1. Introduction
Global population ageing is placing a heavy demand on healthcare systems across the world, especially in hospital emergency departments (EDs), where attendances of older adults with complex frailty syndromes have increased [1]. Over the next 35 years (2015–2050), the number of older people is expected to triple, and there will be a greater proportion living longer with chronic conditions [2,3]. High prevalence rates for frailty, an age-linked state of vulnerability [4], increase the risk of adverse outcomes in this population [4,5]. Identifying frail older adults should be a priority in acute medicine, given the adverse impact frailty can have on health outcomes and the high healthcare utilisation costs that ensue [6].
Although older adults represent 25% of those presenting to the ED [7], those with frailty account for up to 60% of older attendees [8]. Frail older patients have higher ED conversion rates [9] that are associated with prolonged hospital admissions [10], a higher chance of readmission, and a higher rate of inpatient mortality [10,11]. This cohort also constitutes rising numbers of unscheduled care admissions [11] to the perceived main access portal to acute care, the ED [12]. The challenge is that many EDs are not in a position to meet the care needs of frail older adults [13].
Two established frailty assessment methods exist: the Fried Frailty (physical) Phenotype model [14] and the cumulative deficit approach [15]. Both frailty models can predict outcomes in older adults in ED [8], but both have challenges to use in “real life” clinical practice. The Fried Frailty Phenotype requires physical assessments and accruing details that are often not feasible in an urgent care or ED environment, and it defines frailty in ordinal terms [14]. A frailty index, based on the accumulated burden of deficits, can be complex, time-consuming, and difficult to operationalize and take action on in the ED [10]. For example, it is estimated that 30 variables are optimal to generate a frailty index with good predictive validity for adverse outcomes in ED [12]. In tandem with growing awareness of ageing populations, new frailty screening instruments have been created that show an association between older adults with frailty and adverse events [16]. Such instruments are based on deficit accumulation indices or multi-domain physical decline. Instruments validated for use in acute care include the Identification of Seniors at Risk Tool (ISAR) [17], the Clinical Frailty Scale (CFS) [18], and the FRAIL scale [19]. All differ in terms of features, expertise required, and time needed for application. Although most instruments rate better than chance in predicting adverse events, most perform either poorly or very poorly [20,21].
Current obstacles that prevent early recognition of frailty in ED consist of overly complicated frailty pathways and proformas, with assessments that do not occur in real time with the person [13]. In fact, frailty status in the ED is often established after a decision about a patient’s care has been made [13]. Elliott highlighted only four peer-reviewed studies that related to frailty risk categorization instruments in ED [22]. Low rates of completion were recorded for all instruments (52%), with slow rates of completion recorded (1–10 min) [22]. Although a multitude of frailty screening instruments exist, few have been compared with a “gold standard” reference test such as a comprehensive geriatric assessment (CGA) [23] in what are known as diagnostic test accuracy (DTA) studies. Most studies instead examine predictive validity for adverse outcomes such as future healthcare use and mortality; further reporting in such studies is not optimal and requires better standardisation [24]. This systematic review and meta-analysis therefore seeks to synthesise the existing evidence concerning the diagnostic accuracy of current frailty screening instruments used in EDs to identify frailty in older adults. Understanding this will facilitate recommendations on the best available short screen for use in ED and have implications for guidelines in this setting [25].
1.1. Study Design
A systematic review was undertaken to identify research publications that validated frailty screening instruments among older adults in the ED. It conformed to the recommendations laid out in the Cochrane Handbook for Systematic Reviews of Diagnostic Test Accuracy [26] and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA) standardised reporting recommendations [27]. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P) guidelines were adhered to in fulfilling study protocol requirements [28]. The steps taken to produce this systematic review included: (1) clearly stating the research question; (2) recording the protocol on the global register of systematic reviews (PROSPERO number CRD42020216780); (3) conducting database searches; (4) selecting suitable studies as per the inclusion criteria; (5) selecting data from included studies; and (6) amalgamating the results.
1.2. Eligibility Criteria
Studies were chosen based on population, index test, reference test, and diagnosis of interest (PIRD) criteria for diagnostic test accuracy reviews [29]. Details are illustrated in Table 1. The population included older adults aged 60 years and over screened in ED with any available frailty screening instrument (index test). Comparators (reference standards) included the “gold standard” CGA or deficit accumulation index (DAI) frailty model used to diagnose frailty (diagnosis of interest). Descriptive studies, observational studies, and randomised controlled studies were included where frailty screening instruments were utilised to screen older adults and measured against the described reference test to assess the diagnostic accuracy of the instrument.
1.3. Exclusion Criteria
Studies were excluded if the study population mean or median age was <60 years or where data could not be extracted separately on those aged ≥ 60 years. Access to full-text articles was required. Case reports or series, commentary pieces, opinion articles, conference abstracts, editorials, protocol submissions, and review articles were rejected; however, searches of the reference lists of relevant review papers for additional studies were included in the review. Authors of such papers, including abstracts, were contacted to inquire if such papers had been submitted and if these data could be obtained in advance of publication for potential inclusion. If a frailty screening instrument was used in a validation study of another instrument, it was excluded, and research articles that reported on frailty but omitted using or naming a specific instrument were also rejected. Studies where an instrument was used in a specific population or condition (e.g., chronic kidney disease, malignancy, heart failure) were also excluded.
1.4. Information Sources (Search Strategy)
The preliminary database query incorporated publications published from 1 January 2000 to 20 March 2021. These computer searches were updated on 30 November 2021 and 1 September 2022 in advance of submission. The search was limited to studies published after 2000, as prior to this, frailty was not a widely used clinical research term; for example, no publications including the terms “frailty” AND “emergency department” were found in a search of PubMed prior to 2004. The electronic databases searched were PubMed, Cinahl, Cochrane, Embase, Google Scholar, and TRIP. The complete search strategy is shown in Appendix A. Studies in all languages were included and translated where required. All searches were imported into the Endnote citation management tool, with duplicate citations eliminated.
1.5. Study Selection and Data Retrieval
Study titles and abstract content were independently reviewed based on the requirements for inclusion by two researchers (EM, DS). Studies meeting the required criteria were reviewed in full, and final eligibility for inclusion was assessed by the principal investigator (RO’C). Any disagreements were managed by consensus. Data were retrieved from accepted studies by two independent researchers (EM, DS) using standardised proformas to collect the following data, as advised by Campbell et al. (2015) [30]:
Administrative details: Author, title, type of publication, journal, country.
Study features: Design, aim(s), sample size, demographics, clinical location, results.
Components of the instrument(s) used to measure frailty: number of domains/items, what are the domains/items, whether it requires specialised equipment, mode of administration, administration time, details on the development of the tool, and measurement properties.
Which “gold standard” the instruments being studied were compared with (e.g., CGA or other frailty measures).
Prevalence of frailty measured.
Disagreement regarding data retrieval was resolved by consensus. If not directly reported, sensitivity, specificity, positive predictive value, and negative predictive value were calculated based on study findings or (in the case of insufficient data) on additional information obtained from the authors. Missing or incomplete data were dealt with by contacting the study author(s). Missing values were recorded in the data extraction form.
1.6. Risk of Bias
The methodological standards of the accepted publications were evaluated independently by two researchers (EM and RO’C) using the Quality Assessment of Diagnostic Accuracy Studies-Comparative (QUADAS-C) tool, an extension of the QUADAS 2 tool to assess the risk of bias and diagnostic accuracy in comparative studies [31]. Disagreements were resolved by a third reviewer (KM).
