Before evaluating behavioral distrubances of dementia (BDD), several factors should be considered. This includes the patient population, the properties of the rating scales to be used, and the assessor. The frequency of behavioral symptoms appears to vary between countries, making comparisons in multinational studies difficult. These variations may be caused by cultural differences, clinicians' attitudes to behavioral symptoms, and interrater variability. The psychometric properties of a rating scale may be affected by differences in clinicians' attitudes in different countries, which may, in turn, alter the frequency of diagnosis of BDD, causing problems in the interpretation of data in multinational studies. Consideration should be given to all these factors before initiating a multinational clinical trial.