Emergency department: risk stratification in the elderly
Background & aims. The older adults have very frequent access to the Emergency Department (ED). The aim of this study is to explore the ability of some geriatric screening tools validated for the ED to predict outcomes (mortality, hospitalization, ED readmission and institutionalization) at 6 months.
Methods. Older adults consecutively admitted to Cagliari University’s ED between May and December of 2017 were enrolled. In ED older patients were screened with three tools: Identification of Seniors at Risk tool (ISAR); Triage Risk Screening Tool (TRST); International Resident Assessment Instrument Emergency Department Screener (InterRAI ED Screener). At 6 months patients were contacted by phone to verify: mortality, ED readmission, hospital admission, and institutionalization.
Results. Of the 421 patients (median age 77, Interquartile Range 71-83; 55.8% women) enrolled, 72.4% were positive at the ISAR, 50.1% at the TRST; moreover 44.9% of enrolled subjects needed a urgent geriatric evaluation at the InterRAI ED Screener. The dead subjects had ISAR, TRST and InterRAI ED Screener with greater severity compared to the alive ones. The ISAR and the TRST were also more severe in subjects who had ED readmission, while those hospitalized, in addition to the ISAR, had the more severe Inter- RAI ED Screener. However, applying stepwise logistic regression, of the three tools used, only the ISAR was a predictor for hospitalization (OR = 1.23; CI = 1.03- 1.48; P = 0.02; AUC = 0.63).
Conclusions. The association of ISAR and InterRAI ED Screener may be useful in ED to intercept both critical issues typical of the elderly, and the need and priority of the geriatric evaluation.
Older subjects, representing about a quarter of the Italian population, are the greatest consumer of healthcare resources 1. They often access the Emergency Department (ED), whether it reflects a greater burden of multimorbidity, an inappropriate access because of deficiency in outpatient care, or both 2.
They also have a greater length of stay (LOS) as a consequence of their greater complexity (for example, atypical clinical presentation, polypharmacy, and cognitive impairment) 3. A greater LOS has been associated with poorer health outcomes, including missed or incorrect diagnoses 2.
In fact, at 3 months after ED discharge 23% had repeated access to ED, 24% were hospitalized and 10% were institutionalized or died 2.
It has been hypothesized that specific screening procedures and intervention protocols may support ED physician for an appropriate evaluation of older patients, without necessarily having to make use of a time-consuming Comprehensive Geriatric Assessment (CGA) 4. Identification of Seniors at Risk tool (ISAR), Triage Risk Screening Tool (TRST) and International Resident Assessment Instrument Emergency Department Screener (InterRAI ED Screener) have been proposed as valuable and reliable tools to screen high-risk older patients in the ED 5.
The aim of the present study was to compare the ability of ISAR, TRST and InterRAI ED Screening to predict middle-term outcomes (mortality, hospitalization, ED readmission and institutionalization).
MATERIALS AND METHODS
The study population consisted of all older patients admitted to the Cagliari University’s ED in the mornings from Monday to Friday between May and December of 2017. Patients were excluded if younger than 65 years, unable to provide reliable information and to sign the informed consent.
SCREENING TOOLS IN EMERGENCY DEPARTMENT
Emergency color code: is the emergency code used in Italy, it consists of 4 levels: red (immediate life-saving intervention required); yellow (high risk and unstable situation); green (it is not an emergency but the patient needs care); white (it is not an emergency and the patient doesn’t need the ED intervention) 6.
ISAR: is a risk-screening tool designed and validated for use in the emergency setting.
The ISAR is a 6-item self-report screening tool with simple yes/no questions that can be asked to the patient or the caregiver. Well known risk factors for adverse health outcomes in older patients are included among the questions (activities of daily living, visual and cognitive decline, hospital admission history, and polypharmacy). Consequently, it has both immediate clinical relevance and good predictive validity 5. An ISAR score ≥ 2 (in a range from 0 to 6) suggests an increased risk for functional decline, repeated ED visits, hospitalization, institutionalization, and death within 6 months after an ED visit 7.
