771 • Vol 10 • November 2008 Medication Errors and Response Bias It has been estimated that 770,000 patients are injured or die yearly from medication errors and adverse drug events in the United States. Studies have estimated that 2 to 7 of every 100 admissions are affected by adverse drug events. It has further been estimated that approximately half of these events have their source at the drug-ordering stage. The principal types of errors involve missing medication dose[s], incorrect dose or frequency, or incorrect route of administration [1-4]. The consequences of medication errors may include prolonged hospitalization, unnecessary treatment, and death [5]. Although most medication errors do not result in adverse events, the importance of early diagnosis and reporting medication er- rors is well documented. Several studies discuss the extent of medication errors, why they happen, and methods for preven- tion [1-7]. There are sparse data on the rate of physicians who admit to having made a mistake [8]. The purpose of this study was to evaluate the rate of acknowledging medical errors among physicians working in the community and in hospitals. Subjects and Methods An anonymous structured questionnaire was sent to 9320 of the approximately 15,000 member physicians of the Israeli Medical Association. The 9320 recipients are subscribers to Abstract Medical Review, a medical newsletter. Abstract Medical Review® was launched in 1994 by Teva Pharmaceutical Industries Ltd., an Israeli company. The newsletter is sent by mail to some 18,000 physicians, pharmacists, nurses and other health care workers. It includes summaries of the latest and most important referenced papers from a worldwide selection of relevant journals in the fields of medicine, pharmaceuticals, management and marketing. Abstract Medical Review subscribers receive questionnaires two to three times a year on other top- ics. The usual response rate for such questionnaires is 60–70% (Ami Blay, Editor, Abstract Medical Review®, Teva Pharmaceutical Industries Ltd.). The distribution of specialties and subspecial- ties among the recipients of the questionnaire was similar to that of the approximately 15,000 members of the Israeli Medical Association. Two weeks before the questionnaire was sent, the physicians received a postcard announcing that the questionnaire was on its way as part of a survey on medication errors being conducted by Abstract Medical Review and the clinical pharmacology and toxicol- ogy unit of Assaf Harofeh Medical Center. The questionnaire asked a series of general questions about physician use of vari- ous forms of information regarding prescribing, the types of er- rors the physician may have made during his or her professional career, and specific questions about use of a drug consultation center. An "error" was defined as dosage miscalculation, drug- drug interaction, mistaken type of medication, and wrong route of administration. The questionnaire did not address drug-disease interactions or drug-food interactions. The physicians were not asked about the result of their error or how the errors were dis- covered. The respondents were requested to send the completed questionnaire anonymously by mail or fax to the Abstract Medical Review editorial office. The estimated time for filling the question- naire was about 5 minutes. No reminders were sent. Abstract Background: Medication errors are a common cause of morbidity and mortality. Objectives: To evaluate the rate of acknowledgment of medication errors as reported by physicians working in the community and in hospitals. Methods: An anonymous questionnaire was sent to 9320 active physicians (about 48% community physicians, 17% hospital physicians and 35% working in both places), with questions on the rate and type of medication errors that they had encountered during their professional career. The questions specified errors in dosage, type of medicine (wrong indication), route of administration and drug interactions. Results: Only 627 physicians (6.7%) responded. Of these, nearly 79% admitted having made an error in prescribing medication; the majority admitted to more than one error. Physicians with fewer years of experience admitted having made a mistake more than did physicians with more experience (P = 0.019). Pediatricians and geriatricians made more dosage mistakes (P = 0.02), while family physicians and psychiatrists made more mistakes in drug interactions (P = 0.001). Conclusions: It is possible that indifference, fear of identification, or lack of awareness may have contributed to the low response rate despite the fact that the questionnaire was anonymous. Educational programs should be implemented in medical schools to encourage physicians to report errors before the onset of adverse reactions. IMAJ 2008;10:771–774 Medication Errors and Response Bias: The Tip of the Iceberg Benjamin Bar-Oz MD 1 , Michael Goldman MD 2 , Eliezer Lahat MD 3 , Revital Greenberg BSc 4 , Meytal Avgil MSc 4 , Ami Blay 5 , Amir Herman MD 4 and Matitiahu Berkovitch MD 4 1 Department of Neonatology, Hadassah-Hebrew University Medical Centers (Mt. Scopus Campus), Jerusalem, Israel 2 Pediatric Division, 3 Department of Pediatric Neurology and 4 Clinical Pharmacology and Toxicology Unit, Assaf Harofeh Medical Center, Zerifin, and Sackler Faculty of Medicine, Tel Aviv University, Ramat Aviv, Israel 5 Teva Pharmaceutical Industries Ltd., Teva Israel, Netanya, Israel Key words: adverse reactions, adverse drug events, medication error, response bias Toxicology