Translation in Kannada and Cross-Cultural Adaptation of Activities-Specific Balance Confidence Scale

Stage 1: Cross-cultural adaptation of the ABC scale

The cross-cultural adaptation process for the ABC scale into Kannada (ABC-K) involved modifying eight specific items from the original 16-item scale. These modifications were considered essential due to their inappropriateness from cultural, semantic, or geographic perspectives within the Indian context. This process adhered to established guidelines for cross-cultural adaptation of self-reporting measures and involved an expert committee that reviewed and modified the scale to achieve semantic, idiomatic, and conceptual equivalency.

Permission and translation approval were obtained from Dr. Lynda Elaine Powell and Dr. Anita M. Myers, the original developers of the scale. The translation and adaptation procedures adhered to the guidelines for cross-cultural adaptation of self-report measures proposed by Beaton et al.^[11]

For the cultural validation phase, ten healthcare professionals from diverse medical disciplines, all proficient in English and well-versed in the underlying concepts, were selected and provided with the ABC scale. The feedback from these experts was employed to establish semantic, idiomatic, and conceptual equivalence with the original ABC scale. Subsequently, the scale underwent evaluation and refinement to ensure appropriate cultural adaptation.

Stage 2: Forward translation

At this stage, two independent translators, both native or first-language speakers of Kannada, were engaged to translate the original English version of the ABC scale into Kannada. Each translator produced a detailed report of their translation, noting any ambiguous terms or challenging expressions encountered, and provided a written justification outlining the reasoning behind their translation choices.

Stage 3: Reconciliation

At this stage, the two forward translations (T-1 and T-2) were reviewed to identify common elements and reconcile discrepancies. The translations were then compared and merged to produce the preliminary Kannada version of the ABC scale (ABC-Kannada). A detailed written record of the synthesis procedure was prepared to document the process comprehensively.

Stage 4: Back translation

The ABC-Kannada version was subsequently back-translated into English by two professional translators who were native speakers of English. Importantly, these translators had no medical background and were neither briefed on the study’s objectives nor aware of its specific issues. This blinding was implemented to prevent information bias and to reveal potential flaws by allowing unanticipated interpretations of the translated items to emerge. The primary aim of this step was to verify that the translated version corresponded accurately to the content of the original items, as back-translation often helps to identify unclear or imprecise wording.

Stage 5: Back translation review

Following the back-translation process, the original ABC scale was compared with the two back-translated versions (BT1 and BT2). The scale’s developer, Dr. Lynda Elaine Powell, was consulted to confirm the equivalence between the back-translations and the original instrument. Based on this review, adjustments were made to align certain scale parameters with relevant cultural concepts, and any required grammatical refinements were also incorporated.

Stage 6: Expert committee

The Expert Committee comprised ten healthcare professionals from varied medical specialties, including physiotherapy, neurology, and community medicine, all of whom had previously contributed to the adaptation process. Each member was fluent in English and demonstrated a strong grasp of the concepts underlying the ABC scale. The committee’s primary responsibility was to ensure cross-cultural equivalence in the composition of the translated scale. They conducted a meticulous, item-by-item review of correspondences between the original and back-translated questionnaires. Drawing on their expertise, they developed semantic, idiomatic, and conceptual equivalence for the ABC scale and performed evaluation and refinement procedures to achieve optimal cultural adaptation. The collaborative efforts of the committee produced the preliminary version of the instrument for field testing. Additionally, they ensured that the questionnaire would be easily comprehensible to individuals aged ≥15 years.

Stage 7: Test of the pre-final version

A preliminary evaluation of the translated ABC scale was carried out with at least two participants. The primary objective of this initial trial was to assess the clarity and comprehensibility of the adapted instrument prior to large-scale field testing. During this phase, each participant was interviewed to explore their understanding and interpretation of every item in the preliminary questionnaire. Their feedback was carefully recorded, serving a function comparable to cognitive interviewing, to confirm that each question was interpreted as intended and to identify any subtle ambiguities or culturally specific nuances. This stage was essential for detecting items that might still be confusing or challenging for the target population of Kannada-speaking older adults, thereby ensuring that the language was both clear and culturally appropriate before finalising the version for broader testing.

