Reliable change indices for the ruff 2 and 7 selective attention test in older adults.

The assessment of suspected dementia often involves the analysis of change scores from neuropsychological tests administered on two occasions. Frequently, no information is available to allow the significance of a change to be evaluated. Our aim was to prepare data useful for estimating the statistical significance of retest changes for persons older than the age of 65 on the Ruff 2 and 7 Selective Attention Test, one of the few paper-and-pencil tests of attention available to the clinician. The sample tested was composed of 234 healthy adults taking part in a randomized double-blind trial of the effect on cognitive performance of lowering homocysteine using dietary supplements, during which no treatment effects were detected. The test was found to be sensitive to the effects of aging and was well tolerated by our sample of older adults. Retest reliability for the speed of visual search for digit targets (2 and 7) under both the same category (other digits) and different category (letter) distracter conditions was high, and the practice effects during a 12-month period were substantial. Test-retest data with a 1-year interval were used to estimate reliability coefficients and to calculate Reliable Change indices useful for evaluating persons with suspected dementia. The rate of decline in search speed as a function of age was the same for both distraction conditions, suggesting that older persons used a controlled serial search process for targets in both conditions of this test.