Translation, Cross-Cultural Adaptation, and Psychometric Properties of the Malay-Translated Kihon Checklist Among Elderly Patients In The Emergency Department


 BackgroundThe insufficiency of gold standard assessment tools is the current challenge of frailty detection in an emergency department (ED). This study aimed to translate and evaluate the psychometric properties of the Malay-translated Kihon Checklist (KCL) for frailty assessment among Malaysian elderly patients presenting to the ED.MethodThe 25-item English version of the KCL was translated to Malay language through a forward and backward translation procedure. Three expert panels considered the items and pilot-tested on 15 elderly subjects. The final version was administered to 250 elderly patients who presented in the ED. Using confirmatory factor analysis (CFA), we compared the prior factorial models of the KCL. Model fits were determined using the Chi-square test/degree of freedom (df), comparative fit index (CFI), Tucker–Lewis index (TLI), root mean square error of approximation (RMSEA), Akaike information criterion (AIC), Bayesian information criterion (BIC), and expected cross-validation index (ECVI).ResultsAnalysis revealed that none of the prior models (seven-, two-, and one-factorial model) fit the data. After modification, one-factor model with 10 items had a superior fit (Chi-square/df, 54.434/35; CFI, 0.962; TLI, 0.951; RMSEA, 0.047 (95% confidence interval (CI), 0.019–0.071); AIC, 94.434; BIC, 164.863; ECVI, 0.379). The internal consistency reliability for the pooled 10 items was acceptable, i.e., 0.786.ConclusionThe CFA revealed that a one-factor model with 10 items had superior goodness-of-fit than other hypothesized factorial models. The scale demonstrated adequate construct validity and acceptable reliability with caution interpretation of some items.


Introduction
Frailty is a state of heightened vulnerability resulting from an ageing-associated decline in reserve and function across multiple physiologic systems (1). There is a growing awareness for routine assessment of frailty in emergency departments (EDs) (2). Although there is no local data on the prevalence of frailty, the worldwide prevalence of frailty in EDs was reported to be 7-80% (3)(4)(5). Frail patients have been associated with multiple adverse outcomes, including disability (6), falls (7), hospitalization (8), increased healthcare cost (Kojima, 2019), poor quality of life (9), and mortality (8). The added bene t of recognizing frailty in the ED is to prevent unnecessary invasive clinical management, inform a prognosis, and plan for discharge strategies (10).
Although multiple frailty screening tools have been developed, there is no standardization in the measurement of frailty in the ED. The two objective approaches commonly used in recognizing frailty are the phenotype model and the accumulated de cit model. The phenotype model de nes frailty as a distinct clinical syndrome based on the presence of three out of the following ve phenotypic criteria: unintentional weight loss, self-reported exhaustion, weakness, low physical activity, and slow walking speed (11). Meanwhile, the accumulated de cit model views frailty in the form of a frailty index. The index represents the health de cits of an individual to the total number of health de cits evaluated (12). While the phenotype model has a lack of agreed criteria for measuring each phenotype, the accumulated de cit model is too cumbersome to administer in clinical settings.
In busy EDs, self-reported frailty assessment or observed frailty is maybe more practical than objective assessment. Several instruments have been proposed for practical use in the ED, such Clinical Frailty Scale (CFS), PRISMA-7, Identi cation of Seniors at Risk, and Silver Code (5). The CFS, a judgment-based frailty assessment tool, was found to be more appropriate for use in the ED environment given the speed and ease of administration (10). However, the CFS has been criticized for making inappropriate escalation treatment decisions especially in patients with a long-term stable disability (13). The challenges are to identify reliable and validated tool for safe use in the ED. A screening tool intended for use in the ED setting should be based on a simple test, should require a little time, and can be interpreted by nonspecialist professionals.
A self-reported Kihon Checklist (KCL) has been proven to be effective in identifying and screening frail elderly patients in clinical settings (14). The KCL was developed by the Japan Ministry of Health, Labour and Welfare; it assesses seven domains of frailty: the activity of daily living, physical strength, nutrition, oral condition, socialization, memory, and mood (15). It consists of 25 items, and each item is rated either yes (1) or no (0), with a higher score indicating worse functioning (16). Although the KCL has never been validated in EDs, previous studies suggested that the instrument was easy, simple, and sensitive in identifying and screening frail elderly patients (14,17). In this study, we translated the KCL into Malay and assessed the factorial validity and internal consistency in a sample of elderly patients in Malaysia ED.

