Skip to main content
. Author manuscript; available in PMC: 2009 Sep 16.
Published in final edited form as: Ann Intern Med. 2009 Jun 16;150(12):821–W146. doi: 10.7326/0003-4819-150-12-200906160-00005

APPENDIX TABLE 2.

Sensitivity to Missing Values of Estimated Intervention Effects on Change in Body Weight at 2 Years*

Characteristic Calcium (n=168) Placebo (n=167) Difference P Value
“Major Weight Gain” Scenario: Model assuming subjects with major weight gain (2.27 kg) drop out +0.50 (−0.71 to +1.70) +0.77 (−0.37 to +1.91) −0.27 (−1.66 to +1.12) .70
“Best Case for Calcium” scenario: Model assuming subjects taking calcium who drop out gain no weight, but subjects taking placebo who drop out have major (2.27 kg) weight gain −0.01 (−1.22 to +1.19) +0.78 (−0.36 to +1.92) −0.79 (−2.18 to +0.59) .26
“Worst case for Calcium” scenario: Model assuming subjects taking placebo who drop out gain no weight, but subjects taking calcium who drop out have major (2.27 kg) weight gain +0.46 (−0.75 to 1.67) +0.14 (−1 to 1.29) +0.32 (−1.07 to +1.71) .65
*

Estimated marginal means for weight gain in kg (adjusted for covariates including age, race, and sex) with 95th percentile confidence intervals (in parentheses) reported from imputation analyses.