journal article
LitStream Collection
Home
Bennett, Charles L.; Davidson, Charles J.; Raisch, Dennis W.; Weinberg, Peter D.; Bennett, Richard H.; Feldman, Marc D.
doi: 10.1001/archinte.159.21.2524pmid: 10573042
BackgroundOne of the most unusual causes of thrombotic thrombocytopenic purpura (TTP), a life-threatening disease, is ticlopidine hydrochloride, an antiplatelet agent used to prevent strokes in high-risk populations or following coronary artery stent placement. Recently, Hoffman-LaRoche Pharmaceuticals, following reports of 20 deaths from ticlopidine-associated TTP, updated the information about the hematologic adverse effects of the drug.ObjectivesTo review our recent findings on ticlopidine-associated hematologic toxic effects, which served as the impetus for the revised warnings, and to discuss the implications of these findings.MethodsData were obtained from the Food and Drug Administration's MedWatch program, published phase 3 clinical trials and case reports, hematologists, and plasmapheresis centers.ResultsNo cases of TTP have been reported in phase 3 ticlopidine trials. In contrast, postmarketing surveillance has identified serious adverse drug reactions to ticlopidine, resulting in 259 deaths, with TTP accounting for 40 of these deaths. Detailed information was available on 98 cases of ticlopidine-associated TTP. Compared with 42 patients in the coronary artery stent setting, 56 patients with ticlopidine-associated TTP in the stroke prevention setting were more likely to be women (62.5% vs 28.6%; P= .01). Before the onset of TTP in patients receiving stroke prevention therapy and patients with stent placement, ticlopidine had been used for less than 2 weeks in 5.4% and 2.4%, between 2 and 3 weeks in 17.9% and 21.4%, between 3 and 4 weeks in 30.4% and 38.1%, and between 4 and 12 weeks in 46.4% and 38.1%, respectively. Death occurred in almost 60% of all patients not receiving plasmapheresis compared with 21.9% of patients receiving plasmapheresis for stroke prevention and 14.3% of patients receiving plasmapheresis in the stent setting.ConclusionsUse of ticlopidine requires frequent physician visits and laboratory tests for at least 3 months in the stroke prevention setting, while, with short-term use in the coronary artery stent setting, adverse events are less likely to occur. These factors, as well as competition from clopidogrel bisulfate, a new antiplatelet agent, potentially limit the feasibility of ticlopidine as a stroke prevention agent, while having less impact on its use following coronary artery stent placement.THROMBOTIC thrombocytopenic purpura (TTP) is a life-threatening, multisystem disease characterized by thrombocytopenia, microangiopathic hemolytic anemia, neurologic changes, renal failure, and fever.The cause of acute TTP appears to be related to transient immune dysregulation and selective antigenic targeting of a metalloprotease that degrades large multimers of factor VIIIR.Ultralarge factor VIIIR multimers increase platelet adhesiveness in vitro and may be one of the platelet-aggregating agents responsible for the platelet microthrombi that characterize TTP in vivo.An IgG autoantibody against components of the enzyme may account for a lack of metalloprotease activity in patients with TTP. While the reasons for transient immune dysregulation and for the selective antigenic targeting of the protease are unknown, the time course of 2 to 4 weeks following certain drug exposures is consistent with an autoimmune mechanism for some cases of TTP.One example of a drug-associated cause of TTP is ticlopidine hydrochloride, an antiplatelet agent that in 1997 was used by 1 million individuals in the United States for stroke prevention and, more recently, by more than 600 000 individuals following coronary artery stent placement.The estimated incidence of ticlopidine-associated TTP is 1 in 1600, while its mortality rate is 33%.Herein, we review our recent findings about ticlopidine-associated hematologic toxic effects, including TTP; postulate on the potential mechanism of ticlopidine-associated TTP; and outline the implications of our findings for patients and physicians involved in clinical trials or clinical practice with ticlopidine as a stroke prevention agent or following coronary artery stent placement.METHODSTICLOPIDINE-ASSOCIATED ADVERSE DRUG REACTION CASE FINDINGInformation on ticlopidine-associated TTP and other hematologic toxic effects was obtained from clinical trialsfrom the Stent Anticoagulation Regimen Study (STARS), the Intracoronary Stenting and Antithrombotic Regimen (ISAR), the Canadian American Ticlopidine Study in Thromboembolic Stroke, the Ticlopidine Aspirin Stroke Study, and the MedWatch database from 1992 through 1997 through periodic Freedom of Information requests. These data include all Food and Drug Administration (FDA) reports, such as manufacturer's 15-day reports (serious unlabeled adverse drug reactions [ADRs]) and periodic reports (required for all labeled ADRs), and voluntary reports from physicians (conducted through the MedWatch program). All ADR cases were reviewed regarding age, sex, date of TTP diagnosis, and location to eliminate duplicate reports. The MedWatch database is limited to voluntarily reported ADRs; thus, some events may be unreported or reported incorrectly.TICLOPIDINE-ASSOCIATED TTP CASE FINDINGDetailed information on ticlopidine-associated TTP was obtained from published case reports, from personal contact with hematologists, and from plasmapheresis centers that had treated cases of ticlopidine-associated TTP. Some of the cases in this series were described in our initial report of ticlopidine-associated TTP.STATISTICAL ANALYSISComparisons between clinical characteristics and outcomes for patients undergoing coronary artery stent placement and patients undergoing stroke prevention were made using 2-tailed ttests for continuous variables and the Fisher exact test (2-tailed) for dichotomous variables.RESULTSIn phase 3 clinical trials in the stroke prevention setting, significant differences in the rates of neutropenia in the ticlopidine vs aspirin groups were observed in the 2 large stroke prevention studies, the Canadian American Ticlopidine Study (1.0% vs 0.2%) and the Ticlopidine Aspirin Stroke Study (0.9% vs 0.0%). In contrast, STARS and ISAR, the 2 largest phase 3 clinical trials of antiplatelet agents in the setting of coronary artery stents, found no difference in rates of neutropenia in the ticlopidine groups and the control groups (0.5% vs 0.0% in ISAR and 0.2% vs 0.0% in STARS) during the 1-month observation period.No cases of TTP were reported in these phase 3 trials.Data from the FDA's MedWatch program indicated that the most common serious ticlopidine-associated toxic effects were hematologic, reported in 1756 cases, primarily leukopenia, thrombocytopenia, TTP, agranulocytosis, pancytopenia, and aplastic anemia (Table 1). Overall, 259 of the cases resulted in death, with 85.6% of these deaths being associated with hematologic toxic effects of the drug. Despite being infrequently reported, thrombocytopenia (50 deaths) and TTP (40 deaths) were more common causes of death than leukopenia (34 deaths) and agranulocytosis (22 deaths).Most Common Hematologic Adverse Medical Events (All) and Events Associated With Death (D) From FDA MedWatch Reports*Type of Event199219931994199519961997TotalDAllDAllDAllDAllDAllDAllD†All‡Thrombocytopenia8338401247217451169150279TTP61312295906619114340119Leukopenia710719771044354781110034521Agranulocytosis4205303183941932022116Aplastic anemia0832345285102161670Pancytopenia27525216171102181383Hemolytic anemia1146250103591225Anemia0232212262801942810125Marrow depression360829250205735Hypochromic anemia01406090417421561Coagulation disorder14331100010059Disseminated intravascular coagulation001122000314410*FDA indicates Food and Drug Administration; TTP, thrombotic thrombocytopenic purpura.†Number of deaths associated with reaction.‡Total number of reactions reported.In contrast to the absence of reports of ticlopidine-associated TTP cases in the phase 3 clinical trial setting, we identified 56 cases of ticlopidine-associated TTP in the stroke prevention setting and 42 cases of ticlopidine-associated TTP following coronary artery stent placement. The mean age of the patients undergoing stroke prevention was 66.9 years (SD, 11.8 years); for the patients with stent placement, 62.4 years (SD, 11.5 years) (P>.05). Normal platelet counts within 2 weeks of the onset of TTP were documented in most patients in both groups. Before the onset of TTP in patients undergoing stroke prevention and stent placement, ticlopidine had been used for less than 2 weeks in 5.4% and 2.4%, between 2 and 3 weeks in 17.9% and 21.4%, between 3 and 4 weeks in 30.4% and 38.1%, and between 4 and 12 weeks in 46.4% and 38.1%, respectively. Manifestations of TTP in the stroke prevention and coronary artery stent setting were similar, including thrombocytopenia (69.6% vs 76.2% had platelet counts <20 × 109/L); anemia (hemoglobin levels were <0.09 g/L in 26.8% and 26.2%); and neurologic changes, including focal deficits, convulsions, and/or coma (in 75.0% vs 69.1%). Renal insufficiency with a serum creatinine level greater than 221 µmol/L (>2.5 mg/dL) was more common in the stroke prevention vs coronary artery stent setting (35.7% vs 19.1%) (P>.05 for all comparisons). Overall TTP mortality in the stroke prevention setting was 37.5% vs 28.6% in the coronary artery stent setting (P>.05). Plasmapheresis was performed in 57.1% of the patients undergoing stroke prevention and in 66.7% of the patients with stent placement. When both groups of patients were combined to determine overall mortality rates associated with plasmapheresis, the mortality rate for all patients who did not undergo plasmapheresis was 57.9%, while the mortality rate for all patients who underwent plasmapheresis was 18.3% (P<.001).COMMENTWhile ticlopidine is approved by the FDA for the prevention of thrombotic strokes in aspirin-intolerant high-risk individuals, off-label use includes primary stroke prevention therapy in aspirin-tolerant individuals. In 1998, more than 2 million persons received ticlopidine (IMS America, Philadelphia, Pa, oral communication, October 30, 1998). Clinical trials have identified neutropenia as the most common serious adverse effect of ticlopidine use in the stroke prevention setting, with an incidence of 1.0% to 2.4% in most clinical trials.As a consequence, in 1991, the pharmaceutical manufacturer of the drug, concerned about patient safety, provided warnings to physicians and included a "black box" section in the original package insert describing the potential for neutropenia and encouraging physicians to closely monitor complete blood cell counts every 2 weeks for 3 months.Dangerous adverse effects of drugs are commonly discovered after marketing, with more than half of FDA-approved drugs having serious adverse effects that were not detected in clinical trials.Such was the case with ticlopidine. After initial marketing, reports of an additional serious adverse effect began to surface, with 25 TTP cases being identified in the MedWatch database.In 1994, the package insert was amended to include a boldfaced typed statement that TTP can occur in rare circumstances. However, underrecognition of ticlopidine-associated TTP continued, as noted in 1998 publicationsof 60 cases and 20 deaths from ticlopidine-associated TTP, raising additional questions about the safety of ticlopidine. The pharmaceutical manufacturer again revised the ticlopidine package insert in 1998 to include a black box warning section with statements describing an estimated incidence of ticlopidine-associated TTP of 1 in 2000 to 1 in 4000 individuals, the need for extreme vigilance for TTP, the signs and symptoms of TTP, and the importance of early diagnosis and treatment.Furthermore, concern over whether inclusion of TTP in the black box revision would be overlooked by physicians led the pharmaceutical manufacturer to send a "Dear Doctor" letter describing the revisions to all neurologists and cardiologists in the United States.Our results have practical implications for clinical practitioners. Since hematologic adverse effects almost always occur within the first 3 months of ticlopidine therapy, patients who receive longer-term ticlopidine therapy will need weekly or biweekly physician visits and complete blood cell count monitoring during the first 3 months of therapy. Before the publication of our initial reportsof ticlopidine-associated TTP, the black box warnings in the package insert dealt with neutropenia, which was diagnosed by laboratory tests and generally did not require frequent physician visits. Neutropenia cases were rarely fatal if ticlopidine therapy was discontinued and no other action was taken. Ticlopidine-associated TTP is fatal in more than 60% of patients who discontinue ticlopidine use but do not undergo timely therapeutic plasmapheresis. Physician evaluation should include specific attention to skin rashes, which may predate the onset of TTP in some individuals or represent purpura in others, or neurologic changes, which can be easily confused with stroke symptoms and lead to a delay in diagnosis and subsequent initiation of plasmapheresis. The results of laboratory tests, including complete blood cell count and creatinine level determination, should be checked for neutropenia, thrombocytopenia, anemia, or renal insufficiency. An abnormally low hematocrit or platelet count should be followed with a peripheral smear review looking for schistocytes. Cases with a high index of suspicion should be treated emergently with plasmapheresis, which can reduce the expected chances of death by 67%. As noted in several of the case reports, progression of symptoms can occur within hours. The most important factor associated with survival was receipt of plasmapheresis. Patients with ticlopidine-associated TTP in the setting of stroke prevention were less likely to undergo plasmapheresis, a factor that may account for the trend toward a higher mortality rate in this setting.Our findings do not imply that ticlopidine should no longer be used following coronary artery stent placement. Clinical trials have shown that ticlopidine is of added benefit when combined with aspirin. The first randomized trial (ISAR) found that, at 30 days of follow-up, the ticlopidine group had 75% fewer cardiac end points than the phenprocoumon group (1.6% vs 6.2%; P<.001) and had 0% vs 5.0% episodes of stent thrombosis.The largest study (STARS) found that combined antiplatelet therapy with ticlopidine and aspirin had an 80% reduction at 30 days in the combined end point of death, Q-wave myocardial infarction, emergency surgery, target vessel revascularization, and angiographic thrombosis (0.5%) compared with aspirin plus warfarin (2.7%; P= .007) or aspirin alone (3.6%; P<.001).Hall and coworkersfound a 1-month stent thrombosis rate of 2.9% in an aspirin-only group and 0.8% in a ticlopidine-aspirin group, although the results did not reach statistical significance because of a small sample size. Taken together, a recommendation shortening ticlopidine therapy from 4 to 2 weeks seems most prudent and is consistent with patterns of care in many centers in the United States. Most cases of stent thrombosis in the warfarin stenting era occurred within 2 weeks of stent placement. In STARS, stent thrombosis occurred at a mean of 0.7 days.Almost 90% of the cases of TTP following a coronary artery stent procedure have occurred more than 2 weeks after ticlopidine initiation. Also, a preliminary report from the Mayo Clinic (Rochester, Minn) has documented no increase in adverse events with coronary stenting when ticlopidine use was shortened from 4 to 2 weeks.Concerns over hematologic complications may have broad implications for antiplatelet agents. In 1998, the FDA approved an alternative antiplatelet agent, clopidogrel bisulfate, a thienopyridine derivative that differs structurally from ticlopidine by the addition of a carboxymethyl side group, as a potentially less toxic alternative to ticlopidine.Developed because clopidogrel did not show bone marrow toxicity in tissue culture and animal models, the large study of clopidogrel vs aspirin in patients at risk for ischemic events found the incidence of severe neutropenia with clopidogrel about the same as for aspirin (0.1% vs 0.2%), and no cases of clopidogrel-associated TTP were reported.Its antithrombotic activity is similar to that of ticlopidine, requiring its conversion to an active metabolite by the hepatic cytochrome P-450-1A. Clopidogrel-induced agranulocytosis was reported during the trial of clopidogrel vs aspirin in patients at risk for ischemic events (CAPRIE).While many physicians have switched from prescribing ticlopidine to prescribing clopidogrel, vigilance for hematologic complications, including agranulocytosis and TTP, with this new agent is advised. It took 6 years for widespread recognition of ticlopidine-associated TTP to occur. If similar rates of clopidogrel-associated TTP adverse effects occur (ie, 1 in 1600 patients), several years may elapse before similar numbers of cases associated with clopidogrel are identified in the literature.In 1997, ticlopidine was used by more than 2 million individuals as primary stroke prevention therapy and following coronary artery stent placement.In 1998, awareness of ADRs to ticlopidine had increased, yet as many as 1250 cases and 400 deaths from ticlopidine-associated TTP were expected to occur, if prompt recognition and treatment was not undertaken.In practice in the stroke prevention setting, vigilance for ticlopidine-associated hematologic toxic effects will undoubtedly lead to an increase in the number of physician visits and complete blood cell count tests ordered in the first 3 months of ticlopidine therapy or a switch to alternative antiplatelet agents such as clopidogrel. In contrast, in the coronary artery stent setting, ticlopidine plus aspirin is highly effective in minimizing the risk of stent thrombosis and major cardiac complications. With closely monitored use of 2 to 4 weeks of ticlopidine therapy following coronary artery stent placement, serious hematologic complications from ticlopidine should be avoidable. Taken together, the 1998 black box revision in the package insert, the Dear Doctor letter from the pharmaceutical manufacturer, and competition from clopidogrel will almost certainly adversely affect the future use of ticlopidine.EAmorosiJUltmannThrombotic thrombocytopenic purpura: report of 16 cases and review of the literature.Medicine.1996;45:139-159.MFurlanRRoblesMGalbuseravon Willebrand factor–cleaving protease in thrombotic thrombocytopenic purpura and the hemolytic uremic syndrome.N Engl J Med.1998;339:1578-1584.HMTsaiECYLianAntibodies to von Willebrand factor–cleaving protease in acute thrombotic thrombocytopenic purpura.N Engl J Med.1998;339:1585-1594.JMoakeStudies on the pathophysiology of thrombotic thrombocytopenic purpura.Semin Hematol.1997;34:83-89.CLBennettJEKissPDWeinbergThrombotic thrombocytopenic purpura after stenting and ticlopidine.Lancet.1998;352:1036-1037.Not AvailableTiclid (ticlopidine HCl) [ package insert].Nutley, NJ: Roche Laboratories Inc; 1998.CLBennettPDWeinbergKRozenberg-Ben-DrorPRYarnoldHCKwaanDGreenThrombotic thrombocytopenic purpura associated with ticlopidine: a review of 60 cases.Ann Intern Med.1998;128:541-544.SRSteinhublWATanJMFoodyEJTopolIncidence and clinical course of thrombotic thrombocytopenic purpura due to ticlopidine following coronary stenting.JAMA.1999;281:806-810.CLBennettCJDavidsonDGreenPDWeinbergDFeldmanThrombotic thrombocytopenic purpura following ticlopidine and coronary artery stenting.JAMA.In press.MBLeonDSBaimJPopmaA clinical trial comparing three antithrombotic-drug regimens after coronary stenting.N Engl J Med.1998;339:1665-1671.PHallSNakamuraLMaielloA randomized comparison of combined ticlopidine and aspirin therapy versus aspirin therapy alone after successful intravascular ultrasound-guided stent implantation.Circulation.1996;93:215-222.ASchomigFJNeumannAKastratiA randomized comparison of antiplatelet and anticoagulant therapy after the placement of coronary-artery stents.N Engl J Med.1996;334:1084-1089.MGentJDEastonVCHachinskiThe Canadian American Ticlopidine Study (CATS) in thromboembolic stroke.Lancet.1989;1:1215-1220.WKHassJDEastonHPAdamsA randomized trial comparing ticlopidine hydrochloride with aspirin for the prevention of stroke in high-risk patients (TASS).N Engl J Med.1989;321:501-507.Not AvailableTiclid (ticlopidine HCl) [package insert].Nutley, NJ: Roche Laboratories Inc; 1991.TJMooreBMPsatyCDFurbergTime to act on drug safety.JAMA.1998;279:1571-1573.DKWysowskiJBacsanyiBlood dyscrasias and hematologic reactions in ticlopidine users [letter].JAMA.1996;276:952.PBBergerDEGrillSJMelbyMRBellDRHolmes JrCan ticlopidine be safely discontinued two weeks after coronary stent placement?Paper presented at: Proceedings of the American College of Cardiology, March 29-April 1, 1998, Atlanta, Ga.Not AvailableA randomized, blinded trial of clopidogrel versus aspirin in patients at high risk of ischaemic events (CAPRIE): CAPRIE Steering Committee.Lancet.1996;348:1329-1339.DGreenCAPRIE trial correspondence [letter].Lancet.1997;349:354-355.RWinslowHematologist says Ticlid is implicated by data on cases of TTP blood disorder.Wall Street Journal.March 30, 1998:B5.Accepted for publication March 15, 1999.The opinions expressed in this article are solely those of the authors and do not represent opinions of the Food and Drug Administration, the Department of Veterans Administration, or Northwestern University, Chicago, Ill.Presented in part at the Food and Drug Administration Meeting, Rockville, Md, March 1998; the American Society of Hematology Meeting, San Diego, Calif, December 1997; and the American College of Cardiology Conference, Dallas, Tex, November 1998.We thank Steven Fredd, MD, Ray Lipicki, MD, and Diane Wysowski, PhD, for their comments and input before our presentation at the Food and Drug Administration; Attila Kursun, MD, John Godwin, MD, David Green, MD, PhD, and Hau Kwaan, MD, PhD, for comments on this and related articles; and the patients with thrombotic thrombocytopenic purpura whose comments and committed interest in this project were responsible for the continuation of these study efforts for the past 2 years.Reprints: Charles L. Bennett, MD, PhD, Veteran Affairs Medical Science Building, 400 E Ontario St, Suite 204, Chicago, IL 60611 (e-mail: [email protected]).
Armstrong, Gregory L.; Pinner, Robert W.
doi: 10.1001/archinte.159.21.2531pmid: 10573043
BackgroundRecent studies have documented increases in infectious disease mortality and in the proportion of hospitalizations attributable to infectious diseases. To further evaluate trends in the burden of infectious diseases in the United States, we analyzed data from the National Ambulatory Medical Care Survey from 1980 through 1996.ObjectiveTo examine the epidemiology of and recent trends in outpatient visits for infectious diseases.MethodsData were from a national probability sample of patient visits to office-based physicians. Diagnoses reported by the surveyed physicians were coded to indicate whether they were infectious or noninfectious. Infectious diseases were placed into 11 mutually exclusive categories.ResultsDuring the course of the survey, infectious diseases accounted for 19.0% of visits to physicians, or an average of 129 million visits per year. The infectious disease visit rate was higher in females than in males (587 vs 461 per 1000 persons per year) and higher in non-Hispanic whites than in non-Hispanic blacks or Hispanics (538 vs 407 vs 485 per 1000 persons per year). The visit rate for infectious diseases was greatest in 0- to 4-year-olds. Upper respiratory tract infections accounted for the largest proportion of visits (38.0% of infectious disease visits), followed by otitis (15.1%) and lower respiratory tract infections (14.1%). The age-adjusted visit rate for infectious diseases increased from 462 visits per 1000 persons (17.5% of all visits) in 1980 to 575 (20.2%) in 1990. From 1990 to 1996, this rate declined to 483 per 1000 (18.1%).ConclusionsInfectious diseases are responsible for a substantial proportion of outpatient visits to physicians in the United States. Upper respiratory tract infections account for the largest proportion of these visits.FOR MOST OF the 20th century, industrialized societies have experienced a steady decline in the burden of infectious diseases. However, experience in the last decade has shown that this trend may have stopped. Events such as the epidemic of human immunodeficiency virus, the resurgence of tuberculosis, and outbreaks of newly recognized diseases such as cryptosporidiosis and hantavirus pulmonary syndrome have raised the public's awareness of the threat of infectious diseases and have spawned public health efforts to monitor and control emerging and reemerging infections.To examine trends in the overall burden of infectious diseases in the United States, we analyzed 3 national databases maintained by the Centers for Disease Control and Prevention's National Center for Health Statistics (NCHS). The first of these found that the death rate from infectious diseases had increased 58%, from 41 to 65 deaths per 100 000, between 1980 and 1992.The second, a study of hospital discharge data, found that the hospitalization rate for infectious diseases had declined between 1980 and 1994 but that the rate of decline was less than half that for all hospitalizations, resulting in an increase of 27% in the proportion of hospitalizations for infectious diseases.During the same period, the fatality rate for infectious disease hospitalizations had doubled.In this study, we analyzed a third database, the National Ambulatory Medical Care Survey (NAMCS), to better define the burden of infectious diseases in the outpatient medical setting and to evaluate recent trends in ambulatory visits for infectious disease.METHODSDATA SOURCESWe analyzed data from the NAMCS from 1980 through 1996.The NAMCS is a national probability sample survey that has been conducted by the NCHS in each year between 1980 and 1996 except from 1982 to 1984 and 1986 to 1988.The sampling frame of the survey includes physicians in the American Medical Association and American Osteopathic Association master files who are (1) office based, (2) principally engaged in patient-care activities, (3) not federally employed, and (4) not in the specialties of anesthesiology, pathology, or radiology. Within this frame, physicians are chosen according to a multistage probability sampling design. Chosen physicians are then asked to collect information on a random sample of patient visits during the course of a randomly selected week.For each visit, the physician or office assistant fills out a 1-page form that includes questions about the patient's demographic characteristics, the principal diagnosis made at the visit, other diagnoses, tests ordered, and medications prescribed. Information on the forms is then entered into the database by staff at the NCHS, and diagnoses are coded according to the International Classification of Diseases, Ninth Revision(ICD-9).Each visit is assigned a weight equal to the reciprocal of its probability of selection to obtain national estimates.From 1980 through 1996, between 1354 and 2879 physicians participated in the survey (participation rate, 70% to 74%). Approximately 33 000 to 46 000 visits were sampled in each of the years except 1985, in which 71 954 visits were sampled.Population data were obtained from the US Bureau of Census. Intercensal or postcensal estimates of the civilian noninstitutionalized population were used in all calculations of rates. The populations of Hawaii and Alaska were removed from the 1980, 1981, and 1985 data, since these states were not included in the surveys performed in those years.CLASSIFICATION OF ICD-9CODESWe used a previously described protocol to classify ICD-9codes according to whether they were infectious or noninfectious.We restricted our analysis to those diagnoses that are always or almost always caused by infectious agents. Infectious diseases were grouped into 10 categories of related conditions (Table 1) that accounted for 94% of all infectious disease visits. A category of "other infectious diseases" was created for the remaining 6% of visits.Table 1. Classification System for Infectious Diseases*Disease CategoryICD-9CodesUpper respiratory tract infections032.0-032.9, 034.0, 101.0, 460.0-465.9, 473.0-473.9, 474.0, 475.0Otitis media and externa380.1, 382.0-382.9, 383.0-383.2, 384.0Lower respiratory tract infections and influenza022.1, 031.0, 033.0-033.9, 466.0-466.1, 480.0-487.9, 490.0, 510.0-510.9, 511.1, 513.0-513.1, 517.1Skin infections035.0, 053.0-053.9, 078.0-078.1, 079.4, 110.0-111.9, 112.3, 680.0-680.9, 681.0-682.9, 684.0-684.9, 685.0, 686.0-686.9Urinary tract infections590.0-590.9, 595.0, 597.0-597.9, 598.0, 599.0, 646.6Other viral infections045.0-052.9, 054.0, 054.2-066.9, 071.0-076.9, 078.8-079.3, 079.8-079.9, 790.8Eye and eyelid infections054.4, 077.0-077.9, 360.0, 370.0, 372.2, 372.3, 373.0-373.2, 373.4-373.6, 376.0Vaginitis and cervicitis112.1, 616.0-616.1Sexually transmitted diseases054.1, 090.0-099.9, 131.0-131.9, 614.0-614.5, 647.0-647.2Enteric infections001.0-009.9, 022.2, 127.0-127.9, 129.0* ICD-9indicates International Classification of Diseases, Ninth Revision. Infectious diseases not belonging to any of the above categories were placed in an 11th category, "other infectious diseases."DATA ANALYSISData were analyzed with SAS software (SAS Institute Inc, Cary, NC). Standard errors of point estimates of the number of visits were calculated according to algorithms published in the documentation for the public use data tapes.Only principal diagnoses were considered in the analyses. All visit records were weighted by using weights supplied on the public use data tapes. Analyses used either (1) the visit rate, expressed as the number of visits per 1000 persons per year, or (2) the visit proportion, expressed as a percentage of all visits that year attributed to the particular diagnosis. Visit rates and visit proportions were compared by means of ttest or 1-way analysis of variance, assuming no covariance between the groups. When data from several years were combined to give a single estimate, the SE of this estimate was calculated according to algorithms recommended by the NCHS (Iris M. Shimizu, PhD [[email protected]], e-mail, September 26, 1997).Where indicated, rates were age adjusted to the 1990 census with 1-year age groups by means of the direct method.Trends were analyzed by using weighted linear regression with a segmented model. The model included the year as the independent variable and the age-adjusted number of visits in each year as the dependent variable. For each year, the point estimate of number of visits was weighted with the inverse of its variance. The model was segmented, allowing the slope before a designated pivot point to be different from that afterward. To decide where to place this pivot point, the point was varied from 1981 to 1995 and the model was fit to the data. The model that produced the lowest residual sum of squares was chosen.RESULTSVISITS TO PHYSICIANS FOR INFECTIOUS DISEASESDuring the 11 survey years included in the study, there were an estimated 677 million visits to physicians annually, or 2774 visits per 1000 persons per year. Infectious diseases accounted for 19.0% of these visits, with an average rate of 526 visits per 1000 persons per year. When routine visits (visits for physical examinations or for administrative purposes only) were removed from the database, infectious diseases accounted for 21.2% of all visits.The overall rate of outpatient visits varied by age group (Figure 1). This rate was lowest in the 5- to 24-year age group and highest in the oldest 2 age groups. In contrast, the infectious diseases visit rate was highest in the youngest age group and decreased with age until the 65- to 84-year age group, after which the visit rate leveled off. In general, the proportion of all visits attributable to infectious diseases decreased with age. Infectious diseases accounted for 45% of all visits in 0- to 4-year-olds but only 15.9% of visits in all other age groups combined.Figure 1.Average annual outpatient visit rate for all diagnoses, by age group and sex, in the United States during 1980, 1981, 1985, and 1989 through 1996. The differences in rates between males and females were significant (P<.05) in all but the youngest age group.Visit rates for all diagnoses and visit rates for infectious diseases were higher for females than for males (Table 2, and Figure 2). Compared with the visit rate for males, that for females was 41% higher overall and 27% higher for infectious diseases. For infectious diseases, this difference resulted from disparities in visit rates in the age groups between 5 and 84 years. There was no significant difference in infectious disease visit rates between boys and girls aged 0 to 4 years or between men and women 85 years old and older.Table 2. Comparison of Visit Rates by Sex and RaceDisease CategoryRatio of Visit RatesWomen vs MenNon-Hispanic Whites vs Non-Hispanic BlacksNon-Hispanic Whites vs HispanicsUpper respiratory tract infections1.28*1.35*1.11†Otitis media and externa0.91†1.85*1.05Lower respiratory tract infections and influenza1.20*1.53*1.08Skin infections1.041.50*1.48*Other infectious diseases0.940.981.00Urinary tract infections3.67*1.091.22Other viral infections1.101.41*1.02Eye and eyelid infections1.30*1.36†1.08Vaginitis and cervicitisNot applicable0.63*1.07Sexually transmitted diseases2.05*0.37*1.15Enteric infections1.50†0.990.93All infectious diseases1.27*1.32*1.11†All visits to physicians1.41*1.41*1.43**P<.005.