1. INTRODUCTION
Vietnam is among the Asian countries experiencing one of the fastest rates of population aging. In 2015, Vietnam officially entered the category of an aging society. Notably, by 2035, it is projected to become aged society with approximately one-fifth of its population expected to be aged over 60 years [1], In 2019, 12 million people in Vietnam were aged over 60 years, representing 11.9% of the total population. The rapidly aging population presents significant challenges to the healthcare system in Vietnam in delivering care for older adults [1], who often suffer from multimorbidity, polypharmacy, and geriatric syndromes [2]. Frailty is a common geriatric syndrome that increases the vulnerability of older adults to falls, infections, surgery, and hospitalization [3].
However, frailty is a preventable condition when recognized at an early stage [4]. Therefore, screening and early management of frailty are essential for preventing functional dependence, hospitalization, and mortality [5]. Additionally, recognizing frailty has significant implications for older adults, particularly for those who are moderately to severely frail. A range of tools is avilable for evaluating frailty [6,7]. Among them, the Clinical Frailty Scale (CFS), which relies on the clinician’s judgment, is widely used to its validity, simplicity, and convenience [8]. In 2005, Rockwood et al. developed the CFS, which initially included items rated on a seven-point scale, ranging from very fit to severely frail, with brief descriptors and pictographs [9]. Subsequently, the scale was expanded to nine points, with addition of very severely frail and the terminally ill categories. In 2020, the CFS version 2.0 was revised, incorporating minor adjustments to the level descriptions [10]. This version has been translated and adapted into various languages such as Greek [11] and Korean [12]. The CFS has demonstrated high feasibility and accuracy when applied in clinical practice [13]. This instrument has been utilized in the evaluation of patients in intensive care units, hospital admissions, and preoperative monitoring [14].
In Vietnam, the prevalence of frailty is relatively high (18.1%) among older adults living in the community [15] and significantly increases in older inpatients (35%) [16]. Although the CFS has been widely used for evaluating frailty in geriatric clinical practice and research, no studies have yet translated nad validated the CFS into Vietnamese. For that reason, the aim of our study was to translate the CFS into Vietnamese (CFS-VN) and assess its reliability. This will provide a simple and convenient tool for Vietnamese clinicians and researchers to evaluate, manage frailty, thereby improving the quality of care for older adults.
2. MATERIALS AND METHODS
This study involved two processes: 1) translation and cultural adaptation of the CFS and 2) the validation of the reproducibility of the translated instrument. The validation phase was conduted as prospective study among male and female outpatients, who consecutively visited the Geriatric outpatient clinic at Gia-Dinh People’s Hospital, Ho Chi Minh City, Vietnam, from September 2022 to January 2023. The inclusion criteria were individuals aged 60 years and older, able to communicate, and consenting to particiapte in the study. Participants with acute conditions such as infections, respiratory failure, acute coronary syndrome, or other conditions requiring hospital admission were excluded. Thus, a total of 324 participants (out of the 330 participants) were included in the study.
The calculated sample size to detect acceptable kappa (к0) of 0.60 and expected kappa (к1) of 0.75, with 90% power and accounting for a 5% dropout rate, was at least 315 patients. A convenience sampling method was employed for participants selection.
The original CFS is a pictographic scale derived from the accumulated deficit model of frailty including comorbidity, disability, and cognitive impairment, which was used in this study. The English CFS version 2.0 (Appendix 1), which classifies nine distinct levels frailty on a 1–9 scale, was translated into Vietnamese in accordance with the guidelines set by the International Society for Pharmacoeconomics and Outcome Research (ISPOR) Task Force for Translation and Cultural Adaptation [17–19] (Fig. 1).
The authorization for the translation of the CFS was obtained from Dalhousie University, Canada, in February 2022. After receiving permission, two independent translations of the CFS from English to Vietnamese were completed: carried out by a medical doctor with a certified knowledge of the English language (International English Language Testing System-IELTS score of 7.5) and a translation agency, as part of steps 2–3. The two versions were compared, and the authors, along with the geriatrics specialists reached a consensus-based decision on most appropriate translation. Step 4–5: the Vietnamese version of CFS was subsequently back-translated into English by a professional translator and a medical doctor, both native Vietnamese speakers residing in the United States. The two back-translators were blinded to the original scale. The authors compared the two back-translated versions to the original scale, resolving any discrepancies thorugh agreement between the authors and the geriatrics specialists, with the goal of refining the Vietnamese-translated version. After that, the pre-final CFS-VN version was further assessed by 10 medical doctors who had at least 10 years of experience in their specialties with the native language is Vietnamese, in steps 7–8. The scale was revised according to the suggestions of the Vietnamese doctors until the final CFS-VN version was completed in steps 9–10. Finally, the CFS-VN version was published (Appendix 1). The translated instrument was evaluated for conceptual coherence, interpretation, and cultural relevance by testing it on 10 medical doctors from various specialties: geriatrics (three persons), cardiology (two persons), pulmonology (two persons), family medicine (two persons), and neurology (one person).
