Development of a quality of life outcome measure for vestibular rehabilitation

Abstract

Dizziness is a common complaint accounting for a significant proportion of medical consultations. Patients presenting with dizziness frequently report associated psychological consequences and lifestyle restrictions that have implications for quality of life. Vestibular rehabilitation aims to address the symptoms and consequences of dizziness and in most cases is the appropriate approach to management. Performance measures of outcome are inappropriate in this context and self-report measures are considered the most suitable indicators of rehabilitative success. Self-report measures of dizziness are available but have not been developed specifically to measure rehabilitation outcome and no single measure addresses the full range of dizziness impact. A need is identified for a suitable outcome measure for vestibular rehabilitation. In the area of self-report measures, there is debate about the most appropriate method of measuring change. Some favour a before and after 'state' approach while others favour a direct measure of 'change'. Previous research suggests that 'state' measures may be confounded by response shift bias. A suitable outcome measure for vestibular rehabilitation should use the approach to measuring change that is most appropriate in this context The aim of the present study was to develop a validated and responsive self-report measure of outcome from vestibular rehabilitation. It was considered that an appropriate outcome measure should capture the aspects of dizziness impact that are relevant to patients. It was intended that the measure should also be convenient for routine clinical use. A further aim was to compare 'state' and 'change' methods of measuring self-reported change in the context of vestibular rehabilitation, in tenns of responsiveness and bias. A new questionnaire, the Vestibular Rehabilitation Benefit Questionnaire (VRBQ), was developed over three phases of work focusing on i) generation, ii) refinement and iii) validation, using data from a total of 297 patients. A preliminary list of questionnaire items was generated through qualitative analysis of interview data provided by patients undergoing vestibular rehabilitation. FoIIowing this, a cross-sectional study of patients undergoing vestibular rehabilitation provided quantitative data that allowed further refinement of the preliminary item list. Finally, a longitudinal study of patients undergoing vestibular rehabilitation provided data to investigate the validity and responsiveness of the new questionnaire. Data from the longitudinal study also allowed comparison of' state' and 'change' methods of measuring self reported change. Factor analysis revealed an underlying structure of four subscales measuring Dizziness, Anxiety, Motion Provoked Dizziness and Quality of Life. The four subscales were found to be reliable and the construct validity of the new questionnaire was demonstrated. Effect size estimates obtained from the longitudinal study indicated that the VRBQ was more responsive to change than alternative measures of dizziness impact or health-related quality of life. The psychometric profiles of 'state' and 'change' fonnats suggested that direct 'change' measures were affected by social desirability bias in this context. A patient-driven disease-specific quality of life measure of outcome from vestibular rehabilitation has been developed. The questionnaire, the Vestibular Rehabilitation Benefit Questionnaire, appears to be valid, reliable and responsive. The questionnaire is concise, psychometrically robust, and addresses most areas of dizziness impact, providing a convenient tool for a variety of roles in clinic, research and audit. Findings from the present study provide preliminary evidence that response shift biases are minimal in the context of dizziness and vestibular rehabilitation when using the 'state' approach to measuring change. However, direct 'change' measures may suffer from social desirability bias and are not recommended.

More information

Identifiers

Catalogue record

Export record

ASCII CitationAtomBibTeXData Cite XMLDublin CoreDublin CoreEP3 XMLEndNoteHTML CitationHTML CitationHTML ListJSONMETSMODSMPEG-21 DIDLOpenURL ContextObjectOpenURL ContextObject in SpanRDF+N-TriplesRDF+N3RDF+XMLRIOXX2 XMLReferReference ManagerSimple Metadata