TRST: is another screening test for the ED, developed in the United States to be used by nurses, in which the presence of cognitive impairment or a score ≥ 2 in the remaining risk factors suggests an increased risk for functional decline, ED readmission, hospitalization and institutionalization 30 and 120 days after an ED visit 8,9.
The InterRAI ED Screener is a screening tool to identify older adults with increased risk of needing geriatric assessment in the ED or after the discharge. This tool evaluates the patient’s performance and abilities through the analysis of cognitive and physical functions, in particular: mood, understanding, falls, nutritional status and the occurrence of pain or dyspnea. The tool defines six levels of risk: Level 5 and 6 are classified as high risk and identify non-self-sufficient patients who need an urgent geriatric evaluation. Level 3 and 4 indicate intermediate risk and include patients who should refer to a geriatric service after the discharge. Level 2 includes low-risk and self-reliant patients and a referral to geriatric services may not be required, but periodic monitoring by primary care may be warranted. Finally, level 1 is defined as low risk, so the elderly patients are in good health and a geriatric evaluation may not be required 10. In this study we divided into 2 groups the results of this test, the first one including subjects that do not require geriatric evaluation (score 1-2) and the second one including those that require a CGA (score 3-6).
ASSESSMENT OF FUNCTIONAL INDEPENDENCE
Basic Activities of Daily Living (BADL): were used to assess the ability to perform tasks such as taking a bath, using the toilet, walking, urinary and fecal continence, dressing and feeding. For each of these activities carried out independently, two points are assigned, reaching a maximum score of twelve, that means total autonomy 11.
Instrumental Activities of Daily Living (IADL) was used to assess the ability to perform tasks such as using a telephone, doing laundry, and handling finances. The scale measures eight domains: a score of 8 indicating total autonomy, and 0, total dependence 12.
Each patient received a phone call at 6 months after ED discharge. Patients and/or their caregiver were asked to report outcomes (mortality, hospitalization, ED readmission and institutionalization).
Because the variables examined were not normally distributed, data are presented as median and ranges. Data were analyzed using a Mann-Whitney test for independent samples; instead categorical variables were analyzed by chi-square.
Cohen’s kappa was used to evaluate the concordance between TRST and ISAR.
Logistic regression analysis, with each outcome (mortality, hospitalization, ED readmission) separately as dependent variable, except institutionalization due to the low number of institutionalized subjects, was then performed through a stepwise procedure, which eliminates first the least significant association and then the non-significant independent variables (age, gender, color code, waiting time, length of stay, BADL, IADL, ISAR, TRST, InterRAI ED Screener). The results are reported indicating the odds ratio (OR) with 95% confidence intervals (CI) and Area Under the Curve (AUC) (values ≥ 0.7 were considered accurate).
MedCalc software (Version 19.5, Ostend, Belgium) was used for the statistical analysis of the data.
A p-value < 0.05 was considered indicating statistical significance.
For the purpose of this study, 421 patients (median age was 77, Interquartile Range 71-83; 44.2% men) were enrolled.
Their characteristics are summarized on Table I.
Patients at greater risk of unfavorable outcome were 72.4% by ISAR and 50.1% by TRST (Cohen’s kappa coefficient: 0.296). Moreover, 274 (65%) patients resulted in need of a geriatric evaluation according to the InterRAI ED Screener. This need was urgent (score 5-6) for 189 (44.9%) subjects (Tab. I).
Concerning gender differences in the study parameters, women had a worse profile for BADLs (P = 0.0005), ISAR (P = 0.003), TRST (P = 0.04) and InterRAI ED Screener (P = 0.02). (Tab. I).
The events that occurred within 6 months after ED admission were: 42 exitus (10.0%), 89 ED readmissions (21.1%), 77 hospital admissions (18.3%) and only 10 institutionalizations (2.4%) (Tab. II).