Stage 8: Final review and approval of the adapted version

The final stage of the adaptation process involved submitting all reports and documentation to the coordinating committee responsible for overseeing the translated version. This step functioned as an audit, verifying that each phase of the adaptation procedure had been executed rigorously and that all requisite records were complete. The committee’s oversight at this juncture was essential to confirm that the translation was both accurate and appropriate, thereby validating the integrity of the entire cultural adaptation process.

Procedure for content validity

A content validation form was specifically developed for use by an expert review panel to evaluate the ABC-Kannada questionnaire. The experts were provided with the questionnaire along with its linguistically adapted version for thorough content validation. Prior to scoring, they were instructed to examine the domain and all associated items in detail. They were explicitly encouraged to offer verbal feedback aimed at enhancing the relevance of each item according to its perceived importance. This systematic process ensured the appropriateness and contextual relevance of each item in the adapted scale. Insights from the panel contributed directly to the high content validity index (CVI) achieved, with the overall ABC-Kannada scale demonstrating a CVI of 96.3% [Table 1].

Participants for the study were recruited from R.L Jalappa Hospital, Kolar, in addition to community settings and an outpatient physiotherapy facility. Recruitment followed predetermined inclusion criteria, which required individuals to be aged between 60-85 years and to possess the ability to read and write in Kannada. Exclusion criteria included the inability to read or speak Kannada, as well as the presence of psychotic or cognitive impairments.

Prior to participation, all individuals were provided with both an oral explanation of the study and a written information sheet. Written informed consent was then obtained from each participant. A total sample of 50 individuals was targeted, and demographic information was collected for all participants. The Balance Confidence Scale was first administered using the conventional paper-and-pencil format, allowing participants to independently read each item and indicate their level of confidence in maintaining balance in the specified situations. To assess test-retest reliability and language equivalence, the ABC-Kannada version of the scale was re-administered to the same participants after a two-week interval. The two-week duration was deliberately chosen to minimise recall bias from the initial administration and to reduce the likelihood that major life events or health changes would affect participants’ self-reported balance confidence.

The demographic profile of the sample is presented in Table 2. The mean age was 69.48 ± 6.42 years. Of the participants, 60% were male, 68% married, 10% engaged in occupational activities, and 64% had a formal education. Additionally, 20% reported using walking aids, 64% reported a history of falls within the past year, and 16% engaged in regular exercise.

Data Analysis

Statistical analysis was performed using SPSS version 29.0. Descriptive statistics were calculated, including mean ± standard deviation for age and standard error of measurement for each variable. The frequency of sex distribution within the sample was also determined.

The acceptability of the scale was assessed by calculating the proportion of rejected completed questionnaires and incomplete items. Additionally, the acceptability of the scale was gauged by the willingness of the respondents to complete it again.

To evaluate the internal consistency of both the 16-item cross-culturally modified ABC scale and the ABC-K scale, Cronbach’s alpha was calculated. The interpretation of Cronbach’s α coefficient values adhered to the following rules of thumb provided by George and Mallery; ≥0.9, excellent; ≥0.8, good; ≥0.7, acceptable; ≥0.6, questionable; ≥0.5, poor; and ≤0.5, unacceptable.^[16,17] Finally, Pearson correlation coefficient was utilised to test the correlation between each item of the ABC-K scale

Translation and cross-cultural adaptation

Eight items of the original ABC scale were modified to ensure cultural, semantic, and contextual relevance. For example, in item 3, the word “closet” was removed; in item 4, “can off” was replaced with “car”; and in item 8, “car parked in driveway” was adapted to “standing vehicle in front.” Similarly, “mall” was revised to “shops” or “shopping centre” in items 10, 12, and 13, while “icy sidewalks” in item 16 was changed to “slippery pavements.” These modifications ensured linguistic clarity and cultural appropriateness [Table 3]. The ABC-K scale showed excellent internal consistency, with a Cronbach’s alpha of 0.992 (compared to 0.97 for the cross-cultural version). This value exceeds the standard threshold for excellent internal consistency (>0.90), confirming that the items were highly interrelated and measured the same underlying construct (balance confidence). Pearson’s correlation coefficients between items ranged from 0.79-0.95, with the highest correlation observed between items 1 and 2 (r = 0.944). These findings further reinforced the strong inter-item consistency of the instrument [Table 4].