1.7. Analysis
The analysis was performed using R version 4.2.0. Forest plots of the sensitivity and specificity of each of the instruments were generated with a 95% confidence interval (CI) for each estimate. A bivariate random-effects model proposed by Reitsma et al. (2005) [32] was used to calculate pooled sensitivity and FPR (1-specificity) from the True Positive (TP), False Negative (FN), False Positive (FP), and True Negative (TN) of each instrument. Pooled prevalence was calculated from the data by taking the sum of (TP+FN)/(TP+FN+TN+FP). Youden’s Index [33]: positive likelihood ratio (PLR), negative likelihood ratio (NLR), positive predictive value (PPV), negative predictive value (NPV), and accuracy were also calculated using the bivariate random effects model. A summary receiver operating characteristic (ROC) curve (SROC) was graphically displayed for the bivariate random effects model, and the pooled area under the curve (AUC) of the SROC curve was calculated. A 95% CI was generated for this AUC estimate using a parametric bootstrap with 2000 resamples. Descriptive text was used where combined data pooling was not feasible (i.e., less than three data points were available).
2. Results
2.1. Search Results
Database searches returned 21,488 citations. Duplicate records were discarded, and extra hand search results added an additional three papers to give a total of 11,828 citations to the screen. Of these, 11,701 were removed based on title description and abstract content, leaving 127 citations for full text review. A detailed review of these 127 articles led to the rejection of 120 studies for the following reasons: ineligible design (n = 100), ineligible population (n = 15), or ineligible setting (n = 5). A PRISMA flow diagram illustrates the selection details in Figure 1. Accordingly, six studies [8,10,34,35,36,37,38] describing seven frailty screening instruments were included in this systematic review. Some papers included more than one of these instruments, yielding 13 unique data points. Additional details were requested from the authors of one relevant paper, initially found as an abstract publication [37], as the full-text article was unpublished at the time of the literature search (now published [38]). A table providing additional details of the search strategy utilised is included in Appendix A.
2.2. Quality of Methodology of Included Publications
Table 2 illustrates the quality of the included studies according to the QUADAS-C assessment instrument. Reviewer agreement was 90% on all aspects of the quality assessment. Regarding patient selection, the risk of bias was generally considered low, and there were no concerns about the applicability of the selection criteria. Three studies [8,34,35] did not specify whether the results of the frailty screening instrument (index test) and CGA (reference test) were independently scored, and so the risk of bias was unclear in those circumstances. However, the flow and timing of these studies raised few concerns.
2.3. Study Characteristics
All studies were prospective in design. Consecutive sampling was documented in the methods section of four papers [10,34,36,38]. The seven studies included 1663 participants. The mean sample size was 257 patients, ranging from 161 [34] to 498 [36]. The included studies were conducted in Ireland (n = 2) [10,37,38], Italy (n = 1) [8], France (n = 1) [35], Sweden (n = 1) [34], and Canada (n = 1) [36]. All studies were published in English. The majority of studies (n = 6) were published in the latter half of the 2010s [10,34,35,36,37,38]. The age as an inclusion criterion varied from ≥65 years [8] to ≥75 years [36]. The reported mean age of study participants was 79.8 years (range 76–86); 52% were female (range 45–60%). All six studies used CGA as a “gold standard” for measuring frailty. The pooled prevalence rate of frailty was high at 59%.
The instruments included different frailty domains and questions, varying in number from one to nine. Regarding the domains included, all seven instruments evaluated physical deficits such as “slowing up”, loss of activity levels, ability to mobilise, activities of daily living (ADLs), instrumental activities of daily living (IADLs), physical function, falls, communication (vision and hearing), hospitalisation, and excess or inappropriate prescribing (i.e., polypharmacy). Two instruments (the ISAR and BPQ) were screened for both physical and cognitive aspects (memory impairment), and three (the E-TRST, VIP, and PRISMA 7) were screened for both physical and social aspects (social support). A detailed description of all the included studies and frailty indicators is included in Appendix A.
Study authors were contacted regarding administration time if clarification was needed from the research article. Some of the instruments were ultra-short to administer (<2 min), such as the Bergman Paris Question (BPQ) [35] and Variable Indicator of Risk Placement (VIP) [10,37,38], while others were performed in more than one stage, such as the Triage Risk Screening Tool (TRST) [36], with its initial screen, the Emergency-TRST (E-TRST), usually taking less than 2 min [39]. From previous studies in ED, usual administration times for the ISAR vary from 2 to 5 min [40], while the FRESH instrument is recorded as taking ≤5 min [34]. The time for the CFS is more difficult to assess, as ideally, it should not be completed until a more detailed assessment has been completed; nevertheless, an ultra-short time of 24 s has been previously recorded in ED when the CFS is used as a stand-alone screen [41]. The documentation of administration times was variable. Only one study documented an administration time for the completion of all the instruments examined in their study. All the first authors of these studies were contacted regarding the administration times of the screens. Responses were received from three authors: Kajsa (2016) [34] for the FRESH instrument, Lague (2018) [35] for the BPQ instrument, and Salvi (2012) [8] for the ISAR instrument, who confirmed that no timings were taken during the studies and only estimates of the administration times for the screens in those studies could be provided. All three authors reported estimated administration times of between 2 and 5 min to complete these three screens. The highest AUC for diagnosing frailty was reported for the ISAR [8] at 0.92 (95% CI: 0.88–0.96) and the lowest for the BPQ [35] at 0.63 (95% CI: 0.53–0.72). Study characteristics are summarised in Table 3.
2.4. Meta-Analysis
Pooled Estimates for Frailty
In total, seven papers reported on the diagnostic accuracy of frailty screening instruments in ED [8,10,34,35,36,37,38]. Four of these reported results for more than one frailty screening instrument or version thereof, leading to a total of 13 datasets. The numbers TP, FP, FN, and TN were retrieved from the papers, and where unavailable, they were calculated from the sensitivity, specificity, and frailty prevalence values. The sensitivities and specificities for each of the 13 datasets are displayed in Table 4.
Examining the instruments individually in Table 4, the sensitivity values for detecting frailty based on CGA varied widely. The PRISMA-7 was reported as the most sensitive, with a sensitivity ranging between 0.84 and 0.98 but with a moderate specificity of 0.64–0.78. The CFS was the least sensitive instrument but most specific, suggesting it may need to be combined with another frailty instrument that is more sensitive for diagnosing frailty [10]. The ISAR showed high sensitivity, ranging between 0.94 and 0.95, but poor specificity in both studies (0.35–0.63). The geriatrician-assisted TRST (G-TRST) was more sensitive than the ED physician version (E-TRST) (0.93 versus 0.88). There was little difference between the FRESH 4-item and FRESH 5-item screens in terms of sensitivity (0.84 versus 0.81) or specificity (0.75 versus 0.80) [34]. The BPQ had high sensitivity (0.94) but very poor specificity (0.29) [35]. Given these findings, there were insufficient data available to conduct a meta-analysis of the individual instruments. Pooled estimates of diagnostic accuracy data for frailty are presented in Table 5.
The pooled estimate for sensitivity for detecting frailty for all instruments included in the meta-analysis was 0.85 (95% CI: 0.76–0.91), with a pooled estimate for specificity of 0.77 (95% CI: 0.62–0.88). The PPV was 0.85 (95% CI: 0.79–0.91), and the NPV was 0.77 (95% CI: 0.7–0.80). The pooled estimate for accuracy based on the AUC was 0.89 (95% CI: 0.86–0.90). The pooled PLR for frailty was 3.91 (95% CI: 2.41–6.23) with a NLR of 0.2 (95% CI: 0.14–0.27). Summary forest plots for sensitivity and specificity are displayed in Figure 2 and Figure 3, respectively. An SROC curve, a visual representation of the overall diagnostic accuracy of pooled data for the meta-analysis of all available datasets (n = 13), is displayed in Figure 4. The 13 triangles represent each data point, and the black line is the SROC curve.
