La cohabitación de la academia y la industria: ¿qué postura toman las revistas médicas en relación con los artículos de opinión? / The cohabitation of academia and industry: what position do the medical journals take regarding opinion articles?

¿Deben las revistas médicas publicar editoriales y artículos de carácter educativo escritos por autores que presentan conflictos de intereses financieros con las industrias farmacéutica y biotecnológica de cuyos productos (o sus competidores) opinan? En los últimos 18 meses se ha desatado una polémica entre The New England Journal of Medicine y BMJ, que plantean 2 posturas opuestas: la primera entiende que el prejuicio negativo contra autores con conflicto de intereses con la industria es excesivo y, por tanto, acepta artículos de cualquier experto procurando que presenten el mínimo sesgo posible. BMJ, por el contrario, prohíbe publicar ese tipo de artículos por autores que tienen conflicto de intereses financieros con la industria. En este artículo se desarrollan las actitudes de ambas (y otras) revistas y se reflexiona sobre este tipo de conflicto en la profesión médica (AU)

Should medical journals publish editorials and educational articles written by authors who have financial conflicts of interest with pharmaceutical and biotechnology industries on whose products (or their competitors) they discuss? In the last 18 months, a controversy was sparked between The New England Journal of Medicine and BMJ, who took 2 opposite positions: the former stated that the negative bias against authors with conflicts of interest with industry is excessive and therefore accept articles from any expert, ensuring that they have the minimum possible bias. BMJ, in contrast, prohibits the publication of these types of article by authors who have financial conflicts of interest with industry. This article discusses the approaches of the 2 journals (and those of others) and reflects on this type of conflict in the medical profession (AU)

The cohabitation of academia and industry: What position do the medical journals take regarding opinion articles? / La cohabitación de la academia y la industria: ¿qué postura toman las revistas médicas en relación con los artículos de opinión?

Should medical journals publish editorials and educational articles written by authors who have financial conflicts of interest with pharmaceutical and biotechnology industries on whose products (or their competitors) they discuss? In the last 18 months, a controversy was sparked between The New England Journal of Medicine and BMJ, who took 2 opposite positions: the former stated that the negative bias against authors with conflicts of interest with industry is excessive and therefore accept articles from any expert, ensuring that they have the minimum possible bias. BMJ, in contrast, prohibits the publication of these types of article by authors who have financial conflicts of interest with industry. This article discusses the approaches of the 2 journals (and those of others) and reflects on this type of conflict in the medical profession.

Ensayos clínicos con... ¿medicamentos? A propósito de 2 proyectos en enfermedad de Alzheimer y paludismo / Clinical trials with ... drugs? Based on 2 projects in Alzheimer's disease and malaria

La ley de investigación biomédica (LIB) regula la investigación en seres humanos pero no la relativa a los ensayos clínicos con medicamentos. Este artículo describe los fundamentos científicos y normativos por los que 2 proyectos que pueden ser observados como ensayos clínicos pueden seguir los requerimientos de la LIB. Uno es el estudio de tomografía por emisión de positrones con radiofármaco para determinar la presencia de proteína β amiloide en ciertas áreas del cerebro de adultos cognitivamente sanos. El otro es un estudio de infección controlada de paludismo en voluntarios sanos, mediante la inoculación de esporozoitos de Plasmodium falciparum asépticos, purificados y criopreservados. En ambos estudios, al incluir procedimientos invasivos, la LIB exige la autorización del estudio por las autoridades autonómicas competentes. Estos 2 estudios han sido los primeros que han utilizado este procedimiento normativo en Cataluña(AU)

The biomedical research act (BRA) regulates clinical research in humans, but not that related to clinical trials with medicinal products. This article describes the scientific and regulatory foundations supporting 2 projects which could be observed as clinical trials, can follow the BRA requirements. One is a positron emission tomography study with radiopharmaceutical to determine the presence of amyloid-β protein deposition in certain areas of the brain of cognitively healthy adults. The other is a study on controlled malaria infection in healthy volunteers using the inoculation of aseptic, purified and cryopreserved Plasmodium falciparum sporozoites. Since in both studies subjects undergo invasive procedures, the BRA requires the approval of the study by the relevant regional health authorities. These 2 studies have been the first ones that have used this regulatory procedure in Catalonia(AU)

Clinical trials with drugs? Based on 2 projects in Alzheimer's disease and malaria.

