ABSTRACT
BACKGROUND: The Geriatric Depression Scale (GDS) is widely used for screening and assessment of major depressive disorder (MDD). Screening scales are often culture-specific and should be evaluated for item response bias (synonymously differential item functioning, DIF) before use in clinical practice and research in a different population. In this study, we examined DIF associated with age, gender, ethnicity and chronic illness in a heterogeneous Asian population in Singapore. METHODS: The GDS-15 and Structured Clinical Interview for DSM-IV diagnosis of MDD were independently administered by interviewers on 4253 non-institutionalized community living elderly subjects aged 60 years and above who were users of social service agencies. Multiple Indicator Multiple Cause latent variable modelling was used to identify DIF. RESULTS: We found evidence of significant DIF associated with age, gender, ethnicity and chronic illness for 8 items: dropped many activities and interests, afraid something bad is going to happen, prefer staying home to going out, more problems with memory than most, think it is (not) wonderful to be alive, feel pretty worthless, feel (not) full of energy, feel that situation is hopeless. LIMITATIONS: The smaller number of minority Indian and Malay subjects and the self-report of chronic medical illnesses. CONCLUSIONS: In a heterogeneous mix of respondents in Singapore, eight items of the GDS-15 showed DIF for age, gender, ethnicity and chronic illness. The awareness and identification of DIF in the GDS-15 provides a rational basis for its use in diverse population groups and guiding the derivation of abbreviated scales.
Subject(s)
Depressive Disorder, Major/diagnosis , Depressive Disorder, Major/ethnology , Aged , Asian People/statistics & numerical data , Chronic Disease , Culture , Diagnostic and Statistical Manual of Mental Disorders , Factor Analysis, Statistical , Female , Humans , Male , Middle Aged , Severity of Illness Index , Singapore/epidemiology , Social Work/statistics & numerical dataABSTRACT
INTRODUCTION: We compared the salivary immunoglobulin A (IgA) and lysozyme concentration and secretion rates among mild and severe psoriasis patients and controls in Singapore. MATERIALS AND METHODS: Fifty-one psoriasis patients and 24 controls participated in the study. None of the patients were on immunosuppressive therapy. The Psoriasis Area and Severity Index (PASI) was used to assess the severity of psoriasis. Patients were divided into mild and severe groups by the median PASI score. Each subject contributed a 5-minute unstimulated salivary sample. Enzyme-linked immunosorbent assay method was used to determine the salivary IgA and lysozyme levels. RESULTS: Psoriasis patients had lower concentration and secretion rate of IgA (geometric mean [GM], 97.5 micro g/mL and 32.3 micro g/min) and lysozyme (GM, 127.6 micro g/mL and 42.1 micro g/min) than controls (IgA GM 256.3 micro g/mL, 79.1 micro g/min; lysozyme GM 180.9 micro g/mL, 55.8 micro g/min) [P = 0.000 (IgA concentration), P = 0.000 (IgA secretion rate), P = 0.015 (lysozyme concentration) and P = 0.150 (lysozyme secretion rate)]. However, no significant differences were observed between mild and severe patients for both IgA and lysozyme concentrations and secretion rates. PASI score showed negative, but non-significant, correlations with either log salivary IgA (r = -0.22, P = 0.13) or log lysozyme (r = -0.09, P = 0.53) secretion rates. CONCLUSION: Psoriasis patients had lower concentrations and secretion rates of salivary IgA and lysozyme compared to controls. However, among patients, the salivary IgA and lysozyme levels are variable and not related to severity of psoriasis.
