Development and validation of a questionnaire assessing household work limitations (HOWL-Q) in women with rheumatoid arthritis

Abstract

Introduction Rheumatoid arthritis (RA) has female preponderance and interferes with the ability to perform job roles. Household work has 2 dimensions, paid and unpaid. There is not a validated instrument that assesses the impact of RA on limitations to perform household work. We report the development and validation of a questionnaire that assesses such limitations, the HOWL-Q. Methods The study was performed in 3 steps. Step-1 consisted on HOWL-Q conceptual model construction (literature review and semi-structured interviews to 20 RA outpatients and 20 controls, household workers, who integrated sample (S)-1). Step-2 consisted of instructions selection (by 25 outpatients integrating S-2), items generation and reduction (theory and key informant suggestions, modified natural semantic network technique, and pilot testing in 200 household workers outpatients conforming S-3), items scoring, and questionnaire feasibility (in S-3). Step-3 consisted of construct (exploratory factor analysis) and criterion validity (Spearman correlations), and HOWL-Q reliability (McDonald's Omega and test-retest), in 230 household work outpatients integrating S-4. Results Patients conforming the 4 samples were representative of typical RA outpatients. The initial conceptual model included 8 dimensions and 76 tasks/activities. The final version included 41 items distributed in 5 dimensions, was found feasible and resulted in 62.46% of the variance explained: McDonald's Omega = 0.959, intraclass-correlation-coefficient = 0.921 (95% CI = 0.851-0.957). Moderate-to-high correlations were found between the HOLW-Q, the HAQ, the Quick-DASH and the Lawton-Brody index. HOWL-Q score ranged from 0 to 10, with increasing scores translate into increase limitations. Conclusion The HOWL-Q showed adequate psychometric properties to evaluate household work limitations in women with RA.