Skip to main content
. 2017 Dec 15;6:e59. doi: 10.1017/jns.2017.63

Table 5.

Stratified analyses of effects of 100 % fruit juice on the homeostatic model assessment of insulin resistance

Test of heterogeneity Unadjusted (univariate) meta-regression
Parameter Studies (n) Net change 95 % CI Overall P P I2 (%) RC se P
All studies 7 −0·22 −0·5, 0·06 0·13 <0·005 73·9
Fruit juice type
 Apple
 Berry
 Blend
 Citrus 4 −0·11 −0·52, 0·29 0·58 0·02 70·8 0·26 0·44 0·58
 Grape 1 −0·50 −1·21, 0·21 0·17 −0·12 0·64 0·86
 Pomegranate 2 −0·37 −0·57, −0·18 0·00 0·86 0 Reference
Control group
 Sugar bev 5 −0·23 −0·53, 0·07 0·14 0·01 71·9 0·38 0·34 0·33
 Sugar-free bev 1 0·50 −0·16, 1·16 0·14 1·10 0·51 0·10
 No beverage 1 −0·60 −1·03, −0·17 0·01 Reference
Volume of juice             <0·005 <0·005 0·01
 ≤ 250 ml/d 2 −0·34 −1·17, 0·48 0·41 0·86 0
 >250 ml/d 5 −0·21 −0·51, 0·10 0·19 <0·005 82·2
Duration             0·02 0·01 0·08
 2–7 weeks 2 −0·41 −0·59, −0·23 0·00 0·35 0
 ≥8 weeks 5 −0·01 −0·25, 0·23 0·92 0·31 16·9
Baseline FBG
 <5·6 mmol/l 6 −0·22 −0·50, 0·07 0·14 <0·005 78·2 Reference
 ≥5·6 mmol/l 1 −1·60 −15·68, 12·48 0·82 −1·38 7·36 0·86
Location
 Asia 1 −1·60 −15·68, 12·48 0·82 Reference
 Europe 4 −0·28 −0·61, 0·04 0·08 <0·005 83·9 1·31 7·31 0·87
 North America 2 0·01 −0·97, 0·99 0·99 0·04 75·4 1·62 7·31 0·84
Outcome
 Primary 1 −0·37 −0·57, −0·18 <0·005 Reference
 Secondary 6 −0·17 −0·52, 0·18 0·34 0·03 58·8 0·20 0·39 0·64
Study design
 Cross-over 3 −0·22 −0·56, 0·12 0·20 <0·005 85·4 −0·05 0·35 0·90
 Parallel 4 −0·18 −0·83, 0·48 0·60 0·06 59·9 Reference
Jadad score
 <4 4 −0·27 −0·64, 0·10 0·16 0·05 60·8 Reference
 ≥4 3 0 −0·08, 0·09 0·95 0·37 0 0·11 0·34 0·75
Study funding
 Industry support 6 −0·22 −0·50, 0·07 0·14 <0·005 78·2 1·38 7·36 0·86
 No industry support 1 −1·60 −15·68, 12·48 0·82 Reference

RC, regression coefficient; sugar bev, beverage matched for carbohydrate or sugar and/or energy; sugar-free bev, beverage with non-energy-containing or no added sweetener; FBG, fasting blood glucose.