The biomedical research act (BRA) regulates clinical research in humans, but not that related to clinical trials with medicinal products. This article describes the scientific and regulatory foundations supporting 2 projects which could be observed as clinical trials, can follow the BRA requirements. One is a positron emission tomography study with radiopharmaceutical to determine the presence of amyloid-ß protein deposition in certain areas of the brain of cognitively healthy adults. The other is a study on controlled malaria infection in healthy volunteers using the inoculation of aseptic, purified and cryopreserved Plasmodium falciparum sporozoites. Since in both studies subjects undergo invasive procedures, the BRA requires the approval of the study by the relevant regional health authorities. These 2 studies have been the first ones that have used this regulatory procedure in Catalonia.

Harmonisation of ethics committees' practice in 10 European countries.

BACKGROUND: The Directive 2001/20/EC was an important first step towards consistency in the requirements and processes for clinical trials across Europe. However, by applying the same rules to all types of drug trials and transposing the Directive's principles into pre-existing national legislations, the Directive somewhat failed to meet its facilitation and harmonization targets. In the field of ethics, the Directive 2001/20/EC conditioned the way of understanding and transposing the "single opinion" process in each country. This led to a situation in which two models of research ethics committees organisation systems exist, being the model in which the "single opinion" is considered to be the decision made by a single ethics committee more effective and simpler in terms of administrative and logistic workload. METHOD: A survey was conducted in 10 European countries. Members of the European Clinical Research Infrastructures Network working party number 1, with expertise in the field of ethics, responded. RESULTS: There is a major heterogeneity in the composition of ethics committees among the surveyed countries based on the number of members, proportion of experts versus lay members and expertise of the scientific members. A harmonized education of the ethics committees' membership based in common curricula is recommended by the majority of countries. CONCLUSIONS: Despite the efforts for harmonization of the European Clinical Trial Directive, from an ethical point of view, there remains a plurality of ethics committees' systems in Europe. It is important to comprehend the individual national systems to understand the problems they are facing.

Inhibitory quotient as a prognostic factor of response to a salvage antiretroviral therapy containing ritonavir-boosted saquinavir. The CIVSA Study.

BACKGROUND: The addition of a low dose of ritonavir to protease inhibitors (PIs) has become a widespread strategy to improve PI pharmacokinetics. As resistance is a major barrier to long-term suppression, in salvage therapy genotype and/or phenotype scoring is currently used to predict the response. We evaluated the relationship between the saquinavir (SQV) inhibitory quotient (IQ) (virtual and genotypic) and virological response. METHODS: Eligible patients were on a PI-containing highly active antiretroviral therapy (HAART) regimen excluding SQV and had a viral load >5000 HIV-1 RNA copies/mL. The PI was switched to SQV/ritonavir (RTV) 1000/100 mg twice a day (bid) and the same two backbone nucleoside reverse transcriptase inhibitors (NRTIs) were maintained at least until week 4, when the resistance test results became available. Genotype and virtual phenotype were determined at baseline, while the SQV trough plasma concentration was determined at week 4. RESULTS: Fifty-three patients were included in the study. Mean baseline viral load and CD4 count were 137,693 copies/mL and 263 cells/microL, respectively, the mean number of previous PIs was 2.3 and the mean number of protease gene mutations (PGMs) was 4.1. Using an on-treatment analysis, at week 16 the mean increase in CD4 count was 70.9 cells/microL, viral load was <200 copies/mL in 17 out of 37 patients (45.9%), and 30 out of 45 patients (66.7%) were considered virological responders (VRs) (viral load <200 copies/mL or viral load declined > or =1 log(10) at week 16). Median virtual phenotype was 1.3 (0.6-6.9). Baseline differences were detected between VR and non-VR populations: the mean numbers of PGMs were 3.2 and 5.8 (P<0.05), the mean numbers of SQV-associated mutations were 2 and 3.8 (P<0.05), and the mean CD4 counts were 365.9 and 184.3 cells/microL (P<0.05), respectively. Mean SQV trough concentrations at week 4 were 1.1 and 1.0 microg/mL (not significant), and mean virtual IQs were 0.7 and 0.1 (P<0.01), respectively. Multivariate analysis showed that baseline PGMs >5 or SQV-associated mutations>5, virtual phenotype, baseline viral load >50,000 copies/mL, and virtual IQ <0.5, but not genotypic IQ, were the variables independently associated with non-VR. CONCLUSION: In heavily pretreated patients, the use of SQV virtual IQ or alternatively virtual phenotype, as well as PGMs, is a useful tool for the prediction of virological response.