Table III shows the results obtained by comparing the subjects who had an unfavorable outcome (exitus, ED readmission, hospital admission) within 6 months by ED admission. In particular, the dead subjects had more advanced age, access code, ISAR, TRST and InterRAI ED Screener with greater severity and BADL and IADL more compromised, compared to the alive ones. The ISAR and the TRST were also more severe in subjects who had ED readmission, while those hospitalized, in addition to the ISAR, had the more severe InterRAI ED Screener (Tab. III). We applied the logistic regression, considering separately mortality, ED readmission and hospital admission as dependent variable and age, gender, waiting time, length of stay, BADL, IADL, ISAR, TRST and InterRAI ED Screener as independent variable and we found a positive association between mortality and gender (OR = 2.45; CI = 1.16-5.16; P = 0.019; AUC = 0.81) and IADL (OR = 0.66; CI = 0.57-0.77; P < 0.0001; AUC = 0.81). Hospital admission was positively correlated with the ISAR (OR = 1.23; CI = 1.03-1.48; P = 0.02; AUC = 0.63), and negatively with the length of stay (OR = 1.0; CI = 0.996-1.0; P = 0.03; AUC = 0.63). Finally, ED readmission negatively correlated with length of stay (OR = 1.0; CI = 0.996-1.0; P = 0.04; AUC = 0.56) (Tab. IV).
Accesses to the ED by the elderly population increased in the last years, often in relation to inappropriate drug prescription with relevant adverse drug reactions 13. This kind of patients often has atypical signs and symptoms and multimorbidities that amplify difficulty in diagnosis and treatment, with a greater risk of ED return visits, hospitalization and death 5.
A geriatric evaluation of every older patient accessing the ED is not possible, because of time constraint and workload. However, an appropriate screening and decision-making tool may help emergency physicians to prevent adverse outcomes (early mortality, hospitalization, ED readmission, institutionalization, etc.) 5 and to assess the need and priority of a geriatric evaluation. The latter can be identified through the InterRAI ED Screener, available as an application on smartphone, easy and quick to use 10.
Pua and Matchar 14 showed the usefulness of the Short Physical Performance Battery (SPPB) in predicting falls and mobility limitations in the elderly who access ED due to a fall. Regarding SPPB recently 15 it has been found to be predictive in acute care wards of length-of-stay, in-hospital and postdischarge death.
Instead, in our study, the elderly subjects were enrolled regardless of clinical reason of ED access, therefore we considered it more appropriate to use tools with a broader outcome prediction such as ISAR and TRST in accordance with the American guidelines 16.
Certainly, prevalent disability plays a role in determining outcomes. In our study, a high percentage of the 421 enrolled subjects had at least one functional disability (65.6% IADL and 63.9% BADL).
In our study the ISAR positivity at the ED admission was 72.4%, which appears in line with the literature, although a variability is observed from 52.1 to 81.5% 17-22.
This tool at Mann Whitney test was able to predict the risk of mortality, hospital admission and ED readmission ED in line with ISAR validation (excluding institutionalization not considered for the low number of subjects institutionalized – 10 of 421).
In the univariate analysis, TRST was a predictor of mortality and readmission but not for hospitalization. We cannot rule out that different timing for outcomes evaluation may explain the difference between our study and the literature.
InterRAI ED Screener, in addition to indicating the priority for geriatric assessment, was a predictor for death and hospitalization although this tool was not validated for these outcomes. This could be related to the greater number of subjects needing an urgent geriatric examination in our study population as compared to the literature (44.9 versus 27.2 and 14.4%, respectively) 24,25.
However, applying stepwise logistic regression, of the three tools used, only the ISAR was a predictor for hospital admission, with a poor discriminative capacity for predicting this outcome, as evidenced by the AUC of 0.63.
