CONTEMPORARY USE OF ANTIBIOTICS IN 1089 ADULTS PRESENTING WITH ACUTE LOWER RESPIRATORY-TRACT ILLNESS IN GENERAL-PRACTICE IN THE UK - IMPLICATIONS FOR DEVELOPING MANAGEMENT GUIDELINES

Respiratory symptoms are the most common cause of general practitioner (GP) consultation, and hospital-based specialists are often called on to provide management guidelines, particularly in the area of antibio tic prescribing. The present authors have assessed factors associated with antibiotic use by 115 GPs when managing 1089 adults with an acute lower respiratory tract illness, including cough. They prescribed ant ibiotics to three-quarters of patients, but felt antibiotics to be def initely indicated in less than one-third of these cases and Mot needed in one-fifth. Univariate analysis revealed that antibiotics were pres cribed more frequently by older GPs for older patients in the presence of underlying disease, discoloured sputum, shortness of breath, wheez e, fever, signs on chest examination, and 'other factors'. Multivariat e logistic regression confirmed an independent effect for all these fi ndings except for the presence of underlying disease, shortness of bre ath and wheeze. 'Other factors' included patient 'pressure' and social factors, and GP work pressure or prior experience with the patient. T hese factors were an important influence on prescribing, especially if the GP felt an antibiotic was not indicated. Amoxycillin was the firs t choice (58% of total) except where the patient had recently received antibiotics for the same illness. Broader spectrum antibiotics were u sed more commonly in patients with chronic lung disease, discoloured s putum, chest signs on examination and where the GP felt antibiotics we re indicated. However, these antibiotics were also prescribed to 14% o f previously well patients. General practitioners used a wide variety of terms to describe the illness with little consistency or structure. The decision concerning the use and choice of antibiotics and the con fidence with which the GP makes that decision is a complex interaction between patient, doctor and disease, being affected not only by clini cal features but also by the social and psychological elements of the presenting problem. Such issues need to be appreciated by hospital spe cialists when called on to advise on developing relevant guidelines fo r primary care.