Penetration of linezolid into the anterior chamber (aqueous humor) of the human eye after intravenous administration.

We determined the linezolid concentrations in serum samples and aqueous humors (AHs) from 21 patients undergoing cataract extraction. Cataract removal was performed at various times (from 60 to 270 min) after the end of a 30-min infusion of 600 mg of linezolid. Serum samples were obtained 1 h after the end of linezolid administration to determine the maximum concentration of linezolid (C(max)); AHs and a second serum sample were taken simultaneously during the operation, and the concentrations of linezolid in AH (C(AH)) and serum (C(S)) were determined. The mean C(AH) 1 h after linezolid administration was 4.94 micro g/ml, and the mean ratio of C(AH) to C(S) (R = C(AH)/C(S)) was 0.43. All patients had a C(AH) of >2 micro g/ml, which was higher than the MIC at which 90% of Staphylococcus epidermidis strains are inhibited.

Determination of efavirenz in human plasma by high-performance liquid chromatography with ultraviolet detection.

Efavirenz is a non-nucleoside reverse transcriptase inhibitor for the treatment of the HIV infection. A simple, high-performance liquid chromatographic method has been developed and validated for the quantitative determination of efavirenz in human plasma. The method involved solid-phase extraction of the drug and the internal standard (L-737,354) from 300 microl of human plasma. The analysis was via UV detection at 250 nm using a reversed-phase C8 analytical column and a isocratic mobile phase consisting of phosphate buffer (pH 5.75)-acetonitrile that resolved the drug and internal standard from endogenous matrix components and potential coadministered drugs. Within- and between-day precisions were less than 8.6% for all quality control samples. The lower limit of quantification was 0.1 microg/ml. Recovery of efavirenz from human plasma was greater than 83%. This validated assay is being used in pharmacokinetic studies with efavirenz.

Hepatotoxicity in HIV-1-infected patients receiving nevirapine-containing antiretroviral therapy.

OBJECTIVES: To assess the incidence and risk factors for hepatotoxicity associated with nevirapine. DESIGN: A prospective cohort study in a teaching and referral hospital involving all consecutive patients who were prescribed a nevirapine-containing antiretroviral regimen between September 1997 and May 2000. METHOD: Cutaneous and hepatic adverse reactions and clinical hepatitis were assessed. Blood analysis including plasma HIV-1 RNA CD4 cell counts, liver chemistry tests, and serology for hepatitis B and C viruses. Hepatotoxicity was defined as an increase of at least threefold in serum alanine aminotransferase or aspartate aminotransferase levels compared with baseline values. RESULTS: Of a total of 610 patients, 82 (13.4%) were antiretroviral naive when commencing nevirapine, and 46.2 and 8.9% were coinfected with hepatitis C and B viruses, respectively. Median duration of exposure to nevirapine was 8.7 months (interquartile range 3.4--14.3). Hepatotoxicity developed in 76 (12.5%), an incidence of 13.1/100 person-years. Kaplan--Meier estimated incidence of hepatotoxicity at 3, 6 and 12 months was 3.7, 9.7 and 20.1%, respectively. In seven (1.1%) patients, hepatotoxicity was associated with clinical hepatitis, which was reversible upon discontinuation of therapy. Multivariate analysis identified the duration of prior exposure to antiretroviral drugs, hepatitis C virus, and higher baseline levels of alanine aminotransferase as independent risk factors for hepatotoxicity. CONCLUSIONS: Hepatotoxicity but not clinical hepatitis was common in HIV-1-infected patients receiving nevirapine-containing regimens and the incidence steadily increased over time. Prolonged exposure to any antiretroviral therapy, coinfection with hepatitis C virus and abnormal baseline levels of alanine aminotransferase identified patients at a higher risk.

