OBJECTIVES: About two-thirds of stroke survivors develop post-stroke sequelae, including spasticity. The burden of post-stroke spasticity (PSS) is high in terms of treatment costs and the effects of comorbidities. Our objective was to describe the burden of PSS in terms of healthcare resource utilization and costs, and quantify the difference between patients who develop PSS and those who do not. METHODS: This retrospective study used the THIN database. Adult patients with a stroke between 1Jan2007 and 31Dec2012 were included. PSS diagnoses were found to be under-represented; machine learning methodology was applied to identify potentially undiagnosed PSS. Cases were defined as patients with diagnosed or predicted PSS in the 12 months after stroke; for patients without PSS, each stroke acted as a control event. PSS cases were matched to controls on age, gender, prior strokes, socioeconomic status, and comorbidities, using the nearest neighbour algorithm. Direct healthcare resources, including primary care visits, all-cause hospitalisations, and specialist referrals during the year post-stroke, were costed out at 2014 GBP rates, using public sources.  RESULTS: Of the 3,082 PSS cases and 28,753 controls, 56% were female and 49% were 75 years or older. During the first year, 33% of the PSS cases were hospitalised, compared to 9% of the controls. Specialist referrals were recorded for 76% of PSS patients and 64% of controls. Primary care utilisation was similar for both groups. Total average costs per patient were £1,270 (SD: 772) for cases and £631 (SD: 496) for controls. After adjusting for other covariates, a significant increase in cost for the PSS patients was found; on average £635 in the 12 months post-stroke.  CONCLUSIONS: The costs for patients who develop PSS after stroke are twice as high as those for patients who do not develop spasticity, with the major driver being the number of hospitalisations.