Evaluation of a German version of the tonsil and adenoid health status instrument

Abstract

Background: To create and validate a German version of the Tonsil and Adenoid Health Status Instrument (TAHSI) for evaluation of tonsillectomy outcome in adult patients with chronic or recurrent tonsillitis.

Subjects and methods: 46 healthy volunteers were assessed twice in a 6 week interval with the TAHSI questionnaire. Their results were compared with 45 patients suffering from chronic tonsillitis before tonsillectomy and 6 months following surgery. For internal consistency, Cronbach's alpha was calculated; to identify normal score values, the optimum cutoff between healthy and diseased individuals was identified with receiver operating characteristic analysis; and responsiveness was assessed using Guyatt's Responsiveness Index (GRI).

Results: Cronbach's alpha for all questions was 0.92. Test- retest intraclass correlation coefficient was 0.89 (95% confidence interval 0.80-0.94 p < 0.001). Mean score for the healthy individuals was 7.0 (95% confidence interval 4.2-9.7). The optimum cut off score between healthy and diseased was 20 with a sensitivity of 80% and a specificity of 90% to differentiate controls from tonsillectomy patients.

Conclusion: The TAHSI performed well in this validation tests and is considered a favorable instrument to evaluate the effectiveness of tonsillectomy in adults with chronic or recurrent tonsillitis.

Keywords: Chronic tonsillitis; Cronbach’s alpha; Guyatt’s responsiveness Index; Quality of life; Receiver operating characteristic analysis; Test-retest reliability.

Figure 1

Box and whiskers plot illustrating TAHSI scores of healthy controls and scores of patients: box and whiskers plot illustrating TAHSI scores of healthy controls (n = 46) compared with the scores of patients planned for tonsillectomy for chronic tonsillitis (n = 45). Bar: median; boxes: quartiles; whiskers: Tukey hinges; circles: outliers; asterisks: far outliers; Mann–Whitney test: p < 0.001).

Figure 2

Receiver operating characteristic (ROC) curve: receiver operating characteristic (ROC) curve illustrating the true positive rate (sensitivity; y-axis) vs. the false positive rate (1-specificity; x-axis) at increasing TAHSI-score values. The area under the ROC curve was 0.94 (95% CI 0.89 to 0.99; p < 0.001). At a cut-off score of 20 (circle), sensitivity to detect members of the tonsillectomy group with chronic or recurrent tonsillitis was 80%, the specificity 90%.

Figure 3

Box and whiskers plot illustrating TAHSI scores following tonsillectomy: box and whiskers plot illustrating TAHSI scores of patients with chronic tonsillitis (n = 40) before and 6 months following tonsillectomy. (Wilcoxon signed rank test p < 0.001).

Figure 4

Mean scores (minimum 0; maximum 4) of the 9 subscales of the modified TAHSI questionnaire: the scores of the 2 questions of each subscale per participant were averaged and then the mean of the sample was calculated. White bars: control group; grey bars: patients scheduled for tonsillectomy Error bars represent 95% confidence intervals (n.d. = no data).