Psychometric evaluation of the Hospital Anxiety an

Authors: YaFen Chan Doris Y P Leung Daniel Y T Fong ChiMing Leung Antoinette M Lee

Publish Date: 2010/04/07

Volume: 19, Issue: 6, Pages: 865-873

PDF Link

Abstract

The Hospital Anxiety and Depression Scale HADS is widely used in adult populations however its usefulness with adolescents has been explored less This study sought to evaluate the reliability validity and factor structure of the Chinese version of HADS in a community sample of adolescents residing in Hong KongA prospective cohort of 5857 students recruited from 17 secondary schools completed the HADS Internal consistency and concurrent validity were examined Confirmatory factor analysis was applied to test the relative fits of six factor structures of the HADS The best fitting model was further crossvalidated by male female splithalf samples and age subgroupsThe HADS possessed adequate internal consistency especially for the anxiety subscale Significant concurrent intercorrelations with selfreported suicidal thoughts and the Youth Self Report Anxious/Depressed subscale were discovered and found to be stronger for females The crossvalidation supported a twofactor model where anxiety item 7 “I can sit at ease and feel relaxed” was placed in the depression subscaleThe HADS showed satisfactory psychometric properties as a screening instrument in assessing anxious and depressive states as two correlated but distinct factors in adolescents Study implications and recommendations for future research were discussedOnset of anxiety and depression often occurs in adolescence 1 2 3 and episodes of these disorders are likely to persist into adulthood and account for subsequent manifestations of psychosocial and health adversities 4 5 6 7 Yet many youths suffering from anxiety and depression were left unidentified and only a small proportion has received mental health care 8 9 10 This may be attributable in part to the lack of a suitable instrument that could be used in community and school settings to screen for anxiety and depressive symptoms in adolescents Therefore a brief and feasible selfrated screener that can be administered easily is essential to assist with service referral and further indepth assessment and aid for intervention and prevention effortsThe Hospital Anxiety and Depression Scale HADS is a 14item selfadministered instrument developed by Zigmond and Snaith 11 to measure anxiety and depression symptomatology During the past decades the HADS has been extensively validated in a variety of adult populations including clinical and community samples with welldocumented good measurement properties The good reliability and stability of the HADS were also demonstrated in various translated versions across culturally diverse groups eg 12 13 14 15 16 17 18 19 20 However its usefulness in adolescents remains underresearched To our knowledge there have been only two studies that addressed this issue 21 22 Furthermore there have been mixed results in adult studies that support the factor structure in the HADS that underpins the clinical significance of assessing anxiety and depression as two distinct dimensions 13 16 20 23 24 25 26 27 28 For example several studies reported a tridimensional structure in the HADS 13 16 23 24 25 26 While White et al 22 reported acceptable validity and two underlying factors for use with adolescents further examination is needed Additionally although the validation of the Chinese version was reported in several adult studies 15 16 20 26 its psychometric properties in adolescents have not been examined Thus the current study sought to assess the usefulness of the HADS by examining its internal reliability concurrent validity and factor structure using a large community sample of adolescents in Hong KongData were derived from a baseline survey as part of a multiwave schoolbased survey on the development of high risk behaviors among secondary school youths in Hong Kong Seventeen schools were selected each representing a juridical district in Hong Kong except for the schools in the outlying islands The selfreported baseline questionnaire survey was conducted during the second half of 2004 and early 2005After obtaining written informed consent from both students and their parents students were instructed to complete the questionnaire The survey was conducted by a trained research assistant in a classroom setting with no teachers present Students were assured that teachers and school principals had no access to their responses and data were kept strictly confidential and used only for research purposes Students who could not speak and read Chinese were excluded Of the 6926 questionnaires collected 86 consent rate 875 were invalid 82 did not provide age information or were out of the target age range age 10–19 and 112 did not complete the information on the HADS As such the sample for the current study comprised 5857 adolescents Among them 554 were females and the average age was 134 years standard deviation SD = 12 with distribution as follows age 10–11 02 age 12 234 age 13 319 age 14 279 age 15 12 age 16 33 and age 17–19 13 In particular 418 were in the 7th grade 348 in the 8th grade and 234 in the 9th grade Most youths lived with both parents 90 Approximately 75 and 81 reported their father’s and mother’s education was equal to or beyond a high school degree respectively The study was approved by the Institutional Review Board of the University of Hong Kong

Keywords: