Translation and linguistic validation of the Pittsburgh Fatigability Scale for Korean breast cancer survivors: A cognitive interviewing study

Abstract

Background:

Fatigability—the perception of fatigue contextualized to activities of fixed intensity and duration—has received growing attention in oncology research.

Objectives:

This study aimed to translate and linguistically validate a Korean-language version of the Pittsburgh Fatigability Scale.

Methods:

Following the Linguistic Validation Manual for Health Outcome Assessment, we applied a multi-stage cognitive interviewing (CI) method to ensure accurate translation and interpretation of the Pittsburgh Fatigability Scale. After forward- and backward-translation were completed, three rounds of CI regarding the translated instrument were performed with 18 participants having or not having breast cancer.

Results:

The first round of CI showed that seven items of the 10-item Korean-language instrument required revision to reflect the physical abilities of breast cancer patients with similar levels of physical activity. After the second round, two additional items were revised to reflect the cultural context and gender roles. During the third round, all participants exhibited full understanding of the Korean-language instrument.

Conclusions:

The translated instrument, its quality enriched by cross-cultural linguistic validation in combination with CI, can be effectively used to assess perceived physical and mental fatigability.

Implications for Practice:

Use of the Pittsburgh Fatigability Scale can facilitate oncology nurses’ assessment of perceived fatigue levels in cancer patients and can expand understanding of how patients’ fatigue perceptions are related to their exercise capacity.

Foundational:

This study is valuable as an example of how a multi-stage CI process can be effectively applied in cross-cultural oncology research.

Fatigue is composed of both subjective cognitive factors and objective physiological, biochemical, and behavioral factors.¹ The concept of fatigue has been defined as “a suboptimal psychophysiological condition caused by exertion.”² The form, dynamics, and concept of fatigue give rise to its extent and multidimensional characteristics, which are shown by the fact that the meaning and value of fatigue manifest differently in individuals. ² Although fatigue can be experienced by anyone, it is an important and common symptom experienced by cancer patients,³ patients with hypothyroidism,⁴ and patients with respiratory diseases.⁵ In order to measure fatigue as a dynamic and multidimensional concept, it is important to analyze fatigue from multiple perspectives and to evaluate how much fatigue affects an individual’s everyday activities.

However, global fatigue measures are limited in that they do not anchor fatigue perception to intensity and duration of activity. This limitation makes recognition of fatigue problematic because people tend to control their activity levels in order to avoid or minimize feelings of fatigue. To overcome this limitation and allow meaningful comparisons of fatigue in individuals over time, the concept of fatigability—which accounts for self-pacing of activity—has been a topic of interest in the human physiology literature for decades.⁶ Fatigability refers to the relationship between an individual’s perceived fatigue level and their capacity to perform activities. It differs from muscle endurance as discussed in the exercise literature and unlike fatigue, it has a broader meaning in terms of activity levels.⁷ Because it is a normalized measure of fatigue and permits meaningful fatigue comparisons among individuals, fatigability can be employed as an important outcome and predictor in research. With respect to existing fatigability measures, a 2018 integrative review identified such measures in three categories: self-reported fatigability, perceived fatigability, and performance fatigability. ⁷ Perceived fatigability referred to self-reported fatigue measured after a performance test, and performance fatigability was measured in terms of deterioration in performance by quantifying decreased activity speed or prolonged reaction time. Although 17 fatigability measures were identified, the integrative review found that psychometric properties were reported for only one-third of the measures.

One widely used fatigability measure is the Pittsburgh Fatigability Scale (PFS), a 10-item tool that was designed to overcome the methodological weaknesses of global fatigue measures. The PFS was originally developed to measure perceived level of fatigue in performing certain activities in adults aged 60 and older.^8,9 This tool is composed of 10 items dealing with four types of activity: social, sedentary (≤1.5 metabolic equivalents [METS]), lifestyle or light-intensity (1.6–2.9 METS), and moderate to high-intensity (≥3.0 METS). Each of the 10 items assesses physical fatigability and mental fatigability. The PFS measures fatigability by asking individuals how much fatigue they expect or imagine they would have across a range of activities of varying intensity and duration.

