OBJECTIVES: Health economists are being encouraged to use routine datasets for resource-use measurement and costing purposes in economic evaluations alongside clinical trials to reduce research burden.  The Hospital Episode Statistics (HES) dataset, which records all NHS hospital-based activity in England, is one such dataset. However, its validity for research purposes has not been established.  This study aims to assess the validity of the HES outpatient dataset. METHODS: Men who died of, or with, prostate cancer were selected from a prostate-cancer screening trial (CAP, Cluster randomised triAl of testing for Prostate cancer).  Details of visits that took place after 1/4/2003 to hospital outpatient departments for conditions related to prostate cancer were extracted from medical records (MR).  Data from the HES outpatient dataset were obtained for the same men.  Appointments for visits extracted from MR were sought in the HES dataset.  The matching procedure was repeated for periods before and after 1/4/2008 (when the dataset was accredited as a national statistic). RESULTS: p=0.5).  For appointments occurring when the dataset was considered experimental (prior to 1/4/2008), 2194/2754 (79.7%; 95%CI 78.1–81.2) matches were observed, while 1892/2168 (87.3%; 95%CI 85.8–88.6) appointments occurring after 1/4/2008 were identified (p=0.03). CONCLUSIONS:   The HES outpatient dataset appears reasonably valid for research, particularly following accreditation.  The dataset may be a suitable alternative to collecting MR data manually within a trial, although caution should be exercised with earlier data.  Further work is ongoing to establish the nature of the missing data and the implications for cost differences.