Pooled analysis

A pooled analysis is a statistical technique for combining the results of multiple epidemiological studies. It is one of three types of literature reviews frequently used in epidemiology, along with meta-analysis and traditional narrative reviews. Pooled analyses may be either retrospective or prospective.[1] It is often used when the results of individual studies do not allow for a firm conclusion to be drawn.[2] Unlike meta-analyses, pooled analyses can only be conducted if the included studies used the same study design and statistical models, and if their respective populations were homogeneous. If individual-level data from the included studies is available, the result of a pooled analysis can be considered more reliable.[3]

References
  1. Friedenreich, Christine M (2002-02-01). "Commentary: Improving pooled analyses in epidemiology". International Journal of Epidemiology. 31 (1): 86–87. doi:10.1093/ije/31.1.86. ISSN 0300-5771. PMID 11914300.
  2. Taioli, Emanuela; Bonassi, Stefano (2003-01-01). "Pooled analysis of epidemiological studies involving biological markers". International Journal of Hygiene and Environmental Health. 206 (2): 109–115. doi:10.1078/1438-4639-00198. ISSN 1438-4639. PMID 12708232.
  3. Blettner, M (1999-02-01). "Traditional reviews, meta-analyses and pooled analyses in epidemiology". International Journal of Epidemiology. 28 (1): 1–9. doi:10.1093/ije/28.1.1. ISSN 0300-5771. PMID 10195657.


This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.