In recent years, fatigability has received considerable attention internationally, both as a concept and in terms of the development of measurement tools, but no tools are available to measure fatigability in South Korea. It is therefore essential to provide a fatigability tool that can be applied in that country to measure perceived fatigue in relation to various activities of fixed type, intensity, and duration. Furthermore, no studies have yet translated relevant tools following academically rigorous procedures, such as cognitive interviewing (CI). CI is a process that confirms the meaning of each item and how each item is understood by respondents, rather than focusing on the technical meaning of words and connotations from the researcher or professional’s point of view.¹⁰ CI has the advantage of minimizing the errors that can result from translation by asking about respondents’ understanding of each item and eliciting alternative wordings.^11–13

In this study, we translated the PFS into Korean using the CI method based on the Linguistic Validation Manual for Health Outcome Assessment.¹⁴ Approval was received from the original author of the PFS (N. Glynn), and the accuracy of the translation’s meaning was ensured by carrying out a systematic adaptation process. Glynn et al. developed and evaluated the performance, validity, and reliability of the PFS among older adults who were at least 60 years old at the time of its development. As fatigability has been a focus of interest in recent cancer research,^15–17 a fatigability tool that can be widely used in South Korea, including among cancer patients and healthy older adults, is needed. In a collaboration with the original developer of the tool, we conducted a linguistic validation of the Korean version of the PFS that can be used for both cancer patients and healthy individuals; the validation process centered on carefully reviewing the Korean-language tool’s descriptions of activities and tasks in cognitive interviews with Korean subjects. Thus, the aim of this study was to develop a Korean version of the PFS (K-PFS) through a rigorous cultural adaptation process and to validate the K-PFS items using the CI method in order to confirm the accuracy and cultural appropriacy of each item’s meaning.

Methods

Study design: Methodologic study

This study aimed to develop the K-PFS by adapting and translating the PFS and conducting linguistic validation through CI. The adaptation process included four steps, according to the Linguistic Validation Manual for Health Outcome Assessment Guidelines by c et al. (2000), Willis (2004), and Acquadro et al. (2012).^10,14,18

Translation and linguistic validation process

FORWARD TRANSLATION

Two native Korean speakers, one familiar with PFS-related vocabulary and another who was not familiar with that vocabulary, independently translated the PFS from English to Korean. After each translator finished the translation, the two translations were integrated by the team.

BACKWARD TRANSLATION

Two native English speakers independently back-translated the Korean into English. Neither translator had any knowledge about the tool, enabling them to compare the Korean and English versions in an unbiased way and to confirm whether there were differences in the Korean text relative to the original English text or whether there were Korean expressions that may be interpreted differently from the original English.

COGNITIVE INTERVIEWING

CI was conducted in the following four stages: comprehension, information retrieval, judgment formation, and response editing (Figure). To conduct CI, community members in South Korea were recruited by convenience sampling and snowball sampling, according to the following inclusion criteria: Native Korean speakers, over age 20 years, and healthy individuals or cancer survivors who completed active treatments (surgery, chemotherapy, and radiation therapy). The PFS has been used in a mixed sample of healthy adults and chronic conditions (multiple sclerosis, fibromyalgia).¹⁹ Cancer survivors were included because fatigue is a common side effect of treatment and exploring cancer-related experiences in relation to PFS items could allow richer and relevant examples that would further allow the application of the PFS across wider Korean populations. This study was done in preparation for using the K-PFS for a larger study of breast cancer survivors experiencing cancer-related fatigue, and thus recruiting cancer survivors was important.

Figure.

The four steps of the cognitive interviewing process and two techniques

EXPERT COMMITTEE REVIEW

A committee of two translators (one health expert and one linguistics expert) was convened to confirm whether there were discrepancies between the Korean version and the original tool. If the committee did not reach a consensus, the relevant data were shared with the author of the original tool.

The interviews were conducted using the initial items for which the committee reached consensus. The items that elicited diverging responses from participants were double-checked and analyzed again through expert committee review. A new group of participants was sampled to test the items that elicited diverging responses in another round of CI, which was eventually done in three rounds, and convergence in understanding and interpretation was confirmed by the expert committee.

Sample size

The sample size for CI followed the suggestion of 5-10 participants in Willis (2004) and Acquadro et al. (2012).^10,14 For each round of CI, around six participants were recruited. When there were diverging responses to items, another round of approximately six participants was recruited. CI was completed after establishing the validity of all items. Anticipating three to four rounds of participant recruitment and drop-out, the maximum sample size was set at around 25 participants. The final number for the three rounds of CI were 18 participants (six each).

