Adverse Events in Patients With Community-Acquired Pneumonia at an Academic Tertiary Emergency Department Do They Contribute to Hospital Death? Renata Daud-Gallotti, MD, PhD,* Hillegonda Maria Dutilh Novaes, MD, PhD, y Maria Cecı ´lia Lorenzi, MD, PhD, z Jose ´ Eluf-Neto, MD, PhD, y Mirna Namie Okamura, RN, MS, y Vladimir Ribeiro Pinto Pizzo, MD,* and Irineu Tadeu Velasco, MD, PhD* Abstract: A matched case-control study was carried out at a Brazilian academic medical emergency department to identify the occurrence of adverse events (AEs) in patients with community- acquired pneumonia (CAP) and disclose their association with in- hospital death. The cases comprised 101 consecutive deaths and the controls 101 discharged patients admitted for CAP from March 1996 to September 1999. Adverse events were classified according to severity, potential causes, the occurrence of nosocomial in- fections, and the related professional category. A total of 603 AEs were identified, predominating among the deceased patients: 456 events (75.6%) in 85 cases and 147 (24.4%) in 55 controls. Severe AEs also predominated among the cases. Procedures and nursing activities were the main AE-related potential causes. The oc- currence of at least 1 AE was associated with a higher risk of death (3.25-fold). This association was stronger when either severe AEs (8.37-fold) or medical AEs (4.23-fold) were analyzed. (Infect Dis Clin Pract 2006;14:350 – 359) C ommunity-acquired pneumonia (CAP) is defined as pneumonia not acquired in hospitals or other long-term- care facilities, causing nearly 5 million episodes of illness annually in the United States. Approximately 20% of those episodes require hospitalization. 1–3 Mortality rates of CAP range from 5% to 36% in hospitalized patients, representing the sixth leading overall cause of death and the most common infectious cause of death. 4–6 Recently, efforts have been made to evaluate and improve the quality of CAP management, but substantial variations in quality of care still persist. 7–10 A widespread approach to evaluate quality of medical care is the identification of adverse events (AEs), defined as unintended injuries or complications caused by health care management rather than the underlying condition of the patient. 11 Previous important studies described AE rates of 2.9% to 16.6% of all admissions, 12 – 18 and half of them were considered preventable, that is, resulting from errors. Although most of the AEs led to minimal impairments, death was associated to 4.9% to 13.6% of the events, 12 – 18 making AEs the eighth leading cause of death in the United States. 19 Several factors have been associated to AE occurrence, including urgent care. 20 Emergency departments (EDs) are considered error-prone environments, facilitating the occur- rence of AEs. 21,22 Overcrowding, unrestricted access, the presence of young and inexperienced professionals, lack of supervision, insufficient information regarding patient’s previous medical situation, frequent interruptions in medical care, and shortage of nurse staff and resources contribute to this unfavorable scenario. 23 – 25 Considering that EDs are the source of admission and site of initial treatment for almost 75% of patients with CAP, 26 we aimed to evaluate the occurrence of AEs in CAP during ED hospitalizations, describing their degree of severity, potential causes, and related professional categories involved in patient care. We also aimed to identify the association between AEs and hospital death in patients with CAP, identifying their significance when compared with other usual potential risk factors for hospital death, such as severity of illness on admission and demographic and health care characteristics. METHODS Design and Setting This retrospective study with a 1:1 matched case- control design was conducted at the medical ED of a 1100- bed tertiary university hospital in Sa ˜o Paulo, Brazil. The medical ED is responsible for providing clinical, urgent care to individuals older than 12 years (excluding cardiologic and psychiatric emergencies, referred to specific associated 350 Infectious Diseases in Clinical Practice  Volume 14, Number 6, November 2006 Original Article *Medical Emergency Medicine, Department of Hospital das Clı ´nicas, yDepartment of Preventive Medicine and zSleep Laboratory of The Heart Institute of Hospital das Clı ´nicas, University of Sa ˜o Paulo School of Medicine, Sa ˜o Paulo, SP, Brazil. Presented at the XXXII Brazilian Congress of Pneumology 2004, November 13 to 17, Salvador, Bahia, Brazil. Supported by a grant from the Fundac ¸a ˜o de Amparo a ` Pesquisa do Estado de Sa ˜o Paulo (FAPESP), grant no. 2002/09277-0. Address correspondence and reprint requests to Renata Daud-Gallotti, MD, PhD, Alameda Itu, 1420 apt 101, Sa ˜o Paulo, SP, Brazil. E-mail: renatagallotti@terra.com.br. Copyright n 2006 by Lippincott Williams & Wilkins ISSN: 1056-9103/06/1406-0350 Copyr ight © Lippincott Williams & Wilkins. Unauthor iz ed reproduction of this article is prohibited.