Many registered drugs in different countries not always have been tested by clinical assays and their efficacy has no definitively proved. Good clinical practice must assume the prescription of the more efficacy drugs. The aim of this study is to evaluate the quality of the drugs prescribed by spanish general internists. We analyze the treatments prescribed by general internists using their hospital medical records. The records were offered by the Heads of the Internal Medecine Departments of Spanish hospitals with 250 beds or more when required by the Spanish Society of Internal Medecine. A random sample, stratified by the hospital size, was obtained from the whole of the Spanish hospitals. Quality of the drugs was determined using two criteria: A. Their intrinsic value; it was decided it was high if clinical assays showed their efficacy and security, and low of it was not B. The number of different chemical substances in the drug; we considered a high quality criteria if only one chemical substance, or two with proved synergistic efficacy were present, and low if it was more than two, or without proved synergistic efficacy. The sample was 250 clinical records, with 1022 different prescriptions; the records were collected between February and July 1994. We found that 94.06 +/- 1.45 of evaluated treatments have a high intrinsic value, and 96.93 +/- 1.06 have only one chemical component. The drugs without a high intrinsic value were: oral hypoglycaemics, some including several combinations of vit B, drugs acting on the cerebral blood flow and some mucolitics. Spanish general Internists usually prescribed monocomponent drugs with proved efficacy and safety.