Help viewing high resolution images Return to article

(Downloading may take up to 30 seconds.
If the slide opens in your browser, select File -> Save As to save it.)

Figure 3. Best-fitting model for wine effect (R²=0.54 versus R²=0.27 for the linear model with a positive linear term; P=0.34), using dose-response curves in 7 prospective studies. Parameters of the model were ß₁=-9.9±4.4x10^-4 (P=0.042) and ß₂=0.0067±0.0023x10^-4 (P=0.013). The best-fitting model using data from the 3 case-control studies was a quadratic model that was not statistically significant with a positive linear term (P=0.16) and a negative quadratic term (P=0.091). Horizontal lines represent the 95% CI.

Return to article