†P<.05.Figure 2.Average annual outpatient visit rate for infectious diseases, by age group and sex, in the United States during 1980, 1981, 1985, and 1989 through 1996. The differences in rates between males and females were significant (P<.05) in all but the youngest and oldest age groups.Visit rates for all diagnoses and visit rates for infectious diseases were higher in non-Hispanic whites than in non-Hispanic blacks or Hispanics (Table 2, Figure 3). The visit rates for non-Hispanic whites were 41% higher overall and 32% higher for infectious diseases than rates for non-Hispanic blacks. The visit rates among non-Hispanic whites were 43% higher overall and 11% higher for infectious diseases than rates for Hispanics.Figure 3.Average annual outpatient visit rate for infectious diseases, by age group and race and ethnic group, per 1000 persons in the United States during 1980, 1981, 1985, and 1989 through 1996.TRENDS IN VISITS FOR INFECTIOUS DISEASESThe segmented linear regression model fit the data best when the pivot point was placed at 1990. This model fit the data better than the nonsegmented model (residual sum of squares, 15.8 vs 38.5, P=.006 for the difference).Between 1980 and 1990, the visit rate for infectious diseases increased by 2.14% per year (P=.006, Table 3and Figure 4). This trend was seen in all age groups and was statistically significant in all but 0- to 4-year-olds. Overall, the proportion of visits attributable to infectious diseases increased from 17.5% to 20.2% (P=.003 for trend). This increase in proportion of visits attributable to infectious diseases was statistically significant in all but the 2 oldest age groups, in which increases in infectious disease visits were matched by increases in visits for all reasons.Table 3. Trends in Outpatient Visit RatesDisease CategoryVisits per Year (Rate/1000)*1980 to 19901990 to 1996Change in Rate of Visits per Year (95% CI)†% Change‡Change in Rate of Visits per Year (95% CI)†% Change‡Upper respiratory tract infections48,783 (199.6)4.35 (0.98 to 7.72)2.18§−1.48 (−7.96 to 4.99). . .Otitis media and externa19,436 (79.5)2.73 (1.31 to 4.15)3.43∥−3.46 (−6.19 to −0.74)−4.35§Lower respiratory tract infections and influenza18,129 (74.2)1.99 (0.67 to 3.30)2.68∥−2.23 (−4.72 to 0.26). . .Skin infections11,814 (48.3)0.36 (−0.38 to 1.10). . .−1.72 (−3.06 to −0.39)−3.56§Other infectious diseases7181 (29.4)0.40 (−0.21 to 1.01). . .−0.54 (−1.72 to 0.63). . .Urinary tract infections6116 (25.0)0.48 (−0.03 to 0.98). . .−1.01 (−1.95 to −0.06)−4.03§Other viral infections5822 (23.8)0.56 (−0.15 to 1.27). . .−0.49 (−1.86 to 0.89). . .Eye and eyelid infections4716 (19.3)0.29 (−0.06 to 0.65). . .−0.37 (−1.05 to 0.32). . .Vaginitis and cervicitis3928 (16.1)0.10 (−0.30 to 0.49). . .−1.66 (−2.31 to −1.01)−10.31∥Sexually transmitted diseases1349 (5.5)−0.20 (−0.49 to 0.08). . .−0.94 (−1.33 to −0.55)−17.09∥Enteric infections1255 (5.1)0.27 (0.07 to 0.48)5.29§−0.59 (−0.99 to −0.20)−11.57§All infectious diseases128,529 (525.8)11.27 (4.12 to 18.36)2.14∥−15.26 (−28.34 to −2.19)−2.90§All visits to physicians677,870 (2773.6)22.5 (0.22 to 44.79)0.81§−27.21 (−67.16 to 12.75). . .*Thousands of visits per year in which the primary diagnosis was within the given category; numbers in parentheses are visits per 1000 population per year.†Estimated annual increase or decrease (and 95% confidence interval [CI]) in number of visits per 1000 population. Age-adjusted data were used to estimate trends.‡Estimated average annual percentage change in age-adjusted visit rate; data not shown where P>.05 for trend (ellipses).§P<.05.∥P<.01.Figure 4.Trends in outpatient visits for infectious diseases, showing the age-adjusted infectious disease visit rate and 95% confidence interval for each year of the survey. The dotted line represents the fitted regression curve.There was a significant decrease in visits for infectious diseases from 1990 to 1996 (2.9% per year; P=.03; Figure 4), and the proportion of all visits resulting from infectious diseases decreased from 20.2% to 18.1% (P=.02). This decrease was significant only in 25- to 44-year-olds and 45- to 64-year-olds.INFECTIOUS DISEASE CATEGORIESThe top 4 categories of infectious diseases—upper respiratory tract infections, otitis media and otitis externa, lower respiratory tract infections, and skin infections—accounted for 76.4% of all visits for infectious diseases (Figure 5). Within these categories, as in most of the categories, a small number of diagnoses accounted for most of the visits (Table 4).Figure 5.Visit rates for individual categories of infectious diseases, showing the average annual visit rate per 1000 persons in the United States during 1980, 1981, 1985, and 1989 through 1996. URTI indicates upper respiratory tract infection; LRTI, lower respiratory tract infection; UTI, urinary tract infection; and STDs, sexually transmitted diseases.Table 4. Diagnoses in the 4 Largest Disease CategoriesDisease CategoryDiagnoses (ICD-9Code)*PercentUpper respiratory tract infectionsUnspecified upper respiratory tract infection (465.9)35.3Acute pharyngitis (462)21.3Unspecified chronic sinusitis (473.9)19.7Other23.7Otitis media and externaAcute unspecified otitis media (382.9)87.8Infective otitis externa (380.1)10.8Other1.4Lower respiratory tract infections and influenzaUnspecified bronchitis (490)53.4Acute bronchitis (466)14.4Unspecified pneumonia (486)13.7Influenza (487.1)13.3Other5.2Skin infectionsCellulitis and pyogenic skin infections (035, 680.0-680.9, 684-686.9)40.1Warts (78.1)31.9Dermatomycoses (110.0-111.9, 112.3)18.3Zoster (53.0-53.9)7.5Molluscum contagiosum (78.0)1.8Other0.4*ICD-9indicates International Classification of Diseases, Ninth Revision.During the 1980s, significant increases in visit rates were seen in the 3 largest categories of infectious diseases (upper respiratory tract infections [P=.02], otitis media and otitis externa [P=.001], and lower respiratory tract infections and influenza [P=.008]) as well as in enteric infections (P=.02; Table 3). Because visits for all 4 of these categories were increasing faster than all visits, the proportion of visits attributable to diseases in these categories also increased significantly.Of the disease categories, only sexually transmitted diseases declined during this period. The decline in visit rate was not statistically significant, but, because overall visits were increasing during this time, there was a significant decline in the proportion of visits attributable to this category (P=.04).During the 1990s, visit rates for 6 infectious disease categories decreased: otitis media and externa (P=.02), skin infections (P=.02), urinary tract infections (P=.04), vaginitis and cervicitis (P=.001), sexually transmitted diseases (P=.001), and enteric infections (P=.008). Visits for diagnoses in these categories also decreased as a proportion of all visits (P=.009, .03, .06, <.001, <.001, and .01, respectively).COMMENTDuring the years included in this study, infectious diseases accounted for 19.0% of ambulatory visits to office-based physicians in the United States, or approximately 129 million visits each year. These measures provide conservative estimates of the impact of infectious diseases on outpatient medicine, because they include only conditions that are always or almost always caused by infectious agents. Numerous conditions that are often a direct result of infectious diseases, especially chronic and neoplastic diseases such as hepatic cirrhosis and cervical cancer, were not examined.Additionally, certain outpatient visits fall outside the scope of NAMCS. Visits to emergency departments and to hospital-based outpatient clinics are captured in a separate survey, the National Hospital Ambulatory Medical Care Survey, which began in 1992.Examination of the data in the 1995 surveywith the use of criteria from this study shows that infectious diseases accounted for 21.5% of the estimated 96.5 million visits to emergency departments and 14.7% of the estimated 67.2 million visits to hospital-based clinics. Government-sponsored health clinics, such as sexually transmitted disease clinics, military clinics, and prison clinics, fall outside the scope of both surveys.Trends in visits for infectious diseases were not uniform during the course of our study. It is clear that visits for infectious diseases increased slightly from 1980 until the early 1990s and that infectious diseases constituted an increasing proportion of illness seen by outpatient practitioners. But the magnitude of this increase was small compared with that seen for the infectious disease mortality rateor the proportion of hospitalizations for infectious disease.Infectious disease visits have declined slightly since 1990. The reason for this decline and the increase that preceded it is unclear. Infectious disease visit rates depend not only on the incidence of disease but also on factors such as access to care, current diagnostic practices, and the availability of over-the-counter medications.Visits for certain categories of infectious diseases declined during the study, most prominently for vaginitis and cervicitis and for sexually transmitted diseases. The decrease in vaginitis and cervicitis mostly reflects a decline in unspecified vaginitis and candidal vaginitis, which accounted for 82.8% of the visits in this category. The decline in outpatient visits for this disorder may be a result of the availability since 1992 of over-the-counter treatments for fungal vaginitis.The trends in visits for sexually transmitted diseases are more complex and are similar to those documented by the National Notifiable Disease Surveillance System.Visits for gonorrhea declined steadily during the 16 years, a trend that has been attributed to improvements in screening and treatment initiated in the 1970s.Visits for syphilis increased during the late 1980s and declined in the 1990s, a trend that has been attributed by some to the rise and fall in crack cocaine abuse and that, according to national surveillance data, was most prominent in minorities in Eastern cities and in the South.The NAMCS is a useful component in determining the overall burden of infectious diseases in the United States and in monitoring trends in broad categories of infectious diseases. However, 3 factors limit its usefulness in the surveillance of individual diseases. First, the survey is not large enough to provide precise estimates of visits for most specific diseases. Second, the diagnoses used are clinical diagnoses given by the physician at the time of the patient visit and often have not been confirmed by laboratory tests. It is possible that many of the visits attributed to infectious diseases in the survey are actually the result of noninfectious conditions such as allergies that are sometimes clinically indistinguishable from certain infectious diseases. Third, diagnoses in the survey are not assigned according to uniform case definitions.Three indicators of overall burden of infectious diseases increased during the 1980s: the mortality rate attributable to infectious diseases,the proportion of hospitalizations attributable to infectious diseases,and the rate of ambulatory visits for infectious diseases. Many factors, such as the emergence of acquired immunodeficiency syndrome, the reemergence of tuberculosis, and the increase in antibiotic resistance, probably contributed to this increase. The trends in the 1990s have been different. Ambulatory visits for infectious diseases have declined, while infectious disease mortality may have leveled off (G.L.A. and R.W.P., unpublished data, 1998). Interpretation of these trends is complex, given the diverse, sometimes offsetting factors that influence them— eg, the impact of protease inhibitors on improved survival in persons with acquired immunodeficiency syndrome, increased public health efforts to control tuberculosis, and structural changes in health care delivery that have resulted in trends toward fewer hospitalizations and outpatient visits. It is clear, however, that many dynamic factors continue to promote the emergence of infectious diseases and that trends in the infectious disease burden will probably continue to fluctuate.JCButlerCJPetersHantaviruses and hantavirus pulmonary syndrome.Clin Infect Dis.1994;19:387-394.WRMacKenzieNJHoxieMEProctorA massive outbreak in Milwaukee of cryptosporidium infection transmitted through the public water supply.N Engl J Med.1994;331:161-167.Institute of MedicineEmerging Infections: Microbial Threats to Health in the United States.Washington, DC: National Academy Press; 1992.Centers for Disease Control and PreventionAddressing Emerging Infectious Disease Threats to Health: A Prevention Strategy for the United States.Atlanta, Ga: US Dept of Health and Human Services; 1994.RWPinnerSMTeutschLSimonsenTrends in infectious diseases mortality in the United States.JAMA.1996;275:189-193.LSimonsenLAConnRWPinnerSMTeutschTrends in infectious disease hospitalizations in the United States, 1980-1994.Arch Intern Med.1998;158:1923-1928.National Center for Health StatisticsNational Ambulatory Medical Care Survey: Public Use Data Tape Documentation.Hyattsville, Md: National Center for Health Statistics, Centers for Disease Control and Prevention; 1980.EBryantIShimizuSample design, sampling variance, and estimation procedures for the National Ambulatory Medical Care Survey.Vital Health Stat 2.1988;1-39.World Health OrganizationInternational Classification of Diseases, Ninth Revision (ICD-9).Geneva, Switzerland: World Health Organization; 1977.PArmitageGBerryStatistical Methods in Medical Research.Boston, Mass: Blackwell Scientific Publications; 1994:436.LCMcCaigTMcLemorePlan and operation of the National Hospital Ambulatory Medical Survey.Vital Health Stat 1.1994;1-78.National Center for Health StatisticsNational Hospital Ambulatory Medical Care Survey: Public Use Data Tape and Documentation.Hyattsville, Md: National Center for Health Statistics, Centers for Disease Control and Prevention; 1995.JDSobelSFaroRWForceVulvovaginal candidiasis: epidemiologic, diagnostic, and therapeutic considerations.Am J Obstet Gynecol.1998;178:203-211.DGFerrisCDekleMSLitakerWomen's use of over-the-counter antifungal pharmaceutical products for gynecologic symptoms.J Fam Pract.1996;42:595-600.Centers for Disease Control and PreventionSummary of notifiable diseases, 1995.MMWR Morb Mortal Wkly Rep.1997;44:33-59.KFoxWLWhittingtonWCLevineJSMoranAAZaidiAKNakashimaGonorrhea in the United States, 1981-1996: demographic and geographic trends.Sex Transm Dis.1998;25:386-393.AKNakashimaTTRolfsMLFlockPKilmarxJRGreenspanEpidemiology of syphilis in the United States, 1941-1993.Sex Transm Dis.1996;23:16-23.DHMartinRPDiCarloRecent changes in the epidemiology of genital ulcer disease in the United States: the crack cocaine connection.Sex Transm Dis.1994;21(2 suppl):S76-S80.Accepted for publication March 15, 1999.This work was supported by a National Foundation for Infectious Diseases–Merck Postdoctoral Fellowship in Emerging Infectious Diseases, Bethesda, Md (Dr Armstrong).Presented in part at the International Conference on Emerging Infectious Diseases, Atlanta, Ga, March 11, 1998.Reprints: Gregory L. Armstrong, MD, Mailstop G-37, 1600 Clifton Rd NE, Atlanta, GA 30333 (e-mail: [email protected]).
Strand, Vibeke; Cohen, Stanley; Schiff, Michael; Weaver, Arthur; Fleischmann, Roy; Cannon, Grant; Fox, Robert; Moreland, Larry; Olsen, Nancy; Furst, Dan; Caldwell, Jacques; Kaine, Jeffrey; Sharp, John; Hurley, Frank; Loew-Friedrich, Iris
Gleason, Patrick P.; Meehan, Thomas P.; Fine, Jonathan M.; Galusha, Deron H.; Fine, Michael J.
doi: 10.1001/archinte.159.21.2562pmid: 10573046
BackgroundAlthough medical practice guidelines exist, there have been no large-scale studies assessing the relationship between initial antimicrobial therapy and medical outcomes for patients hospitalized with pneumonia.ObjectiveTo determine the associations between initial antimicrobial therapy and 30-day mortality for these patients.MethodsHospital records for 12,945 Medicare inpatients (≥65 years of age) with pneumonia were reviewed. Associations between initial antimicrobial regimens and 30-day mortality were assessed with Cox proportional hazards models, adjusting for baseline differences in patient characteristics, illness severity, and processes of care. Comparisons were made with patients treated with a non-pseudomonal third-generation cephalosporin alone (the reference group).ResultsInitial treatment with a second-generation cephalosporin plus macrolide (hazard ratio [HR], 0.71; 95% confidence interval [CI], 0.52-0.96), a non-pseudomonal third-generation cephalosporin plus macrolide (HR, 0.74; 95% CI, 0.60-0.92), or a fluoroquinolone alone (HR, 0.64; 95% CI, 0.43-0.94) was independently associated with lower 30-day mortality. Adjusted mortality among patients initially treated with these 3 regimens became significantly lower than that in the reference group beginning 2, 3, and 7 days, respectively, after hospital admission. Use of a β-lactam/β-lactamase inhibitor plus macrolide (HR, 1.77; 95% CI, 1.28-2.46) and an aminoglycoside plus another agent (HR, 1.21; 95% CI, 1.02-1.43) were associated with an increased 30-day mortality.ConclusionsIn this study of primarily community-dwelling elderly patients hospitalized with pneumonia, 3 initial empiric antimicrobial regimens were independently associated with a lower 30-day mortality. The more widespread use of these antimicrobial regimens is likely to improve the medical outcomes for elderly patients with pneumonia.EACH YEAR in the United States approximately 4 million adults develop pneumonia, of whom more than 1 million patients are hospitalized.In 1993, in-hospital mortality for pneumonia among patients older than 65 years was 10.7 deaths per 100 discharges, and in that year alone $3.5 billion was spent on inpatient care of Medicare patients with this illness.Because of the substantial mortality of pneumonia, particularly among the elderly, it is essential that initial antimicrobial therapy have activity against the causative organism(s). Unfortunately, the causative organism(s) are often unknown at the time antimicrobial therapy is initiated; bacteriological culture results and other microbiological studies are positive in less than 50% of hospitalized patients, even in carefully conducted prospective studies of pneumonia etiology.Wide variations in antimicrobial prescribing practices exist for the treatment of community-acquired pneumonia.To reduce this variation and improve the appropriateness of antimicrobial therapy, the Infectious Disease Society of America (IDSA) and the American Thoracic Society (ATS) have published guidelines for empirical antimicrobial therapy for this illness.However, these guidelines were derived from a limited number of clinical studies of pneumonia etiology and have not been validated in clinical practice.Consequently, even the authors of these guidelines have advocated caution in their clinical adoption until the implications for patient outcomes are better understood.To better understand current prescribing practices and to assess associations between empirical antimicrobial therapy and patient outcomes, we designed a study with the following aims: (1) to describe the initial antimicrobial regimens most frequently prescribed for all hospitalized patients, including the subsets of community-dwelling patients and those admitted from long-term care facilities (LCFs), and (2) to assess the associations between initial antimicrobial regimens, 30-day mortality, and other relevant medical outcomes. Our primary hypothesis was that initial antimicrobial therapy, which includes coverage for "atypical" in addition to "typical" bacterial pathogens, would be associated with lower 30-day mortality. We made an explicit decision to separately assess the impact of initial antimicrobial therapy on medical outcomes among patients admitted from the community and from LCFs caused by differences in sociodemographic factors, comorbidity, illness severity, and pneumonia etiology across these patient subsets.PATIENTS AND METHODSSTUDY POPULATIONThe Medicare Quality Indicator System is a Health Care Financing Administration–sponsored standardized data collection system developed to assess quality of care for hospitalized patients with specific clinical conditions. The Medicare Quality Indicator System pneumonia module, a national, community-based, retrospective study of pneumonia care in adults aged 65 years and older who were community-dwelling or admitted from an LCF, was initiated in March 1994.SAMPLE SELECTIONSample selection has been described in detail previously.Briefly, from October 1, 1994, through September 30, 1995, potential pneumonia cases were identified by the Health Care Financing Administration from the Medicare National Claims History File if they had a principal discharge diagnosis of pneumonia according to International Classification of Diseases, Ninth Edition, Clinical Modification(ICD-9-CM) codes, or if they had a principal discharge diagnosis of respiratory failure and a secondary diagnosis of pneumonia. With the use of a random selection procedure, 26 000 discharges (500 from each state, the District of Columbia, and Puerto Rico) were selected from approximately 650 000 nonfederal, acute care hospital discharges with pneumonia.After potential cases were identified, data were extracted from the medical records to confirm the diagnosis of pneumonia and to apply exclusion criteria. Of the 26 000 potential cases, medical records were obtained from 25 561 (98.3%). Case confirmation required that the patient have an ICD-9-CMcode for pneumonia,that a clinician document an initial working diagnosis of pneumonia, and that a chest x-ray examination performed within the first 48 hours after hospital presentation be reported as consistent with pneumonia. Patients were excluded if they were younger than 65 years, had experienced acute care hospitalization within the previous 10 days, were infected with the human immunodeficiency virus, had the acquired immunodeficiency syndrome, had a history of organ transplantation (heart, lung, liver, kidney, or bone marrow), had been exposed to chemotherapy or immunosuppressive therapy within the previous 2 months, had been transferred from another acute care facility, or had died or been discharged on the date of hospitalization. Patients in whom we could not document delivery of antimicrobial therapy within 48 hours after hospitalization (n=483), patients whose residence was unknown (n=68), patients whose 30-day mortality could not be verified (n=33), and patients with missing information on the exact time of antimicrobial administration or the exact time of blood culture performance (n=874) were not included in these analyses. For patients with more than 1 pneumonia hospitalization during the study period (n=113), only the initial episode of pneumonia was included.DATA COLLECTIONHospitals were asked to provide copies of the medical records of potential cases. Trained medical record abstractors collected the data from the medical records by means of an electronic data collection instrument. Abstracted data were merged with hospital claims data provided by the Health Care Financing Administration. Reliability testing indicated moderate to excellent interabstractor agreement, with κ statistics ranging from 0.48 to 0.95 for pneumonia confirmation and exclusion criteria, clinical characteristics, antimicrobial therapy before hospitalization, timing of initial antimicrobial therapy, and performance of blood cultures within 24 hours of hospitalization.DATA ELEMENTSFive categories of variables were used in this study: (1) case confirmation and exclusion criteria (listed previously), (2) patient characteristics, (3) processes of care, (4) antimicrobial agents prescribed, and (5) medical outcomes. Patient characteristics included demographic characteristics (age, sex, and coming from an LCF), comorbid illnesses (cerebrovascular disease, congestive heart failure, and neoplastic disease excluding skin cancer, liver disease, and renal disease), physical examination findings (abnormal mental status, temperature, heart rate, respiratory rate, systolic blood pressure), and laboratory or radiographic results (arterial pH, serum urea nitrogen level, sodium level, glucose level, hematocrit, PaO2, and pleural effusion). For purposes of this study, an LCF consisted of a skilled nursing home or a long-term or intermediate care facility.Each patient's risk of 30-day mortality was assessed by means of a validated pneumonia-specific mortality risk index with demonstrated accuracy and discrimination among Medicare patients with pneumonia.Patients were assigned to 1 of 4 severity categories (risk classes II-V) based on the presence of the 3 demographic characteristics, 5 comorbid illnesses, 5 physical examination abnormalities, and 7 laboratory or radiographic findings listed above. The physical examination and laboratory values used were the first recorded findings in the initial 24 hours of hospitalization. All other patient characteristics were taken from abstracted data elements except liver disease and neoplastic disease, which were derived from a combination of abstracted data elements and coded secondary ICD-9-CMdiagnoses, and renal disease, which was assessed from secondary diagnosis codes.Process of care variables included hospital arrival date and time, blood culture collection date and time, and initial antimicrobial administration date and time. Initial intensive care unit (ICU) admission was defined as ICU treatment with the following procedures documented within 24 hours of arriving at the hospital: insertion of an endotracheal tube, respiratory tract intubation, continuous positive airway pressure mechanical ventilation, pulmonary artery catheterization or monitoring, cardiac output determination by thermodilution, or central venous catheter monitoring. Use of all antimicrobial agents was assessed from abstracted data. For each antimicrobial agent prescribed, the date and time of the first dose were recorded; however, the route of administration and discontinuation date were not recorded. The initial antimicrobial regimen was defined as all antimicrobial agents used during the first 48 hours after arrival at the hospital.The cause of pneumonia was defined by the presence of ICD-9-CMdiagnosis codes for Streptococcus pneumoniae, gram-negative bacilli (Klebsiella pneumoniae, Pseudomonas aeruginosa), Haemophilus influenzae, staphylococcal species (Staphylococcus aureus, Staphylococcus epidermidis), viral pneumonia, fungal pneumonia, Pneumocystis carinii, aspiration pneumonia, miscellaneous organisms, and unknown etiology.High-risk pneumonia etiology was defined as gram-negative rod, staphylococcal species, or aspiration pneumonia.Outcome variables were the date of hospital discharge, date of subsequent hospitalization if applicable, and date of death. Mortality data were obtained from the Medicare Enrollment Database, and rehospitalization was assessed by means of Medicare part A claims. Mortality was defined as death within 30 days from the date of the index hospitalization. Length of stay (LOS) was defined as the discharge date minus the admission date. Rehospitalization was defined as any hospitalization within 30 days from the discharge date of the index hospitalization. Assessments of hospital (LOS) and rehospitalization were limited to patients surviving the index hospitalization for pneumonia who were not directly transferred to another acute care hospital.METHODS OF ANALYSISAssociations between patient outcomes (ie, mortality, rehospitalization, and LOS) and initial antimicrobial regimens that accounted for greater than 1% of all initial regimens were assessed in all patients and in 2 patient subsets. These 2 patient subsets consisted of patients who were (1) community-dwelling (n=9751) and (2) LCF-dwelling (n=3194). Associations between patient outcomes and initial antimicrobial regimens were also assessed by stratification according to severity risk classes. Because of the small number of patients in risk class II (n=1189), patients in risk classes II and III were combined (n=4099).For categorical data, proportions were compared by means of the Pearson χ2test. For each antimicrobial regimen, testing for trends in frequency of antimicrobial regimen use by severity risk class was performed with the Mantel-Haenszel χ2test for trend. For continuous data (eg, LOS), we calculated medians with interquartile ranges and means with SDs. All LOS analyses were performed with the log transformation of the actual LOS as the dependent variable.The independent associations between initial antimicrobial regimens and 30-day mortality were assessed by Cox proportional hazards models. Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated for each initial antimicrobial regimen, adjusting for the following independent variables: antimicrobial therapy before hospitalization, 4 baseline pneumonia severity risk classes, arrival from an LCF, initiation of antimicrobial therapy within 8 hours of hospital arrival, performance of blood cultures within 24 hours of hospital arrival, region of enrollment within the United States, ICU treatment on day 1 of hospitalization, change in antimicrobial therapy after the initial 48 hours of hospitalization, and high-risk pneumonia etiology. With the exception of previous antimicrobial therapy and region of enrollment, all of these patient characteristics and process measures have previously been shown to have a significant association with short-term mortality in patients with pneumonia.Significant 2-way interactions between ICU treatment and high-risk etiology, ICU treatment and risk class, and coming from an LCF and high-risk etiology were identified and used as independent variables in the Cox proportional hazards models. The reference category used for initial antimicrobial regimen was therapy with a non-pseudomonal third-generation cephalosporin only (ie, ceftriaxone, cefotaxime, or ceftizoxime), an initial regimen recommended by the ATS and the IDSA guidelines for hospitalized patients with moderate to severe pneumonia.Log-log survival plots were constructed to assess the proportionality assumption underlying the Cox models.The results of all Cox models were confirmed by logistic regression, indicating that our results were insensitive to modeling technique.To further assess the association between initial antimicrobial regimen and 30-day mortality, 2 separate Cox models were performed in community-dwelling patients (n=9751) and patients admitted from an LCF (n=3194). In addition, 3 separate Cox models were performed in patients in risk classes II and III (n=4099), patients in risk class IV (n=5711), and patients in risk class V (n=3135). All of the previously described independent variables used in the overall Cox model were used in these analyses. We varied our definition of initial empirical antimicrobial therapy, specifying it as antimicrobial therapy in the first 8 hours and 24 hours of hospital admission to ensure that our results were insensitive to the time threshold used. We also excluded the 171 patients who died within 48 hours of hospital admission, since initial antibiotic therapy is unlikely to have influenced their mortality.To assess the independent associations between initial antimicrobial regimens and other medical outcomes (ie, LOS and 30-day rehospitalization), linear regression analysis was used when the dependent variable was LOS and Cox modeling was used when the dependent variable was rehospitalization. All independent variables used in the Cox models for mortality were used in these analyses.RESULTSThe study population was composed of 12,945 eligible patients: 9751 (75.3%) community-dwelling and 3194 (24.7%) admitted from an LCF. Study patients had a mean (±SD) age of 79.4 ± 8.1 years; 84.4% were white, and 50.7% were female (Table 1). The majority of patients (58.1%) had at least 1 comorbid illness, and 68.3% were in the 2 highest severity risk classes (IV and V) at initial examination. The most frequently coded bacteriological pathogens were S pneumoniae(6.6%), and H influenzae(4.1%); 10.1% of patients were coded as having aspiration pneumonia and 60.5%, an unknown pneumonia etiology.Table 1. Demographic and Clinical Characteristics of the Study PopulationCharacteristicCommunity Dwelling (n=9751)Long-term Care Facility Dwelling* (n=3194)Total Study Cohort (N=12,945)DemographicsAge, mean, y†78.183.579.4Race, % white84.384.984.4Sex, % female†48.557.250.7Comorbid illness, %Congestive heart failure†25.334.627.6Coronary artery disease†25.530.426.7Cerebrovascular disease†15.835.120.6Neoplastic disease†9.36.88.7Chronic renal disease†2.94.63.3Chronic liver disease0.90.70.8Physical examination findings, %Altered mental status†9.447.718.9Respiratory rate ≥30/min†23.233.925.9Pulse ≥125 beats/min†10.111.410.4Systolic blood pressure <90 mm Hg†2.46.03.3Temperature <35°C or ≥40°C†1.62.21.8Laboratory and radiographic results, %Serum urea nitrogen >10.7 mmol/L (>30 mg/dL)†19.542.225.1Glucose ≥13.9 mmol/L (≥250 mg/dL)†7.39.57.8Hematocrit <0.30†6.310.17.2Sodium <130 mmol/L5.45.25.3PaO2<60 mm Hg†22.019.421.4Arterial pH <7.35†6.99.87.6Pleural effusion19.119.419.2Antimicrobial therapy, %Previous antimicrobial therapy†24.434.026.8Antimicrobial therapy initiated ≤8 h80.980.480.8Blood cultures within 24 h of hospital arrival, %†65.771.167.1Intensive care unit treatment, day 1 of admission, %2.62.82.7Severity of illness risk classification, %‡Risk class II†11.50.38.7Risk class III†28.75.422.9Risk class IV†44.742.444.1Risk class V†15.152.024.2Pneumonia etiology, %Streptococcus pneumoniae†7.53.96.6Haemophilus influenzae†4.72.54.1Staphyloccocal species†2.86.03.6Pseudomonas aeruginosa3.33.83.4Klebsiella pneumoniae1.92.32.0Other†§3.72.53.4Aspiration pneumonia†5.922.910.1Unknown†63.052.860.5*Admission from a long-term care facility consisted of residence at a skilled nursing home, intermediate care, or long-term care facility before hospitalization. All other patients were defined as community dwelling.†Statistically significant differences (P<.05) between patient subsets (ie, community dwelling and long-term care facility dwelling) were identified by the χ2test with the exception of age, where an unpaired ttest was used.‡Risk class was determined according to the methods of Fine et al.22There were no risk class I patients, who by definition are all younger than 50 years.