Background Periodic audits allow monitoring of healthcare quality by comparing performances at different time points. Aims To assess quality of in-hospital stroke care in Catalonia in 2007 and compare it with 2005 (post-/preguidelines delivery, respectively). Methods Data on 13 evidence-based performance measures were collected by a retrospective review of medical records of consecutive stroke admissions (January–December 2007) to 47 acute hospitals in Catalonia. Adherence was calculated according to the ratio (patients with documented performance measures' compliance) (valid cases for that measure). Sampling weights were applied to produce estimates of compliance. The proportions of compliance with performance measures in both audits were compared using random-effects logistic regressions, with each performance measure as the dependent variable and audit edition as the explanatory variable to determine whether changes in stroke care quality occurred along time. Results We analyzed 1767 events distributed among 47 hospitals. In 2007, there was an increase in tissue plasminogen activator administrations (2·8% vs. 5·9%) and stroke unit admissions (16·6% vs. 22·6%) and a reduction in seven-day mortality (9·5% vs. 6·8%). Logistic regression models provided evidence of improved adherences to seven performance measures (screening of dysphagia, management of hyperthermia, baseline computed tomography scan, baseline glycemia, rehabilitation needs, early mobilization, and anticoagulants for atrial fibrillation), but worsening of management of hypertension, dyslipidemia, and antithrombotics at discharge. The remaining three performance measures showed no changes. Conclusions The Second Stroke Audit showed improvements in most dimensions of care, although unexpectedly a few but relevant performance measures became worse. Therefore, periodic stroke audits are needed to check changes in quality of care over time.