Previous polysomnographic (PSG) investigations have reported a rhythmic electromyographic (EMG) pattern (0.5-3.0 cps) of leg movement activity in a subset of patients with neuroleptic-induced akathisia (NIA). It has been suggested that this EMG pattern may represent a pathophysiological correlate of NIA and thus have clinical utility as an objective marker for this condition. We present preliminary measures of sensitivity and specificity for this EMG pattern as a diagnostic marker for NIA for 26 neuroleptic-treated patients. The EMG marker yielded a diagnostic sensitivity of 68.9% and a specificity of 70.0%, falling just short of statistical significance (Fisher's exact test p = 0.06). Quantitative analysis of the EMG pattern revealed a significant positive correlation between the percentage of time the NIA marker occurred during wakefulness and corresponding chlorpromazine equivalent levels. Clinical demographic findings for true-positive, false-positive, true-negative, and false-negative groups are discussed. Overall findings suggest that this particular pattern of EMG marker activity observed in neuroleptic-treated patients during PSG and EMG studies is valuable in facilitating the diagnosis and monitoring treatment.