Procedures

To recruit participants, advertisements were posted in Y Nursing School and community organizations, such as churches in Seoul, and circulated in the S Hospital cancer self-help group network. When potential participants inquired, a trained researcher explained on the phone that the aim of the interview was to minimize errors and increase the semantic accuracy of each item in the process of translating the tool into Korean, and participants initially provided oral consent. On the day of the interview, a written consent form was completed. It was explained again before the interview that all data would be anonymous, participants could withdraw if they ever felt uncomfortable responding, and their responses would only be used for research purposes.

After study approval was received from the S Hospital Institutional Review Board (Y-2017-0095) and informed consent was obtained from all participants, CI was conducted for 3 months after the translation of the tool was completed. Interviews were conducted in locations where participants felt comfortable, such as cafes and meeting rooms. Just before each interview, the participant completed a demographic questionnaire asking about characteristics such as age, sex, residence, and education level. In the face-to-face individual interviews, the researcher (author A) and participants reviewed each of the 10 items of the K-PFS. As illustrated in the Figure, the participants first read an item aloud, and the researcher asked them whether they could understand the item (step 1) and to explain how they interpreted the item’s meaning. Probes included “Do you understand this sentence? If so, how did you understand it?” and “What kind of information did you process as you thought about your answer?” (steps 2, 3). If participants did not understand the item or wished to offer alternative phrasing, the researcher documented which part of the item was vague or awkward, why they thought so, and what expression or term might be more effective (steps 3, 4). The in-depth interviews took about 1 hour to complete, and a small token of appreciation (worth about 10,000 Korean won) was given to participants afterward.

The in-depth interviews were organized according to the CI results for each stage through content analysis. In the analysis process, if there were items that had errors or disagreements, they were first discussed in the research team, organized through the original author and expert committee review, and confirmed in the next round of interviews. The translation was finalized through an iterative process of CI and content analysis in parallel, considering various ways that participants could understand each term, until agreement was reached regarding the most appropriate and accurate phrases.

Results

The 18 participants in this study were mean 56.0 years old (SD, 8.37 years; range, 36–68 years). Most were women (n=16, 88.9%), and in the first two rounds of CI, the majority had been diagnosed with breast cancer (n=14, 77.8%). Seven (38.9%) had graduated from university. CI was conducted sequentially using the fatigability questionnaire that was produced after backward translation. The Table displays the revisions that were made during the three rounds of CI.

Table.

Revisions to the Pittsburgh Fatigability Scale – Korean version through cognitive interviewing

Item	Original English version	Initial translation	Finalized translation
1	Leisurely walk for 30 minutes	Walking (lightly) at a usual pace for 30 minutes.	(Retained)
2	Brisk or fast walk for 1 hour	Quick walking that makes one slightly out of breath for 1 hour.	(Retained)
3	Light household activity for 1 hour (cleaning, cooking, dusting, straightening up, baking, making beds, dishwashing, watering plants)	Light household activity for 1 hour (making beds, washing clothes, cooking, preparing meals, dishwashing, etc.)	Light household activity for 1 hour (cleaning, cooking, straightening up the house, preparing meals, dishwashing, washing clothes, etc.)
4	Heavy gardening or outdoor work for 1 hour (mowing (push), raking, weeding, planting, shoveling snow)	Doing outdoor work for 1 hour (caring for a plot, plowing, pulling weeds, planting, shoveling snow, etc.)	Doing indoor or outside work for 1 hour (wiping/mopping the floor, moving furniture/boxes, weeding, planting, etc.)
5	Watching TV for 2 hours	Watching TV for 2 hours	(Retained)
6	Sitting quietly for 1 hour	Sitting quietly for 1 hour	Sitting comfortably for 1 hour
7	Moderate- to high-intensity strength training for 30 minutes (hand-held weights or machines greater than 5 lbs., push-ups)	Moderate- to high-intensity strength training for 30 minutes (push-ups, hand-held weights or machines greater than 2.5 kg, etc.)	Moderate- to high-intensity strength training for 30 minutes (squats, sit-ups, free gymnastics (chejo), hand-held weights or machines greater than 2.5 kg, push-ups, etc.)
8	Participating in a social activity for 1 hour (party, dinner, senior center, gathering with family/friends, playing cards, bridge)	Participating in a social activity for 1 hour (gathering with family/friends, alumni meetings, kye meetings, dinner, playing cards, Korean chess, etc.)	Participating in a social gathering for 1 hour (gathering with family/friends, alumni meetings, kye meetings, dinner, jang-gi, baduk, etc.)
9	Hosting a social event for 1 hour (not including preparation time)	Organizing a social event for 1 hour (gathering with family/friends, class reunion, kye meetings, etc.; not including preparation time)	Initiating and leading a social gathering for 1 hour (gathering with family/friends, class reunion, kye meetings, etc.; not including preparation time)
10	High-intensity activity for 30 minutes (jogging, hiking, biking, swimming, racquet sports, aerobic machines, dancing, Zumba)	High-intensity activity for 30 minutes (jogging, hiking, biking, dancing, aerobics, badminton, etc.)	High-intensity activity for 30 minutes (jogging, hiking, biking, swimming, badminton, aerobics, dancing, Zumba, etc.)