§Other microbiological etiology included viral pneumonia (n=93), fungal pneumonia (n=53), Pneumocystis carinii(n=2), and pneumonia due to other unspecified organisms (n=263).In comparison with community-dwelling patients, patients admitted from an LCF were older and had a higher prevalence of the most prevalent comorbid illnesses and a substantially higher prevalence of altered mental status, vital sign abnormalities, laboratory abnormalities, and high-risk etiology. More than 94% of all patients admitted from an LCF were in risk classes IV and V.INITIAL ANTIMICROBIAL REGIMENSThe 3 most commonly used initial antimicrobial regimens were a non-pseudomonal third-generation cephalosporin only (ceftriaxone, cefotaxime, ceftizoxime) in 26.5%, a second-generation cephalosporin only (cefuroxime) in 12.3%, and a non-pseudomonal third-generation cephalosporin (as above) plus a macrolide in 8.8% (Table 2). Significant differences in the prevalence of use of virtually all initial antimicrobial regimens existed across the 2 patient subsets (P<.05). Initial therapy with a non-pseudomonal third-generation cephalosporin only, a β-lactam/β-lactamase inhibitor only, a fluoroquinolone only, or an aminoglycoside plus another agent was more common among patients admitted from an LCF. Significant differences in the choice of initial antimicrobial regimens existed between the northeastern, southern, midwestern, and western regions of the country (defined by the US Census Bureau) for 10 of the 12 initial regimens.The absolute differences in use were minimal with the exception of the northeastern region, where more second-generation cephalosporins alone or in combination with a macrolide were used and fewer non-pseudomonal third-generation cephalosporins alone or in combination with a macrolide than the other 3 regions (data not shown).Table 2. Use of Initial Antimicrobial Regimens and 30-Day Mortality by Admission Source*Initial Antimicrobial Regimens†Prevalence of Antimicrobial Regimen, %30-d Mortality, % (95% CI)Community Dwelling (n=9751)LCF Dwelling‡ (n=3194)Total Study Cohort (N=12,945)Community Dwelling (n=9751)LCF Dwelling‡ (n=3194)Total Study Cohort (N=12,945)First-generation cephalosporin only§3.72.43.39.8 (6.9-13.3)19.7 (11.4-30.4)11.6 (8.7-14.9)Second-generation cephalosporin only§13.58.712.39.3 (7.7-10.9)23.5 (18.6-28.9)11.7 (10.1-13.4)Pseudomonal third-generation cephalosporin only§1.62.21.710.6 (6.2-16.6)29.0 (18.6-41.1)16.4 (11.7-21.9)Non-pseudomonal third-generation cephalosporins only (reference group)§25.629.226.510.7 (9.5-11.9)26.2 (23.4-29.1)14.9 (13.7-16.1)Ceftriaxone§17.820.118.49.8 (8.4-11.2)27.3 (23.8-30.8)14.5 (13.1-15.9)Cefotaxime§6.67.76.912.9 (10.4-15.7)24.5 (19.2-30.3)16.1 (13.7-18.7)Ceftizoxime1.00.90.910.8 (5.3-18.8)20.7 (8.0-39.7)13.1 (7.7-20.4)β-Lactam/β-lactamase inhibitors only§6.79.87.514.1 (11.4-16.9)27.7 (22.8-33.0)18.5 (16.0-21.0)Ampicillin/sulbactam§3.95.44.313.4 (10.1-17.2)22.0 (16.0-28.8)16.1 (13.1-19.3)Ticarcillin/clavulanate§2.43.82.814.5 (10.2-19.6)33.9 (25.5-43.0)21.1 (16.9-25.6)Piperacillin/tazobactam§0.30.60.415.2 (5.1-31.8)36.8 (16.2-61.6)23.1 (12.5-36.8)Macrolides only§2.20.81.88.6 (5.2-13.2)20.0 (6.8-40.7)9.8 (6.3-14.3)Erythromycin§1.10.61.010.8 (5.7-18.1)11.1 (1.4-34.7)10.9 (6.1-17.5)Clarithromycin§0.70.10.64.3 (0.9-12.0)25.0 (0.6-80.5)5.4 (1.5-13.2)Azithromycin0.20.10.214.3 (3.0-36.3)50.0 (1.3-98.7)17.4 (5.0-38.7)Second-generation cephalosporin plus macrolide§5.01.74.27.8 (5.6-10.5)13.0 (5.4-24.9)8.4 (6.2-11.0)Non-pseudomonal third-generation cephalosporins plus10.24.68.86.8 (5.3-8.6)24.3 (17.6-32.0)9.1 (7.5-10.9)macrolide§Ceftriaxone + macrolide§7.03.26.17.2 (5.3-9.3)18.8 (11.7-27.8)8.7 (6.8-10.8)Cefotaxime + macrolide§2.51.12.26.5 (3.8-10.8)40.0 (23.8-57.8)10.7 (7.3-14.9)Ceftizoxime + macrolide0.50.30.44.3 (0.5-14.5)37.5 (8.5-75.5)9.1 (3.0-19.9)β-Lactam/β-lactamase inhibitors plus macrolide§1.60.81.417.8 (12.0-24.7)50.0 (29.1-70.8)22.2 (16.2-29.0)Ampicillin/sulbactam + macrolide§0.90.30.814.8 (8.1-23.9)40.0 (12.1-73.7)17.4 (10.4-26.3)Ticarcillin/clavulanate + macrolide0.60.30.524.1 (13.4-37.6)60.0 (26.2-87.8)29.7 (18.9-42.4)Piperacillin/tazobactam + macrolide0.10.10.110.0 (0.3-44.5)33.0 (0.8-90.5)15.4 (1.9-45.4)Fluoroquinolones only§1.72.72.07.1 (3.7-12.0)17.4 (10.1-27.1)10.6 (7.1-14.9)Ciprofloxacin§1.22.11.46.7 (2.9-12.8)15.2 (7.5-26.1)9.7 (5.9-14.9)Ofloxacin0.50.60.57.8 (2.2-18.8)21.1 (6.1-45.5)11.4 (5.1-21.2)Aminoglycosides plus any other antimicrobial agent(s)§5.310.16.518.2 (14.9-21.7)33.1 (28.0-38.5)24.0 (21.0-26.9)All other regimens§∥23.127.124.113.6 (12.2-15.1)30.7 (27.6-33.8)18.4 (17.0-19.7)TotalNANANA11.2 (10.0-11.9)27.5 (26.5-29.1)15.3 (14.6-15.9)*CI indicates confidence interval; LCF, long-term care facility; and NA, not applicable. Percentages for the individual antimicrobial agents within antimicrobial regimens may not sum to the total category percentage because of rounding error.†The specific antimicrobial agents associated with these regimens were cefazolin for first-generation cephalosporin only; cefuroxime for second-generation cephalosporin only; ceftazidime for pseudomonal third-generation cephalosporin only; erythromycin, clarithromycin, or azithromycin for macrolides; and gentamicin, tobramycin, or amikacin for aminoglycosides.‡LCF denotes admission from a long-term care facility. Admission from an LCF consisted of residence at a skilled nursing home, intermediate care facility, or LCF before hospitalization. All other patients were defined as community dwelling.§Statistically significant differences (P<.05) between the 2 patient subsets (ie, community dwelling and LCF dwelling) for comparisons of frequency of use of intial antimicrobial regimens by means of the χ2test.∥The "all other regimens" category comprised a total of 851 unique antimicrobial regimens. The most frequent regimens were ceftriaxone plus cefuroxime (n=97), imipenem/cilastatin only (n=87), and cefotetan only (n=67).Associations between initial antimicrobial regimen and baseline severity of illness existed for 8 of the 12 commonly prescribed regimens (Table 3). Treatment with a first- or second-generation cephalosporin only, a macrolide only, and a second- or non-pseudomonal third-generation cephalosporin plus a macrolide occurred less frequently with increasing illness severity. Treatment with a β-lactam/β-lactamase inhibitor only, an aminoglycoside plus any other antimicrobial agent, and the other antimicrobial agent category occurred more frequently with increasing severity of illness at hospital arrival.Table 3. Use of Initial Antimicrobial Regimens and 30-Day Mortality by Severity Risk Class*Initial Antimicrobial Regimens†Prevalence of Antimicrobial Regimen, %30-d Mortality, % (95% CI)Risk Classes II/III (n=4099)Risk Class IV (n=5711)Risk Class V (n=3135)Risk Classes II/III (n=4099)Risk Class IV (n=5711)Risk Class V (n=3135)First-generation cephalosporin only‡4.13.32.53.6 (1.3-7.7)12.1 (7.8-17.6)27.3 (17.7-38.6)Second-generation cephalosporin only‡13.413.09.63.6 (2.2-5.6)10.3 (8.2-12.8)29.8 (24.7-35.3)Pseudomonal third-generation cephalosporin only1.42.01.53.4 (0.4-11.7)14.2 (8.3-22.0)37.5 (24.0-52.6)Non-pseudomonal third-generation cephalosporins only (reference group)26.426.626.43.3 (2.3-4.6)12.4 (10.8-14.2)34.7 (31.5-38.1)β-Lactam/β-lactamase inhibitors only‡5.97.89.12.9 (1.2-5.9)15.3 (12.1-19.0)36.6 (31.0-42.5)Macrolides only‡2.91.70.63.4 (0.9-8.4)15.2 (8.7-23.7)22.2 (6.4-47.6)Second-generation cephalosporin plus macrolide‡5.44.12.62.2 (0.7-5.2)10.2 (6.6-14.8)19.8 (11.7-30.1)Non-pseudomonal third-generation cephalosporins plus macrolide‡10.49.26.01.2 (0.4-2.7)8.5 (6.3-11.2)28.6 (22.2-35.6)β-Lactam/β-lactamase inhibitors plus macrolide1.61.31.26.0 (1.6-14.6)29.2 (19.0-41.1)37.8 (22.5-55.2)Fluoroquinolones only2.21.91.84.4 (1.2-10.8)9.4 (4.6-16.5)22.8 (12.7-35.8)Aminoglycosides plus any other antimicrobial agent(s)‡4.16.010.46.5 (3.3-11.4)19.6 (15.6-24.3)37.5 (32.3-43.0)All other regimens‡§22.223.128.43.5 (2.4-4.9)15.9 (14.0-18.0)37.2 (34.0-40.5)Total mortality for all regimensNANANA3.3 (2.8-3.9)13.4 (12.5-14.3)34.3 (32.6-35.9)*CI indicates confidence interval; NA, not applicable. Risk classes were determined according to the methods of Fine et al.22There were no risk class I patients, who by definition, are all younger than 50 years.†The specific antimicrobial agents associated with these regimens were cefazolin for first-generation cephalosporin only; cefuroxime for second-generation cephalosporin only; ceftazidime for pseudomonal third-generation cephalosporin only; ceftriaxone, cefotaxime, or ceftizoxime for non-pseudomonal third-generation cephalosporins; ampicillin/sulbactam, ticarcillin/clavulanate, or piperacillin/tazobactam for β-lactam/β-lactamase inhibitors; erythromycin, clarithromycin, or azithromycin for macrolides; ciprofloxacin or ofloxacin for fluoroquinolones; and gentamicin, tobramycin, or amikacin for aminoglycosides.‡P<.05 for comparisons of frequency of use of initial antimicrobial regimens by risk class were calculated by the χ2test for trend.§The "all other regimens" category comprised a total of 851 unique antimicrobial regimens. The most frequent regimens were ceftriaxone plus cefuroxime (n=97), imipenem/cilastatin only (n=87), and cefotetan only (n=67).ASSOCIATIONS BETWEEN INITIAL ANTIMICROBIAL REGIMENS AND MORTALITYThirty-day mortality was 15.3% (95% CI, 14.6%-15.9%) in the entire study population, ranging from 11.2% (95% CI, 10.6%-11.9%) in community-dwelling patients to 27.5% (95% CI, 26.0%-29.1%) among patients admitted from an LCF (Table 2). Mortality ranged from 8.9% for patients who initially received a second-generation cephalosporin plus a macrolide to 24.0% for those who received a β-lactam/β-lactamase inhibitor plus a macrolide. Higher 30-day mortality rates were observed among patients admitted from an LCF for all of the initial antimicrobial regimens, with significantly higher rates observed for 8 regimens.Increasing risk class was strongly associated with increased 30-day mortality, as follows: class II and III, 3.3% (95% CI, 2.8%-3.9%); class IV, 13.4% (95% CI, 12.5%-14.3%); and class V, 34.3% (95% CI, 32.6%-35.9%) (P<.001, χ2for trend) (Table 3). With the exception of initial therapy with a fluoroquinolone only in risk classes II and III, the point estimates for 30-day mortality stratified by severity class for fluoroquinolones only, second-generation cephalosporins plus macrolide, and non-pseudomonal third-generation cephalosporins plus macrolide were consistently lower than the reference category (ie, non-pseudomonal third-generation cephalosporins only), while the stratified point estimates for 30-day mortality for β-lactam/β-lacatamase inhibitors plus macrolide and aminoglycosides plus any other agent were consistently higher than the reference category.Five initial antimicrobial regimens were identified as having an independent association with 30-day mortality (Table 4). Use of a second-generation cephalosporin plus a macrolide (HR, 0.71; 95% CI, 0.52-0.96), a non-pseudomonal third-generation cephalosporin plus a macrolide (HR, 0.74; 95% CI, 0.60-0.92), or a fluoroquinolone only (HR, 0.64; 95% CI, 0.43-0.94) were independently associated with a lower 30-day mortality. Use of a β-lactamase inhibitor plus a macrolide (HR, 1.77; 95% CI, 1.28-2.46) and an aminoglycoside plus another agent (HR, 1.21; 95% CI, 1.02-1.43) were independently associated with a higher 30-day mortality.Table 4. Independent Associations Between Initial Antimicrobial Regimens and 30-Day Mortality Among the Total Study Cohort and Stratified by Source of AdmissionInitial Antimicrobial Regimens†Adjusted Hazard Ratio (95% CI)*Total Study Cohort (N=12,945)Community Dwelling (n=9751)LCF Dwelling‡ (n=3194)First-generation cephalosporin only0.92 (0.69-1.23)0.99 (0.70-1.41)0.79 (0.47-1.34)Second-generation cephalosporin only0.89 (0.75-1.05)0.88 (0.71-1.10)0.90 (0.68-1.19)Pseudomonal third-generation cephalosporin only1.12 (0.80-1.57)0.87 (0.52-1.44)1.43 (0.91-2.27)Non-pseudomonal third-generation cephalosporins onlyReference groupReference groupReference groupβ-Lactam/β-lactamase inhibitors only1.07 (0.91-1.28)1.09 (0.86-1.38)1.07 (0.84-1.37)Macrolides only1.06 (0.69-1.61)1.07 (0.66-1.73)1.06 (0.44-2.58)Second-generation cephalosporin plus macrolide0.71 (0.52-0.96)0.78 (0.56-1.10)0.49 (0.23-1.04)Non-pseudomonal third-generation cephalosporins plus macrolide0.74 (0.60-0.92)0.66 (0.51-0.86)0.95 (0.67-1.34)β-Lactam/β-lactamase inhibitors plus macrolide1.77 (1.28-2.46)1.61 (1.08-2.39)2.24 (1.24-4.04)Fluoroquinolones only0.64 (0.43-0.94)0.64 (0.36-1.14)0.64 (0.38-1.09)Aminoglycosides plus any other antimicrobial agent(s)1.21 (1.02-1.43)1.29 (1.02-1.65)1.16 (0.92-1.46)All other regimens1.12 (0.99-1.27)1.11 (0.94-1.31)1.14 (0.96-1.36)*The hazard ratios shown are the hazards of dying within 30 days of hospitalization among patients who received the antimicrobial regimen listed compared with patients who received a non-pseudomonal third-generation cephalosporin only (ie, ceftriaxone, cefotaxime, or ceftizoxime). All hazard ratios were adjusted for severity risk class, admission from community or LCF (for the overall model), previous antimicrobial use, region of enrollment, intensive care unit treatment on day 1 of hospitalization, performance of blood cultures within 24 hours of hospitalization, initiation of antimicrobial therapy within 8 hours of hospitalization, high-risk pneumonia etiology (gram-negative rod, staphylococcal species, or aspiration pneumonia), and change in antimicrobial therapy after 48 hours of hospitalization. CI indicates confidence interval.†The specific antimicrobial agents associated with these regimens were cefazolin for first-generation cephalosporin only; cefuroxime for second-generation cephalosporin only; ceftazidime for pseudomonal third-generation cephalosporin only; ceftriaxone, cefotaxime, or ceftizoxime for non-pseudomonal third-generation cephalosporins; ampicillin/sulbactam, ticarcillin/clavulanate, or piperacillin/tazobactam for β-lactam/β-lactamase inhibitors; erythromycin, clarithromycin, or azithromycin for macrolides; ciprofloxacin or ofloxacin for fluoroquinolones; and gentamicin, tobramycin, or amikacin for aminoglycosides.‡LCF denotes admission from a long-term care facility. Admission from an LCF consisted of residence at a skilled nursing home, intermediate care facility, or LCF before hospitalization. All other patients were defined as community dwelling.The adjusted mortality associated with initial treatment with a non-pseudomonal third-generation cephalosporin plus a macrolide (P<.05), a second-generation cephalosporin plus a macrolide (P<.05), or a fluoroquinolone only (P<.05) was significantly lower than the mortality in the reference group on days 2, 3, and 7 after hospitalization, respectively, and remained lower throughout the 30-day follow-up period (Figure 1). In contrast, the adjusted mortality for initial treatment with an aminoglycoside plus any other agent (P<.05) or a β-lactam/β-lactamase inhibitor plus a macrolide (P<.05) was significantly higher than the mortality in the reference group on days 4 and 7, respectively, and continued to exceed the mortality in the reference group during the remaining time of the 30-day follow-up.Adjusted mortality within 30 days of hospital admission for the 5 antimicrobial regimens with an independent association with 30-day mortality based on a Cox model for mortality in the overall study population (N=12 945). The Cox model adjusted for severity risk class, admission from a community or a long-term care facility, previous antimicrobial use, region of enrollment, intensive care unit treatment on day 1 of hospitalization, performance of blood cultures within 24 hours of hospitalization, initiation of antimicrobial therapy within 8 hours of hospitalization, high-risk pneumonia etiology (gram-negative rod, staphylococcal species, or aspiration pneumonia), and change in antimicrobial therapy after 48 hours of hospitalization.Two separate Cox models were performed in community-dwelling patients and patients admitted from an LCF (Table 4). The HRs for 30-day mortality were very similar in these 2 subsets and in the overall study population, with the exception of non-pseudomonal third-generation cephalosporins plus a macrolide. This initial regimen had an HR of 0.66 (95% CI, 0.51-0.86) for community-dwelling patients and an HR of 0.95 (95% CI, 0.67-1.34) for patients admitted from an LCF. Treatment with a second-generation cephalosporin plus a macrolide and a fluoroquinolone only was associated with a lower 30-day mortality in both models (not statistically significant). Treatment with a β-lactam/β-lactamase inhibitor plus a macrolide was independently associated with higher 30-day mortality in both models, while treatment with an aminoglycoside plus another agent was associated with a significantly higher (HR, 1.29; 95% CI, 1.02-1.65) 30-day mortality among community-dwelling patients and a nonsignificantly higher mortality among patients admitted from an LCF.Separate Cox models of 30-day mortality performed for patients in severity risk classes II-III, IV, and V were consistent with the findings from the model performed in the total study population. Although few of the associations between initial antimicrobial regimen and mortality were statistically significant because of the relatively small number of patients within cells, the point estimates for the HRs derived in these risk class–specific models had a similar direction and magnitude of effect as those derived from the model in the total study population.IMPACT OF THE DEFINITION OF INITIAL ANTIMICROBIAL THERAPY ON 30-DAY MORTALITYModifications in antimicrobial therapy were observed in 8117 patients (62.7%) after the first 8 hours of hospital admission, 5985 (46.2%) after the first 24 hours, and 4760 (36.8%) after the first 48 hours. From 8 to 48 hours after hospital admission, there was a 38% to 14% absolute decrease in the use of single-agent antimicrobial regimens and a 45% to 74% increase in the use of combination regimens.Our findings of the association between initial antimicrobial regimen and 30-day mortality were insensitive to our definition of the initial regimen. A model that used an 8-hour post–hospital admission threshold to define initial antimicrobial therapy identified a significantly lower 30-day mortality among patients treated with a non-pseudomonal third-generation cephalosporin plus a macrolide (HR, 0.73; 95% CI, 0.55-0.97) and an independent association with higher 30-day mortality among patients treated with a β-lactam/β-lactamase inhibitor plus a macrolide (HR, 1.65; 95% CI, 1.04-2.61) and the "all other antimicrobial regimens" category (HR, 1.24; 95% CI, 1.07-1.43). The HRs for 30-day mortality associated with a second-generation cephalosporin plus a macrolide (HR, 0.77; 95% CI, 0.51-1.15), a fluoroquinolone only (HR, 0.81; 95% CI, 0.56-1.19), and an aminoglycoside plus any other agent (HR, 1.11; 95% CI, 0.90-1.38) were consistent with the corresponding HRs estimated when the 48-hour definition of initial therapy was used (although not statistically significant). The HRs observed with a 24-hour threshold used to define initial therapy were nearly identical to those estimated with the 8-hour definition (data not shown).Our findings were also insensitive to the exclusion of the 171 patients who died within the first 2 days of hospitalization. As in the original model, 30-day mortality was significantly lower with initial treatment with a non-pseudomonal third-generation cephalosporin plus a macrolide (HR, 0.80; 95% CI, 0.64-0.99) and a fluoroquinolone only (HR, 0.64; 95% CI, 0.43-0.97) and significantly higher with a β-lactam/β-lactamase inhibitor plus a macrolide (HR, 1.81; 95% CI, 1.29-2.55) and an aminoglycoside plus another agent (HR, 1.27; 95% CI, 1.07-1.51). The HR for treatment with a second-generation cephalosporin plus a macrolide was consistent with the results of the original model (although not statistically significant), while the "all other antimicrobial regimens" category was associated with a significantly higher 30-day mortality in these analyses (data not shown).ASSOCIATIONS BETWEEN INITIAL ANTIMICROBIAL REGIMENS AND OTHER MEDICAL OUTCOMESAmong the 11 432 patients discharged alive from the index hospitalization and not transferred to another acute care hospital, the median LOS (interquartile range) was 7 (5 to 10) days, ranging from 7 (5 to 9) days among community-dwelling patients to 8 (6 to 11) days among patients admitted from an LCF. The 30-day rehospitalization rate was 15.9% (95% CI, 15.3%-16.6%), ranging from 15.5% (95% CI, 14.8%-16.3%) in community-dwelling patients to 17.2% (95% CI, 15.8%-18.7%) in patients admitted from an LCF.Initial therapy with a pseudomonal third-generation cephalosporin alone, a β-lactam/β-lactamase inhibitor alone, a β-lactam/β-lactamase inhibitor plus a macrolide, an aminoglycoside plus another agent, and the "other" category of antimicrobial agents were all independently associated with a longer LOS for the index hospitalization; no regimen was associated with a significantly shorter LOS. Initial therapy with an aminoglycoside plus another agent was associated with an increased rehospitalization rate within 30 days after discharge from the index hospitalization; no regimen was independently associated with decreased rehospitalization.COMMENTThis national study of antimicrobial therapy for elderly patients hospitalized with pneumonia demonstrated that initial therapy with a non-pseudomonal third-generation cephalosporin plus a macrolide, a second-generation cephalosporin plus a macrolide, or a fluoroquinolone alone was associated with 26%, 29%, and 36% lower 30-day mortality, respectively. Yet, only 15.0% of all patients received 1 of these 3 initial regimens. An additional 7.9% were treated with a β-lactam/β-lactamase inhibitor plus a macrolide or an aminoglycoside plus another agent, which had mortality rates 77% and 21% higher than the reference group, respectively. These findings suggest that opportunities exist to dramatically improve the quality of care for hospitalized elderly patients with pneumonia by modifying existing initial antimicrobial prescribing practices.The 3 initial antimicrobial regimens associated with a decreased mortality in the overall study population were also associated with a 22% to 36% lower mortality among community-dwelling patients and a 5% to 51% lower mortality among patients admitted from an LCF. In the 75.3% of the entire population who were community-dwelling before hospitalization, only initial therapy with a non-pseudomonal third-generation cephalosporin plus a macrolide was independently associated with a lower 30-day mortality despite trends toward lower mortality in community-dwelling patients treated with a second-generation cephalosporin plus a macrolide or a fluoroquinolone alone. The associations between initial antimicrobial therapy and mortality among LCF-dwelling patients were very similar to the associations in the overall study population, with the exception of non-pseudomonal third-generation cephalosporins plus macrolide. We believe the lack of statistically significant associations between initial antimicrobial regimens and 30-day mortality in these 2 subsets of patients was caused by smaller sample sizes reducing statistical power rather than systematic differences in these associations. Alternatively, initial antimicrobial therapy may not influence short-term mortality among patients with severe disease or with a greater prevalence of comorbid illnesses such as those admitted from LCFs.This alternative explanation is not supported by our findings that treatment with a second- or non-pseudomonal third-generation cephalosporin plus a macrolide was associated with an improved survival beginning within 2 to 3 days after hospital admission and persisting throughout the 30-day follow-up period.OUR FINDINGS confirm our hypothesis that initial antimicrobial regimens with activity against the most common "typical" bacterial pathogens (eg, S pneumoniaeand H influenzae) and "atypical" pathogens (eg, Legionellaspecies, Mycoplasma pneumoniae, Chlamydia pneumoniae, and Coxiella burnetti) are associated with a decreased 30-day mortality. These findings are supported by recent studies of pneumonia etiology, which showed a high prevalence of "atypical" pathogens especially among hospitalized elderly.The prevalence of atypical pneumonia resulting in hospitalization is as high as 44%, with M pneumoniaeaccounting for 33% and C pneumoniae, 9%.Moreover, contrary to the historical belief that atypical pathogens are associated with a universally favorable prognosis, a recent study demonstrated that C pneumoniaewas responsible for deadly outbreaks of pneumonia among nursing home residents.The finding of an association between use of fluoroquinolones alone (ie, ciprofloxacin or ofloxacin) and lower 30-day mortality is surprising given reported in vitro resistance of S pneumoniaeto ciprofloxacin and ofloxacin, ranging from to 3% to 32%.However, reported in vitro resistance of S pneumoniaeto fluoroquinolones may be attributed to minimum inhibitory concentrations that overlap susceptibility break points and not to in vivo drug resistance.Our findings may underestimate the actual survival benefit of this class of antimicrobial agents given that many of the newer fluoroquinolones (eg, levofloxacin, gatifloxacin, and gemifloxacin) have greater activity against S pneumoniaeand S aureus, in addition to providing excellent coverage of the atypical bacterial pathogens. However, the potential survival advantage associated with this class of antimicrobial agents must be balanced with the problem of increased bacterial resistance that could result if fluoroquinolones are used widely to treat this common illness.Although many difficult-to-treat pathogens, such as S aureusand P aeruginosa, were exquisitely sensitive to this class of agents when first introduced, within 5 years of introduction, drug resistance began to develop.Therefore, judicious use of fluoroquinolones for treatment of pneumonia is strongly recommended.Our finding of a 77% higher 30-day mortality among patients initially treated with a β-lactam/β-lactamase inhibitor plus a macrolide, an association that remained statistically significant among community- and LCF-dwelling patients, raises concerns for use of this regimen. Although one potential explanation of this finding is that patients treated with a β-lactam/β-lactamase inhibitor plus a macrolide differed from those treated with other initial antimicrobial regimens, patients treated with this initial regimen were not systematically more severely ill at the time of hospital arrival. In fact, with the exception of a slightly higher rate of ICU treatment within 24 hours of hospital arrival (6.3% vs 2.6% for all other regimens), these patients had significantly lower rates of coming from an LCF, less cerebrovascular disease, less altered mental status at hospital arrival, and a significantly greater proportion in the 2 lowest severity risk classes (II and III) at hospital arrival. Another possible explanation for this positive association with mortality is that this regimen was used more often for patients with a presumptive diagnosis of aspiration pneumonia. However, higher use of this regimen was not demonstrated among patients with an ICD-9-CMdiagnosis code for aspiration pneumonia. Alternatively, the increased mortality associated with this initial regimen may be explained by deterioration of patients' conditions because of the β-lactam/β-lactamase inhibitors' lack of activity against penicillin-resistant S pneumoniaeand ampicillin/sulbactam's limited activity against gram-negative Enterobacteriaceae (eg, K pneumoniae).Moreover, it is possible that unrecognized pathogens with differing susceptibility patterns to second- or non-pseudomonal third-generation cephalosporins plus a macrolide compared with β-lactam/β-lactamase inhibitors plus a macrolide exist. Unfortunately, our data cannot be used to either support or refute the latter 2 hypotheses.One puzzling question raised by this study is why initial therapy with β-lactam/β-lactamase inhibitors alone was not associated with an increased 30-day mortality, while the same class of agents in combination with a macrolide was independently associated with an increased 30-day mortality. A likely explanation for this observation is that patients treated with a β-lactam/β-lactamase inhibitor alone during the 48 hours after admission had an adequate clinical response to initial therapy, while patients with a suboptimal clinical response had a macrolide added to their initial therapy. Among the 1069 patients receiving a β-lactam/β-lactamase inhibitor alone at 24 hours, 132 (12.3%) had a second antimicrobial agent added between 24 and 48 hours after admission. Of the 132 patients who had a second agent added to initial therapy, the 30-day mortality was 30.3%, compared with 18.5% among those continuing to receive a β-lactam/β-lactamase inhibitor alone.Overall, these findings are likely to be useful for the refinement of medical practice guidelines for the treatment of pneumonia. The recently released IDSA guidelines and ATS guidelines originally published in 1993 were developed with the use of consensus opinion of an expert panel without a large body of empirical evidence to support their clinical recommendations.Both the IDSA and ATS guidelines recommend use of a second-generation cephalosporin alone, a non-pseudomonal third-generation cephalosporin alone, or agents from either of these drug classes plus a macrolide as appropriate initial therapy for non–severely ill patients. The IDSA guidelines also recommend use of fluoroquinolones with good S pneumoniaeactivity or azithromycin as appropriate initial therapy for non–severely ill hospitalized patients. The ATS guidelines recommend use of a β-lactam/β-lactamase inhibitor alone or in combination with a macrolide for this group of patients. Our findings support the recommendations in the IDSA and ATS guidelines for use of a second-generation cephalosporin plus a macrolide, or a non-pseudomonal third-generation cephalosporin plus a macrolide. Our findings also support the IDSA recommendations for use of a fluoroquinolone alone. In contrast, our findings do not support the use of a β-lactam/β-lactamase inhibitor alone or in combination with a macrolide as initial therapy. Firm conclusions about the use of azithromycin alone cannot be made on the basis of our results because of the small number of patients treated with this newer macrolide.The present study has limitations that warrant further discussion. First, our study was observational in design, and therefore, antimicrobial treatment selection biases were possible. However, we controlled for potential confounding by comorbid disease and illness severity by using a widely validated pneumonia-specific severity model.Second, antimicrobial route of administration, dose, and discontinuation date were not recorded. Consequently, it is possible that antimicrobial therapy was modified during the initial 48 hours of treatment after the patient received only a single dose of an initial agent. Patients treated in this manner would not have received the full benefit of their initial antimicrobial regimen. However, blood and sputum cultures generally require 48 hours for results to be reported, and if an organism is identified, therapy is generally streamlined, not broadened.In addition, study findings on the association between initial antimicrobial regimen and 30-day mortality were insensitive to using either an 8- or a 24-hour posthospitalization threshold to define initial therapy. Third, our definition of initial antimicrobial therapy that consisted of all agents prescribed within 48 hours of hospital arrival does not reflect the influence of subsequent changes in therapy on patient outcomes. Nevertheless, only 36.8% of patients had such a change in therapy after 48 hours, and the results of analyses that considered such changes were nearly identical to the primary results of this study. Fourth, no microbiological culture or sensitivity data were available in this study. As a result, it was not possible to definitively correlate the associations between antimicrobial therapy, microbiological etiology, antimicrobial sensitivity, and patient outcomes. Fifth, "do not resuscitate" orders were not systematically recorded; patients with such orders may have received less aggressive antimicrobial treatment in accordance with patient and family wishes. Finally, because of the small number of patients having an ICU procedure coded as occurring within 1 day of hospital arrival, it was difficult to assess differences in medical outcomes with the 12 separate antimicrobial regimens among patients initially treated in an ICU.In summary, this study of primarily community-dwelling patients demonstrated that initial antimicrobial regimens with a second-generation cephalosporin plus a macrolide, a non-pseudomonal third-generation cephalosporin plus a macrolide, or a fluoroquinolone alone were associated with lower 30-day mortality among all patients hospitalized with pneumonia. The observed higher mortality associated with 3 initial regimens coupled with the wide variation in the use of all initial antimicrobial regimens suggests that potential opportunities exist to reduce the mortality and thereby improve the quality of care for elderly patients with pneumonia. Although these findings are likely to influence medical practice guidelines designed to improve physician prescribing practices, future randomized controlled trials are warranted to confirm these findings before they are adopted in clinical practice.RAGaribaldiEpidemiology of community-acquired respiratory tract infections in adults: incidence, etiology, and impact.Am J Med.1985;78:32-37.PFAdamsMAMaranoCurrent estimates from the National Health Interview Survey, 1994.Vital Health Stat 10.1995;No. 199:95.FMLaForceCommunity-acquired lower respiratory tract infections: prevention and cost-control strategies.Am J Med.1985;78:52-57.EJGravesBSGillum1994 Summary: National Hospital Discharge Survey.Hyattsville, Md: National Center for Health Statistics; 1996. Advance Data From Vital and Health Statistics, No. 278.Health Care Financing Administration1995 Data Compendium.Baltimore, Md: US Dept of Health and Human Services, Health Care Financing Administration; 1995:75. HCFA No. 03364.TJMarrieHDurantLYatesCommunity-acquired pneumonia requiring hospitalization: 5-year prospective study.Rev Infect Dis.1989;11:586-599.GDFangMJFineJOrloffNew and emerging etiologies for community-acquired pneumonia with implication for therapy: a prospective multicenter study of 359 cases.Medicine.1990;69:307-316.JHBatesGDCampbellALBarronMicrobial etiology of acute pneumonia in hospitalized patients.Chest.1992;101:1005-1012.BJMarstonJFPlouffeTMFilefor the Community-Based Pneumonia Incidence Study GroupIncidence of community-acquired pneumonia requiring hospitalization: results of a population-based active surveillance study in Ohio.Arch Intern Med.1997;157:1709-1718.KGilbertPPGleasonDESingerVariations in antimicrobial use and costs in more than 2000 patients with community-acquired pneumonia.Am J Med.1998;104:17-27.THGraselaLSWelageCAWalawaanderA nationwide survey of antibiotic prescribing patterns and clinical outcomes in patients with bacterial pneumonia.DICP.1990;24:1220-1225.IKappsteinFDDaschnerAntimicrobial usage in community-acquired pneumonia: results of a survey in 288 departments of internal medicine in German hospitals.Infection.1991;19:301-304.LGuglielmoRLeoneUMorettiAConfortiASpolaorGVeloAntibiotic prescribing patterns in Italian hospital inpatients with pneumonia, chronic obstructive