An observational study with a cross-sectional design, translation, and adaptation, aimed at validating a measurement instrument that follows the guidelines of the COSMIN (COnsensus-based Standards for the selection of health Measurement INstruments) framework.11 This study was approved by the Scientific Ethics Committee of the Metropolitan Eastern Health Service (December 5, 2023).

Fifty-four HSCT candidates aged ≥17 years with a diagnosis of onco-hematological diseases participated in this psychometric study in a public hospital in Santiago, Chile. Individuals with observed functional or cognitive deficits, or significant disabilities that prevented them from understanding the study, performing simple functional tests, or giving their written informed consent, were excluded. Additionally, individuals with insufficient understanding of Spanish, which hindered comprehension of instructions and evaluator directions, were also excluded.

Patients attending their first consultation with the hematologist in the HSCT program were recruited from the HSCT unit. Those who met the eligibility criteria were invited to participate in the study, with a detailed explanation of the objectives and procedures involved. Those who voluntarily agreed to participate signed an informed consent form prior to enrollment.

The evaluations were conducted between December 2023 and June 2024 by two physical therapists at a physical medicine and rehabilitation clinic.

The original authors authorized the use of the scale, and the study was conducted in two stages:

The scale was first translated into Spanish by two native Chilean speakers who are bilingual in English. They worked independently during the translation process.

The resulting translations were compared and merged into a single version by a test coordinator. Any discrepancies between the versions were analyzed and resolved by the translators and the coordinator.

The scale was then translated back into English by a native English speaker (language teacher) who is bilingual in Spanish and did not participate in the translation stage.

The back-translated version was compared and evaluated in terms of similarities and conceptual equivalence with the version obtained in phase 1.2 and, in parallel, with the original scale.

In a consensus meeting of the researchers and translators, a second unified version was obtained that was consistent with the original version, with minor adjustments made for the Spanish scale tailored for Chile. Finally, through consensus, a final version derived from the previous process was sent to the original authors for review. After some corrections, they approved the final version to be applied in a second pilot testing phase (Figure 1).

Figure 1.

Flowchart of the translation and cross-cultural adaptation phases.

Since the version obtained in the first stage could not be limited to a simple translation, conceptual and semantic equivalence must be ensured between the original version and the adapted version, as well as the understanding of the obtained version by the target population. In this stage, the degree to which the content of the scale adequately reflects the construct to be measured was assessed. For this purpose, a pilot test was conducted with 54 patients, with some guiding questions being applied. The scale was first administered to 27 patients, observations were compiled, and necessary changes were made. Subsequently, the remaining 27 patients were evaluated, and new observations were gathered.

In this stage, the stability of the scale over time was evaluated by administering it at two different timepoints. The scale was applied twice by two physical therapists to a group of 30 patients, with a 24-h interval between assessments. To improve data reliability and facilitate interpretation, the two assessments were conducted within a maximum interval of 24 h, as recommended by the reviewers.

This interval was chosen to ensure a sufficient period of time to minimize the risk of progressive physical changes in the patients.

Content validity assessed whether the scale made sense to the professionals who care for HSCT candidates. Twenty-one professionals from different HSCT care centers nationwide (hematologists, physiotherapists, and nurses) with at least 5 years of experience in hematology and HSCT patient care were consulted. For content validity, an individual method was used, involving a written survey that each participant answered without having contact with the others. The scale was evaluated in terms of “coherence,” “clarity,” and “relevance” for each of the eight items composing the scale. A Likert-type survey with five response alternatives was used: “Strongly agree,” “Agree,” “Neither agree nor disagree,” “Disagree,” and “Strongly disagree” for each statement. An observation section was also included for additional information.

This scale was developed by professionals at the Princess Margaret Cancer Centre and is designed to classify patients who are HSCT candidates. It consists of eight items, which include various subjective and objective tests and scales. These items were carefully selected and appropriately modified based on previous studies conducted in older populations and transplant centers.8,10

The items are: Clinical Frailty Scale (CFS) score12; Instrumental Activities of Daily Living (IADL) 13; Self-Rated Health Questionnaire (SRH-Q)14; Fall Risk Assessment (Falls-test); Grip Strength (Dynamometry)15; Timed Up and Go test (TUGT)16; and Laboratory Tests such as serum albumin17 and C-reactive protein (CRP).18 Each variable is scored as either “normal” (0 points) or “abnormal” (1, 1.5, or 2 points, depending on the specific variable and its defined cut-off value). The total score is derived from the total of individual item scores, yielding a possible range of from 0 to 10.5 points. This allows for the classification of HSCT candidates in three categories: “fit,” “pre-frail,” and “frail,” regardless of age and underlying diagnosis.

In the present study, a hydraulic dynamometer was used for the grip strength test (Jamar®, J A Preston Corporation, New York, USA).

This variable was assessed using the Karnofsky Performance Status (KPS) scale, a numerical scale from 0 to 100. A lower score indicates a worse performance status.19 In this study, the ranges used were: 50–60, 70–80, and 90–100.

