:34–42. doi:10.1159/000255464 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Kempen GI, Yardley L, van Haastregt JC, et al. The Short FES-I: a shortened version of the falls efficacy scale-international to assess fear of falling. Age Ageing. 2008;37:45–50. doi:10.1093/ageing/afm157 [DOI] [PubMed] [Google Scholar]
- 24. Podsiadlo D, Richardson S. The timed “Up & Go”: a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc. 1991;39:142–148. [DOI] [PubMed] [Google Scholar]
- 25. Stuck AE, Walthert JM, Nikolaus T, Büla CJ, Hohmann C, Beck JC. Risk factors for functional status decline in community-living elderly people: a systematic literature review. Soc Sci Med. 1999;48:445–469. [DOI] [PubMed] [Google Scholar]
- 26. Canadian Study of Health and Aging Working Group. Canadian study of health and aging: study methods and prevalence of dementia. CMAJ. 1994;150:899–913. [PMC free article] [PubMed] [Google Scholar]
- 27. McNutt LA, Wu C, Xue X, Hafner JP. Estimating the relative risk in cohort studies and clinical trials of common outcomes. Am J Epidemiol. 2003;157:940–943. [DOI] [PubMed] [Google Scholar]
- 28. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 1988;44:837–845. [PubMed] [Google Scholar]
- 29. Collard RM, Boter H, Schoevers RA, Oude Voshaar RC. Prevalence of frailty in community-dwelling older persons: a systematic review. J Am Geriatr Soc. 2012;60:1487–1492. doi:10.1111/j.1532-5415.2012.04054.x [DOI] [PubMed] [Google Scholar]
- 30. Avila-Funes JA, Helmer C, Amieva H, et al. Frailty among community-dwelling elderly people in France: the three-city study. J Gerontol A Biol Sci Med Sci. 2008;63A:1089–1096. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Cigolle CT, Osfstedal MB, Tian Z, Blaum CS. Comparing models of frailty: the Health and Retirement Study. J Am Geriatr Soc. 2009;57:830–839. doi:10.1111/j.1532-5415.2009.02225.x [DOI] [PubMed] [Google Scholar]
- 32. Strawbridge WJ, Shema SJ, Balfour JL, Higby HR, Kaplan GA. Antecedents of frailty over three decades in an older cohort. J Gerontol B Psychol Sci Soc Sci. 1998;53:S9–S16. [DOI] [PubMed] [Google Scholar]
- 33. Rockwood K, Andrew ML, Mitnitski A. A comparison of two approaches to meassuring frailty in elderly people. J Gerontol A Biol Sci Med Sci. 2005;62A:738–743. [DOI] [PubMed] [Google Scholar]
- 34. Clegg A, Rogers L, Young J. Diagnostic test accuracy of simple instruments for identifying frailty on community-dwelling older people: a systematic review. Age & Ageing. 2015;44:148–152. doi:10.1093/ageing/afu157 [DOI] [PubMed] [Google Scholar]
- 35. Carpenter CR, Griffey RT, Stark S, Coopersmith CM, Gage BF. Physician and nurse acceptance of technicians to screen for geriatric syndromes in the emergency department. West J Emerg Med. 2011;12:489–495. doi:10.5811/westjem.2011.1.1962 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Di Bari M, Salvi F, Roberts AT, et al. Prognostic stratification of elderly patients in the emergency department: a comparison between the “Identification of Seniors at Risk” and the “Silver Code”. J Gerontol A Biol Sci Med Sci. 2012;67:544–550. doi:10.1093/gerona/glr209 [DOI] [PubMed] [Google Scholar]
- 37. Boyd M, Koziol-McLain J, Yates K, et al. Emergency department case-finding for high-risk older adults: the Brief Risk Identification for Geriatric Health Tool (BRIGHT). Acad Emerg Med. 2008;15:598–606. doi:10.1111/j.1553-2712.2008.00157.x [DOI] [PubMed] [Google Scholar]
- 38. Di Bari M, Balzi D, Roberts AT, et al. Prognostic stratification of older persons based on simple administrative data: development and validation of the “Silver Code,” to be used in emergency department triage. J Gerontol A Biol Sci Med Sci. 2010;65:159–164. doi:10.1093/gerona/glp043 [DOI] [PubMed] [Google Scholar]
- 39. Carpenter CR, Shah MN, Hustey FM, Heard K, Gerson LW, Miller DK. High yield research opportunities in geriatric emergency medicine: prehospital care, delirium, adverse drug events, and falls. J Gerontol A Biol Sci Med Sci. 2011;66:775–783. doi:10.1093/gerona/glr040 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Platts-Mills TF, Owens ST, McBride JM. A modern-day purgatory: older adults in the emergency department with nonoperative injuries. J Am Geriatr Soc. 2014;62:525–528. doi:10.1111/jgs.12699. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.