Exploring potential sources of heterogeneity for the INSIG2 rs7566605 association with obesity (Hypotheses 3�C5).

Stated values are ORs (p-values) based on fixed or random effects models and measures of I2 (p-value from Q statistics) for each group and p-values testing for difference between ORs of the two corresponding groups. (Not for HP, OB, NC, or CH studies due to the few number of studies.)aStudies with percentage of subjects with BMI ��30��18% or bStudies with mean age ��50 years or cStudies with BMI assessment after/during year 2000 or before year 2000.dExcluding studies published before the response letter by Herbert et al., December 2006 [1], in which the hypothesis of potential heterogeneity due to study design and a first call for a meta-analysis were stated (i.e. excluding American_Polish, NHS, KORA-S4, Essen_trios, EPIC_Norfolk, MRC_Ely, DESIR, SHIP, OB_adult).eTesting for difference of fixed effect OR [random effects OR].OR = Odds Ratio, All-CA = all Caucasian adult studies combined, GP = general population studies.