OBJECTIVES: To determine what proportion of cases of heterozygous familial hypercholesterolaemia would be identified by cascade screening conducted by a specialist hospital clinic, and by how much this would increase the prevalence of diagnosed cases. SETTING: Hospital clinic serving a population of 605,900 in Oxfordshire, UK. METHODS: A specialist nurse obtained details of living first-degree relatives from 227 adult patients with heterozygous familial hypercholesterolaemia currently or previously attending Oxford lipid clinic after excluding 79 adults without relatives living in Oxfordshire and 48 children. Index cases were asked to invite relatives resident in Oxfordshire for testing. RESULTS: A total of 227 index cases had 1075 first-degree relatives, including 442 adults and 117 children aged < 18 years resident in Oxfordshire. We excluded 171 previously screened adults and 46 for other reasons. Among 225 eligible adult relatives, 28 responders (12%) planned to consult their general practitioner and 52 (23%) attended the clinic for testing. Parents of 113 children (97%) wanted them tested. The positive diagnostic rate was 29% (15/52) in adults and 32% (36/113) in children. Screening increased prevalence by 14.4%, from 0.58/1000 (95% confidence intervals [CI] 0.52-0.65) to 0.67/1000 (95% CI 0.60-0.73), representing 33.5% of predicted cases. CONCLUSIONS: Cascade screening conducted by a specialist hospital clinic within its population catchment area did not substantially increase the prevalence of diagnosed familial hypercholesterolaemia. To maximize response rates, clinic staff need to approach relatives directly. Validated age, sex and country-specific diagnostic criteria should be defined, possibly with access to DNA-based tests, to help resolve diagnostic uncertainty.