3. Discussion
3.1. Overview
This systematic review and meta-analysis examined the diagnostic accuracy of available short frailty screening instruments as determined by an independently conducted assessment of frailty (i.e., DTA studies conducted blind to the scores of the screens). In all, only seven different frailty screening instruments across just six studies were found. Most were not bespoke frailty screens. The TRST, VIP, and ISAR were instead designed as risk-prediction instruments for use in acute care settings such as the ED with a view to predicting the risk of adverse outcomes rather than measuring frailty, which was not screened for when these instruments were first developed. Multiple reviews have shown that such individual instruments have reasonable-to-good predictive validity for future adverse health outcomes among older patients attending the ED, including mortality [42], re-admission [42,43], and prolonged length of hospital stay [44]. This is the first systematic review, to our knowledge, that confirms that they are actually accurate in identifying frail older people in ED who may require more detailed assessment and tailored management.
Based on the data obtained in the systematic review, diagnostic accuracy and properties such as sensitivity and specificity for frailty varied markedly between studies and individual instruments. In particular, the c-statistic (AUC) values for the instruments published varied from 0.63 for the BPQ instrument to 0.95 for the CFS instrument. However, a pooled meta-analysis showed that overall, these instruments had good to excellent diagnostic accuracy with high levels of sensitivity and specificity for frailty, as measured against the CGA, usually regarded as the “gold standard” for measuring frailty [45]. The pooled estimate for PPV for all instruments was also good (85%), confirming that these are useful as screening tests with a high likelihood that they are detecting true frailty. A high PPV is an important characteristic of a screening test [46]. The overall pooled diagnostic accuracy measured using AUC was 0.89, indicating good accuracy. There was insufficient data available to conduct a meta-analysis of the individual instruments. Hence, it is difficult to recommend any single instrument based on this. Based on the data found in the review, the PRISMA-7 had the highest sensitivity and the CFS the highest specificity.
The PRISMA-7, although originally advised by the British Geriatrics Society as the instrument to opportunistically screen older people for frailty in different settings [47] is less studied in ED, though a dedicated systematic review is planned to clarify further [48], The CFS is one of the most widely used instruments, with growing evidence for predictive validity for adverse outcomes in large samples [42,43,49]. However, a recent scoping review of the use of the CFS showed that reporting is not optimal and more standardisation of results is needed to better interpret its utility as a screen for frailty in ED [24]. In addition, its administration time is ill-defined, particularly given that many suggest that the CFS should only be conducted after a CGA when information is gathered to stratify a patient adequately [50]. Of the other instruments, the diagnostic accuracy of the ISAR was only studied in two papers [8,10]; in these studies, it showed high sensitivity but only low-moderate specificity. The ISAR is one of the most widely utilised screening instruments to highlight older persons at risk of functional deterioration in ED, but Galvin et al. (2017) [17] have previously recommended further evaluation of the ISAR rule owing to variations in patient and process outcomes in a systematic review evaluating adverse outcomes of older adults attending ED [17]. It is used to identify frailty but is less accurate than other instruments. Only one study examined the VIP, comparing 3 and 4-item versions [37,38]. The VIP 4-item version had low sensitivity, high specificity, and low false positives [37,38]. Although it is a quick and easy-to-use tool that does not require much training, it predominantly identifies people who are more severely frail. Further evaluation of the validity and reliability of the VIP as a frailty identification tool is recommended [51].
3.2. Results in Context
The diagnostic accuracy of frailty screening instruments in older adults has previously been reported in an umbrella review by Apóstolo et al., (2017) [43]. Using baseline results from the Cardiovascular Health Study (CHS) phenotype model, the Canadian Study of Health and Ageing (CSHA) DAI model, and the frailty index based on a CGA (FI-CGA), short frailty assessments in all settings were examined [43]. The PRISMA-7 was the only screening instrument from our study to be included in this umbrella review, which confirmed high sensitivity and specificity to identify frailty in adults over 65 years. However, no information relating to DTA studies in ED was reported. Apóstolo did note that only a few frailty instruments reviewed seemed to be diagnostically accurate, but there was clear usefulness to using simple risk indicators such as gait speed [43]. Our study differs from existing publications in that it examines DTA studies in the ED setting, a clinical area that lacks previous systematic review publications. To illustrate this, one only needs to review the two most up-to-date frailty evaluation study protocols, which omit the ED as a study location [48,52]. Our study fills an important data vacuum in the ED environment, aligning with EUGMS geriatric emergency medicine service recommendations to consider the use of frailty screening instruments in the ED [6]. The COVID-19 pandemic has shown that this is timely by highlighting the importance of early recognition of frailty as a way to improve the triage of older patients and identify those most likely to benefit from critical and intensive care [53]. The difficulty of staff without geriatric medicine training using frailty screening instruments has also been shown to be challenging [54]. Inaccurate and poorly understood frailty assessments can lead to inadvertent changes to treatment access, while decisions based solely on chronological age or current presentation have concerned older adult advocates worldwide [53,55]. Without clear guidelines on the most appropriate frailty screening tool to use in the ED environment and subsequent training of ED staff, the integration of routine frailty screening into Geriatric Emergency Medicine protocols will be hindered [56].
Lengthy administration time is often identified as a key barrier to the use of frailty screening tools by ED staff, with some tools requiring 30–40 min to complete. Administration times were not clearly documented in any of the studies included in this systematic review. These data were acquired by contacting the study authors, who could only give estimates of completion times for three of the screens. While it was previously reported in another systematic review that similar frailty screening tools have administration times of less than five minutes [22], this would appear to relate predominantly to the ISAR tool. In order to integrate frailty screening into routine ED practice, it is important that the screening process be shown to be feasible so as to be implemented in a busy clinical environment, and hence the rapid application times of frailty screening tools should be clearly highlighted. Thus, we recommend that future studies include mean/median administration times to help clinicians decide which is the optimal tool for their own practice.
3.3. Strengths and Weaknesses
Positive strengths of this work include the comprehensive search strategy that followed Cochrane procedures with no limitation on language, as well as the manual search of references in the included studies. In addition to contacting authors, the research team was able to calculate some missing values based on the data provided. The UN older person classification of older adults as those aged ≥ 60 years was used to avoid limiting the external validity of the studies [57]. As far as we are aware, this is the first review to combine the accrued evidence on the diagnostic accuracy of current frailty screening instruments used in EDs to identify frailty in older adults.
Despite attention being paid by the research team, a risk remains that all appropriate studies may not have been included in maintaining methodological rigour. For example, since the search was completed, the interRAI Emergency Department Screener has been validated against the interRAI Emergency Department Assessment System to identify those experiencing complex health and social concerns and was found to have a mean administration time of <2 min and good sensitivity and specificity for this outcome [58]. While that study did not include frailty as a specified outcome, it identified those in need of CGA, which could be taken as a surrogate for frailty. From the six studies, including seven index tests, four had relatively small numbers of participants and all represented patients recruited from high-income western countries (Europe and Canada) [8,34,35,37,38], which could limit the generalizability of the results. The pooled prevalence of frailty across the studies included was 59%, which is relatively high and suggests that these studies were at risk of spectrum bias. The data were markedly heterogeneous, and the small sample size of studies available for most instruments limited the ability to perform a meta-analysis. Instruments that did not have sufficient numbers to enable this may be more accurate, but this review could not confirm this, suggesting the need for additional studies. Finally, as the search was widened to include additional databases (e.g., TRIP) and was updated on two occasions due to delays with the submission, there is discordance with the published PROSPERO protocol.
3.4. Clinical Implications and Future Research
While there may be a multitude of frailty screening tools currently available, this systematic review and meta-analysis has shown that utilising a two-step assessment process with an index diagnostic screening test and gold standard reference test model allows accurate frailty diagnosis in the ED. It is argued that this may not be feasible in routine practice, but our results reinforce the view of Heeren et al. that incorporating more objectively validated measures into future vulnerability studies may improve instrument accuracy [59]. This would also align with the theory that if risk stratification for common geriatric syndromes is embedded at the beginning of the ED process, and can be part of classic nurse triage or doctor assessment at triage, which has shown improved ED performance [60]. From this research, it is clear that the TRST, VIP, and ISAR were not originally designed as frailty screening instruments but as risk-prediction instruments, with frailty screening retrospectively considered as an additional measurement component.