In conclusion, our study showed the utility of using an instrument capable of intercepting critical issues typical of the elderly, such as the ISAR, in association with InterRAI ED Screener, as the latter can identify the need and priority of the geriatric evaluation. Future large studies are needed to confirm the results of our study.
Conflict of interest
On behalf of all authors, the corresponding author states that there is no financial/conflict of interest or source of funding.
All the authors contributed in the development of this manuscript.
Figures and tables
|Whole population(n = 421)||Male(n = 186)||Female(n = 235)|
|Median (IR)||Median (IR)||Median (IR)||Mann Whitney|
|Age, years||77 (71-83)||76 (71-83)||78 (72-84)||0.10|
|Waiting time (minutes)||58 (18-122.25)||48 (31.16-65.84)||68 (52.99-92.03)||0.0295|
|Length of stay (minutes)||237 (154.75-317.25)||233 (149-316)||238 (158.25-317.75)||0.45|
|BADL||10 (7-12)||11 (8-12)||9(6-12)||0.0005|
|IADL||5 (1-8)||5 (2-8)||5(1-8)||0.17|
|ISAR||3 (2-4)||2 (1-3)||3(2-4)||0.003|
|TRST||2 (1-2)||1 (1-2)||2(1-3)||0.04|
|InterRAI ED Screener||3 (1-6)||3 (1-6)||5(1-6)||0.02|
|BADL||N. (%)||N. (%)||N. (%)||χ2|
|< 12||269 (63.9)||107 (57.5)||162 (68.9)||0.008 *|
|12||132 (31.4)||71 (38.2)||61 (26.0)|
|NA||20 (4.7)||8 (4.3)||12 (5.1)|
|IADL||N. (%)||N. (%)||N. (%)||χ2|
|< 8||276 (65.6)||120 (64.5)||156 (66.4)||0.59*|
|8||125 (29.7)||58 (31.2)||67 (28.5)|
|NA||20 (4.7)||8 (4.3)||12 (5.1)|
|Discharge||N. (%)||N. (%)||N. (%)||χ2|
|Hospitalization||226 (53.7)||109 (58.6)||117 (49.8)||0.17|
|Home||182 (43.2)||71 (38.2)||111 (47.2)|
|Other||13 (3.1)||6 (3.3)||7 (3.0)|
|Caregiver||N. (%)||N. (%)||N. (%)||χ2|
|Presence||234 (55.6)||96 (51.6)||138 (58.7)||0.15|
|Absence||187 (44.4)||90 (48.4)||97 (41.3)|
|Color code at the access||N. (%)||N. (%)||N. (%)||χ2|
|Red||29 (6.9)||15 (8.1)||14 (6.0)||0.59|
|Yellow||296 (70.3)||133 (71.5)||163 (69.4)|
|Green||92 (21.9)||37 (19.9)||55 (23.4)|
|White||4 (0.9)||1 (0,5)||3 (1.3)|
|ISAR||N.(%)||N. (%)||N. (%)||χ2|
|Positive||305 (72.4)||131 (70.4)||174 (74.0)||0.26|
|Negative||116 (27.6)||55 (29.6)||61 (26.0)|
|TRST||N.(%)||N. (%)||N. (%)||χ2|
|Positive||211 (50.1)||86 (46.2)||125 (53.2)||0.12|
|Negative||210 (49.9)||100 (53.8)||110 (46.8)|
|InterRAI ED Screener||N.(%)||N. (%)||N. (%)||χ2|
|1||116 (27.5)||61 (32.8)||55 (23.4)||0.132*§|