Materials And Methods
The study was conducted at the ED, Universiti Kebangsaan Malaysia Medical Centre (UKMMC) in Kuala Lumpur, Malaysia, between June 2020 and December 2020. The study was conducted in three phases.
Phase one includes the translation and cultural adaptation of the English version of the KCL. In phase two, content validation and pilot testing were performed. In phase three, we performed eld testing to determine the reliability and factor structure of the Malay-translated KCL (Fig. 1). The study received ethical approval from the UKMMC research ethic committee (FF-2020-349).
Phase one: translation process The permission to translate the English version of KCL to the Malay language was obtained from the original publisher. Two independent certi ed professional translators performed a forward translation of the English version of KCL to the Malay language; it was followed by backward translation by two different independent translators. The consensus was made on the translation of words, phrases, and items of the Malay version between researchers and translators; the Malay version was then compared with the original KCL. After some slight revisions, the nal version of the Malay-translated KCL was prepared and ready to undergo the validation process.
Phase two: content validation and pilot testing The content validity of the tool was conducted based on the quantitative methods described by Lynn (18). Three expert panels, which consist of a geriatrician, emergency physician, and family medicine specialist with an area of interest in geriatric medicine, were invited to review the Malay-translated KCL. The panels rated each item according to a four-point rating scale: 1 = not relevant, 2 = item needs major revision, 3 = relevant but needs minor revision, and 4 = very relevant; comments that carefully assessed and reworded.
As noted by Lynn (18), the content validity index (CVI) score was estimated both at the item level (I-CVI) and scale level (S-CVI). The I-CVI was calculated by dividing the number of experts that rated an acceptable grade (rating 3 or 4) by the total number of expert panels. The S-CVI was calculated based on two different methods: the average method (S-CVI/Ave) and the universal agreement method (S-CVI/UA). The S-CVI/Ave is the average of the I-CVI score for all items on the scale. Meanwhile, the S-CVI/UA is the proportion of the relevant items on the scale (rating 3 or 4) by all experts. Based on standard recommendations for content analysis, the item with I-CVI > 0.79 and scale with S-CVI ≥ 0.9 was considered to have excellent content validity (19). Additionally, a modi ed Kappa statistic (K) was computed to adjust I-CVI for chance agreement using the formula K = (CVI-Pc)/(1-Pc) (20). The probability of chance occurrence (Pc) was computed using the formula Pc = [N!/A!(N-A)!]0.5N, where N is the number of experts, A is the number agreeing on good relevance, and ! is a mathematical symbol for the product of all positive interfere less than or equal to N. The K of 0.75 and above is considered to have the excellent agreement of relevance (21).
To con rm for face validity, we conducted a qualitative approach using cognitive interviews (22,23). A pilot study was performed, in which the instrument was administered to 15 elderly patients who presented in the ED. The respondents provided feedback if they have di culties answering the items and if any of the items were confusing or containing di cult vocabulary. The feedback was carefully assessed and reworded to give the nal version of the Malay-translated KCL.
Phase three: eld testing Over 4 months between September 2020 and December 2020, a convenient sample of elderly patients aged 60 and above who presented in the ED of UKMMC were invited for the recruitment of the subject.
Patients must be able to read and speak in the Malay language. We excluded patients with preexisting cognitive impairment and underlying psychiatric illness and who are critically ill. All included subjects provided written consent. Recruitment was conducted without interfering with clinical evaluation, investigations, treatments, and interventions by the attending clinician. The following data were documented: age, sex, race, education level, marital status, living support, comorbidity, diagnosis, and disposition.

Frailty assessment
The Malay-translated KCL was self-administered by the patients or surrogates for patients with physical di culties. No time limit was given during the administration. In addition to the Malay KCL, we used the CFS, a clinical judgment-based screening tool that evaluates speci c frailty domains including comorbidity, function, and cognition, to assess frailty. The CFS has been validated and proven to be reliable to identify frailty in EDs (24). The CFS has a total score of 9, ranging from 1 (very t) to 9 (terminally ill), and a total score of 5 or more is considered frail (25). Measurements using either instrument represent trait or baseline at two weeks before assessment as suggested from the previous study (26).