In a clinical context, this suggests that efforts should be made to improve the DTA of validated short frailty screening tools using this two-stage assessment process, where feasible in ED environments, and apply frailty screening in regions of high frailty prevalence, such as sub-Saharan Africa. Scant attention has been paid to ageing populations in low- and middle-income countries, which is concerning given the rapid population growth of older adults in these regions, where life expectancy has also improved [61]. Accurate frailty identification in EDs is a rapid mechanism that selects older adults for adapted trajectories to deliver the “right care to the right person at the right time and place” [62]. Such frailty data should assist public health strategic planning in areas under increasing strain, as highlighted by the WHO Study on Global Ageing and Adult Health Report [63].
4. Conclusions
Overall, although a limited number of heterogenous studies and instruments were available for review, the result of this pooled DTA meta-analysis suggests that the available short screens for frailty had good sensitivity, reasonable specificity, and good to excellent accuracy as measured against CGA. The PRISMA-7 was the most sensitive instrument for identifying frailty, while the CFS had the highest specificity, albeit with insufficient data to conduct a meta-analysis of individual instruments. Further high-quality studies utilising an independent gold standard such as the CGA are required before a specific instrument could be recommended. Despite the relative paucity of studies included, no major methodological concerns were noted. Our work demonstrates that while the use of these short frailty screens in ED appears to be accurate in diagnosing frailty, more research is now required comparing additional instruments in large samples and in different populations and settings to properly examine the psychometric performance of individual instruments among older adults attending ED.
E.M.: conceptualization, methodology, data curation, formal analysis, writing—original draft preparation, review, and editing; M.R.O.: data curation, software, formal analysis, writing—review and editing; D.S.: data curation, writing—review and editing; K.M. and E.F.: writing—review and editing; S.T.: supervision, writing—review and editing; R.O.: conceptualization, methodology, formal analysis, writing—review and editing, supervision. All authors have read and agreed to the published version of the manuscript.
Not applicable as this is a systematic review of existing studies and no participants were included.
Not applicable as this is a systematic review of existing studies and no individuals were included.
On request from the authors.
This research received no specific funding. This work was conducted as part of an MD submission to University College Cork, Ireland.
The authors declare no conflict of interest.
Footnotes
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Figure 1. PRISMA Flow Diagram illustrating selection process for included studies.
Figure 4. SROC curve for the meta-analysis of all available data (n = 13 data points). The summary estimate is the pooled sensitivity/specificity estimate from the bivariate random effects model and the conf. region is a 95% confidence region for this summary estimate.
Question layout for this systematic review using the population, index test, reference test, and diagnosis of interest (PIRD) system for diagnostic accuracy studies.
PIRD Question Layout | |||
---|---|---|---|
Population | Index Test | Reference Test | Diagnosis of Interest |
Persons aged ≥ 60 years presenting to ED utilising an established definition of frailty | Short screens and risk stratifying instruments utilised to highlight older adults with frailty | CGA or components of existing frailty models in the context of independent review | Accurate frailty diagnosis and prediction of defined adverse events |
Quality assessment of included studies (n = 6) in the review using the Quality Assessment of Diagnostic Accuracy Studies-Comparative (QUADAS-C).
Study | Risk of Bias | Applicability Concerns | |||||
---|---|---|---|---|---|---|---|
Patient Selection | Index Test | Reference Standard | Flow and Timing | Patient Selection | Index Test | Reference Standard | |
Salvi et al., 2012 [ |
Low | Low | Unclear | Low | Low | Low | Low |
Kajsa et al., 2016 [ |
Low | Unclear | Unclear | Low | Low | Low | Low |
Lague et al., 2018 [ |
Unclear | Unclear | Low | Low | Low | Low | Low |
O’Caoimh et al., 2019 [ |
Low | Low | Low | Low | Low | Low | Low |
McGuaran et al., 2020 [ |
Low | Low | Unclear | Low | Low | Low | Low |
Piffer et al., 2020 [ |
Low | Unclear | Unclear | Low | Low | Low | Low |
Features of included studies in the systematic review of diagnostic accuracy tests for frailty in the emergency department.
Study |
N = X | Mean |
% F | Inclusion |
Study |
Frailty Screens Used with Cut-Offs | Reference Standard | Time to Complete | AUC for Each Screen |
---|---|---|---|---|---|---|---|---|---|
Salvi et al., 2012 [ |
200 | 80 | 57% | ≥65 years |
Prospective observation cohort, |
ISAR ≥ 2 | CGA | ND | ISAR = 0.92 |
Kajsa et al., 2016 [ |
161 | 82 | 55% | ≥80 years or 65–79 with ≥1 chronic |
Cross-sectional, consecutive sampling | FRESH ≥ 2 | CGA | ND | FRESH = 0.86 |
Lague et al., 2018 [ |
171 | 76.9 | 52% | ≥65 yrs |
Prospective observational cohort, |
BPQ |
CGA | ND | BPQ = 0.63 |
O’Caoimh et al. 2019 [ |
265 | 78 | 54% | ≥70 years |
Prospective cross-sectional, consecutive sampling | CFS ≥ 5 |
CGA | ND | CFS = 0.83 |
McGuaran et al. 2020 [ |
197 | 79 | 45% | ≥70 years |
Prospective cross-sectional, consecutive sampling | VIP-3 ≥ 2 |
CGA | ND | CFS = 0.91 |
Piffer et al., 2020 [ |
498 | 86 | 60% | ≥75-year-olds who |
Prospective, cross-sectional, consecutive sampling | TRST ≥ 2 | CGA | ND | E-TRST = 0.86 (0.83–0.89) |
ADL: Activities of Daily Living; AUC: Area Under the Curve; BPQ: Bergman-Paris Question; CGA: Comprehensive Geriatric Assessment; CFS: Clinical Frailty Scale; ED: Emergency Department; F: Female; ISAR: Identification of Seniors At Risk; PRISMA-7: Programme of Research on Integration of Services for the Maintenance of Autonomy; VIP: Variable Indicator of Placement Risk; E-TRST: Emergency Physician administered TRST; G-TRST: Geriatrician administered TRST; ND = not documented.
Performance of the eight screening instruments with cut-offs for identifying frailty including sensitivity (sens), specificity (spec).
