ADHDgene Database

Hot Results

Quick Search

Keyword:
Type:

Large-scale studies

Data Summary

Published Variant
- SNP: 1391
- CNV: 398
- Others: 173
Published Gene: 359
Published Region: 128
Pathway by PBA: 8
Study: 361

Meta-analysis Report

Basic Info

Reference	Cheuk DK, 2006(c)16676224
Citation	Cheuk D. K. and Wong V. (2006) "Meta-analysis of association between a catechol-O-methyltransferase gene polymorphism and attention deficit hyperactivity disorder." Behav Genet, 36(5): 651-9.
Study Type	Candidate association study
Summary	They perform a meta-analysis of previous studies to assess the overall magnitude and significance of the association. Family-based and case-control studies of the association between the COMT gene polymorphism and ADHD were searched systematically and comprehensively. Odds ratios (OR) of association were pooled by the fixed effects model if no significant heterogeneity was present among different studies. Subgroup analysis by gender and ADHD subtypes were also performed. Eleven family-based and two case-control studies were identified. After pooling the results, no significant association between the COMT Val158/108Met polymorphism and ADHD was found. There was also no significant association when the results were stratified by gender or ADHD subtype. There was no significant statistical heterogeneity although clinical heterogeneity was present in the studies, especially the ethnicity of subjects. Sensitivity analysis demonstrated absence of undue influence of any single study. Standard regression analysis showed no significant publication bias.

Detail Info

Samples	They searched through Pubmed, MEDLINE, EMBASE, BioMed Central, Derwest Biotechnology database and China Journet Net with keywords 'attention deficit hyperactivity disorder' and 'catechol-O-methyltransferase.' A total of 12 studies investigating the association between a COMT gene polymorphism and ADHD were identified in the literature search and were included. All studies examined the Val158/108Met polymorphism in clinical samples. Ten were family-based studies (Barr et al. 1999; Bellgrove et al. 2005; Eisenberg et al. 1999; Hawi et al. 2000; Jiang et al. 2005; Manor et al. 2000; Payton et al. 2001; Taerk et al. 2004; Tahir et al. 2000; Turic et al. 2005) and one was a case-control study (Zhang et al. 2003). The 12th study employed both designs (Qian et al. 2003).
Statistic Method	The odds ratio (OR) of a particular polymorphism at a certain locus was used as the primary outcome measure. The data from different types of design and analysis method were analyzed separately first and then combined. If two analyses methods (TDT and HHRR) were used for the same family-based sample, the odds ratio derived from TDT was used in the combined analysis so that the same sample would not be counted twice in the same meta-analysis. TDT was used in preference to HHRR in the final combined analysis since TDT is robust to population stratification whereas HHRR is not. The Comprehensive Meta-analysis software was used to estimate the combined odds ratios with 95% confidence intervals. They adopted a strategy that the random effects model would be used if there was a statistically significant heterogeneity among different studies. Otherwise the fixed effects model would be employed. Heterogeneity among different studies was analyzed by the X² test. A P-value less than 0.1 instead of the usual 0.05 was taken as the cut-off for significant heterogeneity. They employed the most commonly used inverse variance method for the fixed effects model and the DerSimonian and Laird method for the random effects model (Egger and Smith 2001). Publication bias was assessed by the method described by Egger et al. (1997).
Basic Result	After pooling the results, no significant association between the COMT Val158/108Met polymorphism and ADHD was found (OR 0.99 (95% CI: 0.88-1.12), P=0.87). There was also no significant association when the results were stratified by gender or ADHD subtype. There was no significant statistical heterogeneity (X²=12.27, P=0.2) although clinical heterogeneity was present in the studies, especially the ethnicity of subjects. Sensitivity analysis demonstrated absence of undue influence of any single study. Standard regression analysis showed no significant publication bias.

SNPs reported by this study: 1

Genes reported by this study: 1