To study the usefulness of standard cardiovascular autonomic reflex tests in the elderly, 224 healthy controls and 49 aged diabetic patients were examined. Based on the data obtained from healthy controls, age-related reference values for several autonomic indices were calculated and their usefulness was tested with aged diabetic patients. It was found (1) that in elderly subjects (aged > or = 50 years) the indices based on heart rate differences are more suitable for the assessment the autonomic parasympathetic control than indices based on R/R interval ratios, (2) that the tests (and indices) of choice in the elderly subjects are the Valsalva manoeuvre (Valsalva difference and tachycardia difference) and the active orthostatic test (Max-Rest difference, immediate and later change in systolic blood pressure), (3) that the usefulness of the deep breathing test is limited in the elderly, (4) that such commonly used indices as the Valsalva ratio and the Max/Min ratio in orthostatic test are not useful in the elderly, and (5) that the isometric handgrip test is of little use in the assessment of the autonomic function in the elderly. In conclusion, standard cardiovascular autonomic reflex tests can be used in the assessment of autonomic function to some extent also in the elderly subjects. However, one must bear in mind the limitations in their applicability in that age group.