Simultaneous determination of the HIV-protease inhibitors indinavir, amprenavir, ritonavir, saquinavir and nelfinavir in human plasma by reversed-phase high-performance liquid chromatography.

A rapid, simple and sensitive high-performance liquid chromatographic (HPLC) assay has been developed for the simultaneous quantification of the HIV-protease inhibitors indinavir, amprenavir, ritonavir, saquinavir and nelfinavir in human plasma. The method involved the solid-phase extraction of the five drugs and the internal standard (I.S., verapamil) from 400 microl of human plasma. The HPLC analysis used a reversed-phase C18 analytical column and a mobile phase consisting of a gradient with 15 mM phosphate buffer (pH 5.75)-acetonitrile and UV monitoring. The method was linear over the therapeutic concentration range for the five HIV-protease inhibitors. The accuracy of the method ranged from 98.2 to 106.7% and the precision values ranged from 1.4 to 8.1% for intra-day precision and from 3.1 to 6.4% for the inter-day values.

The use of evidence in pharmacovigilance. Case reports as the reference source for drug withdrawals.

OBJECTIVE: Withdrawal of a drug from the market for safety reasons is a serious and sometimes complex decision. The scientific evidence supporting drug withdrawals in the past years is critically appraised. METHODS: With data provided by the Spanish Medicines Agency, all drugs withdrawn from the Spanish market for safety reasons from January 1990 to December 1999 were identified. The adverse drug reactions (ADRs) were classified by the year of withdrawal, by the organ/system affected and by the alleged type of reaction (Rawlins and Thompson classification). A systematic review of the literature was performed. RESULTS: A total of 22 drugs were withdrawn from the market due to safety reasons. In 18 of 22 cases (82%), the evidence supporting the drug withdrawal came from individual case reports, cases series or the combination of data provided by randomised clinical trials and case reports. Hepatic (eight cases) and cardiac (five cases) reactions accounted for 59% (13 of 22) of the total withdrawals. In 10 of 22 (45%) cases, drug withdrawal was clearly due to type-B reactions. Only four withdrawals were based on evidence from observational studies including a comparison group. CONCLUSION: Case reports are the main source of information used to withdraw a drug from the market for safety reasons. It is necessary to improve the quality of evidence supporting the withdrawal process of drugs linked to unexpected and severe ADRs. The use of large databases to perform cohort or nested case-control analyses is the most efficient and reliable method to study type-A class effect ADRs. The implementation of such databases in different countries could increase the quality of the information on ADRs by allowing researchers to conduct efficiently these type of studies.

Dose-finding study of once-daily indinavir/ritonavir plus zidovudine and lamivudine in HIV-infected patients.