pulmonary disease, and urinary tract infections.Ann Pharmacother.1993;27:18-22.MSNiedermanJBBassGDCampbellAmerican Thoracic Society guidelines for the initial management of adults with community-acquired pneumonia: diagnosis, assessment of severity, and initial antimicrobial therapy.Am Rev Respir Dis.1993;148:1418-1426.JGBartlettRFBreimanLAMandellTMFile JrCommunity-acquired pneumonia in adults: guidelines for management.Clin Infect Dis.1998;26:811-838.AMFeinMSNiedermanGuidelines for the initial management of community-acquired pneumonia: savory recipe or cookbook for disaster?Am J Respir Crit Care Med.1995;152:1149-1153.MJFineMASmithCACarsonPrognosis and outcomes of patients with community-acquired pneumonia: a meta-analysis.JAMA.1996;275:134-141.OLeroyCSantreCBeuscartA five-year study of severe community-acquired pneumonia with emphasis on prognosis in patients admitted to an intensive care unit.Intensive Care Med.1995;21:24-31.TPMeehanMJFineHMKrumholzQuality of care, process, and outcomes in elderly patients with pneumonia.JAMA.1997;278:2080-2084.Not AvailableInternational Classification of Diseases, Ninth Revision, Clinical Modification.Washington, DC: Public Health Service, US Dept of Health and Human Services; 1988.MJFineDESingerBHHanusaJRLaveWNKapoorValidation of a pneumonia prognostic index using the Medisgroups comparative hospital database.Am J Med.1993;94:153-159.MJFineTEAubleDMYealyA prediction rule to identify patients with community-acquired pneumonia at low risk for mortality and other adverse medical outcomes.N Engl J Med.1997;336:243-250.MJFineJJOrloffDArisumiPrognosis of patients hospitalized with community-acquired pneumonia.Am J Med.1990;88:1N-8N.ABCantorExtending SAS Survival Analysis Techniques for Medical Research.Cary, NC: SAS Institute Inc; 1997:208.DWHosmerSLemeshowApplied Logistic Regression.New York, NY: John Wiley & Sons Inc; 1989.Not AvailableA Guide to State and Local Census Geography. Washington, DC: Bureau of the Census Economic and Statistics Administration, US Dept of Commerce; 1993.ATorresJSerra-BatllesAFerrerSevere community-acquired pneumonia: epidemiology and prognostic factors.Am Rev Respir Dis.1991;144:312-318.EWHookCAHortonDRSchabergFailure of intensive care unit support to influence mortality from pneumococcal bacteremia.JAMA.1983;249:1055-1057.JStephensonStudies suggest a darker side of "benign" microbes.JAMA.1997;278:2051-2052.DTaylor-RobinsonInfections due to species of Mycoplasma and Ureaplasma: an update.Clin Infect Dis.1996;23:671-684.MRHammerschlagAntimicrobial susceptibility and therapy of infections caused by Chlamydia pneumoniae.Antimicrob Agents Chemother.1994;38:1873-1878.CJTroyRWPeelingAGEllisChlamydia pneumoniaeas a new source of infectious outbreaks in nursing homes [published correction appears in JAMA.1997;278:118].JAMA.1997;277:1214-1218.JFPlouffeRFBreimanRRFacklamfor the Franklin County Pneumonia Study GroupBacteremia with Streptococcus pneumoniae: implications for therapy and prevention.JAMA.1996;275:194-198.RNJonesFluoroquinolone (Lomefloxacin) International Surveillance Trial: a report of 30 months of monitoring in vitro activity.Am J Med.1992;92(suppl 4A):52-57.FWGoldsteinJFAcarEpidemiology of quinolone resistance: Europe and North and South America.Drugs.1995;49(suppl 2):36-42.DJHobanRNJonesThe North American component (the United States and Canada) of an international comparative MIC trial monitoring ofloxacin resistance.Diagn Microbiol Infect Dis.1993;17:157-161.VGCoronadoJREdwardsDHCulverRPGaynesCiprofloxacin resistance among Pseudomonas aeruginosaand Staphylococcus aureusin the United States: National Nosocomial Infections Surveillance (NNIS) System.Infect Control Hosp Epidemiol.1995;16:71-75.BLTProsserGBeskidMulticenter in vitro comparative study of fluoroquinolones against 25,129 gram-positive and gram-negative clinical isolates.Diagn Microbiol Infect Dis.1995;21:33-45.JFAcarFWGoldsteinTrends in bacterial resistance to fluoroquinolones.Clin Infect Dis.1997;24(suppl 1):S67-S73.JCButlerJHofmannMSCetronThe continued emergence of drug-resistant Streptococcus pneumoniaein the United States: an update from the Centers for Disease Control and Prevention's Pneumococcal Sentinel Surveillance System.J Infect Dis.1996;174:986-993.DMJohnsonGVDoernTAHaugenJHindlerJAWashingtonRNJonesComparative activity of twelve beta-lactam drugs tested against penicillin-resistant Streptococcus pneumoniaefrom five medical centers.Diagn Microbiol Infect Dis.1996;25:137-141.GVDoernABrueggemannHPHolley JrAMRauchAntimicrobial resistance of Streptococcus pneumoniaerecovered from outpatients in the United States during the winter months of 1994 to 1995: results of a 30-center national surveillance study.Antimicrob Agents Chemother.1996;40:1208-1213.JGBartlettLMMundyCommunity-acquired pneumonia.N Engl J Med.1995;333:1618-1624.Accepted for publication April 27, 1999.Dr M. J. Fine was supported as a Robert Wood Johnson Generalist Physician Faculty Scholar. Dr J. M. Fine was supported in part by a grant from the Polly Annenberg Levee Charitable Trust, Washington, DC. The analyses on which this publication is based were performed under contract 500-96-P549, entitled "Utilization and Quality Control Peer Review Organization for the State of Connecticut," sponsored by the Health Care Financing Administration, Department of Health and Human Services, Washington.The content of this publication does not necessarily reflect the views or policies of the US Department of Health and Human Services, nor does the mention of trade names, commercial products, or organizations imply endorsement by the US government.The authors assume full responsibility for the accuracy and completeness of the ideas represented. This article is a direct result of the Health Care Quality Improvement Program initiated by the Health Care Financing Administration, which has encouraged identification of quality improvement projects derived from analysis of patterns of care, and therefore required no special funding on the part of this contractor. Ideas and contributions to the authors concerning experience in engaging with issues presented are welcomed.Reprints: Michael J. Fine, MD, MSc, Department of Medicine, Montefiore Hospital, Room 824 East, 200 Lothrop St, Pittsburgh, PA 15261 (e-mail:[email protected]).
Stahl, James E.; Barza, Michael; DesJardin, Jeffrey; Martin, Rhonda; Eckman, Mark H.
doi: 10.1001/archinte.159.21.2576pmid: 10573047
BackgroundThe choice of antibiotics to treat community-acquired pneumonia (CAP) is primarily empiric, and the effect of this choice on length of stay (LOS) and mortality is largely unknown.ObjectiveTo examine the impact of antibiotic choice on these outcomes in general medical patients hospitalized with CAP.MethodsOne hundred patients hospitalized with CAP were prospectively identified. Seventy-six met inclusion criteria and were entered into the study. After hospital discharge, each medical chart was examined by 2 independent physicians who verified the admitting diagnosis and entered the data for antimicrobial regimens, a CAP mortality prediction tool, a social and disposition index, and other health outcomes. Patients were stratified according to the antibiotic received. Simple regression techniques were used to examine the correlation between initial therapy, specifically, ceftriaxone sodium or a macrolide, and LOS and mortality.ResultsPatients who received macrolides within the first 24 hours of admission had a markedly shorter LOS (2.8 days) than those not so treated (5.3 days; P=.01). This effect diminished as the interval before administering macrolides increased. Including ceftriaxone as part of the initial therapy did not appear to affect LOS. Patients given a macrolide for initial treatment did not differ significantly from those not treated in terms of mean age, mortality prediction tool score, or Social and Disposition Index score. Eleven of the 12 patients who received macrolides also received a β-lactam antibiotic.ConclusionUse of macrolides as part of an initial therapeutic regimen appears to be associated with shorter LOS.THERE ARE more than 5 million cases per year of community-acquired pneumonia (CAP) in the United States, resulting in more than 1 million hospital admissions.The estimated annual cost of care is $34.4 billion.The predominant component of the cost of care is the cost of hospitalization, including nursing and hotel costs. In one recent large studyin the Midwest, mean hospital length of stay (LOS) was found to be 8.5 days for teaching hospitals; in another multistate study,this was a reported 6.3 days. Such durations are in contrast to a goal of 2 days, which was recently suggested by consultants to the managed care industry.In most patients with CAP, the causative agents are not identified during the illness. Therefore, patients are commonly treated with empirically chosen antibiotics. The American Thoracic Society has published guidelinesfor the empiric treatment of CAP, based on the age of the patient and the severity of illness. These guidelines were generated from the recommendations of a consensus panel, and their usefulness has not been prospectively evaluated. The guidelines suggest macrolides as one component of an antimicrobial regimen for hospitalized patients to be used when the clinician suspects that an "atypical" agent (Mycoplasma pneumoniae, Chlamydia pneumoniae, and Legionella pneumophila) may be the cause of pneumonia.We recently conducted a prospective study to determine what factors might explain the variability in LOS for patients with CAP. We examined the correlation between LOS and a series of factors we thought might explain variability in LOS, namely, the Pneumonia Severity of Illness Index(a mortality prediction tool), various social factors (Social and Disposition Index [SDI]), and an elevated respiratory rate 48 to 72 hours after admission. We found that only a minority (<35%) of the variability in LOS could be explained by any combination of these factors, consistent with the results of prior studies.These results are reported elsewhere (J.E.S., M.H.E., M.B., J.D., and R.M., unpublished data, 1997).Having determined that less than half of the variability in LOS could be explained by the factors described herein, we decided to examine the impact of the initial choice of antibiotic treatment on LOS. We were particularly interested in the effects of macrolides, because they are active against "atypical" pathogens, and ceftriaxone sodium, because a previous, unpublished study from this institution had shown a somewhat shorter LOS among patients with CAP treated with ceftriaxone as opposed to other agents. In this article, we report our findings on the effect of the antibiotic regimen on the outcomes of patients hospitalized with CAP.PATIENTS AND METHODSThe New England Medical Center is a 300-bed, tertiary care, teaching hospital serving the metropolitan Boston, Mass, area. Patients with CAP admitted between May 1996 and January 1997 were prospectively identified. Shortly after discharge, each patient's medical chart was examined by 2 researchers (J.E.S. and J.D.) to verify that CAP was the primary diagnosis.Excluded from this study were patients with human immunodeficiency virus, clear evidence of aspiration, those directly admitted to the intensive care unit, and those transferred from another hospital if they had been there for more than 24 hours. Patients admitted for presumed CAP but whose primary reason for hospitalization was determined to be something other than CAP, such as major gastrointestinal bleeding or congestive heart failure, were also excluded. These criteria were intended to ensure that the focus remained on patients in whom CAP was the dominant reason for hospitalization. Patients admitted from nursing homes were included in this study if they met these criteria.During the study, a guideline for the treatment of CAP was in circulation at the Department of Medicine, New England Medical Center, that recommended ceftriaxone sodium, 1 g once daily, for initial empiric treatment, and suggested adding macrolides if an "atypical" pneumonia was suspected. The guideline was voluntary.For each patient, data were extracted for several categories of indicators we thought might influence the outcomes, particularly mortality and LOS (Table 1). Data were entered into a computer database (Access; Microsoft Corp, Redmond, Wash) and analyzed (JMP software; SAS Corporation, Cary, NC). Statistical associations of dichotomous variables were assessed using the χ2test; for continuous variables, the ttest was used. Simple linear regression and logistic regression were used for multivariate analyses.Table 1. Data Extracted for IndicatorsIndicatorComponentsScoring SystemSeverity of illness (mortality prediction tool)Age, sex, mental status changes, vital signs, laboratory testsSee Fine et al(weighted scores)Social and Disposition IndexAdmission from nursing home, No. of hospital or emergency department visits in previous year, substance abuse, homelessness, anticipated compliance problems1 Point per item except No. of visits, which received 1 point per visit; range, 0-8 (mean, 2.6) (not weighted)Clinical courseRespiratory rate >20/min between 48 and 72 hComplications of therapyIntravenous line–related sepsis, pneumothorax after pleurocentesis, Clostridium difficile, adverse antibiotic reactions1 Point per item (not weighted)RESULTSOf 100 patients identified, 20 were excluded for not meeting inclusion criteria, misdiagnosis at admission, or missing charts. Four were excluded as extreme outliers because their LOS was greater than 2 SDs from the mean. Average age of the 76 remaining patients was 68.3 years (range, 26-97 years); 43% were male. Each of the following comorbidities was present in 13% to 21% of patients: neoplastic disease, chronic obstructive pulmonary disease, cerebrovascular disease, coronary artery disease, congestive heart failure, and chronic renal insufficiency. Diabetes mellitus was present in 5 patients. Twenty-three patients were admitted from a nursing home, and 4 had a history of substance abuse (alcohol or intravenous drug abuse), homelessness, or anticipated problems with compliance after discharge.Overall mortality prediction toolscore was 0.17, and average mortality risk class was 4. From the latter value, between 6 and 7 deaths would have been anticipated. In fact, 2 patients died during hospitalization, and 4 others died with pneumonia-related illnesses within 30 days after discharge. Six people were transferred to the intensive care unit during admission. Of the 4 complications that we monitored specifically (Table 1), only Clostridium difficilewas noted, which occurred in 3 patients.Sixty-eight patients received a β-lactam antibiotic in the first 24 hours. Ceftriaxone was given to 51 patients, ticarcillin disodium and clavulanic acid to 13, ampicillin sodium and sulbactam sodium to 3, cefuroxime to 2, and imipenem to 1. Twelve patients received a macrolide within the first 24 hours. In 6 patients, the drug was erythromycin, given intravenously; in another 5, clarithromycin was given orally. One patient received both agents. Other agents given to 1 or 2 patients as part of initial therapy were clindamycin, ciprofloxacin, gentamicin sulfate, vancomycin, and doxycycline. Of the 12 patients given a macrolide as part of initial treatment, 11 also received either ticarcillin–clavulanic acid or ceftriaxone; the last received clarithromycin only.Mean LOS for all 76 patients was 4.9 days (range, 1-15 days). There was no difference in mean LOS between initial treatments that did and did not include ceftriaxone (Table 2). By contrast, there was a striking difference in mean LOS between patients whose initial treatment included a macrolide. Patients who received macrolides within the first 24 hours had a mean LOS of 2.75 days; those who did not had a mean LOS of 5.3 days, ie, about 2-fold longer (P=.01). When the data were analyzed on the basis of macrolide therapy within the first 48 hours of admission, the LOS was 4.05 days for patients who received a macrolide vs 5.2 days for those who did not (P=.19). There was no significant difference in LOS between patients who received a macrolide at any time during hospitalization and those who never received it (Table 2). Thus, the association of macrolide use with a decreased LOS was statistically significant only if the drug was given in the first 24 hours of admission. There were too few deaths, complications, or readmissions within 30 days to evaluate the impact of early macrolide treatment on these outcomes.Table 2. Effect of Choice of Antibiotic on OutcomesOutcomeCeftriaxone Sodium Within First 24 hPMacrolide Within First 24 hPMacrolide at Any TimePYes (n=51)No (n=25)Yes (n=12)No (n=64)Yes (n=27)No (n=49)Length of stay, mean (SD)4.76 (3.3)4.96 (3.4).82.75 (1.8)5.3 (3.4).014.6 (3.5)5.1 (3.3).51Complications, No. (%)1 (2.0)2 (8.0).181 (8.3)2 (3.1).451 (3.7)2 (4.0).19Deaths in hospital, No. (%)1 (2.0)1 (4.0).611 (8.3)1 (1.6).251 (3.7)1 (2.0).67Deaths within 30 d after discharge, No. (%)2 (4.0)2 (8.0).4704 (6.25).231 (3.7)3 (6.1).64We examined the possibility of significant differences in risk factors between patients who did and did not receive macrolide therapy, which might have independently influenced the LOS. We found no statistically significant differences with regard to age, sex distribution, mortality risk, or social factors between patients who did and did not receive macrolide therapy within the first 24 hours (Table 3). There was a trend in the early macrolide group for a lower male-female ratio, but the difference was not statistically significant. Similarly, when patients were segregated according to whether they received a macrolide at any time during hospitalization, there were no differences between the groups in age, sex distribution, or mortality risk. However, patients treated with a macrolide had a lower SDI score, primarily due to a lower rate of admission from a nursing home in the group that received macrolides.Table 3. Comparison of Demographic and Severity of Illness ScoresVariableMacrolide Within First 24 h (n=12)No Macrolide Within First 24 h (n=64)PMacrolide at Any Time (n=27)No Macrolide at Any Time (n=49)PAge, mean, y67.868.5.9167.469.73Male, %2545.184143.86Mortality prediction tool score0.140.17.60.150.18.52Social and Disposition Index score2.43.03.382.13.4.02Predictions of LOS from simple univariate and multivariate regression analyses looking at treatment with macrolides within the first 24 hours and admission from a nursing home were not significantly different. The predictive power of early macrolide use (within the first 24 hours) was not significantly affected by adding the variable that denoted admission from a nursing home, as evidenced by a minimal decrease in the β for early macrolide use from 1.27 (P=.001) in the univariate model to 1.25 (P=.01) in the bivariate model. This did not affect the statistical significance of the model or its correlation with LOS. By itself, admission from a nursing home did not explain a statistically significant amount of the variability in LOS.We also noted a difference in the incidence of identified pathogens between patients who were and were not given a macrolide. This difference was of borderline significance (P=.06) when the groups were stratified by treatment within the first 24 hours (Table 4). In both groups, most patients did not have a pathogen identified. In only 1 patient was an "atypical" pneumonia agent identified, ie, M pneumoniae. A pathogen was identified in 1 (8%) of 12 patients in the early macrolide group, and in 23 (36%) of 64 patients in the group not treated with a macrolide (P<.06). Three patients had 2, and 2 patients had 3 identified pathogens. In addition, the mean LOS for the 52 patients with no identified pathogens was significantly shorter than that of the 24 patients with identified pathogens (4.3 vs 6.3 days, P=.02).Table 4. Microbiologic AgentsCultured AgentsMacrolides Within First 24 h (n=12)No Macrolides Within First 24 h (n=64)No. (%) of isolatesStreptococcus pneumoniae07 (10.9)Staphylococcus aureus06 (9.4)Haemophilus influenzae04 (6.3)Enteric gram negative1 (8.3)10 (15.6)"Atypicals"01 (1.6)Other02 (3.1)No. (%) of patients with an identified pathogen†1 (8.3)23* (35.9)No. (%) of patients without identified pathogen11 (91.7)41 (64.1)*χ2(Homogeneity across treatment categories)=3.56; P=.06.†Three patients had 2 agents cultured, and 2 patients had 3 agents cultured.We examined whether receiving a macrolide early in the admission might be a marker for other factors, which were themselves responsible for shorter LOS. If we controlled for the absence of an identified pathogen (by adding such a variable to the regression model), early use of a macrolide still had significant predictive power on the LOS. The β coefficient for early macrolide use decreased only slightly, from 1.27 (P=.01) to 1.15 (P=.02).Finally, in the broader analysis of factors that might affect LOS (J.E.S., M.H.E., M.B., J.D., and R.M., unpublished data, 1997), we found that an elevated maximum respiratory rate measured between 48 and 72 hours of hospitalization predicted a longer LOS. Therefore, we performed a similar analysis in this study. Comparing subjects given and not given a macrolide within the first 24 hours, the mean respiratory rates were 28.8/min vs 26.3/min on day 1 (P=.3), 22.9/min and 22.6/min on day 2 (P=.85), and 20.4/min vs 22.3/min on day 3 (P=.16). Thus, there was no significant difference in respiratory rate at 48 and 72 hours between patients stratified by macrolide use.COMMENTThe most striking finding in this study is the marked difference in LOS between patients who did and did not receive a macrolide as part of therapy within the first 24 hours of admission. The LOS was about 50% shorter (2.75 vs 5.3 days) for patients who received a macrolide in the first 24 hours. This effect was less evident the longer after admission a macrolide was administered. The lack of a beneficial effect for patients in whom a macrolide was administered after 48 hours is not surprising because the change in treatment presumably reflected concern about a poor response to the initial therapeutic regimen, and such patients would be expected to have a longer LOS.There were too few events in either group to allow us to evaluate whether early use of a macrolide was associated with fewer deaths or with a lower rate of complications or readmission. The overall predicted mortality in this study, based on the criteria of the Pittsburgh Pneumonia Risk Class scale,was 9% compared with an actual mortality of 7.9%.We examined the possibility that differences in age, sex distribution, mortality prediction tool score, or SDI score might explain the difference in LOS between patients given or not given a macrolide in the first 24 hours. There was a lower proportion of males and a slightly lower SDI score (primarily because of a smaller proportion admitted from nursing homes) among patients treated with a macrolide within the first 24 hours compared with those not so treated, but the differences were not statistically significant.More striking was the finding that patients in the "early macrolide" group were more likely to have no pathogen identified as a cause of their CAP than those not given macrolides in the first 24 hours; the difference was of borderline statistical significance (P=.06). Furthermore, there was a significantly shorter LOS in patients with no identified pathogen than in those in whom a pathogen was identified. Taken together, these observations raise the possibility that early macrolide treatment may simply be a marker for patients with no pathogen identified, which, in turn, may be a marker for patients with pneumonia caused by an "atypical" or nonbacterial pathogen.The low overall rate of pathogen identification is not surprising. Even in studies specifically designed to find the causes of CAP, no pathogen has been identified in a high proportion of patients.Our study was not designed to investigate the causes of CAP. No special effort was made to identify a pathogen in most patients,and the serologic studies that are usually the basis for diagnosis of infection by an "atypical" pathogen were rarely done. Indeed, there was no specified protocol for obtaining cultures or conducting serologic studies in these patients. Therefore, caution must be exerted in interpreting the results of these microbiologic studies. Nevertheless, the lower rate of retrieval of an identified pathogen in the macrolide-treated group in the present study raises the possibility that a higher proportion of these patients than those in the control group was infected by an "atypical" pathogen, such as M pneumoniae, C pneumoniae, or L pneumophila. Recent studiessuggest that these agents may play a larger role in CAP than was previously thought, either as primary pathogens or as copathogens with ordinary bacteria. Pneumonia caused by an atypical agent might have a shorter LOS—especially when a macrolide is used for treatment—than bacterial pneumonia treated with a β-lactam antibiotic. In that case, the shorter LOS associated with early treatment with a macrolide would be the result of segregation—perhaps unwitting—of a subgroup of patients with infection caused by "atypical" pathogens, rather than a nonspecific effect of macrolides in CAP. The difference in causative agents, rather than the use of the macrolide per se, could explain the difference in LOS.There are 3 reasons to doubt that the difference between the groups in causative agents is the explanation for our findings. First, physicians would have had to be able to identify patients with infection caused by an "atypical" pathogen at the time of admission in order to assign them preferentially to the macrolide group. Many studieshave shown that clinical and routine laboratory tests have very low predictive power for diagnosing infection by an "atypical" pathogen. Second, the microbiologic studies, as noted herein, were not designed to identify the pathogens in CAP and therefore should be interpreted with great caution. Third, the predictive power of the variable that represents early macrolide use did not decrease significantly when the variable for the unknown pathogen was added to the model predicting LOS.In addition to a direct antibacterial effect, especially against "atypical" pathogens, there is one other potential effect of macrolides that might explain the finding of a shorter LOS in patients with CAP treated initially with a macrolide, namely, a beneficial immunologic or anti-inflammatory effect. There is some evidence that macrolides may inhibit interleukin 8 production and may reduce the proinflammatory effects of various bacterial products.There is also evidence that macrolides may enhance the production of certain inflammatory mediators such as interleukin 1 and interleukin 6.Our data did not allow us to distinguish a role for either of these possibilities.Other reports suggest a beneficial effect of macrolides for the treatment of CAP. One studyof outpatients with CAP found that the use of macrolides was associated with lower rates of mortality and subsequent all-cause hospitalization, although the differences were not statistically significant. In another study,among the factors associated with a significantly lower mortality rate in patients hospitalized for the treatment of CAP was an initial administration of a β-lactam and macrolide combination. A recent studycomparing clarithromycin alone with amoxicillin–clavulanic acid alone in the treatment of patients hospitalized for CAP showed no difference between the 2 agents in clinical response, microbiologic response, or LOS, but no specific data were given for LOS. In contrast to these patients, in whom the macrolide was given as sole therapy, in our study, 11 of the 12 patients who received a macrolide within the first 24 hours of admission also received a potent β-lactam agent. It is possible that macrolides are not the most effective antibacterial agents by themselves but play a useful adjunctive role in CAP.Length of stay is the major determinant of cost for the hospitalized CAP patient. The cost of a typical day in the hospital, with standard nursing care on a medical ward, has been approximated at $640 nationally.Most episodes of CAP never have a specific origin identified and are treated empirically. If the addition of a macrolide to the usual treatment with a β-lactam is able to shorten the LOS substantially, the savings in cost of care could be large. Our findings suggest that a randomized, prospective trial of this hypothesis would be worthwhile.MSNiedermanSPPetersUpdate in pulmonary medicine.Ann Intern Med.1998;128:208-215.GERosenthalDLHarperLMQuinnGSCooperSeverity-adjusted mortality and length of stay in teaching and nonteaching hospitals: results of a regional study.JAMA.1997;278:485-490.Not AvailableHealthcare Management Guidelines (Pneumonia, Community Acquired).Washington, DC: Milliman & Robertson Inc; 1995:2.82-2.83.MSNiedermanJBBassGDCampbellGuidelines for the initial management of adults with community-acquired pneumonia: diagnosis, assessment of severity, and initial antimicrobial therapy, American Thoracic Society.Am Rev Respir Dis.1993;148:1418-1426.LIIezzoniMSchwartzASAshYDMackiernanUsing severity measures to predict the likelihood of death for pneumonia inpatients.J Gen Intern Med.1996;11:23-31.SDHornPDSharkeyJMBuckleJEBackofenRFAverillRAHornThe relationship between severity of illness and hospital length of stay and mortality.Med Care.1991;29:305-317.MJFineBHHanusaJRLaveComparison of a disease-specific and a generic severity of illness measure for patients with community-acquired pneumonia.J Gen Intern Med.1995;10:359-368.MJFineDESingerBHHanusaJRLaveWNKapoorValidation of a pneumonia prognostic index using the Medisgroups comparative hospital database.Am J Med.1993;94:153-159.MJFineJJOrloffDArisumiPrognosis of patients hospitalized with community-acquired pneumonia.Am J Med.1998;88:1N-8N.JHBatesGDCampbellALBarronMicrobial etiology of acute pneumonia in hospitalized patients.Chest.1992;101:1005-1012.TJMarrieHDurantLYatesCommunity-acquired pneumonia requiring hospitalization: a 5-year prospective study.Rev Infect Dis.1989;11:586-599.MAWoodheadJTMacFarlaneJSMcCrackenDHRoseRGFinchProspective study of the aetiology and outcome of pneumonia in the community.Lancet.1987;1:671-674.JPachonMDPradosFCapoteJACuelloJGarnachoAVeranoSevere community-acquired pneumonia: etiology, prognosis, and treatment.Am Rev Respir Dis.1990;142:369-373.ATorresJSerra-BatilesAFerrerSevere community-acquired pneumonia: epidemiology and prognostic factors.Am Rev Respir Dis.1981;144:312-318.British Thoracic Society Research Committee and the Public Health Laboratory ServiceThe aetiology, management and outcome of severe community-acquired pneumonia in the intensive care unit.Respir Med.1992;86:7-13.DLiebermanFSchlaefferIBoldurMultiple pathogens in adult patients admitted with community-acquired pneumonia: a one year prospective study of 346 consecutive patients.Thorax.1996;51:179-184.MTKauppinenPSaikkuPKujalaEHervaHSyrjalaClinical picture of community-acquired Chlamydia pneumoniaepneumonia requiring hospital treatment: a comparison between chlamydial and pneumococcal pneumonia.Thorax.1996;51:185-189.GDFangMFineJOrloffNew and emerging etiologies for community-acquired pneumonia with implication for therapy: a prospective multicenter study of 359 cases.Medicine (Baltimore).1990;69:307-316.TJMarrieRWPeelingMJFineDESingerCMColeyWNKapoorAmbulatory patients with community-acquired pneumonia: the frequency of atypical agents and clinical course.Am J Med.1996;101:508-515.DIsaacsProblems in determining the etiology of community-acquired childhood pneumonia.J Pediatr Infect Dis.1989;8:145-148.MAWoodheadJTMacFarlaneComparative clinical laboratory features on Legionellawith pneumococcal and Mycoplasmapneumonias.Br J Dis Chest.1987;81:133-139.BMFarrDLKaiserBDWHarrisonCKConnollyPrediction of microbial aetiology at admission to hospital for pneumonia from the presenting clinical features.Thorax.1989;44:1031-1035.JTMacFarlaneACMillerWHRSmithAHMorrisDHRoseComparative radiographic features of community-acquired Legionnaires' disease, pneumococcal pneumonia, Mycoplasma pneumonia, and psittacosis.Thorax.1984;39:28-33.JTewLCalenoffBSBerlinBacterial or nonbacterial pneumonia: accuracy of radiographic diagnosis.Radiology.1977;124:607-612.CHSChanMCohenJPangA prospective study of community-acquired pneumonia in Hong Kong.Chest.1992;101:442-446.KMatsumotoNew approaches to Pseudomonas aeroginosalower respiratory tract infections.Verh K Acad Geneeskd Belg.1995;57:109-112.RMizukaneYHirakataMKakuComparative in vitro exoenzyme-suppressing activities of azithromycin and other macrolide antibiotics against Pseudomonas aeroginosa.Antimicrob Agents Chemother.1994;38:528-533.KSakataHYajimaKTanakaErythromycin inhibits the production of elastase by Pseudomonas aeroginosawithout affecting its proliferation in vitro.Am Rev Respir Dis.1993;148:1061-1065.GMolinariCAGuzmanAPasceGCSchitoInhibition of Pseudomonas aeroginosavirulence factors by subinhibitory concentrations of azithromycin and other macrolide antibiotics.J Antimicrob Chemother.1993;31:681-688.GJRasRAndersonGWTaylorClindamycin, erythromycin, and roxithromycin inhibit the proinflammatory interactions of Pseudomonas aeroginosapigments with human neutrophils in vitro.Antimicrob Agents Chemother.1992;36:1236-1240.YHirakataMKakuRMizukanePotential effects of erythromycin on host defense systems and virulence of Pseudomonas aeroginosa.Antimicrob Agents Chemother.1992;36:1922-1927.PPGleasonWNKapoorRAStoneMedical outcomes and anti-microbial costs with the use of the American Thoracic Society guidelines for outpatients with community-acquired pneumonia.JAMA.1997;278:32-39.BJGuglielmoVDudasSTranSLeTMasudaAHopeflTreatment outcomes associated with community-acquired pneumonia (CAP) in US hospitals: a 3000 patient survey.In: Program and abstracts of the 37th Interscience Conference on Antimicrobial Agents and Chemotherapy; September 1997; Toronto, Ontario. Abstract K-146.DGenneHHSiegristLHumairBJanin-JaquatAde TorrenteClarithromycin vs. amoxicillin-clavulanic acid in the treatment of community-acquired pneumonia.Eur J Clin Microbiol Infect Dis.1997;16:783-788.JRLaveMJFineSSSankeyBHHanusaLAWeissfeldWNKapoorHospitalized pneumonia: outcomes, treatment patterns, and costs in urban and rural areas.J Gen Intern Med.1996;11:415-421.Accepted for publication April 14, 1999.Reprints: James E. Stahl, MD, Center for Research on Health Care, University of Pittsburgh Medical Center, 820 E Montefiore University Hospital, Pittsburgh, PA 15213 (e-mail: [email protected]).
