1Universidad de Valparaíso, Departamento de Estadística, Valparaíso, Chile. Profesora auxiliar. Email: monica.catalan@uv.cl
2Universidad de Salamanca, Departamento de Estadística, Salamanca, España. Profesora titular. Email: pgalindo@usal.es
3Universidad de Salamanca, Departamento de Estadística, Salamanca, España. Profesor titular. Email: jmv@usal.es ]]>
Un meta-análisis integra información proveniente de varios estudios con el propósito de generar un resultado común para un problema determinado. En la literatura nos encontramos con varios métodos de integración de resultados, siendo el más básico el método de integración de niveles de probabilidad y, con una complejidad mayor, el método de integración del tamaño del efecto. Este último hace uso de modelos de efectos fijos y aleatorios. En este estudio, comparamos los resultados de dos métodos de estimación del tamaño del efecto basados en un meta-análisis usando modelos de efectos fijos y aleatorios. La medida del tamaño del efecto considerada en este estudio es el odds ratio, debido a que esta medida es usada frecuentemente en revisiones sistemáticas de varios temas de interés en salud pública, tales como cáncer cérvico uterino, colecistectomía laparoscópica, enfermedades cardiovasculares, enfermedad de Parkinson y tabaquismo. Las conclusiones de este trabajo indican las condiciones de aplicabilidad de los estimadores analizados del odds ratio en función de la magnitud del efecto poblacional, de la variabilidad entre estudios, del tamaño del meta-análisis y de los tamaños muestrales de tales estudios.
Palabras clave: bioestadística, ensayos clínicos, medicina, tamañodel efecto.
Meta-analysis integrates information from different studies to generate a common response to a determined problem. In the literature, we find several integration methods of results, with the integration method of levels of probability being the more basic and, with a greater complexity, the integration method of the effect size, which uses fixed and random effect models. In this study, we compare the results of two estimation methods of the effect size based on meta-analysis using fixed and random effect models. The measure of the effect size considered here is the odds ratio, due to this measure is frequently used in systematic reviews of several topics of interest in public health, such as heart diseases, laparoscopic colectomy, Parkinson disease, tobacco addiction and uterine cervical cancer. Conclusions of this work indicate the applicability conditions of the analyzed estimators of the odds ratio in function of the size of the population effect, of the variability among studies, of the size of the meta-analysis and of the sample sizes of such studies.
Key words: Biostatistics, Clinical trials, Effect size, Medicine.
Referencias
1. Abrams, K. & Jones, D. (1995), 'Meta-analysis and the synthesis of evidence', IMA Journal of Mathematics Applied in Medicine and Biology 12, 297-313. [ Links ]
2. Allam, M. F., Del Castillo, A. S. R. & Navajas, F. (2003), 'Enfermedad de Parkinson temprana y tabaco: metanálisis', Revista de Neurología 12, 1101-1103. [ Links ]
3. Bax, L., Yu, L. M., Ikeda, N., Tsuruta, H. & Moons, K. G. M. (2006), 'Enfermedad de Parkinson temprana y tabaco: metanálisis', BMC Medical Research Methodology 6. [ Links ]
4. Berlin, J. A., Laird, N. M., Sacks, H. S. & Chalmers, T. (1989), 'A comparison of statistical methods for combining event rates from clinical trials', Statistics in Medicine 8, 141-151. [ Links ]