BACKGROUND: Strategies for treatment of HIV need to be considered in terms of combining potency, safety, and convenience of dosage. However, regimens including once-daily protease inhibitors are not yet available. We have performed a pilot study to determine an indinavir/ritonavir (IND/RTV) regimen for once-daily dosing, by monitoring plasma levels. METHODS: Antiretroviral-naive HIV-infected adults were eligible. Therapy was zidovudine/lamivudine 1 pill twice daily plus IND/RIT (liquid formulation) 800/100 mg twice daily with food. At 4-week intervals, plasma levels were measured and dosage of IND/RIT switched to 1000/100 mg daily and then 800/200 mg daily. If 12-hour minimum concentrations (Cmin12h) of IND were too low (<0.1 microg/ml) with IND/RIT 1000/100 mg once daily in the first half of the patients, it was planned to switch directly to 800/200 mg once daily in the other half. RESULTS: In all, 27 patients were recruited. Mean baseline CD4+ lymphocyte count was 107 x 106/L (range, 4-623 x 106/L). Eleven patients (40%) discontinued the study medication within the first 4 weeks due to clinical progression (n = 3) or grade 1-2 RTV related side effects (n = 8). Nine patients (group A) switched from 800/100 mg twice daily to 1000/100 mg once daily and then to 800/200 mg once daily. Seven patients (group B) switched directly to 800/200 mg once daily. At week 32, viral load was <5 copies/ml in 15 of 16 patients (94%). RTV levels were always <2.1 microg/ml. The mean and 95% confidence interval for IND Cmin and Cmax in microg/ml was: using IND/RTV 800/100 mg twice daily (n = 16) 1.4 (0.5-2.3) and 6.7 (4.4-9.1), respectively; using IND/RTV 1000/100 mg once daily (n = 9) 0.18 (0-0.41) and 8.6 (3.3-14), respectively; and using 800/200 mg once daily (n = 16) 0.38 (0-0.9), and 7.5 (0.8-14.8). For all 16 patients who received IND/RTV 800/100 mg twice daily, the Cmin value for IND was >/=0.1 microg/ml. Conversely, IND Cmin was <0.1 microg/ml in 4 of 9 receiving 1000/100 mg once daily but in only 1 of 16 receiving 800/200 mg once daily. CONCLUSION: Once-daily regimen of IND/RIT is feasible and deserves further evaluation in larger randomized trials. Liquid formulation of RIT was not well tolerated by our antiretroviral-naive patients despite lower than suggested doses.

Determination of stavudine in human plasma and urine by high-performance liquid chromatography using a reduced sample volume.

Sensitive high-performance liquid chromatographic assays have been developed for the quantification of stavudine (2',3'-didehydro-3'-deoxythymidine, d4T) in human plasma and urine. The methods are linear over the concentration ranges 0.025-25 and 2-150 microg/ml in plasma and urine, respectively. An aliquot of 200 microl of plasma was extracted with solid-phase extraction using Oasis cartridges, while urine samples were simply diluted 1/100 with HPLC water. The analytical column, mobile phase, instrumentation and chromatographic conditions are the same for both methods. The methods have been validated separately, and stability tests under various conditions have been performed. The detection limit is 12 ng/ml in plasma for a sample size of 200 microl. The bioanalytical assay has been used in a pharmacokinetic study of pregnant women and their newborns.

[Vancomycin and teicoplanin use as antibiotic prophylaxis in cardiac surgery: pharmacoeconomic study]. / Vancomicina y teicoplanina como profilaxis antibiótica en cirugía cardíaca: estudio farmacoeconómico.