Before the study was conducted, the two examiners were trained in the use of the CFS-VN for frailty assessment. After the initial assessment, the CFS-VN scores for each patient (CFS-VN1) were obtained by a geriatrician, and a second CFS-VN assessment (CFS-VN2) was performed by another geriatrician who was blinded to the patient’s initial scores to evaluate inter-rater reliability. Four weeks later, the CFS scores were re-assessed by the initial examiner, after reviewing the patients’ entire records (CFS-VN3) to evaluate the test-retest reliability. For both CFS-VN1 and CFS-VN2, the scoring was based on participants’s bseline function two weeks before visiting the geriatric clinic. Participants who scored 1–3 in the CFS-VN were grouped as non-frail, whereas those who scored ≥4 were grouped as frail.
Other information including the patient’s demographics (age, sex, weight, height, previous occupation, educational level, and marital status) and health-related characteristics (comorbidities, medication use, and functional status including activities of daily living [ADL] and instrumental activities of daily living [IADL]) were also obtained. The ADLs and IADLs were evaluated using the Katz [20] and Lawton & Brody [21] scale. The ADLs include six activities: feeding, toileting, bathing, dressing, transferring, and incontinence. The IADLs include eight activities: the ability to use the telephone, shopping, food preparation, laundry, modes of transportation, housekeeping, ability to handle finances, and responsibility for own medications. ADL impairment was defined as a total score of less than six, and IADL impairment was defined as a total score of less than eight. We collected data on participants’ chronic diseases based on their medical records. Multimorbidity was defined as having two or more chronic conditions [22]. Polypharmacy was defined as the concurrent use of five or more medications [23].
All statistical analyses were performed using the Stata version 14.0 software (STATA, College Station, TX, USA). A p-value of <0.05 was considered significant. All tests were two-tailed. Continuous variables were expressed as means and standard deviations, whereas the categorical variables were presented as numbers and percentages. The test-retest reliability and inter-rater reliability were assessed using weighted kappa (к). The к value was interpreted as no agreement (к=0), poor agreement (к=0.01−0.20), slight agreement (к=0.21−0.40), fair agreement (к=0.41−0.60), good agreement (к=0.61−0.80), very good agreement (к=0.81−0.92), and excellent agreement (к=0.93–1) according to standard practice [24]. The correlation of CFS score with health-related characteristics (total medication number, polypharmacy, and multimorbidity) and functional status (ADL impairment and IADL impairment) were assessed using Kendall’s tau.
3. RESULTS
Following the ISPOR guidelines, the CFS was successfully translated into Vietnamese and its reliability was assessed. The process of translation is illustrated in Fig. 1 and can be outlined as follows: During the forward translation steps, there was a high level of agreement in both meaning and wording. The minor discrepancies observed primarily revolved around the use of synonyms for specific terms, which were carefully discussed and resolved during a reconciliation meeting. The back translation aligned well with the forward translation versions, with only a few discrepancies identified. Only slight modifications in some items were made. Despite the thorough evaluation, all the items received a “very good” or “excellent” rating, and only minor changes were made in the Vietnamese terms used in the scale. The adjusted words included: (1) Category 1. Very Fit _ “một trong những người khỏe nhất” was revised to “những người khỏe nhất”; (2) Category 4. Living with Very Mild Frailty_ “giai đoạn đánh dấu sớm” was changed into “giai đoạn chuyển biến sớm”; and (3) in mild dementia _ “tách biệt với xã hội” was revised to “cô lập với xã hội”. In the final steps, the translation underwent proofreading, and subsequently, the CFS-VN version was established (Appendix 1).
A total of 324 outpatient participants were analyzed in the validation cohort; 25.2% were between 60 and 69 years, 37.1% were aged between 70 and 79 years, and 36.7% were aged ≥80 years. Approximately 65% of the participants were women. The mean body mass index was 20.6±2.9 kg/m2, with 25.6% classified as having ADL impairment and 63.9% IADL impairment. Approximately 79% of the participants had hypertension, 31.5% had diabetes, 15.1% had a history of stroke, 17.9% had chronic kidney disease, and 4.6% had cancer. The mean total number of medications was 4.5±3.0 with 60.2% classified as having polypharmacy. Further participant characteristics are presented in Table 1.
For the frailty assessment, none of the participants were classified as “very fit” (1) or “terminally Ill” (9). The more prevalent CFS phenotype was “mildly frail” (5) (94 patients), followed by “managing well” (3) or “moderately frailty” (6) (62 patients). The distribution of the participants by CSF scores is illustrated in Figs. 2 and 3.
