AIMS: To develop the Discharge Planning of Ward Nurses (DPWN), a Japanese self-evaluation instrument for ward nurses' discharge planning practices. DESIGN: Cross-sectional survey. METHODS: Participants were 624 ward nurses from six hospitals in Japan with a discharge planning department. Items about discharge planning practices were collected from literature and interviews with nurses and researchers. Construct validity, concurrent validity, internal consistency and test-retest reliability were tested. RESULTS: Initially, 55 items were collected. Examination of the floor effect, item-total, good-poor analyses and exploratory factor analysis yielded a four-factor model with 24 items ('teaching home-care skills with community/hospital professionals,' 'identifying clients' potential needs early in the discharge process,' 'introducing social resources' and 'identifying client/family wishes and building consensus for discharge'). The four-factor structure was supported by confirmatory factor analysis. The DPWN correlated with scales ascertaining similar concepts, supporting concurrent validity. Internal consistency and test-retest reliability were generally satisfactory.