Objective The ‘GP check’ at 6–8 weeks forms part of the selective surveillance system for developmental dysplasia of the hip (DDH) in the UK. It is imperative to pick up DDH within the first months of life to allow for non-invasive treatment and the avoidance of surgery. We aim to investigate the effectiveness of hip examination at 6–8 weeks.

Methods This is a longitudinal observational study including all infants born in our region in the 5 years following 2006. Early presentation was defined as diagnosis within 14 weeks of birth and late presentation after 14 weeks. Treatment records for early and late DDH as well as referrals for ultrasound (US) following examination at 6–8 weeks were analysed. Attendance of the examination at 6–8 weeks in those patients who went on to present with a late DDH was also analysed.

Results 23 112 live births occurred during the study period. There were 141 confirmed cases of DDH. 400 referrals for US were received following examination at 6–8 weeks; 6 of these had a positive finding of DDH. 27 patients presented after 14 weeks and were classified as late presentations. 25 of these patients had attended examination at 6–8 weeks and no abnormality had been identified.

Conclusions The sensitivity of examination at 6–8 weeks was only 19.4%, its specificity was 98% and it had a positive predictive value of 1.5%. For many years the check at 6–8 weeks has been thought of as a means to identify those children not identified as neonates; however, we found that four out of five children with DDH were not identified by the check at 6–8 weeks. Unfortunately, we conclude that the presumed safety net of the examination in its current form is not reliable.