Introduction
Comprehensive geriatric assessment (CGA) is a multi-dimensional, multi-disciplinary diagnostic and therapeutic process that aims to identify medical, mental, and functional problems in frail older people. The goal is to develop a coordinated and integrated treatment plan and follow-up [1]. In older cancer patients, CGA is crucial for guiding therapeutic interventions and avoiding over- or under-treatment, especially in patients identified as vulnerable or frail [2–4].
One of the challenges of conducting a CGA is the time it takes. In older cancer patients, this procedure can range from 30 to 80 minutes, depending on the components and tools used [5–7]. Although simplified 10-minute versions of the CGA have been developed [8], the greatest benefits are observed in cancer patients classified as vulnerable or frail. For fit patients, especially in areas with a shortage of geriatric specialists or high workloads, this procedure may not be necessary [9].
A two-step frailty assessment strategy in older cancer patients involves using a screening instrument to prevent unnecessary CGA in fit patients. The second step is to perform a CGA in the selected vulnerable or frail patients [10]. This strategy can also facilitate referrals to centers with greater expertise in CGA, particularly in low-income countries. A recent systematic study investigated validated instruments to identify older cancer patients who may benefit from CGA [11]. The study found that two instruments, the Vulnerable Elders Survey (VES-13) [12] and the G8 geriatric screening tool [13], had the most evidence for usefulness. However, most of these studies did not report on the time required to administer each tool [11]. Additionally, a modified G8 has recently been released [14]. Another study using the net benefit approach found that both G8 and the modified G8 failed to demonstrate clinical value in prescreening for frailty across various tumor types, disease stages, and age groups [10].
The Timed Up and Go (TUG) test is used to measure functional mobility of older adults and assess their risk of falls [15]. It has also been studied in a group of older cancer patients, showing a predictive capacity for the risk of early death in onco-geriatric patients receiving chemotherapy [16]. The TUG test can predict the risk of postoperative complications [17] and increased 5-year mortality in older adults undergoing surgery for solid tumors [18]. However, the TUG has not been studied in relation to its ability to identify older adults with cancer who are vulnerable or frail. Therefore, this study aimed to investigate the diagnostic performance of the TUG in identifying vulnerable or frail older adults with cancer who might benefit from CGA.
Material and methods
Setting
An observational and retrospective study was conducted study was conducted at the Geriatric Department of the ESSALUD Almenara Hospital, a tertiary care hospital in Lima, Peru. We reviewed CGA reports stored in the electronic medical records of hospitalized or outpatient adult patients aged ≥ 60 years with a previous cancer diagnosis, who had been evaluated between November 2022 and July 2023. The study followed the Standards for Reporting Diagnostic Accuracy Studies (STARD) recommendations [19].
Comprehensive geriatric assessment
Comprehensive geriatric assessment was performed by two trained geriatricians, who assessed the following domains: function and mobility, nutritional status, cognition, mood, social environment, and comorbidities. Six CGA indicators were selected: functional impairment (Activities of Daily Living score, ADL ≤ 5/6) [20]; cognitive impairment: Mini-Mental State Examination (MMSE, Spanish version) score < 24/30 [21]; malnutrition defined as one or more of the following French National Authority for Health criteria: at least 10% weight loss in 6 months or 5% in 1 month, and/or body mass index less than 21 kg/m2, and/or Mini-Nutritional Assessment (MNA-SF) score less than 12/14, and/or serum albumin level less than 35 g/L [22]; inadequate social environment defined as a score ≥ 10 on the Gijon social family assessment scale (Spanish version) [23]; verification of the diagnosis of depression in the medical history and use of antidepressants or depression diagnosed by a semi-structured interview to identify criteria for a major depressive episode from the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) [24]; and the number of severe (grade 3–4) comorbidities as assessed by the Cumulative Illness Rating Scale for Geriatrics (CIRS-G 0, 1, ≥ 2) [25]. Data was also collected on tumor site, metastatic status, age, sex, and in/outpatient status at the time of the CGA.
Timed Up and Go
The TUG test assesses the time a patient needs to get up from a chair, walk 3 meters, turn around, walk back, and sit down again [15]. This is measured in seconds with a handheld stopwatch. Two measurements were taken, and the average of these measurements was used in the study. The cut-off points for impaired TUG scores in older patients varied between 10 to 20 seconds [26]. The TUG test is an integral component of the CGA procedure in our center. Consequently, the results are routinely documented. However, it is important to note that the TUG test was not performed in patients who were unable to walk due to various reasons, such as being bedridden, dizziness, or knee pain, among others.
Vulnerability and frailty criteria
We used the frailty criteria of the International Society of Geriatric Oncology (SIOG-2) [27]. A patient is considered frail when they meet one of the following criteria: the presence of CIRS-G ≥ 1 grade 4 comorbidity, or ≥ 2 grade 3 comorbidities, or IADL score ≥ 7 of 8, or MMSE score < 24 of 30, or malnutrition (MNA-SF ≤ 7), or ADL score ≤ 3 of 6. A patient is called vulnerable when they meet the following criteria: number of severe (grade 4) comorbidities = 0 (assessed by the CIRS-G), and IADL score > 7 of 8, and MMSE ≥ 24 of 30, and 1 grade 3 comorbidity, or ≥ 1 grade 2 comorbidity, or at risk for malnutrition (MNA-SF < 12), or ADL score 4 or 5 of 6, or depression. Finally, a patient is considered fit when they score > 14 of 17 on the G8 scale. For this study, patients were assigned into two groups: fit vs. non-fit (vulnerable plus frail).