Data was collected on sociodemographic and clinical factors such as age, sex, education level, marital status, employment status, smoking and alcohol drinking habits, weight, height, diagnosis, type of HSCT, treatments received, Disease Risk Index (DRI), and the Hematopoietic Cell Transplantation-Comorbidity Index (HCT-CI).20

The data were tabulated and analyzed using the Statistical Package for Social Sciences (SPSS) version 25.0. Descriptive analyses were conducted, considering the mean, standard deviation, and the 95 % confidence interval. Spearman’s correlation test was used for reliability analysis between the two assessments considering the items of the scale in the test-retest. The following values were considered for interpretation: between 0.00 and 0.10, insignificant correlation; between 0.10 and 0.39, weak correlation; between 0.40 and 0.60, moderate correlation; between 0.70 and 0.89, strong correlation; and between 0.90 and 1.00, very strong correlation.21

Content validity was determined using Kendall's W test, considering the dimensions of “clarity,”, “coherence,” and “relevance” for each item of the scale based on data from expert evaluators. The following interpretation was applied: 0: No agreement; 0.10: Weak agreement; 0.30: Moderate agreement; 0.60: Strong agreement; and 1.0: Perfect agreement.22

After the forward translation process carried out by two independent translators, the versions were compared and deliberations took place to determine which words should be adjusted for better understanding, resulting in a single version (Table 1), which then proceeded to the back-translation process. Subsequently, the back-translated version was compared with the original version, and no significant differences were found, confirming that the translations were similar.

In the section of instructions for applying the scale, there were differences in the translation of the word “test,” where the first translator translated it as “prueba,” and the second translator kept it as “test,” with the final consensus being “prueba.” Similarly, in the application instructions, the word “fit” in the original version was translated into Spanish as “apto.” In the back-translation, it was rendered as “suitable,” but it was consensually accepted as “fit.”

A total of 54 HSCT candidates participated in the psychometric evaluation of the scale. The median age was 36.9 ± 14.6 years, with the majority being women (53.7 %) and multiple myeloma being the most prevalent diagnosis (33.3 %). Twelve patients (12.2 %) were categorized as “fit,” 26 (48.1 %) as “pre-frail,” and 16 (29.6 %) as “frail.” The sociodemographic and clinical background of the participants are shown in Table 2.

The first 27 patients evaluated stated that the scale was easy to understand, except for Item 6, “self-reported health question” which required further explanation for 12 patients. Regarding the functional tests, they mentioned that they were not difficult to perform. They also reported that the instructions were clear and the items were relevant and appropriate for assessing their frailty and functionality before the HSCT. No modifications were made during this stage.

Later, in the second round, the scale was applied to another 27 patients, of whom two also had some difficulty answering Item 6. Four patients experienced some difficulty executing the TUGT. All 54 evaluated patients emphasized the importance of being assessed on their “functional status” as a critical aspect prior to the transplant.

Regarding the reliability analysis between the two assessors (test-retest), positive and strong correlations were identified (Spearman's Rho [ρ]: 1; p-value <0.001) for all items of the scale (Table 3).

In general, all the variables of the scale were evaluated as consistent and relevant (Kendall's W range: 0.13–0.17; p-value <0.05). However, there were discrepancies regarding the clarity of some items (Kendall's W: 0.07; p-value = 0.11; Table 4).

Based on the analysis and observations made by the experts, improvements were incorporated to enhance the clarity and understanding of the scale, and some changes were made to the version from the first stage.

For Item 3, it was agreed to use the TUGT without translation, as this test is widely recognized and accepted, and has been integrated by professionals in the national clinical context.

For Item 4, which evaluates handgrip strength, some experts noted that while the test is coherent and relevant for this population, they inquired about the appropriateness of using the scale with values adjusted for the Chilean population. The original authors argued that the cutoff points (16 kg for women and 26 kg for men) used in the scale’s design methodology, supported by previous studies, were specifically chosen to make the scale applicable to other institutions. Therefore, the original values were retained.

Additionally, to improve comprehension, the phrase “patient classification” was changed to “patient categorization.” Furthermore, for the CFS scoring, the original version mentioned that it should be performed by “a physician,” which was changed to “healthcare professional” to adjust to the national clinical context, providing the option for these assessments to be conducted by other professionals.

A final version of the HCT-Frailty Scale adapted for use in Chile is provided in Supplementary Material 1.

There were also differences regarding the time required for application. The original authors mentioned 5–6 min, which was insufficient, as professionals took between 20–25 min to complete the scale. Additionally, it was consensually deliberated that the most suitable professionals for administering the scale are physiotherapists, as they frequently conduct all the tests that make up the scale in various clinical settings.

Regarding the results of each item on the Frailty-Functionality Scale reported by participants in the test and retest evaluations (Figure 2).

Figure 2.

Results of the test with the Hematopoietic Cell Transplantation Frailty Scale - A: Test; B: Retest.