Study design: Validation study.
Objective: To investigate the most valid, reliable, and comprehensible response scale for spinal trauma patients to compare their current level of function and health with their preinjury state.
Summary of background data: In the context of a main project of the AOSpine Knowledge Forum Trauma to develop a disease-specific outcome instrument for adult spinal trauma patients, the need to identify a response scale that uniquely reflects the degree to which a spine trauma patient has returned to his or her preinjury state is crucial.
Methods: In the first phase, 3 different question formats and 3 different response formats were investigated in a questionnaire, which was administered twice. Based on the results of the first phase, in the second phase, a modified questionnaire was administered once to a second group of patients to investigate 5 different response formats: 0-10 Numeric Rating Scale-11, 0-100 Numeric Rating Scale-101, Visual Analogue Scale, Verbal Rating Scale, and Adjective Scale. All patients were interviewed in a semistructured fashion to identify their preferences. Multiple statistical analyses were performed: test-retest reliability, internal consistency, and discriminant validity.
Results: Twenty eligible patients were enrolled in the first phase and 59 in the second phase. The initial phase revealed the highest preference for 1 specific question format (60.0% and 86.7% after the first and second administration of the questionnaire, respectively). The second phase showed the Verbal Rating Scale as the most preferred response format (35.6%). The semistructured interviews revealed that overall, a subgroup of patients preferred a verbal response format (42.4%), and another group of patients preferred a numerical response format (49.1%). The statistical analysis showed good to excellent psychometric properties for all formats.
Conclusion: The most preferred question and response formats were identified for use in a disease-specific outcome instrument for spinal trauma patients.
Level of evidence: 3.