Author | Year | Tool | TP | FN | FP | TN | N | Sen | Spec | Wt-Sen | Wt-Spec |
---|---|---|---|---|---|---|---|---|---|---|---|
O’Caoimh et al., | 2019 [ |
CFS ≥ 5 | 67 | 64 | 5 | 84 | 220 | 0.51 | 0.94 | 5.551 | 6.218 |
McGauran et al., | 2020 [ |
CFS ≥ 5 | 75 | 22 | 13 | 86 | 196 | 0.77 | 0.87 | 5.88 | 6.131 |
Salvi et al., | 2012 [ |
ISAR ≥ 2 | 110 | 7 | 31 | 52 | 200 | 0.94 | 0.63 | 6.075 | 5.457 |
O’Caoimh et al., | 2019 [ |
ISAR ≥ 2 | 146 | 8 | 72 | 39 | 265 | 0.95 | 0.35 | 6.105 | 5.342 |
O’Caoimh et al., | 2019 [ |
PRISMA-7 ≥ 3 | 130 | 24 | 24 | 87 | 265 | 0.84 | 0.78 | 6.065 | 6.061 |
McGauran et al., | 2020 [ |
PRISMA-7 ≥ 3 | 95 | 2 | 36 | 63 | 196 | 0.98 | 0.64 | 6.051 | 5.423 |
McGauran et al., | 2020 [ |
VIP-3 item ≥ 2 | 57 | 40 | 2 | 97 | 196 | 0.59 | 0.98 | 5.154 | 6.24 |
McGauran et al., | 2020 [ |
VIP-4 item. ≥ 3 | 61 | 36 | 2 | 97 | 196 | 0.63 | 0.98 | 5.263 | 6.23 |
Piffer et al., | 2020 [ |
G-TRST ≥ 2 | 345 | 26 | 41 | 86 | 498 | 0.93 | 0.68 | 6.16 | 5.822 |
Piffer et al., | 2020 [ |
E-TRST ≥ 2 | 326 | 45 | 44 | 83 | 498 | 0.88 | 0.65 | 6.16 | 6.025 |
Kajsa et al., | 2016 [ |
FRESH 4 item | 98 | 19 | 11 | 33 | 161 | 0.84 | 0.75 | 6.03 | 5.486 |
Kajsa et al., | 2016 [ |
FRESH 5 item | 95 | 22 | 9 | 35 | 161 | 0.81 | 0.80 | 6.001 | 5.591 |
Lague et al., | 2018 [ |
BPQ “No” | 32 | 2 | 97 | 40 | 171 | 0.94 | 0.29 | 5.741 | 5.339 |
BPQ: Bergman-Paris Question; CFS: Clinical Frailty Scale; ISAR: Identification of Seniors At Risk; PRISMA-7: Programme of Research to Integrate the Services for the Maintenance of Autonomy 7; VIP: Variable Indicator of Placement; TRST: Triage Risk Screening Tool; E-TRST: Emergency Physician TRST; G-TRST: Geriatrician TRST. N = number; TP = True Positive; TN = True Negative; FP = False Positive; TN = True Negative; Sen = Sensitivity; Spec = Specificity; Wt-Sen = Weighted Sensitivity; Wt-Spec = Weighted Specificity
Pooled diagnostic accuracy estimates with 95% confidence intervals (CI) calculated using a bivariate random-effects model [
Pooled Estimates |
AUC |
FPR |
FNR |
Sensitivity |
Specificity |
Youden’s |
Accuracy |
PPV |
NPV |
PLR |
NLR |
---|---|---|---|---|---|---|---|---|---|---|---|
All available data points |
0.89 |
0.23 |
0.15 |
0.85 |
0.77 |
0.62 |
0.82 |
0.85 |
0.77 |
3.91 |
0.20 |
AUC: Area Under the Curve; FPR: False Positive Ratio; FNR: False Negative Ratio; PPV: Positive Predictive Value; NPV: Negative Predictive Value; PLR: Positive Likelihood Ratio; NLR: Negative Likelihood Ratio.
Appendix A
Description of Included Studies
The first study by Salvi et al. (2012) [
The second study by Kajsa et al. (2016) [
The third study by Lague et al. (2018) [
Caregivers who indicated they would NOT be happy to leave their relative alone constituted a “positive BPQ,”, which pointed to a potential frailty syndrome. If the caregiver indicated they would be happy to leave their relative alone, this constituted a “negative BPQ”, which pointed to that relative not requiring specialist geriatric medicine input. Cognition was evaluated using the Telephone Interview for Cognitive Status-modified (TICS-m), functional ability via the Older American Resources and Services scale (OARS), frailty status was assessed using the Clinical Frailty Scale, and the Confusion Assessment Method (CAM) was utilised to rule out delirium. Data on sociodemographic profile, medication use, co-morbidity burden, and psychological status were also collected in the ED.
The fourth study by O’Caoimh et al. (2019) [
The fifth study by Piffer et al. (2020) [
Positive responses to the first four questions elicited a score of 1. Each basic ADL score was divided into partially dependent (0.5 score) or fully dependent (1 score). Total scores range 0–5. TRST scores of ≥2 indicate frailty. SEGA evaluated thirteen variables: age, habitation, medication use, mood, self-reported standard of health, recent falls, diet, gait and balance, cognition, instrumental ADLs, incontinence, and meals. Scores ranged 0–2. The maximum SEGA tally is 26. In this study, SEGA > 8 indicated frailty.
The sixth study by McGuaran et al. (2020) [
Table search strategy with number of citations according to each database.
Search |
PubMed | Cinahl | Cochrane | Embase | Google |
Trip | Total |
---|---|---|---|---|---|---|---|
Search #1 |
18,514 | 8857 | 3967 | 24,609 | N/A | 11,797 | 43,135 |
Search #2 |
949,076 | 376,975 | 1,529,902 | 1,400,342 | N/A | 1,801,459 | 6,057,754 |
Search #3 |
929,624 | 305,742 | 309,120 | 1,272,570 | N/A | 96,115 | 2,913,171 |
#1 AND #2 | 10,199 | 3008 | 3967 | 18,819 | N/A | 60,259 | 79,322 |
#1 AND #3 | 8056 | 3008 | 1488 | 13,605 | N/A | 1427 | 27,584 |
#1 AND # 2
|
4972 | 3008 | 1480 | 10,684 | 63 | 1281 | 21,488 |
Total citation count | 21,488 |
* refers to truncation, a symbol added to the end of the root of a word to instruct the database to search for all forms of a word. “” refers to quotation marks, a symbol that instructs the database to search for an exact phrase.
Summary of frailty indicators measured in the included studies.
Tool | Reference | Self-Report of |
Function | Weight Loss | Mobility | Falls | Vision/ |
Cognition | Nutrition | Social Support | Medication | Recent Hospitalization |
---|---|---|---|---|---|---|---|---|---|---|---|---|
CFS | McGauaran et al., 2020 [ |
+ | + | - | + | - | - | + | - | - | - | - |
ISAR | O’Caoimh et al., 2019 [ |
- | + | - | - | - | + | + | - | - | + | + |
TRST | Piffer et al., 2020 [ |
- | + | - | + | + | - | + | - | - | + | + |
PRISMA-7 | McGauran et al., 2020 [ |
- | + | - | + | - | - | - | - | + | - | - |
FRESH
|
Kajsa et al., 2016 [ |
+ | + | - | - | + | - | - | - | - | - | - |
VIP | McGauran et al., 2020 [ |
- | + | - | - | - | - | - | - | + | - | - |
Note: (-) refers to absence of frailty indicator; (+) refers to presence of frailty indicator.
References
1. Lowthian, J.; Curtis, A.; Stoelwinder, J.; McNeil, J.; Cameron, P. Emergency demand and repeat attendances by older patients. Intern. Med. J.; 2013; 43, pp. 554-560. [DOI: https://dx.doi.org/10.1111/imj.12061]
2. Wan, H.; Goodkind, D.; Kowal, P. An Aging World: 2015; United States Census Bureau: Washington, DC, USA. Available online: https://www.census.gov/content/dam/Census/library/publications/2016/demo/p95-16-1.pdf (accessed on 21 March 2023).