BACKGROUND: To assess the economical impact of vancomycin use versus teicoplanin use as antibiotic prophylaxis for patients undergoing cardiac surgery for valve replacement (VR) and coronary artery by-pass (CABS) procedures. PATIENTS AND METHODS: This is an ancillary cost minimization analysis of a double blinded, parallel groups, randomised clinical trial (RCT), with the main objective of comparing the safety and efficacy of these antibiotics. 500 patients were included in the study; 267 in the CABS group and 233 in the VR group. The CABS patients received 1 g vancomicin or 400 mg teicoplanin, plus 150 mg netilmicin. The VR group received a second dose of each drug after extracorporeal circulation. In order to calculate the costs we considered the direct cost of the drug, the i.v. mix and the administration costs, together with personnel and structure costs. We considered two different situations: the administration of drugs within the surgical room theatre and in the medical ward. RESULTS: The demographic data of both groups were comparable. The frequency of severe adverse drug reactions (ADR) were similar (0.4%) in both groups, as well as the post-operative infection rates (8.6%). Differences were seen in the frequencies of low severity ADRs: 20.4% in the vancomycin group and 1.6% in the teicoplanin group. When the antibiotics were administered in the surgical room, among CABS patients the costs were 8,265 pts. for the teicoplanin group and 12,005 pts. for the vancomycin group; while among VR patients, costs were respectively 11,661 pts. and 14,528 pts. Administration costs of teicoplanin and vancomycin within a medical ward setting, however, the costs were 6,740 pts. and 2,809 pts. for CABS patients, and 5,308 pts. and 10,140 pts. for VR patients, respectively. CONCLUSIONS: The costs of antibiotic prophylaxis among cardiac surgery patients heavily depends on the setting and circumstances of drug administration. The minimization cost analysis indicates that teicoplanin is the most cost-effective option if the drug is administered within the surgical area, while vancomycin is the less costly option when administered within the medical ward. However, if the second option is to be chosen, it is necessary to assure the right plasmatic drug levels of the antibiotic at the beginning of the surgical procedure.

[Comparative analysis of articles published by Spanish authors (1993-1997) in biomedical journals with high impact factor]. / Análisis comparativo de las publicaciones realizadas por autores españoles (1993-1997) en revistas clínicas con factor de impacto elevado.

BACKGROUND: To identify the Spanish scientific production amongst different areas of clinical knowledge, and to compare it with those of five other European Union countries. METHOD: Review of MEDLINE data base, for the period 1993-1997. Search limited to four journals, selected, for 10 different medical specialties (Cardiology, Endocrinology, Infectious Diseases, Gastroenterology-Hepatology, Haematology, Nephrology, Pneumology, Neurology, Oncology, Rheumatology). Articles published by authors from Germany, France, Italy, The Netherlands, Sweden and Spain. Journals included in the Internal Medicine subject classification were independently analysed. Data were also related with several econometric indexes. RESULTS: A total of 1,763 original articles published by Spanish authors were identified in the journal's sample over the analysed period (2.08 articles per 100 all published articles). Spain contributes to the total achieved by the six European countries analysed with 9.07 articles per 100 published articles. Gastroenterology-Hepatology was the medical specially which has more articles published by Spanish authors (total: 338 articles; 4.15 articles/100 published articles); and Oncology the one with less articles published (1.26 articles/100 published articles). The mean IF value per journal by article is highest for Gastroenterology-Hepatology (4.86 FI/article) and lowest for Pneumology (2.42 FI/article). Spain is the last amongst all six European countries analyzed in Endocrinology, Oncology and Haematology, and second to last in all others except for Gastroenterology-Hepatology (4th place). Mean cost for each article produced by Spanish authors in the analyzed sample was 0.49 US $ according the health expenditures per capita, and 0.07 US $ according the R+D expenditures per capita. Data from the independent analysis of Internal Medicine journals also showed that Gastroenterology and Hepatology is the subspecialty with a higher number of papers published in those journals. CONCLUSIONS: All efforts devoted to improve the quality of Spanish biomedical research, specially in clinical research, had produced positive, but uneven, results, measured by the number and impact factor of original articles published in top ranked biomedical journals. The overall distribution of high impact factor scientific production by specialties is poor when compared to the European Union countries included in the analysis. Those results showed several improvement opportunities. Besides increasing the overall budget for R+D, its is likely that the time has come for backing the highest quality Spanish biomedical research, the one that offers greater and better chances for achieving scientifically valid results, and is published in high impact factor biomedical journals.

Methodological and political issues in clinical pharmacology research by the year 2000.