Camargo, Jr, Carlos A.; Weiss, Scott T.; Zhang, Shumin; Willett, Walter C.; Speizer, Frank E.
doi: 10.1001/archinte.159.21.2582pmid: 10573048
BackgroundObesity and asthma are common disorders, and their prevalence rates continue to rise. Although individuals with asthma may gain weight as a result of activity limitations, the relationship between body mass index (BMI), which is calculated as weight in kilograms divided by the square of height in meters, and risk of developing asthma is not known.MethodsWe performed a prospective cohort study of female US registered nurses in the Nurses' Health Study II. In 1991, after excluding women who died with probable asthma or with incomplete data, there were 85,911 participants, aged 26 to 46 years. The main outcome measure was self-report of physician-diagnosed asthma with recent use of an asthma medication.ResultsFrom 1991 to 1995, we identified 1596 incident cases of asthma. In a multivariate model controlling for 9 potential confounding factors (including age, race, smoking, physical activity, and energy intake), the relative risks of asthma for 6 increasing categories of BMI in 1991 were 0.9, 1.0 (reference), 1.1, 1.6, 1.7, and 2.7 (Pfor trend <.001). Stronger associations were found using stricter definitions for asthma, and the finding was present in a variety of subgroups. In analyses controlling for the same variables, as well as BMI at age 18, women who gained weight after age 18 were at significantly increased risk of developing asthma during the 4-year follow-up period (Pfor trend <.001).ConclusionsThe BMI has a strong, independent, and positive association with risk of adult-onset asthma. The increasing prevalence of obesity in developed nations may help explain concomitant increases in asthma prevalence.ASTHMA AFFECTS at least 5% of the US population and its prevalence continues to rise.In 1990, the management of asthma already accounted for more than $6 billion in medical expenditures.Despite the enormity of the problem, there are relatively few large-scale epidemiological studies on the etiology of asthma, particularly adult-onset asthma. In recent years, investigators have begun to define geneticand environmentalrisk factors in the hope that timely interventions might prevent individuals from developing asthma.The prevalence of obesity in the United States has increased steadily during the past 30 years.In recent years, even lesser degrees of overweight have been linked to a variety of health problems, including premature mortality.Although patients with poorly controlled asthma might gain weight as a result of activity limitations, the relationship between obesity and risk of asthma is not known. We examined the relation of body mass index (BMI), which is calculated as weight in kilograms divided by the square of height in meters, and weight change to risk of adult-onset asthma during 4 years of follow-up in more than 85,000 women.PARTICIPANTS AND METHODSPARTICIPANTSThe Nurses' Health Study II is an ongoing prospective cohort study designed to examine the relation of lifestyle and diet to occurrence of breast cancer and other major illnesses. The participants are 116,678 female registered nurses, aged 24 to 44 years, who were living in 1 of 14 states when they responded to the first questionnaire in 1989. Follow-up questionnaires were sent in 1991, 1993, and 1995, with response rates of 93%, 92%, and 90%, respectively.For the present analysis, we designated July 1991 as the start of prospective follow-up because some risk factors were not collected in 1989. Thus, we excluded women who had died between July 1989 and June 1991 (36); who reported physician-diagnosed asthma anytime before July 1991 (7019); who did not complete the July 1991 follow-up questionnaire (7834); or who were missing baseline data on diet (12,432), age (6), physical activity (296), BMI (2486), or weight at age 18 (658). These exclusions left 85,911 women for analysis.METHODSBaseline Data and Validation StudiesParticipants answered a questionnaire about the following factors: age, race, current weight and height, weight at age 18, smoking status, physical activity, dietary intake, other risk factors, and medical history. Participants also were asked if they had recently undergone a health screening examination (as opposed to a physical examination to evaluate symptoms of asthma or no recent physical examination), or if they currently used any nutritional supplements, such as multivitamins. At the 2-year follow-up in 1993, participants were asked to measure the largest circumference around their hips and their waist circumference at the navel while standing. Measured, not estimated, circumferences were recorded. These measurements allowed calculation of the waist-hip ratio (WHR).The BMI index is minimally correlated with height (r= −0.03) and highly correlated with absolute fat mass in women (r= 0.84-0.91).Body mass index categories for the present analysis were less than 20.0, 20.0 to 22.4, 22.5 to 24.9, 25.0 to 27.4, 27.5 to 29.9, and 30.0 or higher. Expressed as a percentage of desirable weight, according to the 1983 Metropolitan Life Insurance Company tables,these categories correspond approximately to less than 90%, 90% to 100%, 101% to 112%, 113% to 123%, 124% to 134%, and 135% or more of recommended weights.Several validation studies have been conducted in the present cohort and in the original Nurses' Health Study, a cohort established in 1976 of 121,700 female nurses aged 30 to 55 years at enrollment. Validation studies of relevance to the present analysis include self-reported current weight,weight at age 18,and waist and hip circumference.For each of these adiposity variables, self-reported values were found to be highly correlated with measured values.Diagnosis of AsthmaAll women who reported in the 1993 or 1995 follow-up questionnaires that a physician had first diagnosed them as having asthma during the preceding 2 years received supplementary questionnaires on the symptoms, diagnosis, and therapy of asthma. Follow-up from July 1991 to June 1995 was 95% complete. Each participant who reported asthma was categorized based on 3 case definitions of increasingly stricter criteria. Case definition 1 defines an incident case of asthma and was considered confirmed if the nurse did the following: (1) reiterated on the supplementary questionnaire that a physician had diagnosed her as having asthma, and (2) reported using an asthma medication since diagnosis. To meet case definition 2, participants had to fulfill both of the preceding criteria and report use of a prescribed long-term preventive medication (ie, inhaled corticosteroids, cromolyn sodium, nedocromil, salmeterol) in the past year. Finally, to meet case definition 3, participants had to meet all the criteria from case definitions 1 and 2 plus report that their physician-diagnosed asthma was within 1 month of symptom onset. In 1998, we recontacted a random sample of 100 women who met the criteria of case definition 2 to determine their asthma status approximately 5 years after receiving the physician diagnosis of asthma (1991-1995). Further validation was not pursued because of the following: (1) these criteria are considerably stricter than those of most asthma epidemiology studies; (2) we found highly accurate reporting of other chronic diseases among registered US nurses; and (3) the cost of pulmonary function testing in this cohort would be prohibitive.Statistical AnalysisParticipants were grouped according to the BMI, based on height and weight reported in the 1991 questionnaire. Incident cases of asthma were assigned to the BMI category at baseline. Cumulative incidence rates were calculated by dividing the number of new asthma cases by the number of people at risk in a given BMI category. The relative risk (RR), computed as the cumulative incidence in a specific category of BMI divided by the corresponding value in the second category (BMI, 20.0-22.4), was used as a measure of the strength of the association. Multivariate logistic regression models controlled for multiple risk factors simultaneously. A test for linear trend was conducted by assigning a score of 1 through 6 for each available BMI value, and this score was then analyzed as a continuous variable. A similar approach was used for analyses involving weight change since age 18. The absolute excess risks (attributable risks) due to adiposity (using the cumulative incidence for women with a BMI of 20.0-22.4 as the reference) or weight change (using those with <2-kg weight change as the reference) were calculated using the multivariate RR. Group comparisons were made using the χ2test. All RRs are presented with 95% confidence intervals (CIs), and all reported Pvalues are 2-sided.RESULTSIn 1991, at baseline, the mean weight of participants in the Nurses' Health Study II cohort was 3.3 kg (7.3 lb) less than women of similar age in the general US population.The participants' mean ± SD BMI was 24.5 ± 5.2. Table 1shows the baseline characteristics according to BMI in 1991. Women with higher BMIs tended to be older, do less physical activity, and have a larger energy intake. Obese women in 1991 tended to weigh more at age 18 and have gained more weight during the ensuing years. Obese women also were less likely to have recently undergone a health screening examination or to use nutritional supplements.Table 1. Age-Standardized Baseline Characteristics According to Body Mass Index in 1991CharacteristicsBody Mass Index, kg/m2, in 1991<20.020.0-22.422.5-24.925.0-27.427.5-29.9≥30.0No. of participants11,41525,76519,87611,877577511,203Mean age, y353636373737White, %959594949494Current smoker, %131212131313Physical activity, METS*/wk292725232118Total energy intake, kJ/d747774567481761976947878Weight, kg, at age 18525557606269Body mass index, kg/m2, at age 1818.820.021.122.022.925.4Weight change, kg, since age 180.73.47.011.515.926.5Health screening examination, %727373717068Nutritional supplement use, %605756565551*METS indicates metabolic equivalents.During 4 years of observation, from 1991 to 1995, we identified 1596 case definition 1 incident cases of asthma. This yielded a cumulative incidence of approximately 0.5% per year. By definition, in all incident cases the participants reported a physician diagnosis of asthma on 2 separate questionnaires and use of at least 1 asthma medication since diagnosis. The median duration from symptom onset to physician diagnosis was 2 months (interquartile range, 0-19 months). Given the 2-year follow-up cycle, the supplementary asthma questionnaire was sent to participants with incident asthma a median of 21 months after they received their asthma diagnosis (interquartile range, 15-28 months). Thus, almost 2 years after first being told they had asthma, 90% of asthmatic participants reported use of an asthma medication in the past year. Use of inhaled corticosteroids was reported by 59%. Other medications used in the past year included oral or intravenous steroids for asthma (25%), cromolyn or nedocromil (12%), and salmeterol (5%). Parental history of asthma was reported by approximately 21% of incident cases of asthma. Common triggers of asthma exacerbations included respiratory infections (70%) and exposure to environmental allergens (67%), such as pets, house dust, mold, or pollen. Only 2% of nurses reported latex allergy as a precipitant of an asthma exacerbation. Smoking also was uncommon, with 66% never smokers, 23% past smokers, and 11% current smokers at baseline. Approximately 7% of incident cases of asthma reported at least 1 hospitalization for acute asthma since physician diagnosis of asthma.The 1998 validation survey of 100 participants who met the criteria of case definition 2 provides additional evidence that these health care professionals accurately reported their asthma. Only 5% of women who originally met case definition 2 raised issues in 1998 that might lead us to question their diagnosis (eg, asthma symptoms resolved when 1 woman stopped taking β-blockers; another woman received a diagnosis of asthma based on 1 visit to the emergency department and reported mild respiratory symptoms thereafter). Another 4% of cases had symptomatic asthma requiring regular medication use for more than 6 months but reported asthma "resolution" by the time of the 1998 survey. Even if we include these potential noncases, 89% of cases reported use of a β-agonist inhaler in 1998, and 73% reported ongoing use of long-term preventive medication. Furthermore, 82% reported at least 1 respiratory symptom (eg, coughing, wheezing, or shortness of breath) in the preceding month, and 34% reported at least 1 visit to the emergency department or urgent office visit in the past year for treatment of an asthma exacerbation.Returning to the primary analysis, we found that BMI in 1991 had a strong, independent, and positive association with risk of adult-onset asthma. Table 2exhibits RR of adult-onset asthma during 1991 to 1995, according to the BMI and the 3 case definitions of asthma. Restricting cases to participants who used a preventive medication in the past year (n = 1079) yielded almost identical results. Because some cases may have had activity limitations before diagnosis, we included only those with physician-diagnosed asthma within 1 month of symptom onset (n = 453). Again, BMI in 1991 remained a strong, independent risk factor for developing asthma during the 4-year follow-up. All these models were recalculated while controlling for additional factors (eg, diet, menopause, use of oral contraceptives or hormones) without any material change in the BMI-asthma association (data not shown).Table 2. Relative Risk (RR) of Adult-onset Asthma During a 4-Year Follow-up (1991-1995) According to Body Mass Indexand Case Definition*CaseBody Mass Index, kg/m2, in 1991Pfor Trend<20.020.0-22.422.5-24.925.0-27.427.5-29.9≥30.0Definition 1†No. (n = 1596)144359305259131398. . .Age-adjusted RR (95% CI)0.9 (0.7-1.1)1.0 (reference)1.1 (1.0-1.3)1.6 (1.4-1.9)1.7 (1.4-2.0)2.6 (2.3-3.1)<.001Multivariate RR (95% CI)0.9 (0.7-1.1)1.0 (reference)1.1 (1.0-1.3)1.6 (1.3-1.9)1.7 (1.4-2.0)2.7 (2.3-3.1)<.001Definition 2‡No. (n = 1079)10022721716492279. . .Multivariate RR (95% CI)1.0 (0.8-1.2)1.0 (reference)1.3 (1.0-1.5)1.6 (1.3-2.0)1.9 (1.5-2.4)3.0 (2.5-3.6)<.001Definition 3§No. (n = 453)2894876531148. . .Multivariate RR (95% CI)0.7 (0.4-1.0)1.0 (reference)1.2 (0.9-1.6)1.5 (1.1-2.1)1.5 (1.0-2.3)3.8 (2.9-5.0)<.001*Multivariate RRs are adjusted for age, race, US region, smoking status, physical activity, total energy intake, hysterectomy status, birth weight, and duration of breastfeeding. Ellipses indicate not applicable; CI, confidence interval.†Participant-reported physician diagnosis of asthma on 2 separate questionnaires and use of an asthma medication since diagnosis.‡The criteria for case definition 1 (see above) plus use of a preventive asthma medication in past year.§The criteria for case definition 2 (see above) plus 1 month or less between symptom onset and physician diagnosis of asthma.Because obesity represents the complex interplay of both lifestyle and genetic factors, we also report the RR of developing asthma according to levels of physical activity and total energy intake; data on familial predisposition to obesity or asthma were not available. In the primary model (case definition 1 in Table 2), the multivariate RR of asthma for quintiles of increasing physical activity was 1.0 (reference), 1.0, 1.0, 0.9, and 1.0 (Pfor trend = .31). Substituting frequency of vigorous physical activity, as opposed to quintiles of overall physical activity, yielded similar results (data not shown). By contrast, the multivariate RR of asthma for quintiles of increasing total energy intake was the following: 1.0 (reference), 1.0, 1.0, 1.1, and 1.3 (95% CI, 1.1-1.5) (Pfor trend <.001). Results from similar models that used stricter case definitions did not differ materially (data not shown).Weight change since age 18 also had a strong, independent, and positive association with risk of adult-onset asthma (Table 3). Women who lost weight appeared to be at slightly decreased risk, whereas women who gained weight were at increased risk. As before, stronger associations were present for case definition 2 and case definition 3 (Pfor trend <.001). For example, in the analysis using case definition 3, participants who gained more than 25 kg since age 18 had a multivariate RR of 4.7 (95% CI, 3.1-7.0), compared with those participants whose weight had remained stable.Table 3. Relative Risk (RR) of Adult-onset Asthma During 4-Year Follow-up (1991-1995) According to Change in Weight Since Age 18*Case Definition 1Change in Weight, kgPfor Trend<−5−5 to −2.1−2 to 22.1 to 55.1 to 1010.1 to 2020.1 to 25>25No. (n = 1596)5464158220311390145254. . .Age-adjusted RR (95% CI)1.0 (0.7-1.3)0.8 (0.6-1.1)1.0 (reference)0.9 (0.8-1.2)1.1 (0.9-1.3)1.4 (1.2-1.7)2.1 (1.7-2.6)2.7 (2.2-3.4)<.001Multivariate RR (95% CI)0.8 (0.6-1.1)0.8 (0.6-1.1)1.0 (reference)0.9 (0.8-1.2)1.1 (0.9-1.3)1.4 (1.2-1.7)2.0 (1.6-2.5)2.5 (2.0-3.1)<.001*Case definition 1 includes participants who reported a physician diagnosis of asthma on 2 separate questionnaires and the use of an asthma medication since diagnosis. Ellipses indicate not applicable; CI, confidence interval. Multivariate RRs are adjusted for age, race, US region, smoking status, physical activity, total energy intake, hysterectomy status, birth weight, duration of breastfeeding, and body mass index at age 18.To address potential concerns about detection bias, we reexamined the BMI-asthma association of the cohort according to 2 characteristics that might serve as markers of health-oriented behavior: (1) history of undergoing a recent health screening examination (72% of cohort), and (2) baseline use of nutritional supplements (56% of cohort). We also reexamined the BMI-asthma association according to smoking status in 1991. Figure 1shows the strong dose-response association between BMI and RR of asthma among women who reported undergoing a recent health screening examination. We found strong, positive associations in all 7 subgroups: those who had and had not undergone a recent health screening examination; those taking or not taking nutritional supplements; and nonsmokers, past smokers, or current smokers (all Pfor trend <.001, except among current smokers, in whom Pfor trend = .002). Stronger associations again were found when we used stricter case definitions. For example, in the analysis that was restricted to participants who reported undergoing a recent health screening examination that used the criteria of case definition 3 (n = 285), those who had a BMI of 30.0 or higher had a multivariate RR of 4.3 (95% CI, 3.1-6.0) compared with those participants with a BMI of 20.0 to 22.4.Relative risk of adult-onset asthma during 1991 to 1995, according to body mass index, among 61,324 women who reported undergoing a recent health screening examination (n = 1061 cases). Relative risks are adjusted for all 9 potential confounding factors listed in the first footnote of Table 2. The bars represent 95% confidence intervals. Pfor trend <.001.Because WHR data were not available until 1993, the relation of WHR to risk of asthma was examined during 1993 to 1995 only. Before adding WHR to the multivariate model, we confirmed the strong, positive association between BMI and asthma (Pfor trend <.001). There was a moderate correlation between BMI and WHR (Spearman r= 0.37) and substitution of WHR for BMI in the multivariate model revealed a positive association between WHR and asthma risk. However, when both adiposity measures were included in the multivariate model, BMI remained a significant risk factor (eg, RR, 2.8 for BMI ≥30.0 vs 20.0-22.4; 95% CI, 2.2-3.6; Pfor trend <.001), whereas WHR was not associated with asthma risk (RR, 0.9-1.2; all Pvalues >.30). Substituting waist circumference for WHR yielded similar results (data not shown).Of the overall incidence of asthma in the cohort, 38% was accounted for by excessive body weight, defined as a baseline BMI of 22.5 or higher. For women with a BMI of 25.0 or higher, 50% of their increased risk could be accounted for by their excess weight. For women who were obese (BMI ≥30.0), 62% of their increased risk could be accounted for by their excess weight. Alternatively, 26% of the overall incidence of asthma could be accounted for by weight gains of 2 kg or more since age 18. Likewise, 60% of the risk among those gaining more than 25 kg could be attributed to their weight gain after age 18.Finally, we characterized the incident asthma of thin (BMI <22.5) vs obese (BMI ≥30.0) women. Compared with thin women with asthma (n = 503), obese women with asthma (n = 398) were equally likely to report the use of an asthma medication in the past year (88% vs 90%, respectively; P= .50), the use of inhaled corticosteroids (56% vs 59%; P= .49), and a parental history of asthma (20% vs 22%; P= .55). Thin women with asthma were equally likely to report an environmental allergen as an acute asthma trigger (66% vs 63%; P= .29), but they were less likely to report respiratory infections as a past trigger (68% vs 77%; P= .003). In addition, thin women with asthma were less likely than obese women with asthma to report at least 1 hospitalization for asthma since physician diagnosis of asthma (5% vs 12%; P= .001). Overall, the clinical characteristics of incident asthma among thin vs obese women appeared similar.COMMENTIn this large, prospective cohort study of female nurses, BMI had a strong, independent, and positive association with the risk of developing adult-onset asthma. This association was present using several definitions of asthma. Indeed, the use of stricter definitions led to even stronger associations between BMI and asthma risk. Weight gain since age 18 also was strongly associated with increased risk of asthma.Prior epidemiological data on the relationship between adiposity and asthma are sparse. A few cross-sectional studies have reported that individuals with asthma tend to weigh more than those without asthmaand that adiposity is associated with an increased prevalence of wheezingand asthma.Furthermore, cross-sectional studiesof adiposity and pulmonary function show that obese individuals tend to have decreased forced vital capacity or forced expiratory volume in 1 second. Results from one small studyshowed that reduction in expiratory reserve volume due to obesity also showed a tendency of closure of small peripheral airways (<2 mm) in dependent lung zones. However, across a "normal" range of body weights these relatively weak associations may not be apparent, perhaps because of confounding by smoking or the presence of a nonlinear association between body weight and pulmonary function.Regardless, the use of the cross-sectional study design in the studies mentioned herein makes it difficult to determine if obesity preceded or followed asthma onset. Because individuals with poorly controlled asthma may gain weight as a result of activity limitations, the temporal nature of the relationship is critical.The only relevant prospective studies examined the relation of adiposity to level of pulmonary function, not incidence of asthma. Higher baseline weight or BMI predicted subsequent declines in pulmonary function in some,but not all,participants in these studies. Weight gain during follow-up was significantly associated with a decline in pulmonary function,particularly among men.However, because weight gain did not clearly precede the observed decrease in pulmonary function, these results may have simply reflected antecedent declines in pulmonary function followed by activity limitation and weight gain. Marked weight loss among subjects who are morbidly obese led to improved pulmonary function in most,but not all,subjects in these studies. We believe the information from the studies above show sufficient evidence to support the plausibility of the observed BMI-asthma association.Although mechanistic research is understandably limited, there are several explanations for how adiposity might increase asthma risk. Animal models suggest that obesity leads to a variety of histological changes in the rat lung, but the relevance of this work to humans, particularly across the spectrum of BMI, is uncertain.Of greater relevance, weight gain leads to progressive reductions in airway caliber as a result of chest wall restriction. Since airway conductance is a function of the fourth power of airway radius, other asthma risk factors would only need to have a small effect on airway size to have relatively large effects on air flow. By analogy, differences in relative airway size might explain sex differences in asthma prevalence across the age spectrum.Cross-sectional studies also suggest that obese children without asthmaand obese patients with mild chronic obstructive pulmonary diseasehave more bronchial hyperreactivity than their lean counterparts. The mechanism for this hyperreactivity is unclear but may relate to a reduction in airway caliber, which is associated with hyperreactivity,or to alterations in lipid metabolism.The BMI also may be representative of sedentary lifestyles associated with infrequent deep inspirations, which have been shown to increase bronchial hyperreactivity; however, controlling for a variety of physical activity measures in our study did not alter the results. Alternatively, obese individuals may be at higher risk because they tend to spend more time indoors,thereby leading to more sustained exposure to the indoor allergens associated 2with asthma.Obese individuals are at increased risk for gastroesophageal reflux disease, an important risk factor for adult-onset asthma.Another possible explanation is that obese and lean women might have different diets, and a preliminary study has linked intake of specific nutrients with pulmonary disease.We examined the possibility of dietary differences in detail (data not shown) and conclude that nutrient intake did not account for the strong association between BMI and asthma.Obesity also is known to influence estrogen and progesterone levels,and these sex hormones have been loosely linked to risk of asthma and other atopic disorders. For example, asthma prevalence is higher among women than men during the reproductive years, but not earlier.Postmenopausal estrogen use appears to increase asthma risk,although postmenopausal estrogens were not implicated as a confounding factor in the present study given the age range of participants. However, the relation of obesity to estrogen is complex, with increased plasma levels found in obese postmenopausal women,compared with their lean counterparts, but decreased plasma levels found in obese premenopausal women.Interpretation of the relation of obesity to estrogen concentration is further complicated by evidence that obese women may have higher levels of free, biologically active, plasma estradiol because of concomitant decreases in sex hormone–binding globulin.More research is needed to clarify the interrelations between obesity, sex hormones, and asthma.Regardless of the exact mechanism of how adiposity might increase asthma risk, the public health implications of our findings are potentially large. Nationally representative surveys indicate that one third of US adults are "overweight" (BMI ≥27.3).If one applies the 2-fold to 3-fold increased asthma risk of comparable women in the present study to the general population, the population attributable risk would be approximately 25% to 40%.The rising prevalence of overweight individuals in the United States (25% in 1971-1974 to 33% in 1988-1991)may help explain concomitant increases in asthma prevalence. Furthermore, we observed that the unusual sociodemographic distribution of obesity closely resembles that of asthma: both disorders are more common in developed nations, but, paradoxically, the highest prevalences are among the most disadvantaged members of these same societies.Thus, public health campaigns to encourage weight loss may have an additional benefit in that they may prevent onset or expression of asthma.One potential limitation of our findings involves the select nature of the cohort. Because the age range of participants was 26 to 46 years at baseline, the results from this study may not apply to individuals with asthma that begins in childhood or after age 50. Furthermore, in the early phases of any long-term study, participants tend to be healthier than nonparticipants because individuals who are severely ill and chronically disabled are less likely to volunteer for enrollment. Women enrolled in the Nurses' Health Study II were a subgroup of a source population of nurses that differs in many ways from the general population.The registered nurses' level of education, ready access to medical care, and above-average standard of living are all differences that might limit generalizability. All these factors probably contributed to the cohort's high number of health screening examinations and the popularity of nutritional supplement use. The factors also may explain the cohort's lower mean adiposity level compared with that of the general US population. Nonetheless, there is little biological basis for suspecting that the observed relationship between BMI and asthma would be materially different in other groups of women. Furthermore, the group's relative homogeneity and good baseline health help minimize confounding by socioeconomic factors that are associated with asthma.Other potential limitations of our study included use of self-reported adiposity factors, self-reported asthma, and the possibility of detection bias. However, BMI and weight change since age 18 were found to be highly correlated with measured values in prior validation studies.With regard to asthma diagnosis, this remains a formidable challenge.Although we believe that asthma was accurately reported by this select group of registered US nurses, we note that chronic obstructive pulmonary disease would be rare in this age group and that the BMI-asthma association was present in nonsmokers. Furthermore, we used multiple case definitions to address possible misclassification. The strictest definition, case definition 3, required the participant to report the following: (1) physician diagnosis of asthma on 2 separate questionnaires; (2) use of a prescribed long-term preventive asthma medication, such as an inhaled corticosteroid, in the past year; and (3) physician diagnosis of asthma within 1 month of symptom onset. These criteria are considerably stricter than those generally reported in the asthma epidemiology literature.The strong association between BMI and asthma became even stronger when using the case definition 3 case group. We considered that the results may have resulted from bias, particularly detection bias. However, we found strong evidence for an obesity-asthma association in a variety of subgroups, and no evidence that the asthma severity of obese women with asthma was less than that of their thin counterparts. Moreover, we observed that the risk for asthma was elevated even at a BMI of 22.5 to 24.9. This level is below standard clinical criteria for obesity and, therefore, is unlikely to have led to any detection bias. Even women with a weight that is considered "normal" or average were at slightly increased risk.In summary, this large, prospective cohort study demonstrated that high BMI and weight gain since age 18 are associated with increased risk of developing adult-onset asthma. Confirmation of this finding in other populations, particularly among the young and minority groups, is warranted. Nonetheless, the present results are consistent with cross-sectional reports that individuals with asthma tend to weigh more than those without asthma and that obesity may have an adverse effect on pulmonary function. The magnitude of the RR is larger than that of many putative asthma risk factors and, when combined with the high prevalence of obesity in the United States, suggests that obesity may explain much of the current asthma epidemic. These data add to existing evidence that excess body fat is a major cause of human disease and suggest a new avenue for asthma management and prevention.Centers for Disease Control and PreventionAsthma: United States, 1982-1992.MMWR Morb Mortal Wkly Rep.1995;43:952-955.KBWeissPJGergenTAHodgsonAn economic evaluation of asthma in the United States.N Engl J Med.1992;326:862-866.ASandfordTWeirPParéThe genetics of asthma.Am J Respir Crit Care Med.1996;153:1749-1765.MRBecklakePErnstEnvironmental factors.Lancet.1997;350:10-13.RJKuczmarskiKMFlegalSMCampbellCLJohnsonIncreasing prevalence of overweight among US adults: The National Health and Nutrition Examination Surveys, 1960 to 1991.JAMA.1994;272:205-211.JEMansonWCWillettMJStampferBody weight and mortality among women.N Engl J Med.1995;333:677-685.DSpiegelmanRGIsraelCBouchardWCWillettAbsolute fat mass, percent body fat, and body-fat distribution: which is the real determinant of blood pressure and serum glucose?Am J Clin Nutr.1992;55:1033-1044.Not Available1983 Metropolitan height and weight tablesStat Bull Metrop Life Found.1983;64:3-9.WCWillettMJStampferCBainCigarette smoking, relative weight, and menopause.Am J Epidemiol.1983;117:651-658.EBRimmMJStampferGAColditzCGChuteLBLitinWCWillettValidity of self-reported waist and hip circumferences in men and women.Epidemiology.1990;1:466-473.LMTroyDJHunterJEMansonGAColditzMJStampferWCWillettThe validity of recalled weight among younger women.Int J Obes Relat Metab Disord.1995;19:570-572.CACamargo JrLDRichardsonEpidemiology of asthma.In: Brenner BE, ed. Emergency Asthma. New York, NY: Marcel Dekker Inc; 1999:59-80.GAColditzPMartinMJStampferValidation of questionnaire information on risk factors and disease outcomes in a prospective cohort study of women.Am J Epidemiol.1986;123:894-900.GAColditzWCWillettARotnitzkyJEMansonWeight gain as a risk factor for clinical diabetes mellitus in women.Ann Intern Med.1995;122:481-486.WCBaileyJMRichards JrBAManzellaCMBrooksRAWindsorSJSoongCharacteristics and correlates of asthma in a university clinic population.Chest.1990;98:821-828.DRGoldARotnitzkyAIDamokoshRace and gender differences in respiratory illness prevalence and their relationship to environmental exposures in children 7 to 14 years of age.Am Rev Respir Dis.1993;148:10-18.MShohatTShohatRKedemMMimouniYLDanonChildhood asthma and growth outcome.Arch Dis Child.1987;62:63-65.JSchwartzSTWeissDietary factors and their relation to respiratory symptoms: The Second National Health and Nutrition Examination Survey.Am J Epidemiol.1990;132:67-76.RUngerLKreegerKKChristoffelChildhood obesity: medical and familial correlates and age of onset.Clin Pediatr (Phila).1990;29:368-373.DWDockeryJHWareBGFerris JrDistribution of forced expiratory volume in one second and forced vital capacity in healthy, white adult never-smokers in six U.S.cities.Am Rev Respir Dis.1985;131:511-520.LSInselmanAMilaneseADeurlooEffect of obesity on pulmonary function in children.Pediatr Pulmonol.1993;16:130-137.RLazarusGColditzCSBerkeyFESpeizerEffects of body fat on ventilatory function in children and adolescents: cross-sectional findings from a random population sample of school children.Pediatr Pulmonol.1997;24:187-194.EMarcusSBuistDCurbCorrelates of FEV1 and prevalence of pulmonary conditions in Japanese-American men.Thorax.1988;138:1398-1404.LCollinsPHobertyJWalkerThe effect of body fat distribution on pulmonary function tests.Chest.1995;107:1298-1302.IRubinsteinNZamelLDuBarryVHoffsteinAirflow limitation in morbidly obese, nonsmoking men.Ann Intern Med.1990;112:828-832.JSchwartzSTWeissRelationship of dietary vitamin C intake and pulmonary function in the First National Health and Nutrition Examination Survey (NHANES I).Am J Clin Nutr.1994;59:110-114.PLEnrightRAKronmalMHigginsMSchenkerEFHaponikSpirometry reference values for women and men 65 to 85 years of age: cardiovascular health study.Am Rev Respir Dis.1993;147:125-133.MLWangLMcCabeJLHankinsonLongitudinal and cross-sectional analyses of lung function in steelworkers.Am J Respir Crit Care Med.1996;153:1907-1913.FGDouglasPYChongInfluence of obesity on peripheral airways patency.J Appl Physiol.1972;33:559-563.JBSchoenbergGJBeckABouhuysGrowth and decay of pulmonary function in healthy blacks and whites.Respir Physiol.1978;33:367-393.JBandeJClementKPVan De WoestijneThe influence of smoking habits and body weight on vital capacity and FEV1 in male air force personnel: a longitudinal and cross-sectional analysis.Am Rev Respir Dis.1980;122:781-790.DHeederikBGMillerWeak associations in occupational epidemiology: adjustment for exposure estimation error.Int J Epidemiol.1988;17(suppl):970-974.YChenSLHorneJADosmanBody weight and weight gain related to pulmonary function decline in adults: a six year follow up study.Thorax.1993;48:375-380.MLWangLMcCabeELPetsonkJLHankinsonDEBanksWeight gain and longitudinal changes in lung function in steel workers.Chest.1997;111:1526-1532.JECotesJCGilsonCJohnEffects of inactivity and a gain in weight upon lung function and indices of submaximal exercise.J Physiol (Lond).1967;188:36P-38P.ASDontasDRJacobs JrACorcondilasAKeysPHannanLongitudinal versus cross-sectional vital capacity changes and affecting factors.J Gerontol.1984;39:430-438.GABorkanDSparrowCWisniewskiPSVokonasBody weight and coronary disease risk: patterns of risk factor change associated with long-term weight change: the Normative Aging Study.Am J Epidemiol.1986;124:410-419.GNBedellWRWilsonPMSeebohmPulmonary function in obese persons.J Clin Invest.1958;37:1049-1060.ROCrapoTMKellyCGElliottSBJonesSpirometry as a preoperative screening test in morbidly obese patients.Surgery.1986;99:763-767.PSThomasEROwenGHulandsJSMilledgeRespiratory function in the morbidly obese before and after weight loss.Thorax.1989;44:382-386.KHakalaPMustajokiJAittomakiARSovijarviEffect of weight loss and body position on pulmonary function and gas exchange abnormalities in morbid obesity.Int J Obes Relat Metab Disord.1995;19:343-346.RWVaughanRCCorkDHollanderThe effect of massive weight loss on arterial oxygenation and pulmonary function tests.Anesthesiology.1981;54:325-328.LSInselmanLBPadilla-BurgosSTeichbergHSpencerAlveolar enlargement in obesity-induced hyperplastic lung growth.J Appl Physiol.1988;65:2291-2296.TAKaplanEMontanaExercise-induced bronchospasm in nonasthmatic obese children.Clin Pediatr (Phila).1993;32:220-225.REKannerJEConnettMDAltoseGender difference in airway hyperresponsiveness in smokers with mild COPD.Am J Respir Crit Care Med.1995;150:956-961.BGonenPO'DonnellTJPostTJQuinnESSchulmanVery low density lipoproteins (VLDL) trigger the release of histamine from human basophils.Biochim Biophys Acta.1987;917:418-424.GSklootSPermuttATogiasAirway hyperresponsiveness in asthma: a problem of limited smooth muscle relaxation with inspiration.J Clin Invest.1995;96:2393-2403.SLGortmakerAMustAMSobolKPetersonGAColditzWHDietzTelevision viewing as a cause of increasing obesity among children in the United States, 1986-1990.Arch Pediatr Adolesc Med.1996;150:356-362.DLRosenstreichPEgglestonMKattanThe role of cockroach allergy and exposure to cockroach allergen in causing morbidity among inner-city children with asthma.N Engl J Med.1997;336:1356-1363.WGSimpsonGastroesophageal reflux disease and asthma: diagnosis and management.Arch Intern Med.1995;155:798-803.SEHankinsonWCWillettJEMansonAlcohol, height, and adiposity in relation to estrogen and prolactin levels in postmenopausal women.J Natl Cancer Inst.1995;87:1297-1302.NPotischmanCASwansonPSiiteriRNHooverReversal of relation between body mass and endogenous estrogen concentrations with menopausal status.J Natl Cancer Inst.1996;88:756-758.JFDorganMEReichmanJTJuddThe relation of body size to plasma levels of estrogens and androgens in premenopausal women (Maryland, United States).Cancer Causes Control.1995;6:3-8.EMSkobeloffWHSpiveySSSt ClairJMSchoffstallThe influence of age and sex on asthma admissions.JAMA.1992;268:3437-3440.RJTroisiFESpeizerWCWillettDTrichopoulosBRosnerMenopause, postmenopausal estrogen preparations, and the risk of adult-onset asthma: a prospective cohort study.Am J Respir Crit Care Med.1995;152:1183-1188.CHHennekensJEBuringEpidemiology in Medicine.Boston, Mass: Little Brown & Co Inc; 1987.JSobalAStunkardSocioeconomic status of obesity: a review of the literature.Psychol Bull.1989;105:260-275.JGScaddingKMMoserDefinition of asthma.In: Weiss E, Stein M, eds. Bronchial Asthma: Mechanisms and Therapeutics. 3rd ed. Boston, Mass: Little Brown & Co Inc; 1993:3-14.Accepted for publication March 2, 1999.The Nurses' Health Study II is supported by research grant CA-50385 from the National Institutes of Health, Bethesda, Md. Dr Camargo is supported by grant HL-03533 from the National Institutes of Health.We are indebted to Meir Stampfer, MD, for reviewing the manuscript, Gary Chase, Karen Corsano, Eleni Konstantis, Stefanie Parker, and Lori Ward for their technical support, and the participants in the Nurses' Health Study II for their ongoing dedication to the study.Reprints: Carlos A. Camargo, Jr, MD, Channing Laboratory, 181 Longwood Ave, Boston, MA 02115 (e-mail: [email protected]).