Data analysis and sample size
Statistical analysis was performed using the IBM Statistical Package for Social Sciences (SPSS) Statistics for Windows software and SPSS Amos (version 26.0; IBM Corp., Armonk, NY, USA). A descriptive analysis of study variables was summarized using frequency and percentages for categorical variables and mean ± SD or median (interquartile range) for continuous variables.
Con rmatory factor analysis (CFA) was performed to evaluate the factor structure of the Malay KCL. Three prior hypothesized factorial models of KCL were chosen. The rst model was a seven-correlated factor model based on the original KCL (15), and the dimensions are as follows: the activity of daily living (items 1-5), physical strength (items 6-10), nutrition (items 11 and 12), oral condition (items 13-15), socialization (items 16 and 17), memory (items 18-20), and mood/cognition (items 21-25). The second model was a two-correlated factor structure based on the approach adopted by Fukutomi et al. (27). In this model, items 1-20 were classi ed under the lifestyle domain, and items 21-25 were classi ed under the mood or depression domain. The third factor was the one-factor structure where all the KCL items load on a single factor of frailty as proposed by Satake et al. (14). Model t for each structure was assessed using the following t statistics: χ2 goodness-of-t test, comparative t index (CFI), Tucker-Lewis Index (TLI), and root mean square error of approximation (RMSEA). The insigni cant (p-value > 0.05) model χ2 goodness-of-t test indicates model t. The χ2 goodness-of-t is sensitive to the sample size. Therefore, for the well-tting model, other t indices, namely, CFI, TLI, and RMSEA, must be ≥ 0.93, ≥ 0.92, and ≤ 0.08, respectively (28). The models were compared by examining the Akaike information criterion (AIC), Bayesian information criterion (BIC), and expected cross-validation index (ECVI), where smaller values indicated a better t (29,30).
The correlation of the KCL scores with the CFS was determined using Pearson, r correlation analysis. Internal consistency was evaluated using the Cronbach's alpha with values α > 0.7 considered acceptable reliability (31). For all analysis, p < 0.05 was considered signi cant.
In performing factor analysis, we determined the sample size according to the number of items, whereby ve subjects are required per item of the questionnaire (32). With 25 items in the questionnaire, the minimum subjects were 225.

Results
Translation and cross-cultural adaptation The content validity of the Malay-translated KCL showed a high level of panel's agreement ( Table 1). The S-CVI based on the average method and the universal agreement method were 0.97 and 0.92, respectively, indicating excellent internal validity. Individual item was marked as relevant (K ≥ 0.75), except for two items, namely, numbers 14 and 19, which were revised following a recommendation by the panels. The translated item 14 "Have you recently choked on your tea or soup?" was reworded as "tea" or "soup" is not a commonly used drink among Malaysian. The nal version explained the Malay translation of "Have you recently choked on a drink?" Meanwhile, the translated item 19 "Do you make a phone call by nding the phone number?" appeared to be a functional rather than memory assessment. The item was reformulated to explain the original English version "Do you make a call by looking up phone numbers and call on your own?" After the pretest (n = 15), there were no changes made to the response scale. All items were well understood by the respondents. The respondents took 5-10 minutes to complete the tool.
Therefore, the nal version of Malay-translated KCL was accomplished and ready for eld testing.

Subject characteristics
The Malay KCL was tested on 250 elderly patients. The mean age of patients was 71.3 ± 7.2 years old, and 56% (140) were male. About half of the subjects (52.4%) were Malay, and 13.2% had no formal education. Most of the subjects (59%) had severe comorbidity as indicated by Charlson comorbidity score ≥ 5. Cardiac-related conditions were the most common presentation to the ED (n = 89, 35.5%), followed by surgical (n = 24, 9.6%), and trauma (n = 22, 8.8%).
Majority of the subjects were hospitalized (n = 165, 6%). The demographic and clinical characteristics of the subjects are shown in Table 2.

Preliminary analysis
Of the 250 cases, there were no missing values. Table 3 demonstrates the oor effects, ceiling effects, skewness, and kurtosis scores for the total and subdomain KCL. The skewness and kurtosis were within − 2 and + 2, except for subdomain nutrition. Screening for the individual items at the univariate level revealed non-normal distribution for all items. An assessment at multivariate normality revealed the kurtosis coe cient to be 187.470 with a critical ratio of 40.337. As the data were not normally distributed, the ML estimation with bootstrapping (33) was used to generate an accurate estimation of standard errors using the Bollen-Stine bootstrap p values and con dence intervals (34). Our sample size was 250, which was adequate for bootstrapping as recommended by Nevitt and Hancock (35). A cutoff p > 0.05 was considered for model t.

Factor analysis
The three measurement models tested to evaluate the factor structure of the Malay KCL are shown in Fig. 2. The results showed that none of the models was a good t for the data. All models were revised with careful observation of the modi cation indices. The model t indices for seven-, two-, and one-factor models are shown in Table 4   df, degree of freedom; CFI, comparative t index; TLI, Tucker-Lewis index; NFI, normed t index; RMSEA, root mean square error of approximation; CI, con dence interval; AIC, Akaike information criterion; BIC, Bayesian information criterion; ECVI, expected cross-validation index.