3. Beard, J.R.; Officer, A.; De Carvalho, I.A.; Sadana, R.; Pot, A.M.; Michel, J.P.; Lloyd-Sherlock, P.; Epping-Jordan, J.E.; Peeters, G.G.; Mahanani, W.R. et al. The World report on ageing and health: A policy framework for healthy ageing. Lancet; 2016; 387, pp. 2145-2154. [DOI: https://dx.doi.org/10.1016/S0140-6736(15)00516-4] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26520231]
4. Clegg, A.; Young, J.; Iliffe, S.; Rikkert, M.O.; Rockwood, K. Frailty in elderly people. Lancet; 2013; 381, pp. 752-762. [DOI: https://dx.doi.org/10.1016/S0140-6736(12)62167-9]
5. O’caoimh, R.; Sezgin, D.; O’donovan, M.R.; Molloy, D.W.; Clegg, A.; Rockwood, K.; Liew, A. Prevalence of frailty in 62 countries across the world: A systematic review and meta-analysis of population-level studies. Age Ageing; 2021; 50, pp. 96-104. [DOI: https://dx.doi.org/10.1093/ageing/afaa219]
6. Lucke, J.A.; Mooijaart, S.P.; Heeren, P.; Singler, K.; McNamara, R.; Gilbert, T.; Nickel, C.H.; Castejon, S.; Mitchell, A.; Mezera, V. et al. Providing care for older adults in the Emergency Department: Expert clinical recommendations from the European Task Force on Geriatric Emergency Medicine. Eur. Geriatr. Med.; 2022; 13, pp. 309-317. [DOI: https://dx.doi.org/10.1007/s41999-021-00578-1]
7. Leahy, A.; McNamara, R.; Reddin, C.; Corey, G.; Carroll, I. SOLAR team O’Neill, A.; Flannery, D.; Devlin, C.; Barry, L. et al. The impact of frailty Screening of Older adults with muLtidisciplinary assessment of those At-Risk during emergency hospital attendance on the quality, safety and cost-effectiveness of care (SOLAR): A randomised controlled trial. Trials; 2021; 22, 581. [DOI: https://dx.doi.org/10.1186/s13063-021-05525-w] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34465368]
8. Salvi, F.; Morichi, V.; Grilli, A.; Lancioni, L.; Spazzafumo, L.; Polonara, S.; Abbatecola, A.M.; De Tommaso, G.; Dessi-Fulgheri, P.; Lattanzio, F. Screening for frailty in elderly emergency department patients by using the Identification of Seniors At Risk (ISAR). J. Nutr. Health Aging; 2012; 16, pp. 313-318. [DOI: https://dx.doi.org/10.1007/s12603-011-0155-9]
9. Conroy, S.; Turpin, S. New horizons: Urgent care for older people with frailty. Age Ageing; 2016; 45, pp. 579-586. [DOI: https://dx.doi.org/10.1093/ageing/afw135]
10. O’caoimh, R.; Costello, M.; Small, C.; Spooner, L.; Flannery, A.; O’reilly, L.; Heffernan, L.; Mannion, E.; Maughan, A.; Joyce, A. et al. Comparison of frailty screening instruments in the emergency department. Int. J. Environ. Res. Public Health; 2019; 16, 3626. [DOI: https://dx.doi.org/10.3390/ijerph16193626]
11. Ge, L.; Yap, C.W.; Heng, B.H.; Tan, W.S. Frailty and healthcare utilisation across care settings among community-dwelling older adults in Singapore. BMC Geriatr.; 2020; 20, 389. [DOI: https://dx.doi.org/10.1186/s12877-020-01800-8] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33023490]
12. Samaras, N.; Chevalley, T.; Samaras, D.; Gold, G. Older Patients in the Emergency Department: A Review. Ann. Emerg. Med.; 2010; 56, pp. 261-269. [DOI: https://dx.doi.org/10.1016/j.annemergmed.2010.04.015]
13. Latham, L.P.; Ackroyd-Stolarz, S. Emergency department utilization by older adults: A descriptive study. Can. Geriatr. J.; 2014; 17, pp. 118-125. [DOI: https://dx.doi.org/10.5770/cgj.17.108] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25452824]
14. Fried, L.P.; Tangen, C.M.; Walston, J.; Newman, A.B.; Hirsch, C.; Gottdiener, J.; Seeman, T.; Tracy, R.; Kop, W.J.; Burke, G. et al. Cardiovascular Health Study Collaborative Research Group. Frailty in older adults: Evidence for a phenotype. J. Gerontol. Ser. A: Biol. Sci. Med. Sci.; 2001; 56, pp. M146-M157. [DOI: https://dx.doi.org/10.1093/gerona/56.3.M146] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/11253156]
15. Rockwood, K.; Song, X.; MacKnight, C.; Bergman, H.; Hogan, D.B.; McDowell, I.; Mitnitski, A. A global clinical measure of fitness and frailty in elderly people. CMAJ; 2005; 173, pp. 489-495. [DOI: https://dx.doi.org/10.1503/cmaj.050051] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/16129869]
16. Morrison, L.; Costello, M.; Flannery, A.; Small, C.; Nevin, R.; Havelin, A.; O’reilly, L.; Heffernan, L.; Hong, Y.Y.; Waters, R. et al. 279Frailty in an acute hospital: Point prevalence and change in baseline status during hospitalisation. Age Ageing; 2017; 46, pp. 11-12. [DOI: https://dx.doi.org/10.1093/ageing/afx145.51]
17. Galvin, R.; Gilleit, Y.; Wallace, E.; Cousins, G.; Bolmer, M.; Rainer, T.; Smith, S.M.; Fahey, T. Adverse outcomes in older adults attending emergency departments: A systematic review and meta-analysis of the Identification of Seniors At Risk (ISAR) screening tool. Age Ageing; 2017; 46, pp. 179-186. [DOI: https://dx.doi.org/10.1093/ageing/afw233]
18. Stille, K.; Temmel, N.; Hepp, J.; Herget-Rosenthal, S. Validation of the Clinical Frailty Scale for retrospective use in acute care. Eur. Geriatr. Med.; 2020; 11, pp. 1009-1015. [DOI: https://dx.doi.org/10.1007/s41999-020-00370-7]
19. Morley, J.E.; Malmstrom, T.K.; Miller, D.K. A simple frailty questionnaire (FRAIL) predicts outcomes in middle aged African Americans. J. Nutr. Health Aging; 2012; 16, pp. 601-608. [DOI: https://dx.doi.org/10.1007/s12603-012-0084-2]
20. Lewis, E.T.; Dent, E.; Alkhouri, H.; Kellett, J.; Williamson, M.; Asha, S.; Holdgate, A.; Mackenzie, J.; Winoto, L.; Fajardo-Pulido, D. et al. Which frailty scale for patients admitted via Emergency Department? A cohort study. Arch. Gerontol. Geriatr.; 2019; 80, pp. 104-114. [DOI: https://dx.doi.org/10.1016/j.archger.2018.11.002]
21. Wallis, S.; Wall, J.; Biram, R.; Romero-Ortuno, R. Association of the clinical frailty scale with hospital outcomes. Qjm Int. J. Med.; 2015; 108, pp. 943-949. [DOI: https://dx.doi.org/10.1093/qjmed/hcv066]
22. Elliott, A.; Hull, L.; Conroy, S.P. Frailty identification in the emergency department—A systematic review focussing on feasibility. Age Ageing; 2017; 46, pp. 509-513. [DOI: https://dx.doi.org/10.1093/ageing/afx019]
23. Clegg, A.; Rogers, L.; Young, J. Diagnostic test accuracy of simple instruments for identifying frailty in community-dwelling older people: A systematic review. Age Ageing; 2015; 44, pp. 148-152. [DOI: https://dx.doi.org/10.1093/ageing/afu157]
24. Fehlmann, C.A.; Nickel, C.H.; Cino, E.; Al-Najjar, Z.; Langlois, N.; Eagles, D. Frailty assessment in emergency medicine using the Clinical Frailty Scale: A scoping review. Intern. Emerg. Med.; 2022; 17, pp. 2407-2418. [DOI: https://dx.doi.org/10.1007/s11739-022-03042-5]
25. Moloney, E.; Sezgin, D.; O’donovan, M.; Adja, K.Y.C.; McGrath, K.; Liew, A.; Lenzi, J.; Gori, D.; O’connor, K.; Molloy, D.W. et al. The Diagnostic Accuracy and Clinimetric Properties of Screening Instruments to Identify Frail Older Adults Attending Emergency Departments: A Protocol for a Mixed Methods Systematic Review and Meta-Analysis. Int. J. Environ. Res. Public Health; 2022; 19, 1380. [DOI: https://dx.doi.org/10.3390/ijerph19031380]
26. Higgins, J. Cochrane Handbook for Systematic Reviews of Interventions; 2nd ed. Wiley-Blackwell: Hoboken, NJ, USA, 2019.
27. Page, M.J.; McKenzie, J.E.; Bossuyt, P.M.; Boutron, I.; Hoffmann, T.C.; Mulrow, C.D.; Shamseer, L.; Tetzlaff, J.M.; Akl, E.A.; Brennan, S.E. et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. Syst. Rev.; 2021; 89, 105906.