Showing 1 to 10 of 13 Articles
doi: 10.1001/archinte.159.21.2542pmid: 10573044
ContextLeflunomide is a reversible inhibitor of de novo pyrimidine synthesis shown to be effective in a phase 2 trial in 402 patients with active rheumatoid arthritis (RA).ObjectiveTo compare the efficacy and safety of leflunomide treatment with placebo and methotrexate treatment in patients with active RA.DesignRandomized, double-blind, placebo, and active-controlled 12-month study.SettingForty-seven university and private rheumatology practices in the United States and Canada.PatientsDiagnosis of RA by the American College of Rheumatology (ACR) criteria for duration of 6 months or longer and no previous methotrexate treatment.InterventionLeflunomide treatment (20 mg/d), placebo, or methotrexate treatment (7.5-15 mg/wk).Main Outcome MeasuresAmerican College of Rheumatology success rate (completed 52 weeks of treatment and met the ACR ≥20% response criteria), disease progression as assessed by x-ray films, and improvement in function and health-related quality of life using the intent-to-treat population.ResultsThe 482 patients studied were predominantly women (mean age, 54 years; mean disease duration, 6.7 years) for whom a mean of 0.8 disease-modifying antirheumatic drugs had failed. The ACR response and success rates for patients receiving leflunomide treatment (52% and 41%, respectively) and methotrexate treatment (46% and 35%, respectively) were significantly higher than those for patients receiving placebo (26% and 19%, respectively) (P<.001), and they were statistically equivalent, with mean time to initial response at 8.4 weeks for patients receiving leflunomide vs 9.5 weeks for patients receiving methotrexate therapy. X-ray analyses demonstrated less disease progression with leflunomide (P≤.001) and methotrexate (P=.02) therapy than with placebo. Leflunomide and methotrexate treatment improved measures of physical function and health-related quality of life significantly more than placebo (P<.001 and P<.05, respectively). Common adverse events for patients receiving leflunomide treatment included gastrointestinal complaints, skin rash, and reversible alopecia. Asymptomatic transaminase elevations resulted in treatment discontinuations for 7.1% of patients receiving leflunomide therapy, 1.7% of patients receiving placebo, and 3.3% of patients receiving methotrexate therapy.ConclusionsClinical responses following administration of leflunomide, a new therapeutic agent for the treatment of RA, were statistically superior to those with placebo and equivalent to those with methotrexate treatment. Both active treatments improved signs and symptoms of active RA, delayed disease progression as demonstrated by x-ray films, and improved function and health-related quality of life.CURRENT treatments for rheumatoid arthritis (RA) include nonsteroidal anti-inflammatory drugs (NSAIDs), low-dose steroids, and disease-modifying antirheumatic drugs (DMARDs). The active control drug for this study, methotrexate, is considered to be the "gold standard" DMARD for the treatment of RA.No currently available medication is uniformly effective, however, and all may cause significant adverse effects.Leflunomide is an isoxazole immunomodulatory agent with demonstrated prophylactic and therapeutic effects in animal models of autoimmune disease.Following oral administration of leflunomide, the isoxazole ring is rapidly cleaved to form the active metabolite, which binds to the enzyme dihydroorotate dehydrogenase, thereby inhibiting de novo pyrimidine synthesis.In rapidly dividing cell populations, such as activated lymphocytes, this results in cell cycle arrest, which can be reversed in vitro and in vivo by the administration of uridine.Leflunomide treatment was reported to be effective at dosages of 10 and 25 mg daily in a 6-month phase 2 study of 402 patients with active RA.This article describes a 12-month double-blind phase 3 study designed to provide long-term data on the safety and efficacy of leflunomide treatment compared with placebo and methotrexate therapy for patients with active RA.PATIENTS AND METHODSPATIENT POPULATIONMen and women aged 18 years or older were eligible for treatment if they met the American College of Rheumatology (ACR) criteria for having RA for 6 months or longer.Active RA was defined by 3 of the following 4 criteria: 9 or more tender joints, 6 or more swollen joints, morning stiffness lasting 45 minutes or longer, and Westergren erythrocyte sedimentation rate (ESR) of 28 mm/h or greater. Notably, patients could not have previously received methotrexate treatment; treatment with all other DMARDs must have been discontinued for at least 30 days. Prednisone treatment (≤10 mg/d) (or the equivalent) and NSAIDs were permitted if dosages had been stable for at least 30 days before enrollment and remained stable during treatment. Patients were excluded if they had inflammatory joint disease that was not caused by RA, had a history of clinically significant drug or alcohol abuse, or admitted to consumption of more than 1 alcoholic drink per day. Required baseline laboratory values included a hemoglobin concentration of 100 g/L or greater; a hematocrit of 0.30 or greater; a leukocyte count of 3 × 109/L or greater; a platelet count of 100 × 109/L or greater; a serum creatinine level lower than twice the upper limit of normal (ULN) for age and sex; an albumin level greater than or equal to the lower limit of normal (≥35 g/L); and normal liver function test results, defined as 3 or more serial evaluation results that were less than or equal to 1.2 times the ULN for aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase, and bilirubin levels. Sexually active premenopausal women and men were instructed to use a medically accepted form of contraception throughout treatment and for 6 months thereafter.STUDY PROTOCOLThis was a randomized 12-month multicenter double-blind placebo-controlled study designed in 1993 and initiated (with regulatory approval) in 1995. All patients gave informed consent and then provided a medical history and underwent physical examination, laboratory assessment, chest x-ray, and electrocardiography. Baseline clinical assessments included tender and swollen joint counts (28 joints), patient and physician global assessments of disease activity (on a visual analog scale [VAS], 0-100 cm), patient assessment of pain (VAS, 0-100 cm), Modified Health Assessment Questionnaire (MHAQ) score,Westergren ESR, and C-reactive protein (CRP) level. Rheumatologic assessments were performed biweekly during weeks 4 through 12 and monthly thereafter. The Health Assessment Questionnaire (HAQ), Problem Elicitation Technique (PET) questionnaire (a disease-specific instrument that asks patients to identify and rank activities most affected by their RA), and the Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36) were completed by patients at baseline and at 24 and 52 weeks of treatment or at the time of study exit.Radiographs of the hands and feet were taken at baseline and at 52 weeks or at the time of early study exit, using fine-grained, single-emulsion film and following protocol specifications for optimal positioning.The ACR response criteria were used, which required improvement of 20% or greater in both tender and swollen joint counts as well as improvement of 20% or greater in 3 of the following 5 measures: patient self-assessed function/disability (MHAQ), patient global assessment, physician global assessment, patient assessment of pain, and acute-phase reactant value assessed by ESR or CRP level.The ACR response rates for improvement of 50% and greater and 70% and greater were also calculated.TREATMENT ASSIGNMENTPatients were assigned to 1 of 3 treatment groups in a 3:2:3 randomization: leflunomide treatment (20 mg/d), placebo, or methotrexate treatment (7.5 mg/wk). Patients were enrolled in the order of confirmed eligibility. Randomization was stratified according to the time since the patient's last treatment with a DMARD (<8 or ≥8 weeks). Study sites obtained unique patient numbers by calling a randomization center, which used a computerized adaptive algorithm to assign treatment based on stratum and, to maintain balance, the number of patients previously assigned at that site to each arm. To preserve the blind, all patients received an oral dose of leflunomide or matching leflunomide placebo once daily and an oral dose of methotrexate or matching methotrexate placebo once per week.A daily 20-mg dose of leflunomide was selected based on the phase 2 efficacy and pharmacokinetics data. Based on the 14- to 16-day half-life of the active metabolite, a 100-mg loading dose of leflunomide or leflunomide placebo was administered for the first 3 days to allow a steady-state plasma concentration to be reached within 6 to 8 weeks. Consistent with current medical practice, all patients received 1 mg of folate once or twice daily. If active disease (as defined above) was still present at week 6 of protocol treatment, the methotrexate or methotrexate placebo dosage was mandated to be increased to 15.0 mg over weeks 7 through 9 and continued thereafter.If a patient's response failed to meet the ACR response criteria after 16 weeks of treatment or if a patient developed an unacceptable adverse event, the initial study medication could be discontinued and, after a washout period, the patient could elect to receive alternate therapy. Patients who chose alternate therapy and were originally assigned to methotrexate therapy or placebo received leflunomide therapy; those who were originally assigned to leflunomide therapy received methotrexate therapy. Allocations to initial and alternate therapy were done at the time of study entry and remained blinded. This report presents results from the initial therapy phase only.Consistent with the recommended ACR guidelines for monitoring methotrexate therapy, dosage adjustments, discontinuation of treatment, and/or liver biopsies were mandated for persistent or recurrent transaminase elevations.For elevations that were greater than twice the ULN, daily and weekly medication dosages were reduced; for elevations that were greater than 3 times the ULN or repetitive elevations that were greater than twice the ULN despite dosage reduction, initial therapy was discontinued.STATISTICAL ANALYSESThe principal objective of the study was the comparison of leflunomide therapy with placebo. Comparisons of leflunomide therapy with methotrexate therapy and methotrexate therapy with placebo were secondary objectives of the study; therefore, Pvalues are reported without adjustment. The primary measure of efficacy was meeting the ACR 20% response criteria and completing 52 weeks of initial therapy (ACR success). The ACR response rates for improvement of 50% or greater and 70% or greater at 12 months were also determined. Secondary outcome measures included mean changes over time in each of the components of the ACR response criteria, results of hand and feet x-ray films, measures of physical function and health-related quality of life (assessed with the HAQ, PET questionnaire, and SF-36), duration of morning stiffness, and rheumatoid factor titers.In accordance with the US Food and Drug Administration final guidance document for the clinical development of drugs, devices, and biologic agents for the treatment of RA, the retardation of disease progression as demonstrated by x-ray evaluation and improvement in function and health-related quality of life represent potential claims only if the primary end point shows a statistically significant difference.Therefore, these analyses were conditioned on the primary analysis and no multiple comparison adjustment was required.Original sample size requirements were determined based on 4 primary variables: tender joint count, swollen joint count, physician global assessment, and patient global assessment. Later calculations that were performed after the study had begun confirmed that the estimated sample size would be sufficient to demonstrate treatment differences using the ACR response criteria. This sample size provided a 90% power to detect treatment differences.All efficacy statistical analyses were performed on the intent-to-treat population, defined as all randomized patients who received any dosage of study medication with at least one study evaluation after randomization. In addition to descriptive statistics, comparisons of on-treatment values with baseline values for the 3 treatment groups used logistic regression analysis for categorical variables and analysis of covariance for continuous variables. Analyses used the last observation carried forward method. As health-related quality-of-life data were not normally distributed at baseline, the van Elteren extension to the Wilcoxon rank sum test results was applied to continuous variables, and the Cochran-Mantel-Haenszel χ2test results were applied to categorical variables comparing change from baseline.Statistical analyses (analysis of covariance for 95% confidence intervals and Pvalues) included comparisons and possible interactions using duration of disease (≤2 or >2 years), no prior DMARD therapy, concomitant corticosteroid and/or NSAID administration, investigator pool, geographic region, and recent DMARD therapy (≤8 or >8 weeks). The analysis plan defined 5 geographic regions for the investigational sites.Reports of adverse events, physical examination results, standard laboratory assessments, electrocardiogram results, and chest x-ray results were analyzed for safety, including mean changes in individual parameters over time.RESULTSPATIENT CHARACTERISTICSOf 485 patients enrolled at 42 sites, 482 received at least 1 dose of a study drug or placebo and were evaluated for safety; 480 had at least 1 follow-up visit to evaluate efficacy (182 patients received leflunomide therapy, 118 received placebo, and 180 received methotrexate therapy [3:2:3 randomization]). Demographic and disease characteristics were similar across treatment groups (Table 1and Table 2). Two hundred thirty-eight patients completed 52 weeks of initial treatment (leflunomide therapy, 53%; placebo, 31%; and methotrexate therapy, 58%); 108 of 132 eligible patients received alternate therapy (leflunomide therapy, 13%; placebo, 44%; and methotrexate therapy, 18%). In total, 346 patients completed protocol treatment (leflunomide therapy, 66%; placebo, 75%; and methotrexate therapy, 77%). Early discontinuations occurred more often with placebo (69%) than with leflunomide (47%) or methotrexate (42%) therapy. These discontinuations were caused by a lack of efficacy in 53% of patients receiving placebo compared with 17% of patients receiving leflunomide and 24% of patients receiving methotrexate therapy. The number of treatment withdrawals increased most notably in the placebo group on or after the 4-month visit, at which time entry into alternate therapy was allowed (Figure 1and Table 3).Table 1. Demographic and Disease Characteristics*Leflunomide (n=182)Placebo (n=118)Methotrexate (n=182)Female, %72.570.375.3Age, y†54.1 ± 12.054.6 ± 10.753.3 ± 11.8Rheumatoid arthritis duration, y†7.0 ± 8.66.9 ± 8.06.5 ± 8.1Rheumatoid arthritis duration ≤2 y, %39.033.340.1Rheumatoid factor positive, %64.860.259.4No. of DMARDs that failed†0.8 ± 1.00.9 ± 0.90.9 ± 1.0No prior DMARD treatment, %44.539.844.0Taking concomitant NSAIDs, %75.265.269.7Taking concomitant steroids, %53.855.152.7*DMARD indicates disease-modifying antirheumatic drug; NSAID, nonsteroidal anti-inflammatory drug.†Values are mean ± SD.Table 2. Change in Individual Outcome Parameters for the Intent-to-Treat Population*Leflunomide (n=182)Placebo (n=118)Methotrexate (n=180)Joint count (range, 0-28)TenderBaseline15.5 ± 6.416.5 ± 6.315.8 ± 6.9Mean change−7.7 ± 7.8†−3.0 ± 8.4−6.6 ± 7.6†SwollenBaseline13.7 ± 6.014.8 ± 6.213.0 ± 5.7Mean change−5.7 ± 6.5†−2.9 ± 6.1−5.4 ± 5.5†Global assessment of disease activity (VAS)PatientBaseline5.6 ± 2.25.8 ± 2.25.4 ± 2.3Mean change−2.1 ± 2.7†0.1 ± 2.8−1.5 ± 2.9†PhysicianBaseline6.1 ± 1.56.2 ± 1.65.9 ± 1.7Mean change−2.8 ± 2.8†−1.0 ± 2.5−2.4 ± 2.7†MHAQ scoreBaseline0.8 ± 0.60.9 ± 0.50.8 ± 0.5Mean change−0.3 ± 0.5†‡0.1 ± 0.5−0.2 ± 0.5§Patient assessment of pain (VAS)Baseline5.9 ± 2.26.4 ± 1.95.8 ± 2.2Mean change−2.2 ± 2.9†−0.4 ± 2.4−1.7 ± 2.8†ESR, mm/hBaseline38.4 ± 26.837.3 ± 28.733.8 ± 25.4Mean change−6.3 ± 22.9§2.6 ± 19.0−6.5 ± 20.6†CRP, mg/dLBaseline2.08 ± 2.502.47 ± 2.671.88 ± 1.87Mean change−0.62 ± 2.45†0.47 ± 2.14−0.50 ± 1.87†*Values are mean ± SD. VAS indicates visual analog scale (range, 0-100 cm); MHAQ, Modified Health Assessment Questionnaire; ESR, erythrocyte sedimentation rate; and CRP, C-reactive protein.†Leflunomide or methotrexate vs placebo, P≤.001.‡Leflunomide vs methotrexate, P≤.01.§Leflunomide or methotrexate vs placebo, P≤.01.Figure 1.Study profile. The disposition of all randomized patients is shown for the initial and alternate phases of the study.Table 3. Reasons for Study WithdrawalsReason for WithdrawalNo. of PatientsLeflunomide (n=86)Placebo (n=81)Methotrexate (n=77)Adverse event*401019Lack of efficacy316244Protocol violation011Noncompliance101Lost to follow-up102Voluntary11810Other200*Includes 1 death in the methotrexate group.The weekly dosage of methotrexate was increased to 15 mg for 109 patients (61%) receiving methotrexate therapy. Increased dosages of methotrexate placebo were mandated for 95 patients (52%) receiving leflunomide therapy and 81 patients (69%) receiving placebo. Dosage reductions occurred in 3 patients (2%) receiving leflunomide therapy, none in the placebo group, and 4 patients (2%) receiving methotrexate therapy. Early withdrawals caused by adverse events occurred in 22% of the patients receiving leflunomide therapy, 10% of the patients receiving methotrexate therapy, and 9% of the patients receiving placebo (Table 4). In the leflunomide group, most early withdrawals were caused by gastrointestinal events (5.5%) or protocol-mandated discontinuations for asymptomatic transaminase elevations (7.1%).Table 4. Adverse Events Leading to Treatment WithdrawalAdverse EventPatients Withdrawn, %Leflunomide (n=182)Placebo (n=118)Methotrexate (n=182)Total22.08.510.4Elevated transaminase level7.11.74.4Gastrointestinal5.51.71.7Rash2.22.50.0Hypertension1.10.80.0Hyperlipidemia1.60.00.0Myocardial infarction0.50.00.5Pneumonia0.50.00.5Interstitial pneumonitis0.00.00.5Alopecia0.50.81.1Other3.01.01.7COMPARISON OF LEFLUNOMIDE TREATMENT WITH PLACEBOThe ACR success rate was significantly higher in the leflunomide treatment group compared with the placebo group (41% vs 19%; P<.001). Mean changes over time in each component of the ACR response index were significantly better in the leflunomide and methotrexate treatment groups than in the placebo group (P≤.01) (Table 2). The ACR response rates over time are shown in Figure 2. The data for the area under the response curve (AUC) (the number of weeks a patient reported improvement according to the ACR criteria), time to and duration of initial ACR response, time to and duration of sustained ACR response (reported improvement according to the ACR criteria at ≥3 consecutive visits [minimum duration, 8 weeks], and ACR response rates [improvement ≥20%, ≥50%, and ≥70%]) are displayed in Table 5. American College of Rheumatology responses in the leflunomide group occurred earlier and exceeded those in the placebo group at all time points.Figure 2.Intent-to-treat last observation carried forward analysis of the percentage of patients who met the American College of Rheumatology response criteria for improvement of 20% or greater at month 1 and quarterly thereafter by treatment group. Pvalues calculated for the 12-month data are presented in Table 5.Table 5. Patient Responses According to the American College of Rheumatology Rheumatoid Arthritis Success and Improvement Criteria by Treatment Group for the Intent-to-Treat Population*ParameterLeflunomide (n=182)Placebo (n=118)Methotrexate (n=182)Success, % (95% CI)†‡41 (33.5-47.8)§19 (11.5-25.7)35 (29.0-42.0)§≥20% improvement, % (95% CI)∥52 (45.0-60.0)§26 (18.0-34.0)46 (38.0-53.0)§≥50% improvement, % (95% CI)∥34 (27.0-41.0)§8 (3.0-12.0)23 (17.0-29.0)§≥70% improvement, % (95% CI)∥20 (14.0-26.0)§4 (1.0-8.0)9 (5.0-14.0)No. of weeks patients reported ≥20% improvement†23.7 (20.6)§12.6 (17.1)22.7 (19.2)Time to initial response, wk¶8.6 (7.4)10.4 (8.6)9.5 (6.5)Sustained response, %#533457Time to sustained response, wk¶10.7 (9.3)14.7 (11.5)14.0 (10.2)Duration of sustained response, wk¶33.4 (16.1)26.4 (13.6)29.6 (15.0)*CI indicates confidence interval.†According to the American College of Rheumatology criteria for improvement in rheumatoid arthritis.‡Success was defined as completing 52 weeks of initial therapy, with improvement of 20% or greater at end point.§Leflunomide or methotrexate therapy vs placebo, P≤.001.∥Last observation carried forward.¶Values are mean (SD).#Improvement of 20% or greater at 3 or more consecutive visits.Radiographs of the hands and feet at baseline and 12 months (and at the time of early exit) were obtained for 352 (73%) of 482 patients. Analyses were based on 12-month x-ray films, except when those taken at the time of early exit were the only follow-up films available (n=47). Significantly more disease progression occurred in patients treated with placebo compared with leflunomide therapy and approached the estimated progression (defined as the Sharp score at baseline divided by disease duration at baseline) (Table 6).Table 6. Mean Changes From Baseline in Total Sharp Scores and Erosion and Joint Space Narrowing Subscores for the Intent-to-Treat Population*Leflunomide (n=131)Placebo (n=83)Methotrexate (n=138)Total Sharp scoreBaseline23.11 (34.0)25.37 (31.3)22.76 (39.0)Estimated yearly progression†3.303.683.50Change at end point0.53 (4.5)‡2.16 (4.0)0.88 (3.3)§∥Erosion subscoreBaseline8.95 (19.6)9.28 (14.2)8.05 (18.4)Change at end point0.23 (2.2)¶0.84 (1.8)0.48 (1.8)Joint space narrowing subscoreBaseline14.15 (18.9)16.10 (20.8)14.71 (23.3)Change at end point0.31 (2.8)#1.24 (2.7)0.41 (1.8)¶*All x-ray films were read using the Sharp method (Bluhm et al). Total Sharp scores are the sum of the erosion and joint space narrowing subscores. Values are mean (SD).†Estimated as the total Sharp score at baseline divided by the mean disease duration at baseline.‡Leflunomide therapy vs placebo, P≤.001.§Methotrexate therapy vs placebo, P=.02.∥Leflunomide vs methotrexate therapy, P=.05.¶Leflunomide therapy vs placebo, P=.033.#Leflunomide therapy vs placebo, P<.001.Analyses of function/disability and health-related quality of life demonstrated statistically significant improvement in patients treated with leflunomide compared with patients who received placebo, not only for the MHAQ, the HAQ disability index, and the physical component score of the SF-36, but for all scales of the HAQ and the weighted top-5 score of the PET questionnaire (Table 7).Table 7. Mean Changes in Measures of Function/Disability and Health-Related Quality of Life for the Intent-to-Treat Population*†LeflunomidePlaceboMethotrexateHAQ disability index(n=164)(n=99)(n=168)Baseline1.31.31.3Mean change at end point−0.45‡§0.0−0.26‡PET weighted top-5 score(n=166)(n=101)(n=170)Baseline21.222.420.4Mean change at end point−6.91‡−0.66−3.41∥SF-36 physical component(n=157)(n=101)(n=162)Baseline30.028.929.7Mean change at end point7.6‡1.04.6∥*The population includes 438 instead of 480 subjects because 4 did not complete a baseline questionnaire (a validated Spanish translation of the SF-36 was lacking at the time this study was started) and because 20 subjects exited early and did not complete follow-up questionnaires; 18 questionnaires were excluded because of inconsistent responses, 9 at baseline and 9 at follow-up, as calculated by the "response consistency index" developed by the Health Institute, New England Medical Center, Boston, Mass.†HAQ indicates Health Assessment Questionnaire; PET, Problem Elicitation Technique questionnaire; and SF-36, Medical Outcomes Study 36-Item Short-Form Health Survey.