To assess stability, the ABC-K was re-administered to the same participants after a 2-week interval. This timeframe was selected to minimise recall bias while avoiding significant life events that could affect self-reported balance confidence. Strong correlations were observed between the two administrations (r = 0.92-0.96, p <0.0001), indicating high reproducibility over time and confirming the reliability of the adapted Scale. Analysis of inter-item correlations demonstrated very high positive relationships. Items 1-7 consistently showed correlations above 0.79, with the strongest between items 1 and 2 (r = 0.944). Similarly, items 8-16 showed high correlations with one another and with earlier items (e.g., item 8 correlated 0.903 with item 9 and 0.923 with item 10). These results reinforced the strong interrelatedness of all items, consistent with a unidimensional scale structure. The uniformly high correlations suggested the presence of potential item redundancy, with several items measuring highly overlapping aspects of balance confidence. This insight highlights a possible area for streamlining in future versions of the instrument.

Validity analysis

Descriptive analyses also indicated that items 15 and 16 exhibited lower mean and standard deviation values compared to other items as illustrated in Figure 2, suggesting some variability in item performance [Table 5]. All participants (100%) completed the ABC-K without difficulty. No items were reported as confusing or unclear, and the average completion time was 10-15 min, indicating good feasibility in this population. The ABC-K scale demonstrated excellent content validity, with an overall CVI of 96.3%, slightly higher than the cross-culturally adapted version (94.4%). Assessment was carried out by an expert panel using a structured content validation form. Item-wise analysis demonstrated 100% agreement for most items, with 90% agreement for items 6, 7, 8, 10, 13, and 15. The high CVI values affirmed the experts’ consensus on the relevance, cultural appropriateness, and clarity of the translated items following adaptation [Table 2].

Close

Figure 2: Mean and standard deviation of ABC-Kannada scale, STDEV: Standard deviation, ABC: Activities-specific balance confidence.

Export to PPT

Reliability analysis

Construct validity was supported through measures of internal consistency and inter-item correlations. The ABC-K demonstrated excellent internal consistency (Cronbach’s alpha = 0.992), with item correlations mostly above 0.80 and many exceeding 0.90. These results strongly suggested that all items measured a single underlying construct, balance confidence. Convergent and discriminant validity using external measures was not examined in the current study. Concurrent validity was demonstrated by the strong correlation between the ABC-K and the original English scale (r = 0.92-0.96, p <0.0001), as illustrated in Figure 3, confirming consistency between versions. Predictive validity was not evaluated, since the present study primarily focused on translation, cross-cultural adaptation, and initial psychometric testing. Future studies using longitudinal designs are required to investigate whether ABC-K scores predict falls or related outcomes.

Close

Figure 3: Item-wise Pearson correlation coefficients (r) for the Kannada version of the Activities-specific Balance Confidence (ABC) scale. All items demonstrated strong positive correlations (r = 0.90–0.97).

Export to PPT

Although formal analyses such as exploratory factor analysis (EFA) or confirmatory factor analysis (CFA) were not undertaken, the scale demonstrated indications of a unidimensional structure. This was strongly supported by the very high internal consistency (Cronbach’s alpha = 0.992) and consistently strong inter-item correlations (generally >0.80, many >0.90). These results suggest that all items reflect a single underlying construct, balance confidence, while also highlighting areas where potential redundancy among items may be considered in future refinement. The ABC-K was designed as a subjective self-report measure of balance confidence rather than as a diagnostic tool. Consequently, indices such as sensitivity and specificity, which are primarily applicable to diagnostic instruments, were not considered necessary in this context. While this study emphasised establishing reliability and validity at a single time point, the ABC-K is conceptually well-suited for detecting meaningful changes in balance confidence over time, such as in response to rehabilitation or fall-prevention interventions. Future longitudinal studies could further explore and validate the scale’s responsiveness as a clinical outcome measure.

Preliminary normative data were generated from the study sample of 50 Kannada-speaking older adults. Descriptive statistics provided initial reference values for this population, though broader standardisation across demographic and functional subgroups was not addressed. These early findings support the scale’s applicability in this community.