5. Burguillo, F. J., Martín, F. J., Barrera, I. & Bardsley, W. G. (2010), 'Meta-analysis of microarray data: The case of imatinib resistance in chronic myelogenous leukemia', Computational Biology & Chemistry 34, 184-192. [ Links ]
6. Catalán, M. (2003), Los modelos multinivel como herramienta de análisis de datos biomédicos con estructura jerárquica, Tesis doctoral, Universidad de Salamanca, Departamento de Estadística. [ Links ]
7. Catalán, M. & Galindo, M. P. (2003), 'Utilización de los modelos multinivel en investigación sanitaria', Gaceta Sanitaria 17(3), 35-52. [ Links ]
8. Claros, N., Manterota, C., Vial, M. & Sanhueza, A. (2007), 'Efectividad de la profilaxis antibiótica en el curso de la colecistectomía laparoscopica electiva. Revisión sistemática de la literatura', Revista Chilena de Cirugía 59, 353-359. [ Links ]
9. Coomarasamy, A., Papaioannou, S., Gee, H. & Khan, K. S. (2001), 'Aspirin for the prevention of preeclampsia in women with abnormal uterine artery doppler: A meta-analysis', Obstetrics and Gynecology 98, 861-866. [ Links ]
10. Cornelissen, V. A. (2007), 'Incidence of cardiovascular events in white-coat, masked and sustained hypertension versus true normotension: a metaanalysis', Journal of Hypertension 25, 2193-2198. [ Links ]
11. Dentali, F., Douketis, J. D., Lim, W. & Crowther, M. (2007), 'Combined aspirin-oral anticoagulanttherapy compared with oral anticoagulant therapy alone among patients at risk for cardiovascular disease: A meta-analysis of randomized trials', Archives of Internal Medicine 167, 117-124. [ Links ]
12. DerSimonian, R. & Laird, N. (1986), 'Meta-analysis in clinical trials', Controlled Clinical Trials 7, 177-188. [ Links ]
13. Egger, M., Smith, G. D. & Phillips, A. N. (1997), 'Meta-analysis: Principles and procedures', British Medical Journal 315, 1533-1537. [ Links ]
14. Engels, E. A., Schmid, C. H., Terrin, N., Olkin, I. & Lau, J. (2000), 'Heterogeneity and statistical significance in meta-analysis: An empirical study of 125 meta-analyses', Statistics in Medicine 19, 1707-1728. [ Links ]
15. Glass, G. V. (1976), 'Primary, secondary and meta-analysis of research', Educational Researcher 6, 3-8. [ Links ]
16. Greenland, S. & Salvan, A. (1990), 'Bias in the one-step method for pooling study results', Statistics in Medicine 9, 247-252. [ Links ]
17. Hedges, L. V. & Vevea, J. L. (1998), 'Fixed and random effects in meta-analysis', Psychological Methods 3, 486-504. [ Links ]
18. Hedges, L. & Olkin, I. (1985), Statistical Methods for Meta-Analysis, Academic Press, New York. [ Links ]
19. Jiménez-Ruiz, C., Riesco, J. A., Ramos, A. & Barrueco, M. (2008), 'Recomendaciones para el tratamiento farmacológico del tabaquismo. Propuestas de financiación', Archivos de Bronconeumología 44, 213-219. [ Links ]
20. Leyland, A. H. & Goldstein, H. (2001), Multilevel Modelling of Health Statistics, Wiley, New York. [ Links ]
21. Martín, J. (1995), Métodos estadísticos en meta-análisis, Ph.D. Thesis, Universidad de Salamanca, España. [ Links ]
22. Martín, J., Donaldson, A. N. A., Villarroel, R., Parmar, M. K. B., Ernst, E. & Higginson, I. J. (2002), 'Efficacy of acupuncture in asthma: Systematic review and meta-analysis of published data from 11 randomised controlled trials', European Respiratory Journal 20, 846-852. [ Links ]
23. Moher, D., Cook, D. J., Eastwood, S., Olkin, I. & Rennie, D. (1994), 'Improving the quality of reporting of meta-analysis of randomized controlled trials: the quorom statement', Lancet 354, 1896-1900. [ Links ]