Table 2 shows the results of the validation tests of the CFS-VN. Inter-rater reliability was achieved with a weighted kappa of 0.808 (p<0.001). Test-retest reliability was achieved with a weighted kappa of 0.869 (p<0.001). The CFS-VN scores significantly positively correlated with polypharmacy, multimorbidity, ADL impairment, and IADL impairment; however, they did not correlate positively with the medication number (Table 3).
| Kendall’s τ coefficient | p-value | |
|---|---|---|
| Total medication number | 0.0665 | 0.124 |
| Polypharmacy | 0.1151 | 0.021 |
| Multimorbidity | 0.1810 | <0.001 |
| ADL score | –0.6307 | <0.001 |
| IADL score | –0.8159 | <0.001 |
4. DISCUSSION
A cultural adaptation process is conducted to determine whether a measurement tool remains effective when applied in a culture different from that in which the original scale was developed. In this study, we successfully translated and adapted the CFS version 2.0 into Vietnamese (CFS-VN), with minor changes in some of the terms used, following the comments of the 10 senior medical doctors from various specialties. Furthermore, the translation and review steps based on highly cited, expert consensus guidance [17,18] were significantly followed, thus contributing to the high quality of the revised scale.
The Vietnamese version of the CFS demonstrated satisfactory inter-rater (weighted kappa: 0.808) and test-retest (weighted kappa of 0.869) reliabilities, with good agreements between raters. These outcomes were similar to those reported in French study, where the inter-rater variability of the CFS-French version was 0.73, whereas the test-retest variability was 0.86 [25]. Interestingly, none of the participants in the study were rated as “very fit” (category 1). This may be reflect the context in Vietnam where population is aging rapidly, but overall health status has not improved correspondingly. Moreover, as noted in a previous study [26], older adults in Vietnam typically develop multiple chronic conditions, with an average of seven types of diseases, which serves as significant risk factors for frailty. Therefore, the CFS-VN may allow frailty assessment, imporoving the quality of care and preventive strategies among Vietnamese older adults. In our study, the proportion of females was nearly double that of males (64.2% vs. 35.8%), which is different from the result of the Korean study [12]. Indeed, according to the results of the Population and Housing Census 2019 in Vietnam [27], a widening gender gap was observed in older population , with more females than males at advanced ages. The participants in our study were outpatient, which may reflect the gender disparity among older people living in the community in Vietnam. A similar gender ratio was reported in the previous study on frailty in Vietnam [15]. Otherwise, the Korean study included outpatients and inpatients, which may account for the difference in the gender ratio in the Korean study compared to ours.
Several domains were incorporated into the CFS, including comorbidity, cognition, functional disability, and physical activities. A significant correlation was found between the CFS-VN and several geriatric conditions, such as polypharmacy, multimorbidity, ADL impairment, and IADL impairment, which aligns with these components. Among these components, the IADLs and ADLs showed the highest correlation with the CFS-VN, as the functional status is one of the most important criteria for classifying the CFS. This finding is consistent with previous studies [11,28].
In this study, the CFS successfully distinguish between age groups. As shown in Fig. 2, frailty severity increases with age. Participants in the age group ≥80 years had the highest level of frailty. These findings align with the literature and a large body of studies. Age is a strong predictor of frailty because older age is associated with several negative outcomes, including multimorbidity, impaired cognitive function, and poor physical status [29–31].
This study had some limitations. There is no validated Vietnamese frailty tool that can be compared to the CFS-VN, as a reference tool for assessing frailty. Although the Fried phenotype is widely used in research in Vietnam [15,32], it is not suitable for evaluating frailty in older patients due to the requirement of instruments and is time-consuming [33–35]. Additionally, the study only analyzed older outpatients; healthy older adults living in the community or severely ill people in the hospital were not included. As a result, this study lacked paricipants who are “very fit” with a CFS score of 1 and “severely frail” or “terminally ill” patients with a CFS score of 9, who might receive home care or hospice service. Older inpatients were not included in our study because frailty status in older inpatients could be influenced by acute conditions and during hospitalization [36]. However, our study developed the CFS-VN with good reliability, providing a translated and cultural-adapted tool for frailty assessment in older Vietnamese adults. Further studies should include community-dwelling older adults and outpatients as participants to further validate the efficacy of the frailty assessment, particularly in “very fit” and “terminally ill” patients.
5. CONCLUSION
Our study demonstrated that the Vietnamese version of the CFS version 2.0 (CFS-VN) is a reliable tool for evaluating frailty among older adults in Vietnam. The application of this tool can aid in developing preventive strategies, ensuring appropriate management, and ultimately inproving the quality of health care for Vietnamese elderly population.
