The purpose of this study was to determine the feasibility of using Massachusetts workers' compensation data for passive surveillance of occupational carpal tunnel syndrome (OCTS). Workers' compensation claims for OCTS (n = 358) and for possible cases of OCTS (n = 1,121) active during the first 6 months of 1989 were identified. The availability and distribution of demographic and employment descriptors were assessed. Medical records on a sample of the claims were reviewed to validate the diagnosis of OCTS. Age, gender, and occupation were available for less than 47% of the reported cases of OCTS. The majority (88%) of cases on whom medical record review was performed had a physician's diagnosis of carpal tunnel syndrome (CTS), and most of this group had confirmatory nerve conduction studies or electromyography. However, there were fundamental limitations to workers' compensation based disease surveillance in Massachusetts, including underascertainment of cases, potential ascertainment biases, delayed case reporting, limited access to specific diagnostic information, and incomplete and sometimes inaccurate information. These limitations are likely to be applicable in many, if not most, states and must be made clear in any analyses based on workers' compensation data.