Statistical analysis
In the descriptive analysis, measures of central tendency, dispersion, and absolute and relative frequencies were used. Categorical variables were described as counts and percentages, and quantitative variables as means [standard deviation (SD)] or medians (range) depending on distribution. The performance of the TUG test was evaluated using sensitivity, specificity, receiver operating curve (ROC), and area under the ROC curve (AUC). Confidence intervals (95% CI) were reported. For sensitivity and specificity analysis, patients who did not undergo the test due to being bedridden or wheelchair-bound were timed with the maximum TUG time detected in the study.
Ethical approval
This study was approved by the Research Ethics Committee of Almenara Hospital in Lima, Peru (approval number 80-CIEI-OIyD-GRPA-ESSALUD-2023, March 27, 2023). Necessary strategies were implemented to maintain confidentiality of patient information.
Results
A total of 283 patients were included in the study, with 54.4% of them being hospitalized at the time of the CGA. The mean age was 76.8 ± 15.8 years, and the sample comprised 154 men (54.4%) and 129 women (45.6%). The prevalence of malnutrition, depression, and cognitive disorders was 71.7%, 27.2%, and 39.8%, respectively. Furthermore, 51.6% of the patients had severe comorbidities (grade 3–4 CIRS-G), and 46.0% had functional impairment (Katz < 5/6). The ten most frequent types of tumors were colorectal (19.4%), stomach (15.2%), prostate (9.9%), bile ducts (8.1%), hematologic malignancy (lymphoma, leukemia) (8.1%), breast (4.6%), lung (4.6%), liver (4.2%), skin (4.2%), and pancreas (3.9%). The frequency of patients with metastases and those with two tumors of different origin were 26.9% and 6.4%, respectively (Tab. 1). According to the SIOG-2 classification, the prevalence of fit, vulnerable, and frail patients was 21.9%, 50.9%, and 21.2%, respectively.
|
Variable |
Total patients |
Fit |
Non-fit (vulnerable+ frail*) n = 222 (78.5%) |
|||
|
|
n |
% |
n |
% |
n |
% |
|
Sex |
|
|
|
|
|
|
|
Male |
154 |
54.4% |
40 |
65.6% |
114 |
51.4% |
|
Female |
129 |
45.6% |
21 |
34.4% |
108 |
48.7% |
|
Indicators |
|
|
|
|
|
|
|
Inadequate social environment |
13 |
4.6% |
2 |
3.3% |
11 |
5.0% |
|
Malnutrition |
203 |
71.7% |
6 |
9.8% |
197 |
88.7% |
|
Depression (DSM IV criteria) |
77 |
27.2% |
1 |
1.6% |
76 |
34.2% |
|
Cognitive impairment (MMSE < 24/30) |
112 |
39.6% |
0 |
|
112 |
50.5% |
|
No. of severe comorbidities (grade 3–4 CIRS-G) |
||||||
|
0 |
137 |
48.4% |
61 |
100.0% |
76 |
34.2% |
|
1 |
120 |
42.4% |
0 |
|
120 |
54.1% |
|
≥ 2 |
26 |
9.2% |
0 |
|
26 |
11.7% |
|
Functional impairment (Katz; |
130 |
45.9% |
0 |
|
130 |
58.6% |
|
Outpatient at time of CGA |
129 |
45.6% |
28 |
45.9% |
101 |
45.5% |
|
Tumor site |
|
|
|
|
|
|
|
Colorectal |
55 |
19.4% |
8 |
13.1% |
47 |
21.2% |
|
Stomach |
43 |
15.2% |
5 |
8.2% |
38 |
17.1% |
|
Prostate |
28 |
9.9% |
8 |
13.1% |
20 |
9.0% |
|
Bile ducts |
23 |
8.1% |
7 |
11.5% |
16 |
7.2% |
|
Hematologic malignancy |
23 |
8.1% |
5 |
8.2% |
18 |
8.1% |
|
Breast |
13 |
4.6% |
5 |
8.2% |
8 |
3.6% |
|
Lung |
13 |
4.6% |
3 |
4.9% |
10 |
4.5% |
|
Liver |
12 |
4.2% |
5 |
8.2% |
7 |
3.2% |
|
Skin |
12 |
4.2% |
4 |
6.6% |
8 |
3.6% |
|
Pancreas |
11 |
3.9% |
3 |
4.9% |
8 |
3.6% |
|
Kidney |
10 |
3.5% |
2 |
3.3% |
8 |
3.6% |
|
Head and neck |
7 |
2.5% |
2 |
3.3% |
5 |
2.3% |
|
Brain |
6 |
2.1% |
0 |
|
6 |
2.7% |
|
Endometrium |
3 |
1.1% |
0 |
|
3 |
1.4% |
|
Bladder |
3 |
1.1% |
0 |
|
3 |
1.4% |
|
Ovary |
2 |
0.7% |
1 |
1.6% |
1 |
0.5% |
|
Other/unknown primary sites |
19 |
6.70% |
3 |
4.9% |
16 |
7.2% |
|
Two tumor sites |
18 |
6.36% |
3 |
4.9% |
15 |
6.8% |
|
Metastatic status |
76 |
26.86% |
10 |
16.4% |
66 |
29.7% |
*Classification of the International Society of Geriatric Oncology (SIOG-2); CGA — comprehensive geriatric assessment; CIRS-G — Cumulative Illness Rating Scale-Geriatrics; DSM-IV — Diagnostic and Statistical Manual of Mental Disorders; MMSE — Mini-Mental State Exam
Regarding the performance of the screening tool, the prevalence of fit and non-fit patients was 21.9% and 78.1%, respectively. When the TUG test results were equal to or greater than 15.5 seconds, sensitivity, specificity, positive predictive value, and AUC were 68.5% (95% CI 61.9–74.5), 88.5% (77.8–95.3), 95.6% (91.1–98.2), and 84.8% (0.80–0.90), respectively (Fig. 1). When the TUG analysis was conducted with 217 patients (excluding 66 of 283 who were unable to walk during the examination), the optimal cut-off point remained at 15.5 seconds. Sensitivity, specificity, positive predictive value, and area under the curve (AUC) were as follows: 55.1% (47.0–63.1), 88.5% (77.8–95.3), 92.5% (84.8–94.5), and 0.72 (0.66–0.77), respectively.
