OBJECTIVE: Generalised anxiety disorder (GAD) is the most common anxiety disorder in the general population and has been associated with high economic and human burden. However, it has been neglected in the health services literature. The objective of this study is to assess whether GAD leads to hospital admissions using data from the European Prospective Investigation of Cancer-Norfolk. Other aims include determining whether early-onset or late-onset forms of the disorder, episode chronicity and frequency, and comorbidity with major depressive disorder (MDD) contribute to hospital admissions. DESIGN: Large, population study. SETTING: UK population-based cohort. PARTICIPANTS: 30 445 British participants were recruited through general practice registers in England. Of these, 20 919 completed a structured psychosocial questionnaire used to identify presence of GAD. Anxiety was assessed in 1996-2000, and health service use was captured between 1999/2000 and 2009 through record linkage with large, administrative health databases. 17 939 participants had complete data on covariates. MAIN OUTCOME MEASURE: Past-year GAD defined according to the Diagnostic and Statistical Manual of Mental Disorders, fourth edition. RESULTS: In this study, there were 2.2% (393/17 939) of respondents with GAD. Anxiety was not independently associated with hospital admissions (incidence rate ratio (IRR)=1.04, 95% CI 0.90 to 1.20) over 9 years. However, those whose anxiety was comorbid with depression showed a statistically significantly increased risk for hospital admissions (IRR=1.23, 95% CI 1.02 to 1.49). CONCLUSION: People with GAD and MDD comorbidity were at an increased risk for hospital admissions. Clinicians should consider that meeting criteria for a pure or individual disorder at one point in time, such as past-year GAD, does not necessarily predict deleterious health outcomes; rather different forms of the disorder, such as comorbid cases, might be of greater importance.