|2||12 (2.9)||3 (1.6)||9 (3.8)|
|3||73 (17.3)||37 (19.9)||36 (15.3)|
|4||12 (2.9)||4 (2.2)||8 (3.4)|
|5||26 (6.2)||12 (6.4)||14 (6.0)|
|6||163 (38.7)||62 (33.3)||101 (43.0)|
|NA||19 (4.5)||7 (3.8)||12 (5.1)|
|Outcomes||N. of events at 6 months(N. 421)||N. of events at 6 months in males(N. 186)||N. of events at 6 months in females(N. 235)||χ2|
|Yes||42 (10.0%)||24 (12.9%)||18 (7.7%)||0.075|
|No||379 (90.0%)||162 (87.1%)||217 (92.3%)|
|Yes||89 (21.1%)||45 (24.2%)||44 (18.7%)||0.17|
|No||332 (78.9%)||141 (75.8%)||191 (81.3%)|
|Yes||77 (18.3%)||38 (20.4%)||39 (16.6%)||0.31|
|No||344 (81.7%)||148 (79.6%)||196 (83.4%)|
|Yes||10 (2.4%)||1 (0.5%)||9 (3.8%)||*|
|No||411 (97.6%)||185 (99.5%)||226 (96.2%)|
|Variables||DeadMedian (IR)||AliveMedian(IR)||Mann Whitney||Returned to EDMedian(IR)||Not returned to EDMedian(IR)||Mann Whitney||HospitalizedMedian(IR)||Not HospedalizedMedian(IR)||Mann Whitney|
|Length of stay||226(147-295)||246(165.3-317.8)||0.54||232(144.75-293.5)||246(170.5-335)||0.12||232(140.8-283.3)||246.5(171.5-343.5)||0.046|
|InterRAI ED Screener||6(5-6)||3(1-6)||< 0.0001||3(3-6)||3(1-6)||0.35||5(3-6)||3(1-6)||0.043|
|Enter logistic regression|
|Variables||Mortality||ED readmission||Hospital admission|
|OR||95% CI||P||AUC||OR||95% CI||P||AUC||OR||95% CI||P||AUC|
|Length of stay||1.00||1.00-1.001||0.29||1.00||0.99-1.00||0.04||1.0||1.0-1.001||0.12|
|InterRAI ED Screener||1.15||0.81-1.64||0.43||1.05||0.84-1.32||0.69||1.16||0.92-1.47||0.21|
|Stepwise logistic regression|
|Variables||Mortality||ED readmission||Hospital admission|
|Length of stay||*||*||*||*||1.0||0.996-1.0||0.04||0.56||1.0||0.996-1.0||0.03||0.63|
- ISTAT (2020). Report Statistiche. Indicatori demografici anno. 2019.
- Salvi F, Morichi V, Grilli A. The elderly in the emergency department: a critical review of problems and solutions Intern Emerg Med. 2007; 2:292-301. DOI
- Beauchet O, Launay CP, Fantino B. Screening for elderly patients admitted to the emergency department requiring specialized geriatric care. J Emerg Med. 2013; 45:739-745. DOI
- Graf CE, Zekry D, Giannelli S. Efficiency and applicability of comprehensive geriatric assessment in the emergency department: a systematic review. Aging Clin Exp Res. 2010; 23:244-254. DOI
- Samaras N, Chevalley T, Samaras D. Older patients in the emergency department: a review. Ann Emerg Med. 2010; 56:261-269.