28. PRISMA-P Group Moher, D.; Shamseer, L.; Clarke, M.; Ghersi, D.; Liberati, A.; Petticrew, M.; Shekelle, P.; Stewart, L.A. Preferred reporting items for systematic review and meta-analysis protocols (prisma-p) 2015 statement. Syst. Rev.; 2015; 4, 1. [DOI: https://dx.doi.org/10.1186/2046-4053-4-1]
29. Munn, Z.; Stern, C.; Aromataris, E.; Lockwood, C.; Jordan, Z. What kind of systematic review should I conduct? A proposed typology and guidance for systematic reviewers in the medical and health sciences. BMC Med. Res. Methodol.; 2018; 18, 5. [DOI: https://dx.doi.org/10.1186/s12874-017-0468-4]
30. Campbell, J.M.; Klugar, M.; Ding, S.; Carmody, D.P.; Hakonsen, S.J.M.; Jadotte, Y.T.; White, S.; Munn, Z. Diagnostic test accuracy: Methods for systematic review and meta-analysis. Int. J. Evidence-Based Health; 2015; 13, pp. 154-162. [DOI: https://dx.doi.org/10.1097/XEB.0000000000000061]
31. Yang, B.; Mallett, S.; Takwoingi, Y.; Davenport, C.F.; Hyde, C.J.; Whiting, P.F.; Deeks, J.J.; Leeflang, M.M. The QUADAS-C Group† QUADAS-C: A Tool for Assessing Risk of Bias in Comparative Diagnostic Accuracy Studies. Ann. Intern. Med.; 2021; 174, pp. 1592-1599. [DOI: https://dx.doi.org/10.7326/M21-2234]
32. Reitsma, J.B.; Glas, A.S.; Rutjes, A.W.; Scholten, R.J.; Bossuyt, P.M.; Zwinderman, A.H. Bivariate analysis of sensitivity and specificity produces informative summary measures in diagnostic reviews. J. Clin. Epidemiology; 2005; 58, pp. 982-990. [DOI: https://dx.doi.org/10.1016/j.jclinepi.2005.02.022]
33. Youden, W.J. Index for rating diagnostic tests. Cancer; 1950; 3, pp. 32-35. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/15405679][DOI: https://dx.doi.org/10.1002/1097-0142(1950)3:1<32::AID-CNCR2820030106>3.0.CO;2-3]
34. Kajsa, E.; Katarina, W.; Sten, L.; Synneve, I.-D. Screening for frailty among older emergency department visitors: Validation of the new FRESH-screening instrument. BMC Emerg. Med.; 2016; 16, pp. 1-7. [DOI: https://dx.doi.org/10.1186/s12873-016-0087-0] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27449526]
35. Laguë, A.; Voyer, P.; Ouellet, M.-C.; Boucher, V.; Giroux, M.; Pelletier, M.; Gouin, É.; Daoust, R.; Berthelot, S.; Morin, M. Using the Bergman-Paris Question to screen seniors in the emergency department. Can. J. Emerg. Med.; 2018; 20, pp. 753-761. [DOI: https://dx.doi.org/10.1017/cem.2017.412] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29032788]
36. Piffer, I.; Goetz, C.; Zevering, Y.; Andre, E.; Bourouis, Z.; Blettner, N. Ability of Emergency Department Physicians Using a Functional Autonomy-Assessing Version of the Triage Risk Screening Tool to Detect Frail Older Patients Who Require Mobile Geriatric Team Consultation. J. Nutr. Health Aging; 2020; 24, pp. 634-641. [DOI: https://dx.doi.org/10.1007/s12603-020-1378-4] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32510117]
37. McGuaran, J.; Alcock, M.; O’Donovan, M.; Moloney, E.; Walsh, C.; O’Hea, A.; Hayes, M.; O’Connor, K.; O’Caoimh, R. Comparison of the variable indicative of placement risk to short frailty screens in the emergency department. Eur. Geriatr. Med.; 2020; 11, pp. 1-309. [DOI: https://dx.doi.org/10.1007/s41999-020-00428-6]
38. O’caoimh, R.; McGauran, J.; O’donovan, M.R.; Gillman, C.; O’hea, A.; Hayes, M.; O’connor, K.; Moloney, E.; Alcock, M. Frailty Screening in the Emergency Department: Comparing the Variable Indicative of Placement Risk, Clinical Frailty Scale and PRISMA. Int. J. Environ. Res. Public Health; 2023; 20, 290. [DOI: https://dx.doi.org/10.3390/ijerph20010290]
39. Lalla, R.; Sammy, I.; Paul, J.; Nunes, P.; Maharaj, V.R.; Robertson, P. Assessing the validity of the Triage Risk Screening Tool in a third world setting. J. Int. Med. Res.; 2017; 46, pp. 557-563. [DOI: https://dx.doi.org/10.1177/0300060517714935]
40. Salvi, F.; Morichi, V.; Grilli, A.; Spazzafumo, L.; Giorgi, R.; Polonara, S.; De Tommaso, G.; Dessì-Fulgheri, P. 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, pp. 69-75. [DOI: https://dx.doi.org/10.1007/BF03324901]
41. Oviedo-Briones, M.; Laso, R.; Carnicero, J.A.; Cesari, M.; Grodzicki, T.; Gryglewska, B.; Sinclair, A.; Landi, F.; Vellas, B.; Checa-López, M. A Comparison of Frailty Assessment Instruments in Different Clinical and Social Care Settings: The Frailtools Project. J. Am. Med. Dir. Assoc.; 2021; 22, pp. 607.e7-607.e12. [DOI: https://dx.doi.org/10.1016/j.jamda.2020.09.024]
42. Carpenter, C.R.; Shelton, E.; Fowler, S.; Suffoletto, B.; Platts-Mills, T.F.; Rothman, R.E.; Hogan, T.M. Risk factors and screening instruments to predict adverse outcomes for undifferentiated older emergency department patients: A systematic review and meta-analysis. Acad. Emerg. Med.; 2015; 22, pp. 1-21. [DOI: https://dx.doi.org/10.1111/acem.12569]
43. Apóstolo, J.; Cooke, R.; Bobrowicz-Campos, E.; Santana, S.; Marcucci, M.; Cano, A.; Vollenbroek-Hutten, M.; Germini, F.; Holland, C. Predicting risk and outcomes for frail older adults: An umbrella review of frailty screening tools. JBI Évid. Synth.; 2017; 15, pp. 1154-1208. [DOI: https://dx.doi.org/10.11124/JBISRIR-2016-003018]
44. Drubbel, I.; Bleijenberg, N.; Kranenburg, G.; Eijkemans, R.J.; Schuurmans, M.J.; de Wit, N.J.; E Numans, M. Identifying frailty: Do the Frailty Index and Groningen Frailty Indicator cover different clinical perspectives? A cross-sectional study. BMC Fam. Pr.; 2013; 14, 64. [DOI: https://dx.doi.org/10.1186/1471-2296-14-64]
45. Parker, S.G.; McCue, P.; Phelps, K.; McCleod, A.; Arora, S.; Nockels, K.; Kennedy, S.; Roberts, H.; Conroy, S. What is comprehensive geriatric assessment (CGA)?. An umbrella review. Age Ageing; 2018; 47, pp. 149-155. [DOI: https://dx.doi.org/10.1093/ageing/afx166]
46. Trevethan, R. Sensitivity, Specificity, and Predictive Values: Foundations, Pliabilities, and Pitfalls in Research and Practice. Front. Public Health; 2017; 5, 307. [DOI: https://dx.doi.org/10.3389/fpubh.2017.00307]
47. British Geriatrics Society. Fit for Frailty—Consensus Best Practice Guidance for the Care of Older People Living in Community and Outpatient Settings. British Geriatrics Society 2014. Available online: https://www.bgs.org.uk/sites/default/files/content/resources/files/2018-05-14/fff2_short_0.pdf (accessed on 28 June 2021).
