The objective of this study was to examine the effect of a nonsteroidal anti-inflammatory drug (NSAID) plaster for knee osteoarthritis among Japanese patients. An open-labeled, randomized, controlled, multiclinic trial was performed involving outpatient clinic groups. Two comparative groups-plaster NSAIDs and oral NSAIDs-were randomly allocated. The drugs used were limited to the current top three in both groups in Japan. Treatments were assessed after four weeks and compared with the baseline scores. Outcomes were evaluated by two psychometric measures: Japanese knee osteoarthritis measure, and pain with the visual analogue scale. The total number of patients included in the final evaluation was 165 (87 for the plaster group and 78 for the oral group). Between these two groups there were no significant differences in gender, age, body height and weight, body mass index, and X-ray grading. The subjects in both groups showed improvements in both scores at the end of intervention. The differences in the improvements in scores between the two groups were not significant, though the mean rank score and the 95% CI of the plaster group were slightly better than those of the oral group. In conclusion, the local application of a plaster with NSAIDs leads to the same level of improvement in knee osteoarthritis as oral NSAIDs.