Cross-cultural adaption, translation and validation of the Toronto extremity salvage score (TESS) for patients in German-speaking countries

Patients from the Orthopedic Sarcoma Outpatient Clinic of the Department of Orthopedics and Trauma Surgery Vienna were asked to participate in this study from October 2018 to April 2019. Patients were asked to participate if they met the following inclusion criteria: i) 18 years or older; ii) diagnosed with sarcoma of the upper or lower extremities; iii) at minimum 3 months post-limb salvage surgery; and iv) no disease recurrence or other serious disease in the past. Patients unable to fill out the questionnaire on their own, e.g. did not speak German, were not asked to complete the questionnaire. Both the TESS and the SF-36 were administered during the waiting time in a separate, quiet room to preserve privacy and create a calm atmosphere. After finishing, patients were asked to fill out the TESS test again alone at home and send it back to the clinic. In summary, 50 patients completed the questionnaires, 18 for the upper extremity and 32 for the lower extremity. Response rates for the at-home evaluation were 77.8% (n = 14) and 62.5% (n = 20) for the upper and lower extremity groups, respectively. If a questionnaire was not returned on time, the authors called the patient and sent the questionnaire again; after the second reminder and no response or a negative response, it was noted in the source data that the patient did not or would not send back the questionnaire. Primarily the validation and reliability of the test was examined, so the authors decided not to differ between the evaluation time points asked on the title page of the TESS.

Five fundamental steps of translation and adaption were carried out for the intercultural adjustment process (Fig. 1; [3, 4]): (1) translation of the original version into the target language; (2) synthesis of both translations; (2a) linguistic feedback; (3) back translation by an independent translator; (4) discussion and approval by the authors and (5) pretesting and evaluation. In the first step (1), two independent persons translated the original English language TESS, resulting in two separate transcripts with a third for comparison. The translators were German native speakers fluent in English with university level experience in the language: a sociologist (CT), a medical student (CH), and a professor of statistics and outcome measures (TS, the comparison version) [9]. Second (2), the orthopedic surgeon (GH), who was not included in the initial process, merged the two translations. Between steps two and three (2a), two German-speaking individuals were interviewed for linguistic feedback. Third (3), the back translation was performed by a blinded independent translator without a medical background and with no further interaction in this process. Fourth (4), the committee, composed of the authors—surgeon (GH), patient reported outcome measurement scores (PROMS) scientist (TS), sociologist (CT), and medical student (CH)—and a translation expert, discussed all transcripts, translations and comments, and decided on the final wording of the items. These were then used for the pretest phase (5). Six patients with sarcoma in the upper or lower extremities were randomly selected then interviewed by two authors; one conducted the interview and the other observed and took notes. The direct translations for two questions were too long and confusing for the pretest patients to answer without support from the interviewer [15‐17]. Therefore, these questions (UE 25, 26 and LE 26, 27) were changed so the TESS can be completed by patients themselves without any assistance from or explanation of questions by healthcare providers (12–14). Furthermore, because distinct transcripts from two authors (CH and CT) were merged into one final version during the translation process, minor changes were made to most questions. See Table 1 for further details: “wording shortened”—the length of the question was reduced; “word changed”—the wording was simplified; “sentence structure”—the sentence structure was changed; and “equal”—the translations for upper and lower extremities were equivalent and the question was the same in the upper and lower TESS German versions. Lastly, Swiss-German and German-German native speakers were involved to minimize language barriers and adjust for different dialects. This process resulted in the final version of the German TESS. The SF-36 was used to anchor the validity assessment.

Spearman rank correlation coefficients were calculated between the SF-36 dimension summary scale scores and the TESS scores. The SF-36 is a commonly used instrument for evaluating aspects of health-related quality of life and functioning [13, 18, 19].

Cronbach’s alpha was used to determine internal consistency [14]. For the test-retest analysis, the intraclass correlation coefficient (ICC) was calculated between baseline and repetition for responses to each question and for the total TESS score [20]. The sample size calculation was based on the reliability analysis: for an expected ICC of 0.89, a confidence interval of 0.23, and with 2 measurements per individual, a sample size of 14 individuals was required [5‐11]. Bland-Altman plots were used to visually assess the fluctuation range of the deviations [21].