Reliability
The analysis of the reliability of the Cronbach's alpha for the pooled 10 items showed acceptable internal consistency, i.e., 0.786. Most items appeared to have a good correlation with other items (item-total correlation > 0.3), except for items 5 and 13 (range, 0.187-0.647) ( Table 5).

Discussion
In this study, we translated the English version of the KCL into Malay and evaluated the validity and reliability in a sample of elderly patients in the ED. The psychometric evaluation showed satisfactory validity and reliability, supporting the use of the one-factor model with 10 items. In the rst part of the study, we translated and evaluated the content validation of the instrument. We adopted the Lynn method (18), an accepted approach for establishing the content validity of instruments. The ndings suggested a high degree of agreement beyond chance in most of the individual item as measured by the modi ed Kappa. A small number of items had a minor cultural and syntactical issue, and both items were revised through discussions with the research team. The instrument was pretested using cognitive interview to enhance the face validity of the instrument (36). Together with face validity, the instrument provides a robust foundation for future validation of the Malay-translated KCL.
There is limited consensus regarding the dimension of the KCL. The multidimensional factor structure of the KCL was proposed based on the original version, in which the items are divided across seven structured domains (15). Meanwhile, Satake et al. proposed for unidimensional factor structure based on the signi cant correlation between the total KCL score with frailty phenotypes (14). To date, only one study examined the factor structure of the KCL (37). The Spanish version of the KCL suggested the onefactor model with 15 items in a sample of community-dwelling Spanish older adults (37). In this study, we conducted a CFA to con rm whether the prior hypothesized factor models of the KCL were applicable to Malaysian elderly patients in ED settings.
We did not perform an exploratory factor analysis as the CFA provides stronger evidence in support of the validity of the instrument with regard to the model dimension (38). After carrying CFA, none of the models con rmed the factor structure of the Malay KCL, comparable with the ndings reported in the Spanish sample (37). The proposed one-factor model of the Malay KCL t well after extensive revisions with many of the items were deleted. However, only six items had factor loading larger than 0.5, which is the recommended minimum factor loading in CFA (39). Most of the remaining items predominantly relate to function and physical concern, which might re ect ndings from previous studies that have suggested that the physical function remains the most recognized features of frailty in the elderly (11,40). In addition, a previous study suggested that the physical component of the original KCL had the highest in uence on frailty score (41). As suggested by Awang (Awang, 2015), the items were maintained as the t statistics have achieved the required level.
The internal consistency reliability of the 10-item Malay KCL was found to be acceptable. The item-total correlation for most items was seen to be within 0.30-0.70, indicating a good relationship of items with the construct (42). The total score for the 10-item Malay KCL was signi cantly correlated with the total score of the 25-item KCL, suggesting that both versions shared a similar construct. The 10-item Malay KCL also showed a strong correlation with the CFS, indicating that these two tools measure similar concepts. However, because a large number of items (> 20%) were removed during the model revision, the model needs a revalidation study on a new sample of older patients in the ED (43).
The ndings of our study did not support the KCL de nition of frailty as there was insu cient cognitive and nutrition evaluation in the instrument. These ndings should be interpreted in the context of potential limitation. First, the current study excluded patients with an underlying cognitive impairment since they are often viewed as barriers to the completion of self-reported questionnaires. Patients with dementia may exhibit a wide range of cognitive and behavioral symptoms that potentially interfere with their ability to answer self-reported questionnaires. However, several studies emphasized the inclusion of patients with cognitive impairment in frailty assessment (44). It is also argued the nutrition component is underdeveloped. The cutoff body mass index (BMI) of 18.5 kg/m2 to de ne frailty have seen con icting evidence, in which study by Crow and colleagues has suggested of a poor relationship between frailty and BMI-de ned obesity (45). Second, although KCL was self-administered, in some instances, we asked relatives as a surrogate to ll in the questionnaire. Family members might not recognize or addressed subtle changes in frailty features among patients. In some instances, caregivers might overestimate the actual functional de cits than did patients (46).

Conclusion
This study supports the acceptable psychometric properties of the Malay-translated KCL in a sample of an elderly patient in the ED. The CFA revealed that the one-factor model with 10 items had superior goodness-of-t than other hypothesized models. The scale demonstrated adequate convergent validity and acceptable reliability with caution interpretation of some items. Flowchart of the translation process and cross-cultural adaptation.  The revised one-factor model of the Malay KCL with 10 items.