‡Leflunomide or methotrexate vs placebo, and leflunomide vs methotrexate, P≤.001.§Leflunomide vs methotrexate, P≤.01.∥Methotrexate vs placebo, P<.05.A geographic analysis of site-to-site variations in responses revealed no significant interactions, and no site enrolled more than 5% of the total patient population. Subanalyses of responses according to disease duration, prior DMARD therapy, and concomitant corticosteroid or NSAID administration indicated no significant differences, with the exception of a better ACR response rate in the placebo population with a disease duration of more than 2 years compared with a disease duration of 2 years or less.COMPARISON OF LEFLUNOMIDE WITH METHOTREXATE THERAPYThe ACR success rates in the leflunomide and methotrexate treatment groups (41% and 35%, respectively) were statistically equivalent. Responses from patients receiving methotrexate treatment were significantly better than those for patients receiving placebo. The ACR greater than or equal to 20% response rates over time for patients receiving leflunomide and methotrexate therapy were 52% and 46%, respectively. Onset of effect occurred at a mean of 8.6 weeks for patients in the leflunomide treatment group compared with 9.5 weeks for those in the methotrexate treatment group.Because increases in the dosage of methotrexate therapy from 7.5 to 15 mg/wk occurred only for patients who had no response to treatment, the ACR success rates did not differ significantly for patients receiving 15 mg/wk vs those whose dosage remained at 7.5 mg/wk (34% and 37%, respectively). This was also true for the leflunomide and placebo groups; the ACR success rates in patients whose dosage of methotrexate placebo was increased were 40% for the leflunomide group and 18% in the placebo group, compared with 42% and 20%, respectively, for those whose dosage was not increased.Significantly less disease progression by x-ray analysis occurred with methotrexate treatment than with placebo. Results were better for the leflunomide group than for the methotrexate group (P=.05). Patients receiving leflunomide therapy reported more improvement than those receiving methotrexate therapy as assessed by the HAQ disability index, weighted top-5 score of the PET questionnaire, 5 scales of the HAQ, and 2 of 8 subscores of the SF-36 (Table 7).SAFETYSerious adverse events that were considered to be treatment-related by the investigator were reported for 2 patients receiving leflunomide therapy, 2 patients receiving placebo, and 5 patients receiving methotrexate therapy. These included asymptomatic transaminase elevations not requiring hospitalization (leflunomide therapy, n=1; placebo, n=1; and methotrexate therapy, n=2); nonfatal sepsis (leflunomide therapy, n=1; and placebo, n=1); and 1 death caused by sepsis, 1 case of interstitial pneumonitis, and 1 case of pneumonia in patients who were treated with methotrexate therapy. Withdrawals caused by adverse events occurred more frequently in the leflunomide group than in the placebo or methotrexate groups (Table 4). The higher incidence of withdrawals for patients receiving leflunomide therapy was primarily owing to gastrointestinal complaints and protocol-mandated withdrawals caused by asymptomatic transaminase elevations.Gastrointestinal complaints, specifically diarrhea, were more commonly reported by patients receiving leflunomide treatment (Table 8). Gastroenteritis and oral ulcers were more frequently reported by patients receiving methotrexate treatment. Mild to moderate allergic reactions, predominantly rash and pruritus, were more common in patients receiving leflunomide treatment than in patients receiving methotrexate treatment or placebo; there were no cases of anaphylaxis or angioedema. Infections, predominantly upper respiratory infections, bronchitis, and pneumonia, were most common in patients receiving methotrexate treatment and least common in patients receiving placebo; this was caused in part by the longer protocol exposure times in the active treatment groups. One case of sepsis occurred in each treatment group. No opportunistic infection or case of disseminated herpes was reported. The most frequent cardiovascular adverse event was mild to moderate hypertension, which responded readily to therapy. Hypertension was overrepresented in the leflunomide group at baseline (8.8%, compared with 5.1% for the placebo group and 1.1% for the methotrexate group). New-onset hypertension occurred in 2.1% of patients receiving leflunomide therapy compared with 0% of patients receiving placebo and 1.6% of patients receiving methotrexate therapy, all of whom were receiving concomitant NSAIDs. Excluding those patients with preexisting hypertension, mean increases in systolic and diastolic blood pressure were mild and similar across treatment groups (leflunomide group, 2.2 and 1.9 mm Hg, respectively; placebo group, 5.0 and 1.2 mm Hg; and methotrexate group, 1.9 and 1.3 mm Hg). Reversible alopecia was reported more frequently with leflunomide treatment than with methotrexate treatment; it caused 1 patient in the leflunomide group, 1 patient in the placebo group, and 2 patients in the methotrexate group to withdraw. Reversible renal failure occurred in 1 patient 29 weeks after beginning methotrexate treatment.Table 8. Summary of Reported Adverse EventsAdverse Event% of Patients (95% Confidence Interval)Leflunomide (n=182)Placebo (n=118)Methotrexate (n=182)Treatment-related serious adverse events1.1 (−13 to 16)1.7 (−16 to 20)2.7 (−12 to 17)Withdrawals because of adverse events22.0 (9 to 35)8.5 (−9 to 26)10.4 (−3 to 24)Total gastrointestinal60.4 (51 to 70)41.5 (28 to 55)51.6 (42 to 62)Diarrhea33.5 (22 to 45)16.9 (1 to 33)19.8 (7 to 33)Nausea/vomiting20.9 (8 to 34)18.6 (2 to 35)19.2 (6 to 32)Abdominal pain13.7 (0 to 27)6.8 (−11 to 24)15.4 (2 to 29)Dyspepsia13.7 (0 to 27)11.9 (−5 to 19)13.2 (0 to 27)Gastroenteritis2.2 (−12 to 16)1.7 (−16 to 20)5.5 (−9 to 20)Oral ulcers6.0 (−8 to 20)5.9 (−12 to 23)9.9 (−4 to 24)Allergic reactions24.2 (12 to 37)14.4 (−2 to 31)17.0 (4 to 30)Infections56.6 (47 to 66)48.3 (35 to 61)59.9 (51 to 69)Hypertension*11.0 (−3 to 25)5.1 (−13 to 23)2.7 (−12 to 17)New-onset hypertension†2.1 (−12 to 16)01.6 (−13 to 16)Alopecia9.9 (−4 to 24)0.8 (−17 to 18)6.0 (−8 to 20)*Hypertension reported as an adverse event (systolic blood pressure [BP] ≥160 mm Hg and/or diastolic BP ≥90 mm Hg 2 or more times during the treatment phase).†Patients without either a diagnosis of hypertension at baseline or systolic BP of 160 mm Hg or greater and/or diastolic BP of 90 mm Hg or greater at screening or baseline, prior to treatment.There were no adverse effects on hematologic parameters caused by leflunomide administration. One case of thrombocytopenia (platelet count <100 × 109/L) occurred in the methotrexate group. Treatment with leflunomide resulted in elevated ALT and/or AST levels that were greater than twice the ULN in 11.0% of patients, compared with 3.4% in the placebo group and 9.3% in the methotrexate group. In both active treatment groups, ALT levels were affected more frequently than AST levels (Table 9). All elevations for patients receiving leflunomide treatment (n=20) reverted to less than twice the ULN while treatment continued (n=10) or after treatment discontinuation (n=10). Mild elevations in alkaline phosphatase levels were infrequent; those elevations that were greater than twice the ULN were not related to treatment but were associated with concurrent illness. Liver biopsies were performed in accordance with ACR guidelines for 2 patients, 1 receiving leflunomide therapy (after 102 weeks) and 1 receiving methotrexate therapy (after 54 weeks); the biopsy specimens showed no evidence of fibrosis.Table 9. Summary of Transaminase Elevations*% of PatientsLeflunomide (n=182)Placebo (n=118)Methotrexate (n=182)Reported as an adverse event14.82.511.5Withdrawals7.11.74.4ALT level>2 times ULN and ≤3 times ULN6.606.6>3 times ULN4.42.52.7Reversible to ≤2 times ULN11.02.59.3AST level>2 times ULN and ≤3 times ULN6.01.76.0>3 times ULN2.21.70.5Reversible to ≤2 times ULN8.23.46.5*ALT indicates alanine aminotransferase; AST, aspartate aminotransferase; and ULN, upper limit of normal.COMMENTThis study demonstrates that leflunomide treatment results in sustained improvement in the signs and symptoms of RA over a 12-month period: 41% of patients treated with leflunomide fulfilled the ACR success definition and 52% reported improvement of 20% or greater in accordance with the ACR response criteria—results that were statistically equivalent to those for patients receiving methotrexate treatment (35% and 46%, respectively) and superior to those for patients receiving placebo (19% and 26%, respectively). Mean changes over time in all components of the ACR response index were significantly better for patients in the active treatment groups than for patients receiving placebo (P≤.01) and supported the composite ACR response analysis. The measures of the signs and symptoms of RA were not statistically better when analyses were adjusted for multiple testing.The ACR success definition may underestimate the true effect of active treatment, since subjects who discontinued treatment prior to week 52 were then considered treatment failures even if they met the ACR response criteria. In contrast, an intent-to-treat last value carried forward analysis for improvement according to the ACR response criteria may overestimate the treatment effect because subjects can be considered responders even if they discontinue treatment because of an adverse effect. The trueeffect of active treatment is therefore intermediate between the ACR success and response rates.The time to initial response was shorter with leflunomide treatment than with placebo or methotrexate treatment. Sustained responses occurred earlier and were of longer duration in the leflunomide group than in the placebo or methotrexate groups. An AUC analysis of the time during which an ACR response occurs offers a better estimate of clinical effect than a single time point. The mean AUC for ACR response was statistically superior for patients receiving leflunomide treatment compared with patients receiving placebo, confirming the robust clinical effect of leflunomide therapy. This response over time may best explain why leflunomide treatment resulted in significant improvement in physical function, prevention or decrease in disability, and improvement in health-related quality of life.The primary outcome criterion for this clinical trial was improvement of 20% or greater according to the ACR response criteria, and general acceptance of this definition of improvement by rheumatologist investigators and regulatory authorities has been reaffirmed since the design and implementation of this trial. Additionally, as shown in Table 4, the ACR response rates for improvement of 50% or greater and 70% or greater compared favorably with recently published data regarding new biologic agents.After 12 months, the leflunomide group reported an ACR response rate for improvement of 70% or greater of 20%, similar to response rates at 3 and 6 months among patients receiving 16 or 25 mg/m2of recombinant tumor necrosis factor receptor (p75)–Fc fusion protein (20% and 15%, respectively) or soluble tumor necrosis factor receptor (p55)–IgG protein (13%). Among patients receiving methotrexate therapy (7.5-15.0 mg/wk), the ACR response rate for improvement of 70% or greater was 9.4%.Leflunomide treatment also retarded disease progression as measured by x-ray analysis. To our knowledge, this is the first 12-month placebo-controlled trial that documents a similar disease modification effect for methotrexate therapy compared with placebo.The improvements observed in health-related quality-of-life measures are clinically meaningful. A decrease in the HAQ disability index of 0.22 is considered a minimum clinically important difference, one that is apparent to patients.This study showed decreases of 0.45 in the leflunomide therapy group and 0.26 in the methotrexate therapy group. Minimum clinically important differences have yet to be defined for the PET questionnaire and SF-36.In previous clinical trials of shorter duration (18 weeks to 9 months), retrospectively applied ACR response rates following methotrexate therapy of 40.3% and 64.7% have been reported compared with 8.4% for patients receiving placebo and 28.8% for patients receiving auranofin treatment, respectively.Recently, an ACR success rate of 39% for methotrexate therapy compared with 12% for patients receiving placebo was reported in a 6-month trial that also examined cyclosporine treatment.Several factors may account for these observed differences in response rates following methotrexate treatment: active vs placebo comparators, dosage of methotrexate administered (7.5 vs >7.5 mg/wk), and concomitant administration of folate, despite reports to the contrary.Methotrexate has become the most widely used DMARD; a recent study demonstrated that 38% of patients with RA who were receiving second-line agents received methotrexate therapy.Leflunomide has a different mechanism of action than methotrexate. As a result, although adverse events commonly reported for leflunomide treatment are generally similar to those reported for methotrexate treatment administered with folate, the incidence rates of specific toxic effects differ; therefore, it is possible that patients who are unable to tolerate methotrexate therapy will be able to tolerate leflunomide therapy and vice versa. Additionally, preliminary data suggest that combination therapy with leflunomide plus methotrexate may offer benefit to patients with long-standing RA, with acceptable tolerability and no pharmacokinetic interactions.The principal adverse effects of leflunomide therapy include diarrhea, mild to moderate allergic reactions, reversible alopecia, and transaminase elevations. In this study, patients were commonly taking NSAIDs, which may have contributed to some of the gastrointestinal complaints. Oral ulcers, associated with methotrexate treatment, were less common with leflunomide treatment. As with methotrexate therapy, leflunomide therapy requires monitoring of ALT levels so that dosage adjustments can be made if necessary. Dosage reduction is recommended for repeated elevations that are greater than twice the ULN. Although data are limited, there is no evidence to suggest that leflunomide treatment is associated with the development of clinically significant liver disease. No cases of interstitial pneumonitis or renal dysfunction have been reported in patients treated with leflunomide.As with methotrexate therapy, patients receiving leflunomide therapy must be cautioned against becoming pregnant or fathering children. Despite the long serum half-life of leflunomide, administration of cholestyramine will rapidly reduce blood levels of leflunomide in the event of an adverse event or in preparation for pregnancy.In summary, patients receiving leflunomide therapy derived an important benefit that was demonstrated by improvement in the clinical signs and symptoms of RA; this was first evident at 1 month and was maintained over a 1-year period, with significant retardation of disease progression confirmed by x-ray analysis and improvement in the performance of physical activities important to patients and in health-related quality of life. The results of this trial suggest that leflunomide therapy is as effective as methotrexate therapy and represents an important addition to the treatment armamentarium for patients with active RA.PAAndersenSGWestJRO'DellCSViaRGClaypoolBLKotzinWeekly pulse methotrexate in rheumatoid arthritis.Ann Intern Med.1985;103:489-496.DEFurstRKoehnkeLFBurmeisterJKohlerICargillIncreasing methotrexate effect with increasing dose in the treatment of resistant rheumatoid arthritis.J Rheumatol.1989;16:313-320.MEWeinblattJSCoblynDAFoxEfficacy of low-dose methotrexate in rheumatoid arthritis.N Engl J Med.1985;312:818-822.HJWilliamsRFWillkensCOSamuelson JrComparison of low-dose oral pulse methotrexate and placebo in the treatment of rheumatoid arthritis.Arthritis Rheum.1985;38:721-730.EHSChoyDLScottDrug treatment of rheumatic diseases in the 1990s: achievements and future developments.Drugs.1997;53:337-348.HTSilvaREMorrisLeflunomide and the malanonitriloamides.Am J Med Sci.1997;313:289-301.JPDavisGACainWJPittsRLMagoldaRACopelandThe immunosuppressive metabolite of leflunomide is a potent inhibitor of human dihydroorotate dehydrogenase.Biochemistry.1996;35:1270-1273.MHerrmannCGGrangouBKirschbaumCell cycle control of the de novo pyrimidine synthesis inhibitor leflunomidelunomide through the p53 and p21 pathways [abstract].Arthritis Rheum.1997;40:S177.HTSilvaWCaoRShorhouseREMorrisMechanism of action of leflunomide: in vivo uridine administration reverses its inhibition of lymphocyte proliferation.Transplant Proc.1996;28:3082-3084.VMladenovicZDomljanBRozmanSafety and effectiveness of leflunomide in the treatment of patients with active rheumatoid arthritis.Arthritis Rheum.1995;38:1595-1603.FCArnettSMEdworthyDABlochThe American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis.Arthritis Rheum.1988;31:315-324.TPincusJASummeySASorraciAssessment of patient satisfaction in activities of daily living using a modified Stanford Health Assessment Questionnaire.Arthritis Rheum.1983;26:1346-1353.JFFriesPWSpitzDYYoungThe dimensions of health outcomes: the health assessment questionnaire, disability and pain scales.J Rheumatol.1982;9:789-793.CBombardierRBuchbinderPTugwellEfficacy of cyclosporin A in rheumatoid arthritis: long-term follow-up data and the effect on quality of life.Scand J Rheumatol Suppl.1992;95:29-33.JEWareCDSherbourneThe MOS 36-item short-form health survey (SF-36), I: conceptual framework and item selection.Med Care.1992;30:473-483.GBBluhmJTSharpBCTilleyRadiographic results from the minocycline in rheumatoid arthritis (MIRA) trial.J Rheumatol.1997;24:1295-1302.DTFelsonJJAndersonMBoersThe American College of Rheumatology preliminary definition of improvement in rheumatoid arthritis clinical trials.Arthritis Rheum.1995;38:725-735.JMKremerGAAlarconRWLightfoot JrMethotrexate for rheumatoid arthritis, suggested guidelines for monitoring liver toxicity.Arthritis Rheum.1994;37:316-328.Food and Drug AdministrationGuidance for Industry: Clinical Development Programs for Drugs, Devices, and Biological Products for the Treatment of Rheumatoid Arthritis (RA).Final guidance issued February 1999. Available at: http://www.fda.gov/cder/guidance/1208dft.pdf. Accessed August 18, 1999.LWMorelandSWBaumgartnerMHSchiffTreatment of rheumatoid arthritis with a recombinant tumor necrosis factor receptor (p75)–Fc fusion protein.N Engl J Med.1997;337:141-147.LWMorelandMHSchiffSWBaumgartnerEtanercept therapy in rheumatoid arthritis.Ann Intern Med.1999;130:478-486.JMcKayRRauMWeismanfor the Lenercept 362 Study GroupA double-blind, randomized, 6 arm, parallel-group, dose finding, double dummy, multi-center comparison of sTNFr55-IgG (Ro45-2081-lenercept) subcutaneous to reference treatment with oral methotrexate (MRX) and their combination in patients with rheumatoid arthritis [abstract].Arthritis Rheum.1998;41:S132.CGoldsmithMBoersCBombardierPTugwellfor the OMERACT CommitteeCriteria for clinically important changes in outcomes: development, scoring and evaluation of rheumatoid arthritis patients and trial profiles.J Rheumatol.1993;20:561-565.GWellsPTugwellGKraagPBakerJGrohDRedelmeierMinimum important difference between patients with rheumatoid arthritis: the patient's perspective.J Rheumatol.1993;20:557-560.JGuzmanAMaetzelPPelosoMYeungCBombardierDisability scores in DMARD trials: what is a clinically important change?Arthritis Rheum.1996;39:S208.DTFelsonJJAndersonMLMLangeGWellsMPLaValleyShould improvement in rheumatoid arthritis clinical trials be defined as 50% or 70% improvement in core set measures, rather than 20%?Arthritis Rheum.1998;41:1564-1570.SCohenJRutsteinMLuggenComparison of the safety and efficacy of cyclosporine A and methotrexate in refractory rheumatoid arthritis.Arthritis Rheum.1993;36:S56.Not AvailableNeoral.In: Physicians' Desk Reference.Montvale, NJ: Medical Economics Co Inc; 1998:1182-1184.AOrtizBSheaMSuarez-AlmazorDMoherGWellsPTugwellThe efficacy of folic acid and folinic acid in reducing methotrexate gastrointestinal toxicity in rheumatoid arthritis: a meta-analysis of randomized controlled trials.J Rheumatol.1997;25:36-42.MMWardJFFriesTrends in antirheumatic medication use among patients with rheumatoid arthritis, 1981-1996.J Rheumatol.1998;25:408-416.MEWeinblattJMKremerJSCoblynPharmacokinetics, safety, and efficacy of combination treatment with methotrexate and leflunomide in patients with active rheumatoid arthritis.Arthritis Rheum.1999;42:1322-1328.Accepted for publication May 27, 1999.This study was funded by Hoechst Marion Roussel (HMR), Bridgewater, NJ. Dr Strand serves as a consultant to HMR in clinical and regulatory affairs. Drs Cohen, Schiff, Weaver, Fox, Moreland, Olsen, and Furst were investigators for and have served as consultants to HMR. Dr Hurley serves as a statistical consultant to HMR and is an employee of Quintiles Transnational Corp, Arlington, Va. Dr Sharp serves as a consultant to HMR in x-ray analysis. Dr Loew-Friedrich is an employee of HMR.Reprint requests: Iris Loew-Friedrich, MD, Hoechst Marion Roussel, Route 202-206, PO Box 6800, Bridgewater, NJ 08807-0800.Leflunomide Rheumatoid Arthritis Investigators GroupStanley Cohen, MD, and Roy Fleischmann, MD, Metroplex Clinical Research Center, Dallas, Tex; Michael Schiff, MD, Denver Arthritis Clinic, Denver, Colo; Arthur Weaver, MD, Arthritis Center of Nebraska, Lincoln; Grant Cannon, MD, Veterans Affairs Medical Center and University of Utah, Salt Lake City; Robert Fox, MD, Scripps Clinic, La Jolla, Calif; Larry Moreland, MD, University of Alabama at Birmingham, Birmingham; Nancy Olsen, MD, Vanderbilt University, Nashville, Tenn; Dan Furst, MD, Virginia Mason Arthritis Research Center, Seattle, Wash; Jacques Caldwell, MD, Halifax Clinical Research Center, Daytona Beach, Fla; Jeffrey Kaine, MD, Sarasota Arthritis Center, Sarasota, Fla; Elizabeth Tindall, MD, Portland, Ore; Howard Offenberg, MD, Arthritis and Allergy Institute, Gainesville, Fla; Jeffrey Poiley, MD, Arthritis Associates, Orlando, Fla; Joel Rutstein, MD, Arthritis Center, San Antonio, Tex; Frederick Dietz, MD, Internal Medicine Rheumatology, Rockford Clinic, Rockford, Ill; Alan Brodsky, MD, Dallas; Robert Harris, MD, Whittier, Calif; Mitchell Lowenstein, MD, Arthritis Center, Palm Harbor, Fla; Andrew Baldessare, MD, Arthritis Consultants, St Louis, Mo; Paul Howard, MD, Paradise Valley, Ariz; Marshall Sack, MD, Center For Clinical Research, Austin, Tex; John Tesser, MD, Phoenix Center for Clinical Research, Phoenix, Ariz; Nathan Wei, MD, Arthritis and Osteoporosis Center of Maryland, Frederick; Robin Dore, MD, Anaheim, Calif; Robert Ettlinger, MD, Tacoma, Wash; Larry Anderson, MD, Rheumatology Associates, Portland, Me; Barry Bockow, MD, Seattle; Michael Liebling, MD, Division of Rheumatology, Harbor–University of California, Los Angeles, Research and Education Institute, Torrance; Paul Romain, MD, Department of Clinical Research, Lahey Hitchcock Medical Center, Burlington, Mass; Scott Baumgartner, MD, Spokane, Wash; Joel Silverfield, MD, Tampa, Fla; Andrew Chubick, MD, Arthritis Centers of Texas, Dallas; Charles Franklin, MD, Rheumatic Disease Associates, Willow Grove, Pa; Selwyn Cohen, MD, Trumbull, Conn; Gordon Senter, MD, Salisbury, NC; Richard Furie, MD, Division of Rheumatology, Northshore University Hospital, Manhasset, NY; Sanford Hartman, MD, Decatur, Ga; Robert Levy, MD, Olympia Arthritis Clinic, Olympia, Wash; David Yocum, MD, Rheumatology and Immunology, University of Arizona, Health Science Center, Tucson; Don Cheatum, MD, Texas Medical and Surgical, Dallas; Sicy Lee, MD, Hospital for Joint Diseases, New York, NY; Matthew Heller, MD, Arthritis Associates Inc, Peabody, Mass.Corresponding author: Vibeke Strand, MD, Division of Immunology, Stanford University, Palo Alto, CA 94304 (e-mail: [email protected]).