24. Moher, D., Liberati, A., Tetzlaff, J. & Altman, D. J. (2009), 'The prisma group', PLoS Medicine 6, 1-6. [ Links ]
25. Moores, L., Jackson, W., Shorr, A. & Jackson, J. (2004), 'Meta-analysis: Outcomes in patients with suspected pulmonary embolism managed with computed tomographic pulmonary angiography', Annals of Internal Medicine 141, 866-875. [ Links ]
26. Petitti, D. B. (1994), Meta-Analysis, Decision Analysis, and Cost-effectiveness Analysis, University Press, Oxford. [ Links ]
27. Peto, R. (1987), 'Why do we need systematic overviews of randomized trials?', Statistics in Medicine 6, 233-240. [ Links ]
28. Pocock, S. T. & Hughes, M. D. (1990), 'Estimation issues in clinical trials and overviews', Statistics in Medicine 9, 657-671. [ Links ]
29. Rodríguez, G. (2002), 'Entendiendo los diagramas de odds-ratio de las revisiones sistemáticas', CES Medicina 16, 66-72. [ Links ]
30. Rosenberg, M., Adams, D. C. & Gurevitch, J. (2000), MetaWin 2.0: Statistical software for meta-analysis, Sinauer Associates, Sunderland. [ Links ]
31. Rosenthal, R. (1984), Meta-Analytic Procedures for Social Research, Sage, Beverly Hills. [ Links ]
32. Rydzewska, L., Tierney, J., Vale, C. L. & Symonds, P. R. (2010), 'Neoadjuvant chemotherapy plus surgery versus surgery for cervical cancer', Cochrane Database of Systematic Reviews 1. CD007406. [ Links ]
33. Schmidt, F. L., Oh, I. & Hayes, T. L. (2009), 'Fixed- versus random-effects models in meta-analysis: Model properties and an empirical comparison of differences in results', British Journal of Mathematical & Statistical Psychology 62, 97-128. [ Links ]
34. Stowe, R. L., Ives, N. J., Clarke, C. & van Hilten, J. (2008), Tratamiento con agonistas dopaminérgicos para la enfermedad de Parkinson en sus etapas iniciales, Wiley, Oxford. (Revisión Cochrane traducida). [ Links ]
35. Thompson, S. G. & Pocock, S. J. (1991), 'Can meta-analysis be trusted?', The Lancet 338, 1127-1130. [ Links ]
36. Turner, R. M., Omar, R. Z., Yang, M., Goldstein, H. & Thompson, S. G. (2000), 'A multilevel model framework for meta-analysis of clinical trials with binary outcomes', Statistics in Medicine 19, 3417-3432. [ Links ]
37. Urrutia, G., Tort, S. & Bonfill, X. (2005), 'Metaanálisis (quorum)', Medicina Clínica 125(1), 32-37. [ Links ]
38. Vamvakas, E. C. (2011), Meta-analysis: A statistical method to integrate information provided by different studies, 'Evidence Based Pathology and Laboratory Medicine', Springer, New York, p. 149-171. [ Links ]
39. Yusuf, S., Peto, R., Lewis, J., Collins, R. & Sleight, P. (1985), 'Beta blockade during and after myocardial infarction: An overview of randomized trials', Progress in Cardiovascular Diseases 27, 335-371. [ Links ]
40. Zhou, H., Zhang, J., Wang, Q. & Hu, Z. (2009), 'Meta-analysis: Antibiotic prophylaxis in elective laparoscopic cholecystectomy', Alimentary Pharmacology and Therapeutics 29, 1086-1095. [ Links ]
Este artículo se puede citar en LaTeX utilizando la siguiente referencia bibliográfica de BibTeX:
@ARTICLE{RCEv35n2a02,
AUTHOR = {Catalán, Mónica and Galindo, M. Purificación and Martín, Javier and Leiva, Víctor},
TITLE = {{Métodos de integración de odds ratio basados enmeta-análisis usando modelos de efectos fijos y aleatorios útiles en saludpública}},
JOURNAL = {Revista Colombiana de Estadística},
YEAR = {2012},
volume = {35},
number = {2}, ]]>
pages = {205-222}
}