48. Higginbotham, O.; O’Neill, A.; Barry, L.; Leahy, A.; Robinson, K.; O’Connor, M.; Galvin, R. The diagnostic and predictive accuracy of the PRISMA-7 screening tool for frailty in older adults: A systematic review protocol. HRB Open Res.; 2020; 3, 26. [DOI: https://dx.doi.org/10.12688/hrbopenres.13042.1]
49. Church, S.; Rogers, E.; Rockwood, K.; Theou, O. A scoping review of the Clinical Frailty Scale. BMC Geriatr.; 2020; 20, 393. [DOI: https://dx.doi.org/10.1186/s12877-020-01801-7]
50. Rockwood, K.; Theou, O. Using the Clinical Frailty Scale in Allocating Scarce Health Care Resources. Can. Geriatr. J.; 2020; 23, pp. 254-259. [DOI: https://dx.doi.org/10.5770/cgj.23.463]
51. Sayers, K.; Pender, C.; Maher, J.; Ryan, S.; Cooney, E.; O’Reilly, A.; Donnellan, C.; Pillay, I. 140 The Use of the Variable Indicative of Placement risk (VIP) as a Frailty Identification Tool. Age Ageing; 2019; 48, pp. iii17-iii65. [DOI: https://dx.doi.org/10.1093/ageing/afz103.84]
52. Checa-López, M.; Oviedo-Briones, M.; Pardo-Gómez, A.; Gonzales-Turín, J.; Guevara-Guevara, T.; Carnicero, J.A.; Alamo-Ascencio, S.; Landi, F.; Cesari, M.; Grodzicki, T. et al. FRAILTOOLS study protocol: A comprehensive validation of frailty assessment tools to screen and diagnose frailty in different clinical and social settings and to provide instruments for integrated care in older adults. BMC Geriatr.; 2019; 19, 86. [DOI: https://dx.doi.org/10.1186/s12877-019-1042-1]
53. O’Caoimh, R.; Kennelly, S.; Ahern, E.; O’Keeffe, S.; Ortuño, R.R. Letter to the Editor: COVID-19 and the Challenges of Frailty Screening in Older Adults. J. Frailty Aging; 2020; 9, pp. 185-186. [DOI: https://dx.doi.org/10.14283/jfa.2020.30]
54. Warren, N.; Gordon, E.; Pearson, E.; Siskind, D.; Hilmer, S.N.; Etherton-Beer, C.; Hanjani, L.S.; Young, A.M.; Reid, N.; Hubbard, R.E. A systematic review of frailty education programs for health care professionals. Australas. J. Ageing; 2022; 41, pp. e310-e319. [DOI: https://dx.doi.org/10.1111/ajag.13096] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35801297]
55. Rosenbaum, L. Facing COVID-19 in Italy—Ethics, Logistics, and Therapeutics on the Epidemic’s Front Line. N. Engl. J. Med.; 2020; 382, pp. 1873-1875. [DOI: https://dx.doi.org/10.1056/NEJMp2005492]
56. Carpenter, C.R.; Mooijaart, S.P. Geriatric Screeners 2.0: Time for a Paradigm Shift in Emergency Department Vulnerability Research. J. Am. Geriatr. Soc.; 2020; 68, pp. 1402-1405. [DOI: https://dx.doi.org/10.1111/jgs.16502] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32402103]
57. UNHCR Emergency Handbook (2015). Available online: https://emergency.unhcr.org/entry/43935/older-per-sons#:~{}:text=An%20older%20person%20is%20defined,or%20age%2Drelated%20health%20conditions (accessed on 6 November 2021).
58. Mowbray, F.I.; Heckman, G.; Hirdes, J.P.; Costa, A.P.; Beauchet, O.; Eagles, D.; Perry, J.J.; Sinha, S.; Archambault, P.; Wang, H. et al. Examining the utility and accuracy of the interRAI Emergency Department Screener in identifying high-risk older emergency department patients: A Canadian multiprovince prospective cohort study. J. Am. Coll. Emerg. Physicians Open; 2023; 4, e12876. [DOI: https://dx.doi.org/10.1002/emp2.12876] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36660313]
59. Heeren, P.; Devriendt, E.; Slt, N.I.W.; Rn, M.D.; Flamaing, J.; Sabbe, M.; Milisen, K. Comment on: Geriatric Screeners 2.0: Time for a Paradigm Shift in Emergency Department Vulnerability Research. J. Am. Geriatr. Soc.; 2020; 68, pp. 2414-2415. [DOI: https://dx.doi.org/10.1111/jgs.16691]
60. Abdulwahid, M.A.; Booth, A.; Kuczawski, M.; Mason, S.M. The impact of senior doctor assessment at triage on emergency department performance measures: Systematic review and meta-analysis of comparative studies. Emerg. Med. J.; 2016; 33, pp. 504-513. [DOI: https://dx.doi.org/10.1136/emermed-2014-204388]
61. Teguo, M.T.; Kuate-Tegueu, C.; Dartigues, J.-F.; Cesari, M. Frailty in sub-Saharan Africa. Lancet; 2015; 385, 2151. [DOI: https://dx.doi.org/10.1016/S0140-6736(15)61021-2]
62. Ellis, B.; Carpenter, C.; Lowthian, J.; Mooijaart, S.; Nickel, C.; Melady, D. Statement on minimum standards for the care of older people in emergency de-partments by the Geriatric Emergency Medicine Special Interest Group of the International Federation for Emergency Medicine. Can. J. Emerg. Med.; 2018; 20, pp. 368-369. [DOI: https://dx.doi.org/10.1017/cem.2017.426]
63. Hoogendijk, E.O.; Rijnhart, J.J.; Kowal, P.; Pérez-Zepeda, M.U.; Cesari, M.; Abizanda, P.; Ruano, T.F.; Schop-Etman, A.; Huisman, M.; Dent, E. Socioeconomic inequalities in frailty among older adults in six low- and middle-income countries: Results from the WHO Study on global AGEing and adult health (SAGE). Maturitas; 2018; 115, pp. 56-63. [DOI: https://dx.doi.org/10.1016/j.maturitas.2018.06.011]
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Abstract
Early identification of frailty can prevent functional decline. Although multiple frailty screens exist for use in Emergency Departments (EDs), few are validated against diagnostic standards such as comprehensive geriatric assessment. To examine the diagnostic accuracy of ED screens for frailty, scientific databases were searched for prospective diagnostic accuracy test studies from January 2000 to September 2022. Studies were assessed for risk of bias using QUADAS-C. Psychometric properties were extracted and analysed using R. Six studies involving 1,663 participants describing seven frailty screening instruments (PRISMA-7, CFS, VIP, FRESH, BPQ, TRST, and ISAR), representing 13 unique data points, were included. The mean age of participants ranged from 76 to 86 years. The proportion that was female ranged from 45 to 60%. The pooled prevalence rate of frailty was high at 59%. The pooled estimate for sensitivity was 0.85 (95% CI: 0.76–0.91) versus 0.77 (95% CI: 0.62–0.88) for specificity. Pooled accuracy based on area under the ROC curve was 0.89 (95% CI: 0.86–0.90). Although few studies were found, limiting the ability to conduct a meta-analysis of individual instruments, available frailty screens can accurately diagnose frailty in older adults attending the ED. As specificity was comparatively low, additional assessment may be required to identify those requiring inpatient management or onward community referral. Further study is therefore required.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
Details


1 HRB Clinical Research Facility, Mercy University Hospital, University College Cork, Grenville Place, T12 WE28 Cork, Ireland;
2 HRB Clinical Research Facility, Mercy University Hospital, University College Cork, Grenville Place, T12 WE28 Cork, Ireland;
3 School of Nursing and Midwifery, University of Galway, H91 TK33 Galway, Ireland;
4 Department of Geriatric Medicine, Mercy University Hospital, Grenville Place, T12 WE28 Cork, Ireland;
5 Department of Geriatric Medicine, Mercy University Hospital, Grenville Place, T12 WE28 Cork, Ireland;