Table 4

Statistically significant findings for the subgroup analysis based on evidence assessment unit conclusions and type of review. Data from 51 evidence assessment units reported in 33 reviews

OR (95% CI)P value
Evidence assessment units concluding a neutral effect or association of LES on BW (n=7)
Main message of cited articles
 Neutral (no directional effect or association)1 (ref)
 Unable to draw a conclusion from the article1.67 (1.07 to 2.54)0.03
Cited article type
 Systematic review with meta-analysis1 (ref)
 Observational study1.45 (1.06 to 2.02)0.03
Evidence assessment units concluding insufficient evidence to draw a conclusion about the effect of LES on BW (n=26)
Cited article type
 Systematic review with meta-analysis1 (ref)
 Systematic review without meta-analysis1.97 (1.12 to 3.45)0.02
Cited article population
 Adults1 (ref)
 Children1.84 (1.43 to 2.37)<0.001
Cited article journal impact factor, current (2018)*1.10 (1.00 to 1.20)0.049
Systematic reviews (evidence assessment units) without meta-analysis (n=14)
Cited article type
 Systematic review with meta-analysis1 (ref)
 Randomised controlled trial0.61 (0.38 to 1.00)0.04
Cited article population
 Adults1 (ref)
 Both1.86 (1.20 to 2.82)0.004
 Children1.66 (1.20 to 2.29)0.002
  • Logistic mixed-effects regression. The analysis of neutral reviews is additionally adjusted for overdispersion.

  • Bold value indicates result is statistically significant with p<0.05 or lower.

  • *Journal impact factor was base 2 log-transformed, so OR is the change per twofold change in journal impact factor.

  • BW, body weight; CI, confidence interval; LES, low-energy sweeteners; n, sample size; OR, odds ratio; ref, reference variable.