First round of CI

Six female participants who had experienced breast cancer participated. The participants clearly understood and agreed with translations for three of the 10 items (items 1, 2, and 5; 30%) (Table). The remaining seven items (70%) were not understood accurately or required examples, and are described in detail below. An expert review was performed based on participants’ level of understanding and suggestions was performed, and in order to revise the examples to activities that have a similar level of energy given the characteristics of the fatigability items, the data were shared with the author of the original tool to confirm that the revised phrasing accurately reflected the corresponding energy level before revision.

ITEM #3

Initial translation:

Light household activity for 1 hour (making beds, washing clothes, cooking, preparing meals, dishwashing, etc.)

Revised version:

Light household activity for 1 hour (cleaning, cooking, straightening up the house, preparing meals, dishwashing, washing clothes, etc.)

Interpretation:

All six participants (100%) responded that they could understand item #3 easily. Three participants (50%) suggested that for light household activities for Koreans, rather than making beds, more frequent household activities such as cleaning and straightening up the house, would be easier to understand.

ITEM #4

Initial translation:

Doing outdoor work for 1 hour (caring for a plot, plowing, pulling weeds, planting, shoveling snow, etc.)

Revised version:

Doing outside work for 1 hour (shopping, getting groceries, etc.)

Interpretation:

All six participants (100%) understood the concept of outdoor work, but they indicated that the examples were not culturally or temporally appropriate. The examples mainly pertained to people who have a field, plot, or garden, which are uncommon in urban settings and for Koreans in general. Thus, it was difficult to estimate fatigability for those examples. Moreover, the expression “outdoor work” was felt to imply a stronger sense of purpose than “outside work,” so the participants suggested using “outside work” instead. Specifically, participants who had experienced breast cancer explained that they would likely report a higher level of fatigue than intended by the original author since these activities require using their hands extensively.

ITEM #6

Initial translation:

Sitting quietly for 1 hour

Revised version:

Sitting comfortably for 1 hour

Interpretation:

Three participants (50%) expressed that they did not understand “quietly” very clearly. They reported it was difficult for them to determine their activity level when they tried to answer the question due to the vagueness of “quietly.” In particular, they were unsure whether this term meant not moving and staying still, not conversing, not doing anything, or meditating or praying. After consulting with the author of the original scale, it was understood that the item referred to sitting in a comfortable manner, and the terminology was revised in the expert committee review process.

ITEM #7

Initial translation:

Moderate- to high-intensity strength training for 30 minutes (push-ups, hand-held weights or machines greater than 2.5 kg)

Revised version:

Moderate- to high-intensity strength training for 30 minutes (squats, sit-ups, push-ups)

Interpretation:

After reviewing the examples in item 7, all six participants (100%) responded that push-ups were not appropriate since those who had experienced breast cancer have concerns about developing lymphedema in the arms and actively avoid push-ups or positions that involve weight loading on the arms, such as the plank position. The provided examples, such as weights or machines, were considered to be activities of very high intensity by those who experienced breast cancer, so participants expressed concern that the measured fatigue level would be higher than responses indicated by healthy individuals. They suggested that muscle-based exercises with similar energy levels that use lower body or core muscles instead of just upper body muscles be included.

