OBJECTIVES: Spasticity is one of the well-recognized complications of stroke which may give rise to pain and limit patients’ ability to perform daily activities. The predisposing factors and direct effects of post-stroke spasticity (PSS) also involve high management costs in terms of healthcare resources and case-control designs are required for establishing such differences. In ‘The Health Improvement Network’ (THIN) database, such a study was difficult to provide reliable estimates since the prevalence of post-stroke spasticity was found to be substantially below the most conservative previously reported estimates. The objective of this study was to use predictive analysis techniques to determine if there were a substantial number of potentially under-recorded patients with PSS. METHODS: This study used retrospective data from adult patients with a diagnostic code for stroke between 2007 and 2011 registered in THIN. Two algorithm approaches were developed: 1) a statistically validated data-trained algorithm using machine techniques and 82 potential predictors; and 2)  a clinician-trained algorithm based on the review of 200 stroke cases by two expert neurologists. The algorithm with the better performance was used to identify PSS cases. RESULTS: In THIN data, 45,613 stroke events were identified, with 660 having a diagnosis PSS. A data-trained algorithm using Random Forest showed better prediction performance than the clinician-trained algorithm, with higher sensitivity and only marginally lower specificity.  Overall accuracy was 84% and 72%, respectively. The data-trained algorithm predicted an additional 3,912 records consistent with patients developing spasticity in the 12 months following a stroke. CONCLUSIONS: Using machine learning techniques, additional unrecorded post-stroke spasticity patients were identified, increasing the condition’s prevalence in THIN from 2% to 13%. This work shows the potential for under-reporting of PSS in primary care data, and provides a method for improved identification of cases and control records for future studies.