Turkish Version of the Ankle Fracture Outcome of Rehabilitation Measure (A-FORM-TR): A Multi-Centre Validation Study (A-FORM-TR)

Background and Rationale: Ankle fractures are among the most common lower-limb injuries in orthopedic practice. Patient-reported recovery, including physical function, sleep, mood, and social participation, is not adequately captured by traditional clinician-based outcome measures such as the AOFAS Ankle-Hindfoot Score or the Olerud-Molander Ankle Score. The A-FORM, developed by McPhail et al. (2014), is a 15-item, Rasch-calibrated patient-reported outcome measure designed specifically to evaluate the rehabilitation experience after ankle fracture. No validated Turkish-language version currently exists.

Methodological Framework: The study follows the eight-stage cross-cultural adaptation and validation protocol described by Cruchinho et al. (2024), which incorporates the recommendations of Beaton et al. (2000) and the COSMIN framework (Mokkink et al., 2010). Written permission to translate and use the instrument has been granted by Prof. Steven McPhail (Queensland University of Technology).

Translation and Adaptation: Forward translation is carried out by two independent bilingual translators (one faculty member from the Department of English Language and Literature, providing linguistic expertise; one bilingual orthopedic surgeon, providing clinical content expertise). A reconciliation committee produces a synthesised Turkish version, which is then back-translated by two independent translators blinded to the original. A harmonisation expert committee compares the source, forward, and back versions for semantic, idiomatic, experiential, and conceptual equivalence. The harmonised pre-test version is evaluated through cognitive debriefing interviews with 10 to 15 ankle-fracture patients before field testing.

Field Testing: In the field testing phase, 150 adult patients with a radiographically confirmed unilateral ankle fracture (Weber A, B, or C; treated surgically by open reduction and internal fixation or conservatively with cast, walker boot, or functional brace) are enrolled across three orthopedic centres. Patients between 3 weeks and 12 months post-injury are eligible. The A-FORM-TR is administered together with the AOFAS Ankle-Hindfoot Score and the OMAS-TR. A sub-cohort of approximately 50 patients is re-administered the A-FORM-TR at 7 to 14 days, with a Global Rating of Change item used to identify clinically stable patients for the test-retest reliability analysis.

Statistical Analysis Plan: The analysis plan was locked before any field data were collected; all decision rules are pre-specified. Developed in consultation with the institutional biostatistics team. Internal consistency is assessed by Cronbach's alpha (target greater than or equal to 0.80). Test-retest reliability is assessed by the intraclass correlation coefficient on stable patients, reported with the K-index (greater than or equal to 0.70 acceptable for group-level comparison, greater than or equal to 0.90 desirable for individual-level decisions). Measurement error is evaluated separately from reliability as the standard error of measurement and the smallest detectable change. Rasch measurement analysis is the primary framework and examines item fit, threshold ordering, person separation reliability, person-item targeting, local independence, unidimensionality, and differential item functioning by sex, age group, treatment modality, recruiting centre, and time since injury; pre-specified decision rules govern the handling of item misfit, disordered thresholds and differential item functioning. Confirmatory factor analysis provides a complementary structural test of the prespecified single-factor model, with exploratory factor analysis run as a supplementary check. Construct validity is assessed via pre-specified hypotheses about correlation patterns with the AOFAS and OMAS-TR, and by known-groups comparison across time-since-injury strata. Criterion validity is assessed in a bilingual sub-study of approximately 30 participants completing both the A-FORM-TR and the original English-language A-FORM in counterbalanced order, with agreement quantified by correlation, intraclass correlation and Bland-Altman analysis. Responsiveness is assessed at 3 months against a Global Rating of Change anchor using the standardised response mean (SRM), effect size and receiver operating characteristic (ROC) analysis, with the minimal important change compared against the smallest detectable change. Cross-cultural measurement invariance is examined against the original English calibration, conditional on the source data available from the instrument developer. Content validity is assessed via item-level and scale-level Content Validity Index and the Kappa coefficient from the expert committee. Floor and ceiling effects are examined at the scale level. Missing data are reported per item; the primary analysis is complete-case, with multiple imputation as a sensitivity analysis where item-level missingness exceeds 5 percent.