ITEM #8

Initial translation:

Participating in a social activity for 1 hour (gathering with family/friends, alumni meetings, kye meetings, dinner, playing cards, Korean chess, etc.)

Revised version:

Participating in a social gathering for 1 hour (gathering with family/friends, alumni meetings, kye meetings, dinner, jang-gi, baduk, etc.)

Interpretation:

Five participants (83.3%) suggested that “social activity” is not an expression used comfortably in everyday language and should be changed to expressions such as “social gathering,” “meetings with acquaintances,” or “social interaction.” Kye is a common social gathering that often serves a purpose (e.g., mutual aid) and used by women. Instead of playing cards or Korean chess, participants suggested specifying jang-gi and baduk (Korean version of Go).

ITEM #9

Initial translation:

Organizing a social event for 1 hour (gathering with family/friends, class reunion, kye meetings, etc.; not including preparation time)

Revised version:

Initiating and leading a social gathering for 1 hour (gathering with family/friends, class reunion, kye meetings, etc.; not including preparation time)

Interpretation:

Similar to item #8, participants expressed that rather than “social event,” “social gathering” would fit the examples better and would be easier to understand. The word “organizing” connotes a major event and is not used often to refer to individuals leading a social gathering. They suggested changing the word to “initiating” or “leading.”

ITEM #10

Initial translation:

High-intensity activity for 30 minutes (jogging, hiking, biking, dancing, aerobics, badminton, etc.)

Revised version:

High-intensity activity for 30 minutes (jogging, hiking, biking, swimming, badminton, aerobics, dancing, Zumba, etc.)

Interpretation:

Swimming and Zumba were initially excluded from the translated version as these appeared to be practiced by few Korean. However, all six participants (100%) understood the item and suggested that including more examples of exercise with a similar energy level would help in comprehending the intended level, and swimming and Zumba were activities familiar to women. After consulting the author of the original scale, swimming and Zumba were added back.

Second round of CI

Six female participants who had experienced breast cancer and did not participate in the first round of CI were invited to CI for the seven items that needed additional review after the first round. Suggestions were made to change the examples used for two items. Through the expert committee review and consultation with the author of the original scale, items 4 and 7 were revised again.

ROUND 2: ITEM #4

Translation reviewed:

Doing outside work for 1 hour (shopping, getting groceries, etc.)

Finalized translation:

Doing indoor or outside work for 1 hour (wiping/mopping the floor, moving furniture/boxes, weeding, planting, etc.)

Interpretation:

The author of the original scale suggested maintaining the examples from the original instrument and adding examples of a similar MET level appropriate for the cultural context and gender roles that were suggested in the first round of CI. Since the item should be able to examine the fatigue level of both breast cancer patients and healthy individuals, it was suggested that the item should include a variety of examples. Therefore, the changes from the first round of CI were applied and examples were added. The word “indoor” was also added.

ROUND 2: ITEM #7

Translation reviewed:

Moderate- to high-intensity strength training for 30 minutes (squats, sit-ups, push-ups, etc.)

Finalized translation:

Moderate- to high-intensity strength training for 30 minutes (squats, sit-ups, free gymnastics (chejo), hand-held weights or machines greater than 2.5 kg, push-ups, etc.)

Interpretation:

To create an item appropriate for both breast cancer survivors and healthy men, exercises using hand-held weights heavier than 2.5 kg or machines were added back from the original item. After consulting with the author of the original scale, it was determined that “free gymnastics (chejo)” is an expression easily understood by Koreans, even for older adults or children, with a similar MET level. Therefore, “free gymnastics” was added after expert committee review. CI only needs to be repeated for items that had remaining disagreements, but since the first two rounds of CI were focused on women who had experienced breast cancer, in order to expand the use of the scale, CI for all 10 items was conducted in the third round of CI with both men and women who had not experienced breast cancer. The interviews sought to confirm whether each item was understood and whether the examples were appropriate and suitable for estimating activity levels.

Third round of CI

Six participants were recruited for the third round of CI, including two breast cancer patients, two healthy women, and two healthy men for generalizability. For the items for which expert committee review was performed to revise or add appropriate examples in the second round of CI, all six participants (100%) understood them accurately and expressed the content of the items using words with similar meaning and usage. The final 10 items were back-translated again to receive confirmation from the author of the original scale, and CI was concluded.

