Age-stratified norms for Raven's standard progressive matrices for Sri Lankan adults.

OBJECTIVE: The aim of this study was to create age-stratified norms for the Raven's Standard Progressive Matrices (SPM) for Sri Lankan adults. METHODS: A sample of 610 adults (age: 18-72 years; education: 1-19 years), underwent the 60-item version of the SPM under individual supervision of a test administrator. The sample was stratified into 5-year age bands, and the norms are presented as percentile tables and percentile curves. RESULTS: The age-related changes were more accurately predicted by a curvilinear model (overall R2 = 0.961) than a linear regression model (R2 = 0.639). The SPM norms are presented as age-stratified percentile tables, as well as sex-, age- and education-adjusted multiple regression equations. The highest percentiles in the younger end of the age spectrum showed a ceiling effect. In the context of age-stratified US (1993) and British (1992) norms, older individuals in the Sri Lankan sample scored much lower than their Western counterparts. However, the difference narrowed in the younger age bands, showing no difference among the 18-to-22-year age bands in the three countries. CONCLUSIONS: This age-by-country interaction can be partly explained by poorer education in the older individuals in the present sample compared to those in the US and UK standardization samples. SPM norms presented in this paper fill a hiatus in assessment of general intellectual ability in Sri Lankan adults. Given that Sri Lanka improves its educational, socioeconomic and health standards faster than the nations who have already reached higher standards, these norms would require re-standardization in the coming decades.

Psychometric properties of the Sinhala version of the Swanson, Nolan, and Pelham rating scale (SNAP-IV) Parent Form in healthy children and children with ADHD.

OBJECTIVE: The present study aimed to translate and adapt the Parent Form of the MTA Version of the Swanson, Nolan, and Pelham ADHD/ODD rating scale (SNAP-IV) into Sinhala and examine its psychometric properties. METHOD: The MTA version of the SNAP-IV was translated into the Sinhala language by adhering to the WHO translation and adaptational process. The final Sinhala SNAP-IV (S-SNAP-IV) was administered to parents of 412 healthy children and adolescents (age: 6-18 years), and 272 children and adolescents with ADHD (age: 4-19 years). Internal consistency and factor structure of the S-SNAP-IV were determined in each sample. RESULTS: Conforming to that of the original English version, the S-SNAP-IV showed a three-factor structure (inattention, hyperactivity/impulsivity, and oppositional defiant disorder) in each sample, except for three item deviations (10,11, and 13) in the healthy sample. The S-SNAP-IV showed excellent internal consistency in the total sample (Cronbach's alpha = 0.97), healthy sample (Cronbach's alpha = 0.90), and the clinical sample (Cronbach's alpha = 0.94). The S-SNAP-IV also showed good discriminative validity: Only 1.5% of the healthy sample exceeded the cut-off scores in any domain, whereas all newly diagnosed, unmedicated children were S-SNAP-IV positive in at least one domain. CONCLUSIONS: The findings suggest that the three-factor model fits the S-SNAP-IV, and it clearly distinguishes children with ADHD from non-ADHD children. We conclude that the S-SNAP-IV possesses satisfactory psychometric properties-consistent with the MTA English version and its translations into other languages-making it a reliable and valid instrument for screening children with ADHD.

Regression-based, demographically adjusted norms for Victoria Stroop Test, Digit Span, and Verbal Fluency for Sri Lankan adults.

OBJECTIVE: Lack of test norms is a major limitation in neuropsychological assessment in many non-English-speaking countries. Our objective was to generate sex-, age-, and education-adjusted, regression-based norms for three pen-and-paper-based executive function/working memory tests, viz. Victoria Stroop Test (VST), Digit Span, and Verbal Fluency, for Sinhala-Speaking Sri Lankan adults. METHOD: Six-hundred and six healthy, community-living, Sinhala-speaking adults (273 men) aged 19-83 years completed VST, Digit Span (forward and backward), and Phonemic (Sinhala phonemes "N," "P," and "S") and Category (animals and vegetables) Fluency tests. We conducted multiple linear regression analyses with sex, age, and years of education fitted as predictors to model the test outcome variables. RESULTS: Younger age was associated with better performance in all tests except in Phonemic Fluency. Longer education was associated with better performance in all outcomes except VST errors and interference. Women had a significant but small advantage over men in VST Neutral and Color Word tests, and Phonemic Fluency. We report regression equations to predict norms for the test performance measures based on sex, age, and years of education; and the test variances accounted by these factors. CONCLUSIONS: We propose demographically adjusted norms for VST, Digit Span, and Verbal Fluency for Sinhala-speaking Sri Lankans aged 19-83 years, and supplement the regression formulae with a calculator that produces predicted and standard scores. These norms would help in interpreting the results of clinical samples in the future, taking into account the variability introduced by sex, age, and education.

