Medical audit in general practice. I: Effects on doctors' clinical behaviour for common childhood conditions. North of England Study of Standards and Performance in General Practice.

Abstract

OBJECTIVE: To estimate the effects of medical audit, particularly setting clinical standards, on general practitioners' clinical behaviour. DESIGN: Before and after study strengthened by a replicated Latin square. SETTING: 62 training general practices in the north of England. SUBJECTS: 92 general practitioner trainers, 84 (91%) of whom completed the study; random sample of 3500 children consulting one of these trainers for any of five conditions--acute cough, acute vomiting, bedwetting, itchy rash, and recurrent wheezy chest--stratified by doctor consulted, condition, and age. INTERVENTIONS: Clinical standard set by each of 10 small groups of general practitioner trainers for one randomly selected childhood condition. Each group also experienced a different type of medical audit, randomly selected, for each of the four other study conditions (receiving a clinical standard set by another trainer group, tabulated data comparing clinical performance with that of all other groups, tabulated data from only their own group, and nothing ("control" condition)). MAIN MEASURES: Content of initial consultation divided into: history, examination, investigation, diagnosis, and management (abstracted from medical records and "enhancement forms" completed by doctors). RESULTS: There was increased prescribing of bronchodilators for acute cough, oral rehydration fluids for acute vomiting, antibiotics for itchy rash, and bronchodilators and oral steroids for recurrent wheezy chest and reduced prescribing of antibiotics for acute cough and recurrent wheezy chest and tricyclic antidepressants for bedwetting. Fewer children were "discharged." Each change was consistent with the standard and either limited to doctors who set a standard for that condition or significantly greater for them than all other doctors. CONCLUSION: Setting clinical standards improved prescribing and follow up.