Abstract Background In 2012, the National Institute for Health and Care Excellence (NICE) recommended (public health guidance 43) hepatitis B virus (HBV) testing of people born in countries with a prevalence of more than 2%. Country of birth is not captured in routine HBV surveillance so prevalence of HBV infection in immigrants to the UK by country of birth is unclear. Also, the proportion of migrants tested for HBV is unknown. This cross-sectional study aimed to estimate the proportion of migrants tested for HBV, and the prevalence of HBV infection in HBV-tested migrants by country of birth. Methods All people living in Bristol and registered with a general practitioner (GP) in 2006–13 were included. Country of birth was obtained from the national patient demographic system. We defined a person as not HBV tested if no test result was recorded in the Public Health England Laboratory Bristol records (2006–13) and as HBV infected if HBsAg was positive or if HBV DNA was detected. The sensitivity of this method in detecting HBV-infected people was checked by review of primary care records. Findings Of 687 483 patients registered with a GP in 2006–13, 82 561 (12%) were born in a country with an HBV prevalence of more than 2%, and for 194 025 patients (28%) country of birth was unknown. Of the 82 561 eligible patients, 72 934 (88%) were not HBV tested. Of those who were HBV tested, first tests were requested antenatally in 40% (3875/9627) (1·0%, 39/3857 HBV infected) and by primary care in 42% (4078/9627) (6·1%, 249/4078 HBV infected). The period prevalence of HBV infection by UN subregion of birth was: Eastern Asia 14·8% (63/427) of all tested patients versus 3·1% (5/160) of those tested antenatally, Western Africa 11·6% (68/588) versus 4·2% (8/191), South-Eastern Asia 6·8% (35/515) versus 2·0% (4/205), and Eastern Africa 5·6% (133/2363) versus 1·4% (11/809). In all the remaining subregions the period prevalence in those tested antenatally was less than 1%. 179 054 (92%) of individuals with unknown country of birth were not HBV tested, and among those tested, 0·5% (80/14 971) were HBV infected. Interpretation In a large UK city, most GP-registered migrants, for whom a HBV test was recommended by NICE, had not been tested. The proportion not HBV tested in all migrants in Bristol might be higher because testing could be less likely in those not registered with a GP. Those migrants born in Eastern and South-Eastern Asia and West Africa were at greatest risk of infection. Funding This study was supported with an educational grant via the Gilead UK and Ireland Fellowship Programme. The funder had no role in design of the study, data collection, analysis, interpretation, abstract preparation, or decision to submit.