Background As renal damage can occur in the longer term, urinary tract infection (UTI) in childhood requires follow up, particularly for those aged under five years. No simple method exists of identifying those most at risk, so it is recommended that all children with proven UTI are referred for routine diagnostic imaging. Previous audits indicate poor compliance by general practitioners with the Royal College of Physicians 1990 guidelines on management of urinary tract infection. Aim To improve the care of children with urinary tract infections (UTI) in general practice and improve their long-term outcome. Method Three-month prospective study in 1998: three hospital laboratories within the Eastern Health and Social Services Board (EHSSB), Northern Ireland, identified children aged less than five with a positive mid-stream sample of urine (MSU) (single organism > 105 ml). Hospital records were reviewed after six months to determine if they were referred. Unplanned was that general practitioners of children not referred were contacted to obtain further information. Changes to the medical climate during the audit prompted this decision. Results One hundred and seven subjects were identified. Five were subsequently excluded. Hospital records showed 62 of the 102 cases (58%) had been referred. Of the 45 remaining cases, 40 had an identifiable general practitioner. Of the 40 general practitioners contacted, 35 responded to an anonymised questionnaire. Collation of hospital and general practice information showed 71 of the 102 cases (70%) had been referred to hospital. Conclusion Medical ethics, Good Medical Practice and clinical governance now influence the design of an audit and action points required. This paper shows how these issues influenced what would previously have been considered a simple audit of hospital records. The lessons learned warrant consideration by healthcare professionals involved in practice-based and larger primary care group audits. Patient follow-up should be considered at the audit design stage and in discussion of results.