BackgroundAdverse drug events (ADEs) are common in hospitalized patients, but few empirical data are available regarding the strength of patient risk factors for ADEs.MethodsWe performed a nested case-control study within a cohort that included 4108 admissions to a stratified random sample of 11 medical and surgical units in 2 tertiary care hospitals during a 6-month period. Analyses were conducted on 2 levels: (1) using a limited set of variables available for all patients using computerized data available from 1 hospital and (2) using a larger set of variables for the case patients and matched controls from both hospitals. Case patients were patients with an ADE, and the matched control for each case patient was the patient on the same unit as the case patient with the most similar preevent length of stay. Main outcome measures were presence of an ADE, preventable ADE, or severe ADE.ResultsIn the cohort analysis, electrolyte concentrates (odds ratio [OR], 1.7), diuretics (OR, 1.7), and medical admission (OR, 1.6) were independent correlates of ADEs. Independent correlates of preventable ADEs in the cohort analysis were low platelet count (OR, 4.5), antidepressants (OR, 3.3), antihypertensive agents (OR, 2.9), medical admission (OR, 2.2), and electrolyte concentrates (OR, 2.1). In the case-control analysis, exposure to psychoactive drugs (OR, 2.1) was an independent correlate of an ADE, and use of cardiovascular drugs (OR, 2.4) was independently correlated with severe ADEs. For preventable ADEs, no independent predictors were retained after multivariate analysis.ConclusionsAdverse drug events occurred more frequently in sicker patients who stayed in the hospital longer. However, after controlling for level of care and preevent length of stay, few risk factors emerged. These results suggest that, rather than targeting ADE-prone individuals, prevention strategies should focus on improving medication systems.ADVERSE DRUG events (ADEs) are an important and costly problem.The Medical Practice Study, which examined patients hospitalized in New York State in 1984, found that almost 1% of patients experienced an injury due to drugs that resulted in disability or prolongation of their hospital stay.Other studies,most of which used the adverse drug reaction as the outcome, have also shown that injuries due to drugs are common among hospitalized patients. The term "adverse drug reaction," as defined by the World Health Organization,excludes events associated with errors, while the term "ADE" includes preventable and nonpreventable events.In earlier reportsfrom the ADE Prevention Study, we found an average of 6.5 ADEs per 100 admissions, of which 28% were preventable.One potentially attractive strategy for preventing ADEs is to identify prospectively those patients at high risk of an ADE and to target additional resources toward this group. An example of this approach might be that when a patient is determined to be high risk, the patient would be identified so that the pharmacy could pay extra attention to all medications given. Several patient attributes that may make an ADE more likely have been suggested by various investigators.However, little empirical data exist on their predictive ability.Some risk factors for adverse reactions to drugs that have been proposed to date include age,number of drugs the patient is receiving,and factors that alter drug distribution or metabolism, such as renal or hepatic insufficiency, congestive heart failure, anemia, and alcoholism.It has also been suggested that a patient who is receiving specific drugs or drugs of a certain class may be prone to having an ADE; in fact, substantial work has been done in evaluating which drugs are most often associated with ADEs or adverse drug reactions.When Karch and Lasagnacombined several studies, they found that antibiotics caused 42% of adverse drug reactions, but no other group of drugs was responsible for more than 10%. In the Medical Practice Study, antibiotics were also the most common class of drugs associated with drug-related injuries, despite accounting for only 15% of ADEs.Several models have been developed for predicting events in patients receiving specific drugs, for example, Landefeldindexes for predicting bleeding in patients beginning anticoagulation therapy. However, to our knowledge, no empirical data exist that allow stratification of hospitalized patients according to likelihood of an ADE across drugs.Our aims in this study were (1) to develop a practical, efficient method of identifying patients at increased risk of an ADE based solely on information readily available on the hospital's computer database, either at admission or up to the time of the event; and (2) to evaluate a larger spectrum of potential patient risk factors by comparing case patients who experienced an ADE with a matched control group. In the first portion of the study, electronic access to information was available at only 1 of the 2 hospitals we studied, so the other hospital was excluded from this portion of the study. Using this set of limited, yet online information, we were able to compare patients who experienced an ADE against the entire population of patients who were admitted to the surveyed units during the study period. For the second portion of the study, we collected highly detailed data for each patient at 2 hospitals through extensive daily medical record review and verbal inquiry. We were then able to compare this detailed, time-intensive form of data collection against the larger cohort, about whom less information was available, to weigh the relative efficacy of each approach.PATIENTS AND METHODSPATIENT POPULATIONSPatients surveyed included all adults at 2 large tertiary care hospitals, Brigham and Women's Hospital (726 beds) and Massachusetts General Hospital (846 beds), both in Boston, Mass, admitted to any of 11 units during a 6-month period between February and July 1993, as previously described.Study units were a stratified random sample of medical, surgical, and intensive general care units. These units included 5 intensive care units (ICUs) (3 surgical and 2 medical) and 6 general care units (4 medical and 2 surgical). To identify more events, we intentionally oversampled ICUs and excluded obstetric units, because we previously found that ADEs were more common in ICUs than in general care units and that obstetric units had almost no ADEs.DEFINITIONS AND CLASSIFICATIONS OF EVENTSThe primary outcome of the study was the ADE, which we defined as an injury resulting from medical intervention related to a drug.For example, if a patient with first-degree atrioventricular block were given a β-blocker and then developed complete heart block requiring temporary pacing, this reaction would be an ADE. The secondary outcomes we studied were preventable ADEs and severe ADEs. For example, the ADE would have been recorded as a preventable ADE if the patient in the example who developed atrioventricular block had already been taking a calcium channel blocker, which depressed the atrioventricular node before receiving the β-blocker. An ADE was considered a severe ADE if the consequences were serious, life threatening, or fatal. In prior reports from our group,we also presented data on potential ADEs, which were defined as incidents in which an error was made but no actual harm occurred to the patient; however, we did not attempt to assess risk for potential ADEs in this report.Two physician reviewers evaluated all incidents independently and classified them according to whether an ADE or potential ADE was present, whether it was preventable, and how severe it was.When there were disagreements that affected classifications about the presence of an event, its severity, or its preventability (eg, 1 reviewer scored it as preventable, but the other did not), the reviewers met and reached consensus. If consensus could not be reached, a third reviewer evaluated the incident and made the final decision. Kappa statistics (κ) were calculated and were found to be 0.81 to 0.98 for whether an ADE was present and 0.92 for whether the ADE was preventable; κ statistics were lower, at 0.32 to 0.37, for assessments of severity.DATA COLLECTION STRATEGIES AND DATA ELEMENTSWe collected 2 sets of data: (1) detailed information on the case patients and controls at both hospitals and (2) a limited amount of data available electronically on all patients in the study cohort at 1 of the 2 hospitals. Case patients were defined as all patients with an ADE, and the 2 subcategories of ADEs were preventable ADEs and severe ADEs. Controls were patients on the same unit as the case patient with the most similar preevent length of stay. Thus, controls had the same level of care as the case patients (ICU vs non-ICU), and almost all were on the same service (medicine vs surgery).Three mechanisms were used for identifying incidents. First, nurses and pharmacists were asked to report incidents to the nurse investigators. Second, a nurse investigator visited each unit at least twice daily on weekdays and solicited information from nurses, pharmacists, and clerical personnel concerning all actual or potential drug-related incidents. Third, a nurse investigator reviewed all medical records at least daily on weekdays.In the cohort analysis, we compared patients experiencing an ADE in 1 hospital with all patients admitted to any of the surveyed units at that hospital during the study period. Data available online included demographic and administrative information, such as age, sex, race, admission, and subsequent level of care (eg, whether the patient was in an ICU at admission or on the date of the event); admission source (eg, home or nursing home); primary insurer; and diagnosis related group (DRG) weight. In addition, we were able to obtain some medication data, including the number and names of drugs a patient was receiving at admission. Drugs were classified into American Hospital Formulary System (AHFS) categories, including cardiovascular, antiasthmatic, sedative or hypnotic, antisecretory, antitumor, antidepressants, analgesics, diuretics, antibiotics, anticoagulants, antiarrhythmics, muscle relaxants, antipsychotics, electrolyte concentrates such as potassium, and antihypertensives. Laboratory data were also obtained, including serum total bilirubin level as a proxy for hepatic insufficiency, serum urea nitrogen (SUN) and creatinine levels to identify patients with renal failure, serum albumin level, and platelet count. A SUN level of 5.7 mmol/L (16 mg/dL) was used as a cutoff to separate patients with a low SUN level from those with slightly elevated to high SUN levels. Similarly, a platelet count of 50 × 109/L was used as a cutoff to identify patients with a low platelet count. A serum creatinine level of 133 µmol/L (1.5 mg/dL) was chosen as a cutoff point to separate patients with a low serum creatinine level from those with elevated levels, and a serum bilirubin level of 34 µmol/L (2.0 mg/dL) was designated as the cutoff for patients with low vs elevated bilirubin levels. Serum albumin was stratified into 4 levels: less than 20 g/L, 20 to 24 g/L, 25 to 36 g/L, and more than 36 g/L.In the case-control study, the reviewer was blinded to case status for all risk data obtained from medical record review. We accomplished this by having 1 reviewer identify the case patient and control and then refer the case patients to a second reviewer for data collection but not reveal which individual was a case patient. Data collected included all the information available electronically for the cohort analysis as described earlier and, in addition, clinical data, including severity of illness at the time of the event (measured using the Therapeutic Intervention Scoring System score), comorbidity (measured using the Charlson comorbidity index), and extensive drug exposure information, including the number and names of drugs that case patients and controls were receiving at the time of the event. Drugs were classified into AHFS categories, as they were in the cohort study.ANALYSISComparisons between categorical variables were made using the χ2test as well as a 2-sided trend test for comparisons in which a variable had multiple ordered categories. Comparisons between normally distributed variables were made using the ttest, and nonparametric comparisons were made using the Wilcoxon rank sum test. In the multivariate analyses of the entire cohort, stepwise logistic regression was used, while in the case-control multivariate analyses, comparisons were made using conditional logistic regression to retain the advantages of the paired nature of the data. Analyses were performed using SAS statistical software (SAS Institute Inc, Cary, NC)except for trend tests, which were performed using StatXact software (CYTEL Software Corp, Cambridge, Mass).RESULTSCOHORT STUDYIn this analysis, there were 139 ADEs and 42 preventable ADEs during the study period, among 2379 total admissions. Among the admissions, 360 lacked initial drug exposure information; while these admissions were retained in the demographic and clinical studies, they were excluded from the drug exposure analyses. We, therefore, had a final number of 113 ADEs and 32 preventable ADEs among 2019 admissions for this portion of the study.Univariate analyses of demographics and clinical characteristics in the cohort study revealed 4 significant correlates of an ADE (Table 1): hospital service, SUN level of 6.1 mmol/L (17 mg/dL) or greater, platelet count below 50 × 109/L, and albumin category.Table 1. Demographics and Clinical Characteristics of Patients With ADEs and Preventable ADEs in the Cohort Analysis*VariablePatients With ADEs (n = 139)Cohort (n = 2240)PPatients With Preventable ADEs (n = 42)Cohort (n = 2337)PAge, mean (± SD), y54.3 (19.6)51.8 (18.7).1457.9 (17.6)51.8 (18.8).03Hospital serviceMedical82 (59.0)1014 (45.3)<.0129 (69.0)1079 (46.2)<.01Surgical57 (41.0)1226 (54.7)13 (31.0)1258 (53.8)Male60 (43.2)1102 (49.2).1719 (45.2)1149 (49.2).45Nonwhite race33 (23.7)557 (24.9).7713 (31.0)579 (24.8).46SUN level >5.7 mmol/L (>16 mg/dL)69 (49.6)832 (37.1).0223 (54.8)877 (37.5).02Platelet count <50 × 109/L8 (5.8)57 (2.5).025 (11.9)60 (2.6)<.01Albumin category, g/L<203 (2.2)17 (0.8).02†1 (2.4)19 (0.8).02†20-247 (5.0)61 (2.7)4 (9.5)64 (2.7)25-3647 (33.8)598 (26.7)14 (33.3)631 (27.0)>3682 (59.0)1564 (69.8)23 (54.8)1623 (69.5)*Data are given as number (percentage) of each group, unless otherwise indicated. ADE indicates adverse drug event; SUN, serum urea nitrogen.†χ2test for trend.For preventable ADEs, a comparison of the demographics and clinical characteristics between case patients and the cohort revealed 5 significant correlates (Table 1). Patients experiencing a preventable ADE were older than the rest of the patient population. In addition, presence of a platelet count below 50 × 109/L, a SUN level of 6.1 mmol/L (17 mg/dL) or greater, and higher albumin levels were all correlated with the presence of an ADE (Table 1). There was also a trend toward finding more preventable ADEs on the medicine service.In the drug exposure analysis of the cohort group, we found 7 significant correlates of presence of an ADE (Table 2). Individual drug types associated with higher rates of ADEs (Table 2) include diuretics, electrolyte concentrates, antitumor agents, anticoagulants, and ulcer medications. Two summary categories were also correlated with ADE frequency: the average number of different drugs received (7.3 for case patients vs 6.0 for controls; P= .02) and the average number of AHFS classes of drugs received (3.8 for case patients vs 3.3 for controls; P= .02).Table 2. Univariate Analyses of Drug Exposure Levels and Patients With ADEs and Preventable ADEs in the Cohort Analysis*Type of DrugPatients With ADEs (n = 113)Cohort (n = 1906)PPreventable ADEs (n = 32)Cohort (n = 1987)PDiuretic28 (24.8)252 (13.2)<.0110 (31.3)270 (13.6)<.01Electrolyte concentrate44 (38.9)423 (22.2)<.0117 (53.1)450 (22.6)<.01Antisecretory43 (38.1)559 (29.3).0518 (56.3)584 (29.4)<.01Antidepressant11 (9.7)112 (5.9).107 (21.9)116 (5.8)<.01Antihypertensive19 (16.8)223 (11.7).1010 (31.3)232 (11.7)<.01Cardiovascular47 (41.6)702 (36.8).3122 (68.8)727 (36.6)<.01Antiseizure4 (3.5)71 (3.7).923 (9.4)72 (3.6).09Antitumor14 (12.4)127 (6.7).024 (12.5)137 (6.9).22Anticoagulant35 (31.0)412 (21.6).0210 (31.3)437 (22.0).21*Data are given as number (percentage) of each group, unless otherwise indicated. ADE indicates adverse drug event.Univariate analysis of drug exposure information for preventable ADEs also revealed 7 significant correlates (Table 2). Four of these were also significant in our analysis of all ADEs: a higher frequency of preventable ADEs was associated with use of ulcer medications, use of antidepressants, use of electrolyte concentrates, and the average number of AHFS classes of drugs received (4.3 for case patients vs 3.3 for controls; P= .05). Also associated with preventable events was the use of cardiovascular medications, antihypertensive agents, and diuretics. There was a trend toward an association for the use of antiseizure medications as well.In multivariate analyses to look for significant correlates of ADEs and preventable ADEs in the cohort study, independent correlates of ADEs were use of electrolytes, use of diuretics, and whether the patient was admitted to a medical ward (Table 3). Multivariate correlates of preventable ADEs were platelet category, use of antidepressants, use of antihypertensive agents, whether the patient was admitted to a medical ward, and use of electrolyte concentrates (Table 3).Table 3. Independent Predictors of ADEs and Preventable ADEs in the Cohort Analysis*PredictorβOdds Ratio (95% Confidence Interval)All ADEsElectrolyte concentrate.511.7 (1.1-2.5)Diuretic.491.7 (1.0-2.6)Medical ward admittance.491.6 (1.1-2.3)Preventable ADEsPlatelet category1.514.5 (1.6-12.9)Antidepressant1.183.3 (1.3-7.9)Antihypertensive agent1.082.9 (1.4-6.4)Medical ward admittance.782.2 (1.1-4.4)Electrolyte concentrate.752.1 (1.1-4.1)*ADE indicates adverse drug event.CASE-CONTROL STUDYIn the case-control study, there were 247 ADEs, 70 preventable ADEs, and 106 severe ADEs among 4108 admissions during the study period. If a patient had more than 1 ADE during an admission, only the first episode in the admission was counted. Length-of-stay outliers were identified using Studentized residuals for length of stay.If an ADE had a residual of −2 or lower or 2 or greater, it was examined. For example, some patients were in the hospital for up to a year often awaiting nursing home placement; in such cases, it was judged that length of stay was not influenced by the occurrence of an ADE. When ADEs in which more than 1 event occurred per admission (n = 40) and length-of-stay outliers (n = 17) were excluded, these figures became 190 ADEs, 60 preventable ADEs, and 84 severe ADEs. Each case patient was paired with a control from the same unit with the most similar preevent length of stay.When we compared the demographics of the case patients and the controls, there were no significant differences (Table 4). However, when we compared the case patients with all patients admitted to the study units at either of the 2 hospitals, the case patients had a much longer length of stay. In addition, the case patients were much more likely to be admitted to an ICU. On comparing the clinical characteristics of all patients with an ADE with controls in univariate analyses (Table 5), we found no significant correlation between any of the following and presence of an ADE: age, mean comorbidity or severity scores, number of drugs received in the 24 hours before the incident, number of drugs received since admission, number of psychoactive drugs received since admission, presence of altered mental status, or presence of an elevated bilirubin or creatinine level. The same was true for preventable ADEs. For severe ADEs, case patients were older than controls, but no other factors were significantly correlated with presence of an ADE.Table 4. Demographic Characteristics of Case Patients, Controls, and the Entire Cohort in the Case-Control Analysis*CharacteristicCase Patients (n = 190)Controls (n = 190)P†Entire Cohort (n = 3848)P‡Age, yMean (± SD)55.8 (19.7)55.6 (18.4).9156.8 (18.8).50Categories≤5068 (35.8)74 (39.0).891421 (36.9).9751-6028 (14.7)28 (14.7)532 (13.8)61-7041 (21.6)41 (21.6)800 (20.8)>7053 (27.9)47 (24.7)1095 (28.5)Male92 (48.4)104 (54.7).221866 (48.5).99Nonwhite race42 (22.1)40 (21.1).80806 (21.0).70Uninsured or Medicaid insured71 (37.4)62 (32.6).331209 (31.4).09Hospital serviceMedical118 (62.1)121 (63.7).752312 (60.1).29Surgical72 (37.9)69 (36.3)1489 (38.7)Other0 (0.0)0 (0.0)47 (1.2)Admitted to an intensive care unit at any time during the hospital stay73 (38.4)70 (36.8).751071 (27.8).002Length of stay, median, d§12 (6, 24)12 (7, 25).696 (3, 12).001*Data are given as number (percentage) of each group, unless otherwise indicated.†Reflects χ2statistic for case patients vs controls.‡Reflects χ2statistic for case patients vs the entire cohort.§Data in parentheses are the 25th and the 75th percentiles.Table 5. Univariate Correlates of Clinical Characteristics for ADEs, Preventable ADEs, and Severe ADEs in the Case-Control Analysis*CharacteristicAll ADEs (n = 190)Controls (n = 190)PSevere ADEs (n = 84)Controls (n = 84)PPreventable ADEs (n = 60)Controls (n = 60)PAge, mean (± SD), y55.8 (19.7)55.6 (18.4).9162.5 (18.7)55.8 (18.8).0260.3 (17.6)54.9 (19.5).12No. of drugs received in the 24 h before the incident, mean (± SD)10.9 (5.3)9.9 (5.5).0712.2 (5.0)11.4 (5.5).3012.3 (5.1)11.6 (5.9).53No. of drugs received since admittance mean (± SD)18.5 (15.9)16.8 (12.0).2323.0 (20.5)20.6 (15.0).4021.4 (19.5)16.7 (15.1).58>15 Drugs received in the 24 h before the incident48 (25.3)38 (20.0).2229 (34.5)24 (28.6).4120 (33.3)15 (25.0).32>25 Drugs received since admittance41 (21.6)33 (17.4).3028 (33.3)25 (29.8).6217 (28.3)13 (21.7).40No. of psychoactive drugs received since admittance, mean (± SD)1.4 (0.8)1.2 (0.9).031.4 (0.8)1.4 (0.9).931.6 (0.8)1.6 (1.0).92Confused33 (17.4)27 (14.2).4320 (23.8)16 (19.1).6912 (20.0)10 (16.7).29Elevated creatinine level (≥133 µmol/L [≥1.5 mg/dL])43 (22.6)40 (21.1).7125 (29.8)20 (23.8).3819 (31.7)12 (20.0).14Elevated bilirubin level (≥34 µmol/L [≥2.0 mg/dL])7 (3.7)10 (5.3).466 (7.1)7 (8.3).777 (11.7)9 (15.0).59Comorbidity score, median, mean (± SD)†2.3 (2.5)2.5 (2.7).442.5 (2.3)2.4 (2.4).872.6 (2.3)2.5 (2.8).72Severity, mean (± SD)‡12.2 (12.7)11.2 (13.0).4416.5 (14.6)14.9 (15.0).4913.2 (12.1)14.0 (17.1).79*Data are given as number (percentage) of each group, unless otherwise indicated. ADE indicates adverse drug event.†Indicates the Charlson comorbidity score.‡Indicates the Therapeutic Intervention Scoring System.A comparison of drug exposure levels between case patients and controls (Table 6) revealed that exposure to 1 or more psychoactive drugs was the only class significantly associated with an ADE. This was the case even though most patients in both groups were exposed to these drugs. Dividing patients into 3 categories based on number of psychoactive drugs received (0, 1, or ≥2) resulted in a more specific high-risk group: 47% of the case patients fell into the group exposed to 2 or more psychoactive drugs vs 38% of the controls (P= .01). For severe ADEs, there was a significant association between cardiovascular drugs. Some examples of drugs included in this group are digoxin, atenolol hydrochloride, diltiazem, nifedipine, and propranolol hydrochloride. For preventable ADEs, there were no significant associations across drugs.Table 6. Correlates and Potential Correlates of ADEs, Preventable ADEs, and Severe ADEs in the Case-Control Analysis: Drug Exposure Data*Type of DrugAll ADEs (n = 190)Controls (n = 190)PSevere ADEs (n = 84)Controls (n = 84)PPreventable ADEs (n = 60)Controls (n = 60)PPsychoactive†162 (85.3)139 (73.2).0171 (84.5)68 (81.0).5454 (90.0)50 (83.3).28Cardiovascular81 (42.6)64 (33.7).0752 (61.9)34 (40.5).0133 (55.0)23 (38.3).07Antiasthmatic45 (23.7)36 (19.0).2622 (26.2)22 (26.2)>.9918 (30.0)15 (25.0).54Sedative or hypnotic91 (47.9)76 (40.0).1249 (58.3)46 (54.8).6432 (53.3)36 (60.0).46Antisecretory100 (52.6)89 (46.8).2648 (57.1)46 (54.8).7633 (55.0)35 (58.3).71Antitumor16 (8.4)7 (3.7).056 (7.1)6 (7.1)>.994 (6.7)3 (5.0).70Antidepressant23 (12.1)15 (7.9).177 (8.3)6 (7.1).7713 (21.7)9 (15.0).35Analgesic134 (70.5)120 (63.2).1355 (65.5)54 (64.3).8741 (68.3)43 (71.7).69Antiseizure8 (4.2)11 (5.8).485 (6.0)5 (6.0)>.994 (6.7)5 (8.3).73Diuretic57 (30.0)47 (24.7).2531 (36.9)23 (27.4).1922 (36.7)20 (33.3).70Diabetes27 (14.2)29 (15.3).7719 (22.6)12 (14.3).1611 (18.3)9 (15.0).62Antibiotic135 (71.1)126 (66.3).3261 (72.6)55 (65.5).3239 (65.0)33 (55.0).26Anticoagulant75 (39.5)80 (42.1).6039 (46.4)47 (56.0).2227 (45.0)29 (48.3).71Antiarrhythmic16 (8.4)15 (7.9).8514 (16.7)7 (8.3).107 (11.7)6 (10.0).77Muscle relaxant29 (15.3)26 (13.7).6616 (19.1)10 (11.9).209 (15.0)11 (18.3).62Antipsychotic16 (8.4)16 (8.4)>.994 (4.8)10 (11.9).097 (11.7)6 (10.0).77Electrolyte concentrates109 (57.4)111 (58.4).8448 (57.1)56 (66.7).2038 (63.3)39 (65.0).85Antihypertensive50 (26.3)51 (26.8).9127 (32.1)25 (29.8).7419 (31.7)17 (28.3).69*Data are given as number (percentage) of each group, unless otherwise indicated. ADE indicates adverse drug event.†Includes antidepressant, antipsychotic, sedative or hypnotic, or analgesic drugs.Multivariate analyses of significant correlates of ADEs, severe ADEs, and preventable ADEs in the case-control study revealed few independent predictors. There was 1 independent correlate of ADEs: exposure to 1 or more psychoactive drugs (odds ratio, 2.1; 95% confidence interval, 1.3-3.6). Use of cardiovascular drugs was an independent predictor of severe ADEs (odds ratio, 2.4; 95% confidence interval, 1.3-4.5). There were no independent correlates of preventable ADEs.COMMENTSubstantial data suggest that ADEs are an important public health problem. In previous studies,we found that these events occurred at a rate of 6.5 per 100 admissions and that 28% were potentially preventable. One attractive approach to preventing events involves risk stratification, in which patients are stratified using prospectively gathered information according to their risk of an event.This has been an effective clinical approach for reducing event rates in other areas, for example, for risk of cardiac events in patients undergoing noncardiac surgery.In this study, we tried to develop a risk stratification model for patients likely to experience an ADE using 2 approaches: a cohort analysis using limited information readily available electronically and a case-control study using more detailed information gathered on case patients and a set of matched controls. However, while the cohort analysis identified a few independent predictors of ADEs and preventable ADEs, they had relatively little power. In fact, almost none of the proposed "risk factors" for ADEs were actually associated with a substantially elevated risk of having an ADE. While such factors as age, multiple drug therapy, and impaired renal function almost certainly increase the risk of an ADE, the magnitude of these risks is probably smaller than has been suggested.In a number of recent studies, investigators have suggested 3 factors in particular that may predispose patients to ADEs: age,polypharmacy,and impaired renal function.Our data suggest that the effect of advancing age may be modest at best. While age emerged as a significant correlate in our cohort analysis, it was not retained in the multivariate studies because of lack of predictive power. One explanation may be that while drug metabolism is clearly altered by age, the drugs and dosages used in older patients are adjusted to account for the patient's age.Regarding polypharmacy as a potential predictor of adverse events, we found only a borderline association between number of drugs received and risk of an ADE. In the case-control analysis, there was no correlation, and in the cohort study, neither average number of drugs received nor average number of AHFS classes of drugs received was an independent predictor in the multivariate analysis. In a previous study,we found a correlation of ADE rate with drug exposure. Specifically, a significantly higher rate of preventable and potential ADEs was observed in ICU patients vs non-ICU patients, but after adjusting for number of drugs administered, this difference disappeared. It may be the case that being in an ICU is correlated with receiving an increased number of drugs, while polypharmacy is not an independent predictor of having an ADE. Regarding impaired renal function, the cohort study found a significant univariate correlation between SUN levels above 5.7 mmol/L (16 mg/dL) and ADEs and preventable ADEs compared with the entire cohort population at 1 hospital. However, this was not retained as an independent predictor in multivariate analysis. In the case-control analysis, while we had a relatively small number of patients and thus limited power, there was a slight trend toward an effect of renal impairment for preventable ADEs but it was not statistically significant.The drug exposure data in the case-control analysis suggest that no major drug class was responsible for a disproportionate share of the ADEs, with the possible exception of analgesics. The association between cardiovascular drugs and serious events in the case-control study may be a chance finding given that these drugs were rarely believed to be responsible for individual events in several comparisons, or it may be that use of these drugs was a marker for an underlying condition (eg, cardiovascular instability).If nonpreventable ADEs occur relatively randomly across the hospitalized population, and a large fraction of preventable ADEs occur as a result of system problems that do not occur substantially more often in one patient group, then the clinical and drug exposure data we obtained in this study are what would be expected. The subgroup of events that might be expected to be predictable are those that occur in the group of patients with impaired drug clearance for some reason. For example, older patients with impaired renal function or impaired liver function may indeed be at higher risk of an ADE. However, our data suggest that this group of patients is relatively small.The data analyzed in this study suggest that prevention strategies that focus on improving the systems by which drugs are ordered, dispensed, and administered will prevent more events than patient risk stratification strategies. According to this view, the focus for improvement should be on the system involved rather than on individual patients.These approaches will often involve the use of patient-specific data, for example, in adjusting aminoglycoside dosages based on the presence of renal failure. However, they often will not use patient-specific data, for example, in developing standardized labeling for tubing in situations in which multiple intravenous medications are being administered simultaneously.This study has several limitations. It was performed at only 2 tertiary care institutions, which may hinder the generalizability of the results to other care settings. Another limitation, specific to the cohort analysis that looked only at information available electronically, is that the specific data elements available online from site to site are likely to vary, and clearly more elements will become available over time. However, the elements we chose are reasonably standard, and even if they are not available within some organizations, they are likely to become available over time. Another issue is that an alternative approach—not our focus in this study—would be to take specific patient groups and look for specific types of events; this way of addressing the problem may prove useful in future studies.We conclude that risk stratification approaches to identifying patients at high risk of experiencing an ADE in the hospital are unlikely to be productive. Instead, strategies that attempt to reduce the incidence of ADEs by focusing on the systems by which drugs are given, and that incorporate specific suggestions regarding medication choices and dosing, are more likely to be successful in battling this important and costly problem.DWBatesLLLeapeSPetryckiIncidence and preventability of adverse drug events in hospitalized adults.J Gen Intern Med.1993;8:289-294.MTKoskaDrug errors: dangerous, costly and avoidable.Hospitals.1989;63:24.JKorinCost implications of malpractice and adverse events.Hosp Formulary.1993;28(suppl 1):59-61.TSLesarLLBricelandKDelcoureJCParmaleeVMasta-GornicHPohlMedication prescribing errors in a teaching hospital.JAMA.1990;263:2329-2334.TABrennanLLLeapeNLairdIncidence of adverse events and negligence in hospitalized patients: results from the Harvard Medical Practice Study I.N Engl J Med.1991;324:370-376.LLLeapeTABrennanNMLairdThe nature of adverse events in hospitalized patients: results from the Harvard Medical Practice Study II.N Engl J Med.1991;324:377-384.DCClassenSLPestotnikRSEvansJPBurkeComputerized surveillance of adverse drug events in hospital patients.JAMA.1991;266:2847-2851.GAFaichNational adverse drug reaction reporting, 1984-1989.Arch Intern Med.1991;151:1645-1647.GAFaichAdverse-drug-reaction monitoring.N Engl J Med.1986;314:1589-1592.KLMelmonPreventable drug reactions: causes and cures.Semin Med Beth Israel Hosp.1971;284:1361-1367.MRKeithRABellanger-McCleeryJEFuchs JrMultidisciplinary program for detecting and evaluating adverse drug reactions.Am J Hosp Pharm.1989;46:1809-1812.LGSeidlGThorntonJWSmithStudies on the epidemiology of adverse drug reactions, III: reactions in patients on a general medical service.Bull Hosp Johns Hopkins Hosp.1966;119:299-315.HJickAdverse drug reactions: the magnitude of the problem.J Allergy Clin Immunol.1984;74:555-557.BCHoddinottCWGowdeyWKCoulterDrug reactions and errors in administration on a medical ward.CMAJ.1967;97:1001-1006.RIOgilvieJRuedyAdverse drug reactions during hospitalization.CMAJ.1967;97:1450-1457.NHurwitzOLWadeIntensive hospital monitoring of adverse reactions to drugs.BMJ.1969;1:531-536.LLBerryRSegalTPSherrinKAFudgeSensitivity and specificity of three methods of detecting adverse drug reactions.Am J Hosp Pharm.1988;45:1534-1539.Not AvailableInternational drug monitoring: the role of the hospitalWHO Tech Rep Serv.1969;425:5-24.AKJhaGJKupermanJMTeichIdentifying adverse drug events: development of a computer-based monitor and comparison to chart review and stimulated voluntary report.J Am Med Inform Assoc.1998;5:305-314.DWBatesDCullenNLairdIncidence of adverse drug events and potential adverse drug events: implications for prevention.JAMA.1995;274:29-34.LLLeapeDWBatesDJCullenSystems analysis of adverse drug events.JAMA.1995;274:35-43.WTschepikRSegalTPSherrinDNSchneiderRLHammondTherapeutic risk-assessment model for identifying patients with adverse drug reactions.Am J Hosp Pharm.1990;47:330-334.RJMichockiPPLamyA "risk" approach to adverse drug reactions.J Am Geriatr Soc.1988;36:79-81.REGrymonprePAMitenkoDSSitarDrug-associated hospital admissions in older medical patients.J Am Geriatr Soc.1988;36:1092-1098.TAHutchinsonKMFlegelMSKramerDGLeducHPKKongFrequency, severity and risk factors for adverse drug reactions in adult outpatients: a prospective study.J Clin Epidemiol.1986;39:533-542.JKJonesAssessing potential risk of drugs: the elusive target [editorial].Ann Intern Med.1992;117:691-692.AOkweraJLJohnsonMJVjechaRisk factors for adverse drug reactions during thiacetazone treatment of pulmonary tuberculosis in human immunodeficiency virus infected adults.Int J Tuberc Lung Dis.1997;1:441-445.EJThomasHRBurstinEJOravIncidence of and risk factors for adverse events and negligent care in Colorado and Utah in 1992 [abstract].J Gen Intern Med.1997;12(suppl):81.JTHanlonKESchmaderMJKoronkowskiAdverse drug events in high risk older outpatients.J Am Geriatr Soc.1997;45:945-948.CAColleyLMLucasPolypharmacy: the cure becomes the disease.J Gen Intern Med.1993;8:278-283.MChesterLChenJCKaskiIdentification of patients at high risk for adverse coronary events while awaiting routine coronary angioplasty.Br Heart J.1995;73:216-222.FEKarchLLasagnaAdverse drug reactions: a critical review.JAMA.1975;234:1236-1241.KSteelPMGertmanCCrescenziJAndersonIatrogenic illness on a general medical service at a university hospital.N Engl J Med.1981;304:638-642.Health and Public Policy Committee of the American College of PhysiciansImproving medical education in therapeutics.Ann Intern Med.1988;108:145-147.Medical Practice StudyPatients, Doctors and Lawyers: Studies of Medical Injury in New York.Boston, Mass: Harvard University; 1990.CSLandefeldLGoldmanMajor bleeding in outpatients treated with warfarin: incidence and prediction by factors known at the start of outpatient therapy.Am J Med.1989;87:144-152.CSLandefeldEMcGuireMWRosenblattA bleeding risk index for estimating the probability of major bleeding in hospitalized patients starting anticoagulant therapy.Am J Med.1990;89:569-578.DCullenARNemeskalTherapeutic Intervention Scoring System (TISS).Problems Crit Care.1989;3:545-562.MECharlsonPPompeiKLAlesCRMacKenzieA new method of classifying prognostic comorbidity in longitudinal populations: development and validation.J Chronic Dis.1987;40:373-383.SAS Institute IncRelease 6.11.Cary, NC: SAS Institute Inc; 1992.Not AvailableStatXact User Manual, Version 2.Cambridge, Mass: CYTEL Software Corporation; 1991.DGKleinbaumLLKupperKEMullerApplied Regression Analysis and Other Multivariable Methods.Boston, Mass: PWS-Kent Inc; 1988.JHWassonHCSoxRKNeffLGoldmanClinical prediction rules: applications and methodologic standards.N Engl J Med.1985;313:793-799.RLGoldmanDevelopment of a Veterans Administration occurrence screening program.Qual Rev Bull.1989;15:315-319.PEAshburnPolypharmacy in skilled-nursing facilities.Ann Intern Med.1993;118:649-650.EAGormleyDJGriffithsPNMcCrackenGMHarrisonPolypharmacy and its effect on urinary incontinence in a geriatric population.Br J Urol.1993;71:265-269.KKroenkeEMPinholtReducing polypharmacy in the elderly: a controlled trial of physician feedback.J Am Geriatr Soc.1990;38:31-36.DAMcMillanPMHarrisonLJRogersNTongAJMcLeanPolypharmacy in an Australian teaching hospital: preliminary analysis of prevalence, types of drugs and associations.Med J Aust.1986;145:339-342.JLeSagePolypharmacy in geriatric patients.Nurs Clin North Am.1991;26:273-290.TJMeyerDVan KootenSMarshAVProchazkaReduction of polypharmacy by feedback to clinicians.J Gen Intern Med.1991;6:133-136.JHGurwitzJAvornThe ambiguous relation between aging and adverse drug reactions.Ann Intern Med.1991;114:956-966.DJCullenBJSweitzerDWBatesEBurdickAEdmondsonLLLeapePreventable adverse drug events in hospitalized patients: a comparative study of intensive care units and general care units.Crit Care Med.1997;25:1289-1297.Accepted for publication February 18, 1999.This study was supported by grant R01-HS07107-01 from the Agency for Health Care Policy and Research, Rockville, Md; and the Risk Management Foundation, Cambridge, Mass.We thank the nurses, pharmacists, physicians, and other personnel on the study units for their support in carrying out the study.The ADE Prevention Study GroupBoston, Mass:Lucian L. Leape, MD; Deborah Servi; Nan Laird, PhD; David W. Bates, MD, MSc; Michael Cotugno, PharmD; Mairead Hickey, RN, PhD; Patricia Hojnowski-Diaz, RN; Sharon Kleefield, PhD; Heather Patterson, PharmD; Stephen Petrycki, RN; Brian Shea, PharmD; Martha Vander Vliet, RN; Jeffrey Cooper, PhD; David J. Cullen, MD, MSc; Harry Demonaco, MS, RPh; Margaret Dempsey Clapp, MS, RPh; Theresa Gallivan, RN; Robert Hallisey, MS, RPh; Jeanette Ives, RN, MSN; Ellen Kinneally, RN; Kathy Porter, RN, MSN; Steven D. Small, MD; Bobbie J. Sweitzer, MD; Taylor Thompson, MD; J. Richard Hackman, PhD; Amy Edmondson. Houston, Tex:Laura A. Petersen, MD, MPH. New York, NY:Glenn Laffel, MD, PhD.Corresponding author: David W. Bates, MD, MSc, Division of General Medicine and Primary Care, Brigham and Women's Hospital, 75 Francis St, Boston, MA 02115 (e-mail: [email protected]).