The final Korean version of the PFS is available upon request from the developer at https://publichealth.pitt.edu/epidemiology/research-practice/faculty-research/pittsburgh-fatigability-scale.

Discussion

The PFS was translated and linguistically validated with careful attention to its suitability for Korean cancer patients as well as the Korean general population. This study used CI in three phases with separate participants to comprehensively identify and clarify potential linguistic ambiguities and ensure cultural relevance for Koreans with a diagnosis of cancer as well as healthy Korean adults. With respect to translation of the instrument, we used CI with think-aloud and probing techniques^12,13 to identify several subtle issues of interpretation, which were then discussed and resolved through close communication with the original instrument developer. For example, “sitting quietly” use double quotes was eventually translated to “sitting comfortably” (item 6) as the imagery related to “quietly” appeared to align with not doing anything at all, while the original item meant to contrast with active motion. Also, the initial translation for “social activity” was finalized as “social gathering” (item 8) to portray the activity as a means of personal interaction rather than relations on a wider social scale. The description “hosting a social event” (item 9) was eventually translated as “initiating and leading a social gathering” as this scale was more understandable.

The multi-phase CI process also allowed for identifying suggestions on examples that could be more closely linked with cultural and/or daily life contexts, especially for use with women who had experienced cancer. For example, as most Korean homes do not have a garden/lawn requiring personal upkeep, the examples for “outdoor work” (item 4) on mowing and raking were not relevant to participants. The question stem was eventually changed to “indoor or outside work” and examples such as mopping the floor and moving furniture/boxes were added to better reflect life and work patterns. Examples were also added to illustrate activities of the intended level that would be more familiar, e.g., adding examples of squats, sit-ups, and free gymnastics (chejo) to indicate ‘moderate- to high-intensity’ activities (item 7), and using jang-gi, baduk (the Korean word for ‘Go’- which is the more widely known term) substituting ‘playing cards, bridge’ (item 8).

Limitations

A limitation of the study is that only a subgroup of cancer (breast cancer) was represented and participants were mostly women. It is possible that male cancers (prostate and penile cancers, etc.) may require further specification of activity examples. However, at the final CI round, we purposively recruited healthy men and women, as they were expected to have less restrictions in activity dimensions than cancer survivors, and found all items to be well received and understandable. The CI techniques and steps illustrated in this study can be used with different cancer and illness groups in the future to explore whether the K-PFS should be expanded and updated. We strongly suggest direct and regular communication with the original developer, as this allows clarity of intended item meaning and use.

Implications for Nursing

Considering that the number of cancer survivors continues to increase, that their main symptom is fatigue, and that international instrument translation and linguistic validation constitute a growing trend, this study is valuable as an example of how a multi-stage CI process can be effectively applied in cross-cultural oncology research. In the international literature, CI is recognized as an essential part of the linguistic validation process.^10,12 Rather than administering a translated tool directly to a target population, with CI, five to ten respondents are asked individually about each item in in-depth interviews to assess whether the items are easy to understand and culturally appropriate. However, in nursing research, CI is often not used during the process of tool translation and linguistic validation. With the application of CI, an instrument to be used in a cross-cultural context can be accurately translated and linguistically validated so as to detect linguistic ambiguities and culturally inappropriate expressions; the result is a tool that oncology nurses and researchers can employ with confidence. Our use of this approach generated a Korean-language PFS that can be effectively used to assess the activity level of Korean cancer patients and plan interventions accordingly. Specifically, the PFS enables clinicians to characterize patients’ susceptibility to fatigue in particular activity contexts, accounts for self-pacing of activity, and is a relatively objective means of assessing how much fatigue limits activity. Thus, the instrument can facilitate oncology nurses’ assessment of perceived fatigue levels in cancer patients and can expand understanding of how patients’ fatigue perceptions are related to their exercise capacity.

Conclusion

This study describes how participants in three rounds of CI interpreted translated PFS items, retrieved and processed information relating to their situation (e.g., noting activities that did not pertain to their situation and/or clinical condition), and offered suggestions for item revision. As Korean breast cancer survivors were heavily involved in the CI process, the finalized K-PFS appears to be a sound and stable measure of activity-anchored fatigue for use with this population. Additional research is needed to determine the validity and reliability of the K-PFS for other populations of cancer patients and healthy individuals in Korea.