Effect of ethnicity on glycaemic index: a systematic review and meta-analysis

Abstract

Objectives: Low glycaemic index (GI) foods are recommended to improve glycaemic control in diabetes; however, Health Canada considers that GI food labeling would be misleading and unhelpful, in part, because selected studies suggest that GI values are inaccurate due to an effect of ethnicity. Therefore, we conducted a systematic review and meta-analysis to compare the GI of foods when measured in Caucasians versus non-Caucasians.

Methods: We searched MEDLINE, EMBASE and Cochrane databases for relevant articles. GI differences were aggregated using the generic inverse variance method (random effects model) and expressed as mean difference (MD) with 95% confidence intervals (95% CI). Study quality was assessed based on how well studies complied with official international GI methodology.

Results: Review of 1288 trials revealed eight eligible studies, including 28 comparisons of GI among 585 non-Caucasians and 971 Caucasians. Overall, there was borderline significant evidence of higher GI in non-Caucasians than Caucasians (MD, 3.3 (95% CI, -0.1, 6.8); P=0.06) with significant heterogeneity (I(2), 46%; P=0.005). The GI of eight types of rice was higher in non-Caucasians than Caucasians (MD, 9.5 (95% CI, 3.7, 23.1); P=0.001), but there was no significant difference for the other 20 foods (MD, 1.0 (95% CI, -2.5, 4.6); P=0.57). MD was significantly greater in the four low-quality studies (nine comparisons) than the four high-quality studies (19 comparisons; 7.8 vs 0.7, P=0.047).

Conclusions: With the possible exception of rice, existing evidence suggests that GI values do not differ when measured in Caucasians versus non-Caucasians. To confirm these findings high-quality studies using a wide range of foods are required.

Figure 1

Flow of literature search.

Figure 2

Forest plots of the effect of ethnicity on GI in participants without diabetes. Data are expressed as mean differences with 95% confidence intervals using the generic inverse variance random effects model. Pooled estimate effects are shown as diamonds. Inter-study heterogeneity was tested by Cochrane's Q statistic (χ²-test) at a significance level of <0.10 and quantified by I², where I²⩾50% is considered to be an evidence of substantial heterogeneity. The top 8 foods are rice, the bottom 20 foods are other types of foods.

Figure 3

Funnel plot.