Details

Autor(en) / Beteiligte

• Norton, Sam
• Done, John
• Sacker, Amanda
• Young, Adam
• Cox, Nigel
• Treharne, Gareth J.
• McGavock, Zoe C.
• Tonks, Anna
• Kafka, Sarah A.
• Hale, Elizabeth D.
• Kitas, George D.
• Fletcher, Debbie
• Sanderson, Tessa
• Baker, Gillian
• Street, Phil
• Hewlett, Sarah
• Stynes, Siobhán
• Peat, George
• Myers, Helen
• Croft, Peter
• Bosworth, Ailsa M.
• Crake, Diane
• Hurley, Mike
• Patel, Anita
• Walsh, Nicola
• Mitchell, Helene
• Kumar, Kanta
• Gordhan, Chandrika
• Situnayake, Deva
• Raza, Karim
• Bacon, Paul
• May, James
• Bingham, Clifton O.
• March, Lyn
• Alten, Rieke
• Pohl, Christoph
• Woodworth, Thasia
• Bartlett, Susan
• Stevenson, Kay
• Roddy, Edward
• Jordan, Kelvin
• Waldron, Nicola
• Brown, Sue
• McCabe, Candy
• McHugh, Neil
• Shelmerdine, Jo
• Ferenkeh-Koroma, Ada
• Breslin, Ann
• Sawyer, Sally
• Haas, Miriam
• Elliott, Barbara
• Law, Rebecca-Jane
• Breslin, Anne
• Oliver, Emily
• Mawn, Lauren
• Markland, David
• Peter, Maddison
• Thom, Jeanette
• Cliss, Alena
• Morris, Marianne
• Ambler, Nicholas
• Knops, Beverly
• Hammond, Alison
• Almeida, Celia

Titel

BHPR: Research [278–290]

Ist Teil von

• Rheumatology (Oxford, England), 2010-04, Vol.49 (suppl-1), p.i139-i143

Ort / Verlag

Oxford University Press

Erscheinungsjahr

2010

Quelle

Oxford Journals 2020 Medicine

Beschreibungen/Notizen

• Background: The Hospital Anxiety and Depression Scale (HADS) is a commonly used measure of psychological distress in patient populations. The HADS was designed to measure two correlated anxiety and depression factors, however previous research is inconclusive finding support for several alternative models. In addition, some items may be biased by the somatic features of a disease. Currently there are no published studies considering item bias in rheumatological patients. The objective of this study was to (1) examine the factor structure of the HADS and (2) assess for the presence of item bias. Methods: The sample consisted of 160 patients attending one of the Early Rheumatoid Arthritis Study centres, for whom individual HADS item responses were available. All completed questionnaires were pooled resulting in a total sample of 1728. Alternative models were compared using confirmatory factor analysis. Item bias was assessed by examining relation between individual HADS items and covariates (disease duration, age, sex, HAQ, DAS and pain) whilst controlling for underlying level of psychological distress. Results: Superior fit was observed for a bifactor structure (χ2(20) = 87, RMSEA = 0.04, CFI = 0.97, TLI = 0.99), indicating that the HADS taps into a general psychological distress factor as well as specific anxiety and depression factors. The HADS anxiety and depression scores correlated highly with the respective specific factor but even more so with the general factor (HADS-A: r(anxiety) = 0.59, r(general) = 0.86; HADS-D: r(depression) = 0.64, r(general) = 0.83). Item bias was observed for item D8 “I feel slowed down” with individuals with same underlying level of psychological distress but worse disability (b = 0.34, P < 0.001) and higher disease activity (b = 0.13, P < 0.001) more likely to respond positively to the item. Further item bias was observed for item D14 “I can enjoy a good book or radio or TV programme” with females less likely to respond positively to the item than males with the same level of psychological distress (b = -0.50, P < 0.001). However, the magnitude of the bias was small and controlling for it made no substantive difference in the association between the covariates and psychological distress factors. Conclusions: The HADS total score may be used as an assessment of general psychological distress, incorporating specific features of anxiety and depression. However, factor scores, which can be calculated using a simple tool provided by the author, may provide greater insight into an individual’s psychological well-being, by allowing for the assessment of anxiety and depression separately. Although item bias was observed, it is unlikely to have a substantive effect. Disclosure statement: All authors have declared no conflicts of interest.