Health outcome measurement is a growth industry. Thousands of behavioural assessment instruments, developed for neurological populations alone, are available for diagnosis, prediction and evaluation of interventions. The task of selecting the best instrument for the purpose at hand is thus a daunting one for the clinician and researcher. Fortunately, there are guides that make the task easier. This presidential address covers three interrelated themes that inform assessment in neurorehabilitation: First, it reviews current concepts and the status of behavioural assessment in neurorehabilitation. It then examines evidence-based clinical practice as applied to assessment of function, along with methods to benchmark the scientific quality of assessment instruments. Finally, the article considers the need to move beyond outcome measurement in the neurorehabilitation setting.

Performance on some neuropsychological tests is best expressed as an intra-individual measure of association (such as a parametric or non-parametric correlation coefficient or the slope of a regression line). Examples of the use of intra-individual measures of association (IIMAs) include the quantification of performance on tests designed to assess temporal order memory or the accuracy of time estimation. The present paper presents methods for comparing a patient's performance with a control or normative sample when performance is expressed as an IIMA. The methods test if there is a significant difference between a patient's IIMA and those obtained from controls, yield an estimate of the abnormality of the patient's IIMA, and provide confidence limits on the level of abnormality. The methods can be used with normative or control samples of any size and will therefore be of particular relevance to single-case researchers. A method for comparing the difference between a patient's scores on two measures with the differences observed in controls is also described (one or both measures can be IIMAs). All the methods require only summary statistics (rather than the raw data from the normative or control sample); it is hoped that this feature will encourage the development of norms for tasks that use IIMAs to quantify performance. Worked examples of the statistical methods are provided using data from a clinical case and controls. A computer program (for PCs) that implements the methods is described and made available. (JINS, 2003, 9, 989–1000.)