- I codici colore gravità (triage). 2013. Publisher Full Text
- McCusker J, Bellavance F, Cardin S. Detection of older people at increased risk of adverse health outcomes after an emergency visit: the ISAR screening tool. J Am Geriatr Soc. 1999; 47:1229-1237. DOI
- Meldon SW, Mion LC, Palmer RM. A brief risk-stratification tool to predict repeat emergency department visits and hospitalizations in older patients discharged from the emergency department. Acad Emerg Med. 2003; 10:224-232. DOI
- Aminzadeh F, Dalziel WB. Older adults in the emergency department: a systematic review of patterns of use, adverse outcomes, and effectiveness of interventions. Ann Emerg Med. 2002; 39:238-247. DOI
- Gray LC, Peel NM, Costa AP. Profiles of older patients in the emergency department: findings from the interRAI Multinational Emergency Department Study. Ann Emerg Med. 2013; 62:467-474. DOI
- Katz S, Downs TD, Cash HR. Progress in the development of the index of ADL. Gerontologist. 1970; 10:20-30. DOI
- Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist. 1969; 9:179-186. DOI
- Onder G, Bonassi S, Abbatecola AM, on behalf of the Geriatrics Working Group of the Italian Medicines Agency (AIFA). High prevalence of poor quality drug prescribing in older individuals: a nationwide report from the Italian Medicines Agency (AIFA). J Gerontol A Biol Sci Med Sci. 2014; 69:430-437. DOI
- Pua Y H, David B, Matchar DB. Physical performance predictor measures in older adults with falls-related emergency department visits. J Am Med Dir Assoc. 2019; 20:780-784. DOI
- Villani ER, Vetrano DL, Franza L. Physical performance measures and hospital outcomes among Italian older adults: results from the CRIME project. Aging Clin Exp Res. 2021; 33:319-327. DOI
- American College of Emergency Physicians, The American Geriatrics Society, Emergency Nurses Association, the Society for Academic Emergency Medicine. Geriatric Emergency Department Guidelines. Ann Emerg Med. 2014; 63:e7-25. DOI
- Salvi F, Morichi V, Lorenzetti B. Risk stratification of older patients in the emergency department: comparison between the identification of seniors at risk and triage risk screening tool. Rejuvenation Res. 2012; 15:288-294. DOI
- Salvi F, Morichi V, Grilli A. Predictive validity of the Identification of Seniors At Risk (ISAR) screening tool in elderly patients presenting to two Italian Emergency Departments. Aging Clin Exp Res. 2009; 21:69-75. DOI
- Slankamenac K, Haberkorn G, Meyer O. Prediction of emergency department re-visits in older patients by the Identification of Senior at Risk (ISAR) screening. Geriatrics. 2018; 3:33. DOI
- Suffoletto B, Miller T, Shah R. Predicting older adults who return to the hospital or die within 30 days of emergency department care using the ISAR tool: subjective versus objective risk factors. Emerg Med J. 2016; 33:4-9. DOI
- Tavares JPA, Sá-Couto P, Boltz M. Identification of Seniors at Risk (ISAR) in the emergency room: a prospective study. Int Emerg Nurs. 2017; 35:19-24. DOI
- Graf CE, Giannelli SV, Herrmann FR. Identification of older patients at risk of unplanned readmission after discharge from the emergency department. Swiss Med Wkly. 2012; 141:w13327. DOI
- Rizka A, Harimurti K, Pitoyo CW. Comparison between the identification of seniors at risk and triage risk screening tool in predicting mortality of older adults visiting the emergency department: results from Indonesia. Geriatr Gerontol Int. 2019; 20:47-51. DOI
- Taylor A, Broadbent M, Wallis M. The predictive validity of the interRAI ED screener for predicting re-presentation within 28 days for older adults at a regional hospital emergency department. Austr Emerg Care. 2019; 22:149-155. DOI
- Michalski-Monnerat C, Carron PN, Nguyen S. Assessing older patients’ vulnerability in the emergency department: a study of InterRAI ED screener accuracy. J Am Geriatr Soc. 2020; 68:2914-2920. DOI
- Hustey FM, Mion LC, Connor JT. A brief risk stratification tool to predict functional decline in older adults discharged from emergency departments. J Am Geriatr Soc. 2007; 55:1269-1274. DOI
- Fan J, Worster A, Fernandes CMB. Predictive validity of the triage risk screening tool for elderly patients in a Canadian emergency department. Am J Emerg Med. 2006; 24:540-544. DOI
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
© JOURNAL OF GERONTOLOGY AND GERIATRICS , 2021
How to Cite
- Abstract viewed - 511 times
- PDF downloaded - 165 times