Table 3 Regression models according to pre-pregnancy BMI categories.

From: The pattern of gestational weight gains among Chinese women: a repeated measure analysis

Estimate

Regression model

Underweight

Mean

−0.211 + 0.132*point1 + 0.604*point2 + 0.388*point3

SD

0.359 + 0.206*point1 + 0.088*point2 + 0.061*point3

Normal weight

Mean

−0.259 + 0.134*point1 + 0.590*point2 + 0.392*point3

SD

0.325 + 0.187*point1 + 0.093*point2 + 0.069*point3

Overweight

Mean

−0.246 + 0.151*point1 + 0.546*point2 + 0.434*point3

SD

1.529–0.054*point1 + 0.105*point2–0.051*point3

  1. Note: Using equations of mean and SD one can compute any desired percentiles using relation Pth percentile = mean + K*SD, where K is normal equivalent deviate (z score) corresponding to particular percentile, for example, K = 1.64 for 95th percentile and −1.64 for 5th percentile, and SDs in this equation are predicted estimates from the regression model.
  2. For example, to calculate GWG at 24 weeks for normal weight pregnant woman:
  3. p50 for GWG = (−0.259 + 0.134*5 + 0.590*10 + 0.392*0) + (0 × (0.325 + 0.187*5 + 0.093*10 + 0.069*0)).
  4. p5 for GWG = (−0.259 + 0.134*5 + 0.590*10 + 0.392*0) + (−1.64 × (0.325 + 0.187*5 + 0.093*10 + 0.069*0)).
  5. p95 for GWG = (−0.259 + 0.134*5 + 0.590*10 + 0.392*0) + (1.64 × (0.325 + 0.187*5 + 0.093*10 + 0.069*0)).
  6. Abbreviations: BMI, body mass index; SD, standard deviation.
Back to article page