Background. The effective use of knowledge related to gene-disease associations relies on the optimal conduct and reporting of research studies. Transparent reporting enables readers to identify the strengths and weaknesses of research studies and subsequently to determine the quality of evidence that they offer.
Objective. To investigate the relation between methodological characteristics and the direction and magnitude of effects observed in case-control studies of gene-disease associations.
Methods. Articles were randomly selected from a database of published studies on genetic associations and other epidemiologic research pertaining to the human genome (the HuGE Literature Finder). The analysis evaluated 511 articles indexed in HuGENet in 2007. Gene-gene interaction studies and gene-environment interaction studies were excluded. Univariate and multivariate meta-regression analyses using a random effects model were used to assess the relationship between methodological characteristics and the direction and magnitude of genetic associations.
Results. The studies included in the analysis had been conducted in a total of 52 countries and were published in 220 journals (median impact factor 5.1). The multivariate meta-regression model of methodological characteristics was able to account for 17.2% of the between-study variance in the magnitude of gene-disease associations. Of the factors included in the multivariate regression model, the following tended to be associated with a smaller magnitude of effect: replication of a previously conducted study; nested case-control study design; individual matching of study participants; and reporting of sample size or of statistical power calculation. Moreover, the magnitude of effect tended to decrease as the number of controls increased. By contrast, studies that did not report the process of selecting control participants or that reported that the genotype distribution departed from Hardy-Weinberg equilibrium tended to be associated with larger magnitudes of effect. Studies conducted in North America, Europe, and China had similar magnitudes of effect.
Conclusion. Within the constraints of limitations in reporting, a number of methodological features of gene-disease association studies are associated systematically with the magnitude of effect observed. This provides evidence to direct efforts to improve the reporting of research on genetic associations.
| Identifer | oai:union.ndltd.org:uottawa.ca/oai:ruor.uottawa.ca:10393/28046 |
| Date | January 2009 |
| Creators | Aljasir, Badr Abdulrahman |
| Publisher | University of Ottawa (Canada) |
| Source Sets | Université d’Ottawa |
| Language | English |
| Detected Language | English |
| Type | Thesis |
| Format | 136 p. |
Page generated in 0.0016 seconds