The clinical assessment of patients thought to be suffering from hand-arm vibration syndrome (HAVS) requires the use of multiple vascular and sensory tests. In a family physician's office, Adson's, Allen's and cold water immersion of the hands are the only feasible vascular tests, while the sensory tests have to be limited to assessing impairment of skin sensitivity and manipulative dexterity. This paper reviews the laboratory tests deemed to be useful in a hospital or clinic facility, and reports on the investigation of 364 patients exposed to hand-arm vibration who were examined in Toronto, Canada during the period 1989-92. A statistical clustering algorithm was used to categorise 138 male subjects according to the results of their diagnostic tests. From the cluster analysis, four vascular and four sensorineural categories of impairment were recognised in patients suffering from HAVS. The Stockholm vascular classification stages and the four vascular clusters were found to correspond. The Stockholm sensorineural classification (Stages 1, 2, and 3) correlated with clusters formed from the sensory tests evaluating the sensitivity of the nerve endings and the distal digital branches of the median and ulnar nerves. When the myelinated nerve fibres were affected, as detected by abnormal Tinel's, Phalen's, and nerve conduction tests, an additional cluster group emerged. The subjects with abnormal nerve conduction test results constituted a distinct group with increased impairment, so there is a need for them to be categorised separately i.e. as a Stage 4. It is suggested that a Stage 4 be included in the Stockholm sensorineural classification.