OBJECTIVE: There are no disability instruments that have specifically been validated for gout. The aim of this study was to determine the construct validity of the Health Assessment Questionnaire Disability Index (HAQ-DI) in gout and the internal validity using Rasch analysis. METHODS: An observational cohort study of two groups of clinic patients with gout (n=20, n=53), in which clinical and functional measures were correlated with HAQ-DI scores. Rasch analysis was used to determine the internal validity of summated scores as a measure of physical disability. RESULTS: The HAQ-DI items fitted a Rasch measurement model, confirming internal validity of the scale, although there was evidence of disordered thresholds and rescoring items as a 3-option response rather than a 4-option response improved model fit and resolved the disordered thresholds. HAQ-DI scores showed a bimodal distribution and evidence of floor effects. Clinical indices correlated highly with HAQ-DI scores in gout patients, particularly other measures of physical function. A strong relationship between days of sick leave and HAQ-DI was observed in gout patients (r2=0.44, p<0.001). CONCLUSIONS: It is concluded that HAQ-DI has good construct and internal validity in gout but a modified scoring approach better fits a Rasch model.
OBJECTIVE: There are no disability instruments that have specifically been validated for gout. The aim of this study was to determine the construct validity of the Health Assessment Questionnaire Disability Index (HAQ-DI) in gout and the internal validity using Rasch analysis. METHODS: An observational cohort study of two groups of clinic patients with gout (n=20, n=53), in which clinical and functional measures were correlated with HAQ-DI scores. Rasch analysis was used to determine the internal validity of summated scores as a measure of physical disability. RESULTS: The HAQ-DI items fitted a Rasch measurement model, confirming internal validity of the scale, although there was evidence of disordered thresholds and rescoring items as a 3-option response rather than a 4-option response improved model fit and resolved the disordered thresholds. HAQ-DI scores showed a bimodal distribution and evidence of floor effects. Clinical indices correlated highly with HAQ-DI scores in goutpatients, particularly other measures of physical function. A strong relationship between days of sick leave and HAQ-DI was observed in goutpatients (r2=0.44, p<0.001). CONCLUSIONS: It is concluded that HAQ-DI has good construct and internal validity in gout but a modified scoring approach better fits a Rasch model.
Authors: Jasvinder A Singh; Will J Taylor; Lee S Simon; Puja P Khanna; Lisa K Stamp; Fiona M McQueen; Tuhina Neogi; Angelo L Gaffo; Michael A Becker; Patricia A MacDonald; Omar Dabbous; Vibeke Strand; Nicola D Dalbeth; Daniel Aletaha; N Lawrence Edwards; H Ralph Schumacher Journal: J Rheumatol Date: 2011-07 Impact factor: 4.666
Authors: Eveline M Kool; Marieke J Nijsten; Annelies E van Ede; Tim L Jansen; William J Taylor Journal: Clin Rheumatol Date: 2016-06-14 Impact factor: 2.980
Authors: Sarah Stewart; Keith Rome; Alastair Eason; Meaghan E House; Anne Horne; Anthony J Doyle; Julie Knight; William J Taylor; Nicola Dalbeth Journal: Clin Rheumatol Date: 2018-04-21 Impact factor: 2.980
Authors: Peter M ten Klooster; Harald E Vonkeman; Martijn A H Oude Voshaar; Christina Bode; Mart A F J van de Laar Journal: Clin Rheumatol Date: 2014-08 Impact factor: 2.980
Authors: Priyanka Chandratre; Edward Roddy; Lorna Clarson; Jane Richardson; Samantha L Hider; Christian D Mallen Journal: Rheumatology (Oxford) Date: 2013-08-11 Impact factor: 7.580