Heat-pain threshold and stimulus response characteristics can be evaluated with graduated heating pulses from a radiant heat source or a contact thermode. Results may be used to: (1) evaluate differences in sensation among anatomical sites, sides of the body, and with development and aging; and (2) provide an end-point for the study of the efficacy of drugs; or to follow the course of sensory alteration in disease (medical practice, epidemiologic studies, and controlled clinical trials). Because there is great variability in how tests of this kind are performed and scored, comparisons of results among medical centers are difficult. To meet this need, we have developed, and here describe, a standardized and validated test of heat-pain. We use both pyramidal and trapezoid-shaped stimuli. The range of stimulus magnitudes we recommend is sufficient to test heat-pain at a sensitive region (the face) of young people and an insensitive region (the foot) of healthy old people. From tests on healthy subjects and patients, we find that neither our previously published forced-choice or 4, 2, and 1 stepping algorithms are suitable for testing heat-pain sensation. We, therefore, introduce the Non-Repeating Ascending with Null Stimuli (NRA-NS) algorithm which performs satisfactorily. The graphed data points of responses to increasingly stronger heat pulses were made up of two components-the no pain (0) response line and the heat-pain response line (> or = 1 numerical scaling of the pain responses graded from 1 [least] to 10 [greatest]). For the pain responses, we found that usually a curve could be fit using a quadratic equation. Using this equation, or interpolation where necessary, it is possible to compute the heat-pain detection threshold (HPDT or HP:0.5), an intermediate heat-pain response (HP:5.0), and the difference between the two (HP:5.0-0.5). Our studies show that a certain time is needed between successive stimuli and tests to minimize changing basal skin temperature or threshold. We also demonstrated that low or high baseline skin temperatures can affect heat-pain responses, therefore, we advocate specific testing conditions. Based on a study of 25 healthy subjects, the reproducibility of the test falls within +/-1 stimulus steps 88% of the time for HP:5.0 and 76% of the time for HP:0.5. The precise approaches employed to make the test standard and reproducible are described. We illustrate that the algorithm and testing system is able to document altered pain threshold with skin abrasion, with intradermal injection of nerve growth factor, and with diabetic polyneuropathy.