The objective was to determine whether symptomatic workers with an abnormal sensory nerve conduction study consistent with carpal tunnel syndrome differed, in terms of electrophysiologic measures, psychosocial, demographic, anthropometric, or ergonomic variables, from workers with an asymptomatic median mononeuropathy. This was a cross-sectional study of active workers at six different work sites. Cases were defined as workers with electrodiagnostic findings of a median mononeuropathy in either hand, based on a 0.5-msec prolongation of the median sensory evoked peak latency compared to the ulnar latency. This group was stratified on the basis of symptoms of numbness, tingling, burning or pain in the hand. The two groups were compared in terms of demographic, anthropomorphic, psychosocial, electrophysiologic, and ergonomic risk factors. Active workers from six different sites were tested; five sites involved manufacturing workers, and one site represented clerical workers. One hundred eighty-four active workers with a median mononeuropathy were documented on nerve conduction studies. These workers represented a subset of more than 700 workers screened at six different locations. The main outcome measure was the patient's report of symptoms of pain, numbness, tingling or burning in the hand or fingers that lasted more than 1 week or occurred three or more times at the initial screening. Workers with a median mononeuropathy who complained of hand symptoms were more likely to be female, to have jobs with higher hand repetition levels, to have higher ratings of job security, not to have a history of diabetes, to use more force in their job with more abnormal postures of their wrist and fingers, and to have a trend toward a more prolonged median sensory distal latency. Most logistic regression models explained less than 15% of the variance (pseudo R2). Women with jobs that have higher ergonomic risks and no history of diabetes were more likely to have reported symptoms associated with carpal tunnel syndrome compared to other workers with a documented median mononeuropathy. Psychosocial variables were not particularly discriminatory. None of the models allows enough precision to predict on an individual basis.