Psychometric validation of a pre-existing questionnaire used to measure patient satisfaction following orthodontic treatment in a UK population

Abstract

Objective: To determine the psychometric properties of a pre-existing orthodontic satisfaction questionnaire and to assess patient satisfaction of their orthodontic treatment. Design: Psychometric assessment of a pre-existing orthodontic satisfaction questionnaire. Setting: Two sites at King’s College Hospital NHS Foundation Trust. Participants: Individuals aged 12–15 years who had completed fixed orthodontic treatment. Methods: An expert panel assessed a pre-existing orthodontic treatment satisfaction questionnaire for validity and readability. As a result, revisions were made to the questionnaire and it was distributed to 103 participants aged 12–15 years, on completion of their fixed orthodontic treatment (T1). Test–retest reliability of the questionnaire was assessed through completion of a second questionnaire by 17 participants, at a two-week interval (T2). The questionnaire was assessed for reliability using item-total correlations (I-TC) and Cronbach’s alpha. Factor analysis allowed exploration of the underlying factor structure of the questionnaire. Test–retest reliability was assessed using Cohen’s kappa coefficient. Multiple regression analysis was used to find out the significant demographic variables that predicts the satisfaction level of treatment. Results: Following validity and readability assessment by the expert panel, the questionnaire was revised. Eleven items were removed following item analysis (with I-TC ⩽ 0.3). Factor analysis was deemed uninterpretable. The overall scale demonstrated greater reliability than the underlying sub-scales. Therefore, the sub-scales were removed, resulting in one scale which assessed overall orthodontic satisfaction, comprising 37 items, with a Cronbach alpha of 0.92. The test–retest reliability of the revised 37-item scale was deemed to be poor (kappa coefficient = 0.39). Multiple regression analysis identified ethnicity as a significant predictor of orthodontic treatment satisfaction (P = 0.01). Conclusion: This study provides a valid measure to assess orthodontic treatment satisfaction for use in a UK population aged 12–15 years on completion of fixed orthodontic treatment.