Parallel groups in a large, multicenter, phase III "pivotal" randomized clinical trial (RCT) with clinically relevant end-points are seen by the medical community as the "gold standard" of clinical research. However, there are limitations, some methodological and others political. The main one is the external validity of the method because a treatment, as studied in RCTs, does not necessarily reflect how it is used in clinical practice. Also, the method, as it stands, is not really predictive of the success in a particular patient of a certain intervention studied in a trial. To overcome these methodological drawbacks, different options have been implemented. The most important ones are: (1) the performance of pragmatic RCTs intended to address effectiveness rather than efficacy; (2) meta-analysis; and (3) the use of observational studies, with or without a comparison group. Recent experience has shown that type-A adverse drug reactions (ADRs) related to a specific class of drugs have been successfully characterized throughout cohort- or population-based case-control studies, whereas the evidence linking a specific drug entity to a type-B ADR, apparently severe enough to withdraw the drug from the market, has come mainly from case reports or case series. Other limitations of the RCT are more of a political nature. These large "pivotal" trials are mostly sponsored by the pharmaceutical industry, and to guarantee the scientific and ethical integrity of data produced, they are performed following standard operating procedures (SOPs) and good clinical practice (GCP) guidelines. Sometimes industry is not interested in sponsoring trials; thus, RCTs performed are in practice highly biased because of their potential economical profits. Furthermore, applying SOPs and GCPs is expensive and difficult to implement, and it is hard to find funding in public institutions. As a result, there is an urgent need to create a network of independent, skilled groups interested in sponsoring and performing institutional RCTs following "user friendly" GCP when the profits are low, but scientific interest high.

[Analysis of the cost-effectiveness relationship in the empirical treatment in patients with infections of the lower respiratory tract acquired in the community]. / Análisis de la relación coste-efectividad del tratamiento antibiótico empírico en pacientes con infecciones del tracto respiratorio inferior adquiridas en la comunidad.

BACKGROUND: Cost of treatment of community-acquired infections in Spain is an important factor in overall health expenditures. The aim of this study was to assess the direct health costs related with the treatment of patients with community-acquired pneumonia (CAP) and acute exacerbations of chronic bronchitis (AECB) due to infection, using different antibiotic options, and to identify main cost drivers. METHODS: A basic decision analysis model was developed, including probabilities estimation derived from the literature review, supplemented when needed by the opinion of a panel of 8 Spanish physicians (Delphi technique). Four groups of antibiotics were included (macrolides, beta-lactam, fluoroquinolones and cephalosporins) in two different groups: patients with CAP without hospital admission criteria and patients with AECB due to respiratory infection. The analytic horizon and the perspective used were those of the Spanish National Health Service. Direct cost were assessed (drugs, outpatient visits, hospital admissions, diagnostic tests). Indirect cost were not included in the model. Final costs uses as main outcome measure the average cost per patient treated. All results were calculated following a fold-back technique. Sensitivity analysis were included allowing for variations in several clinically relevant parameters. RESULTS: 1. Patients with CAP: Hospital admissions, directly related to the effectiveness rate of initial empirical antibiotic therapy, were the main cost driver (50%-70%). Acquisition costs of initial antibiotic therapy only account for 2%-13% of total costs. 2. Patients with AECB: Outpatient visits are the main cost driver for these group of patients (49% of total costs). Hospital admission costs are also an important cost driver (40%-51% of total costs). Acquisition costs of initial antibiotic therapy account for 4%-28% of total costs. Clinical effectiveness of first antibiotic option is the main variable regarding the cost-effectiveness rate. CONCLUSION: The model here presented showed that acquisition costs of first empirical antibiotic therapy are only a small proportion of total costs related with the management of community acquired lower respiratory tract infections in Spain. The clinical effectiveness rate of the first antibiotic used is the main variable which determines the final average cost per patient cured. For patients with lower respiratory tract infections the therapeutic option with a better cost-effectiveness ratio must be chosen, in order to minimize the risk of therapeutic failure after first line therapy, and should not be selected only by its lower acquisition costs.