Psychometric Evaluation of the Hungarian Version of Oxford Capabilities - Mental Health Questionnaire
Author(s)
Balázs P1, Brodszky V2
1Corvinus University of Budapest, Doctoral School of Business and Management & Department of Health Economics, Budapest, PE, Hungary, 2Corvinus University of Budapest, Budapest, PE, Hungary
Presentation Documents
OBJECTIVES: Our aim is to evaluate the validity and reliability of the Hungarian version of Oxford CAPabilities - Mental Health questionnaire (OxCAP-MH).
METHODS: A cross-sectional study, with online self-administrated data collection was carried out among Hungarian adult general population in 2021 July. Central tendencies of the 16-item OxCAP-MH was measured as means, standard deviations, variance, skewness, ceiling and floor effect. Methods of classical test theory were applied to evaluate construct and convergent validity given as items correlation and correlation with other health measurement instruments: ICECAP-A, PHQ-9, GAD-7, SWLS. Reliability was measured by Cronbach’s alpha complemented with Cronbach’s alpha if item deleted. Agreement measurement was done by two-way random intraclass correlation coefficient (ICC) of the items.
RESULTS: Overall, N=2000 responded the questionnaire. The sample population had a mean age of 46.3 years, majority being female (n=1145), most completed secondary education (n=909). Item central tendency measures show that mean item scores ranged between 2.8-4.3, lowest at item 9, highest at item 7. The mean standardized OxCAP-MH score was 65.5 (SD:14.4), the median was 69.5 (IQR: 57.8-79.7) in the total sample. Elevated ceiling effect (38.7%-59.4%) was observed at items: 11, 3, 13, 1, 11, 8 and 7 (in ascending order). Item 9 - that is first out of eight likert matrix question of the measurement tool - showed outlying floor effect (14.6%). Item 9 did not correlate with items 1,7,8, other items showed significant slight to moderate correlation (p<0.01; r=0.08-0.62). OxCAP-MH score moderately correlated with ICECAP-A, PHQ-9, GAD-7 (p<0.01; r=0.62, 0.61, 0.58) while showing no correlation with SWLS. The Cronbach’s α test (0.848) revealed high level of internal consistency, none of the items deleted would increase reliability. ICC suggests strong absolute agreement (0.817).
CONCLUSIONS: Validity and reliability assessment of the Hungarian OxCAP-MH questionnaire confirms the good psychometric properties of the instrument.
Conference/Value in Health Info
Value in Health, Volume 25, Issue 12S (December 2022)
Code
EPH195
Topic
Epidemiology & Public Health
Topic Subcategory
Public Health
Disease
SDC: Mental Health (including addition)