Discussion
Our study demonstrated that the TUG test, with an optimum cut-off value of 15.5 seconds, could serve as a valuable screening tool to identify vulnerable or frail older adults with cancer who could benefit from a CGA.
To our knowledge, this study is the first to use the TUG test as a screening tool before CGA in cancer patients, but it can be compared with other studies that used similar strategies. For example, gait speed (GS) measures the time needed for older patients to walk a certain distance at their usual speed [28]. Pamoukdjian et al. [29] assessed the diagnostic performance of GS for assessing vulnerability in older cancer patients and found that a GS < 1 m/s had sensitivity of 79.4%, specificity of 64.7%, and AUC of 82.0% (74.0– –90.0%) [29]. However, GS faces challenges in clinical practice due to the lack of a standardized protocol and variations in measurement methods (e.g. distance walked, starting and deceleration procedures, timing, and type of testing surface) [30]. In contrast, the TUG test is a more internationally standardized option.
The G8 index and its modified version have also been used as screening instruments in older cancer patients. The G8 index showed sensitivity ranging from 76.5% to 87.2% and specificity from 17% to 65% in different studies [13, 14, 31], while the modified version had sensitivity from 89.2% to 89.3% and specificity from 64.7% to 79.0% [14, 29]. Additionally, the VES-13, used for the same purpose, showed sensitivity ranging from 39.0% to 67.8% and specificity from 64.4% to 84.4% [31, 32]. The mean time to complete the G8 or VES-13 is approximately five minutes [31].
Previous evidence supports the usefulness of the TUG test in older cancer patients, as it has been correlated with survival, treatment-related complications, cognitive function, global health decline, disability in activities of daily living, and sarcopenia in various studies [26, 33–35].
However, our study has some limitations. The criteria used to select fit and non-fit patients (vulnerable plus frail) and evaluate TUG’s performance were based on SIOG-2 criteria [27, 36], whereas studies evaluating G8 and VES-13 used different cut-off points for each CGA scale [13]. Additionally, our study was conducted in a group of patients with a high prevalence of frailty, and further research is needed in patients with a lower prevalence of frailty. This is because diagnostic test studies in high-prevalence disease groups may lead to variations in predictive values, increasing the positive predictive value. In addition, the cut-off of > 15.5 is internally valid to our sample and not necessarily externally generalizable, further research is needed in different settings before an international TUG cut-off value can be recommended.
Conclusions
In conclusion, the TUG test with a cut-off of > 15.5 seconds showed promising sensitivity, specificity, positive predictive value, and AUC in identifying older adult cancer patients who may require CGA. This test could be beneficial, especially in hospitals with high demand for geriatric evaluation or a limited number of specialists.
Article Information and Declarations
Data availability statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restriction.
Ethics statement
This study was approved by the Research Ethics Committee of Almenara Hospital in Lima, Peru (letter 80-CIEI-OIyD-GRPA-ESSALUD-2023, March 27, 2023). The necessary strategies were implemented to maintain the privacy of patient information.
Author contributions
T.J.O.: concept and design, drafting of the manuscript, critical revision of the manuscript for important intellectual content; E.C-M.: acquisition, analysis, interpretation of the data, critical revision of the manuscript for important intellectual content; X.V.: acquisition, analysis, interpretation of the data, critical revision of the manuscript for important intellectual content; R.R.-O.: drafting of the manuscript, critical revision of the manuscript for important intellectual content, supervision.
Funding
None.
Acknowledgments
None.
Conflict of interest
The authors have no conflicts of interest to disclose.