OBJECTIVE: To analyse the concordance of published information on probable drug interactions and the results of clinical analyses. DESIGN: Bibliographical review of the 50 most commonly prescribed drug products at the Primary Health Centre of L'Hospitalet de Llobregat, Barcelona, Spain, in 2002 and 2003. DATA SOURCE: Six bibliographical sources consisting of drug product catalogues and reference books. MEASUREMENTS: Interference with the 56 drug products studied. RESULTS: Most theoretical interference belonged to the therapeutic group of anti-inflammatory drugs. The drug products with most interference were hydrochlorothiazide (diuretic) and gliclazide (lipid lowering). The most frequent interference was in the increase of transaminases, thrombocytopaenia, leukopaenia, hyperglycaemia, and hyperuricaemia. Only in 12% of the points of interference analysed was the rate of concordance among the various bibliographical sources reviewed higher than 50%. The highest rate of concordance was found in the diuretics. CONCLUSIONS: Consultation of a single bibliographical source does not necessarily ensure the obtaining of reliable data on possible drug interference. This study demonstrates the need to standardize information on drug interference and to include the medication that the patient follows on the analysis application form.
OBJECTIVE: To analyse the concordance of published information on probable drug interactions and the results of clinical analyses. DESIGN: Bibliographical review of the 50 most commonly prescribed drug products at the Primary Health Centre of L'Hospitalet de Llobregat, Barcelona, Spain, in 2002 and 2003. DATA SOURCE: Six bibliographical sources consisting of drug product catalogues and reference books. MEASUREMENTS: Interference with the 56 drug products studied. RESULTS: Most theoretical interference belonged to the therapeutic group of anti-inflammatory drugs. The drug products with most interference were hydrochlorothiazide (diuretic) and gliclazide (lipid lowering). The most frequent interference was in the increase of transaminases, thrombocytopaenia, leukopaenia, hyperglycaemia, and hyperuricaemia. Only in 12% of the points of interference analysed was the rate of concordance among the various bibliographical sources reviewed higher than 50%. The highest rate of concordance was found in the diuretics. CONCLUSIONS: Consultation of a single bibliographical source does not necessarily ensure the obtaining of reliable data on possible drug interference. This study demonstrates the need to standardize information on drug interference and to include the medication that the patient follows on the analysis application form.