Sex-, age- and education-adjusted norms for the WHO/UCLA version of the Rey Auditory Verbal Learning Test for Sinhala-speaking Sri Lankan adults.

OBJECTIVE: The aim of this study was to create sex-, age- and education-adjusted norms for the WHO/UCLA version of the Rey Auditory Verbal Learning Test (RAVLT) for Sinhala-speaking Sri Lankan adults. METHODS: Five-hundred and sixty-one healthy, community-living adults (252 men), aged 19-83 years, and had 0-23 years of education completed the WHO/UCLA RAVLT in Sinhala language. We conducted multiple linear regression analyses with sex, age and years of education to predict RAVLT list A1-A5 individual trial scores; trials A1-A5 total learning; list B score; immediate and delayed recall and recognition trial scores; and retroactive interference. RESULTS: We report regression equations to predict RAVLT norms based on sex, age and years of education; and the test variances accounted by those variables. Accordingly, all measures, except retroactive interference had a significant age-related decline. All measures, except the recognition trial hits, significantly improved with more years of education. Women had significantly higher scores in all measures except in trial B and retroactive interference. Proactive interference, learning rate, learning over trials were not associated with sex, age or education. A confirmatory factor analysis loaded the RAVLT outcome measures into two factors: acquisition and retention. CONCLUSIONS: We report sex-, age- and education-adjusted WHO/UCLA RAVLT norms for Sinhala-speaking Sri Lankans aged 19-83 years; and supplement the regression formulae with a calculator that produces predicted and standard scores for given test participant. These norms would help clinicians accurately interpret individual test results, accounting for the variability introduced by sex, age and education.

Sex-, age-, and education-adjusted norms for Cambridge Neuropsychological Test Automated Battery in literate Sri Lankan adults.

Objective: Neuropsychological test batteries validated for Sri Lankan population are extremely scarce. Cambridge Neuropsychological Test Automated Battery (CANTAB) is a language-independent test battery used in many countries, but the original UK norms may not be representative in the local setting due to sociocultural differences. Our aim was to generate age-, sex-, and education-adjusted norms for the CANTAB for Sri Lankan adults.Method: Three-hundred and eleven healthy, community-living adults aged 20-64 years completed CANTAB subtests of visual attention (Reaction Time and Rapid Visual Information Processing), visual memory (Paired Associates Learning and Delayed Matching to Sample); and executive functions (Stockings of Cambridge, Stop Signal Task, and Spatial Working Memory). We conducted multiple linear regression analyses with sex, age and years of education fitted as predictors to model the CANTAB outcome variables.Results: Younger age and longer education were associated with better performance in most CANTAB measures. Men outperformed women in few measures of psychomotor speed and visuospatial skills, but sex explained only a small proportion of test variance. We report regression equations to predict CANTAB norms based on sex, age and years of education; and the test variances accounted by these factors.Conclusions: We propose sex-, age- and education-adjusted CANTAB norms for Sri Lankans aged 20-64 years and supplement the regression formulae with a calculator that produces predicted and standard scores of a given test subject. These norms would help in interpreting the results of clinical samples in future studies, taking into account the variability introduced by sex, age and education.

Prevalence of major depressive disorder among spouses of men who use alcohol in a rural community in Central Sri Lanka.

AIMS: To estimate the prevalence of major depressive disorder (MDD) among spouses of men who use alcohol in two rural areas in Sri Lanka, and to examine whether the severity of alcohol-related problems (ARPs) in men and presence of alcohol-related domestic violence are associated with MDD among these women. METHOD: In a cross-sectional study, ARPs among men were assessed using Alcohol Use Disorders Identification Test (AUDIT) questionnaire filled in by men, and domestic violence and husbands' drinking pattern data obtained from the women. MDD among the women was ascertained using the Structured Clinical Interview for DSM-IV Disorders for major depression. Using logistic regression we examined whether age, past history of depression, different indices of ARPs and domestic violence were associated with current MDD among the women. RESULTS: Point prevalence of MDD in the sample was 33.3% (95% CI: 25.93, 40.73%). Once adjusted for other factors, morning drinking of the spouse (odds ratio = 4.11, 95% CI: 1.25, 13.47; P = 0.019) and increasing age (odds ratio = 1.05, 95% CI: 1.01, 1.09; P = 0.003) significantly increased the odds of MDD. Being subjected to domestic violence/arguments also had a trend to be associated with MDD among women, but was not significant (odds ratio = 2.29, 95% CI: 0.95, 5.48; P = 0.062). CONCLUSION: The prevalence of MDD among spouses of men who use alcohol is markedly higher than that has been observed among Sri Lankan women in previous studies. The prevalence of MDD in women seems to increase when their husbands are morning drinkers, and with increasing age.