Table 4.
Sensitivity analyses assessing potential impact of missing data for recognized risk factors on aaPAF estimates for isolated, simple cases of HLHS and TOF*, NBDPS1997–2011
| Risk factors | aaPAF (%), complete case data† | aaPAF (%), MAR‡ | aaPAF (%), NMAR scenario 1§ | aaPAF (%), NMAR scenario 2|| |
|---|---|---|---|---|
| HLHS | ||||
| Maternal nonHispanic White race | 13.4 | 12.1 | 12.3 | 12.0 |
| Male sex | 29.4 | 28.5 | 28.8 | 28.2 |
| First-degree family history of CHD | 2.8 | 2.9 | 2.9 | 2.9 |
| Pregestational diabetes¶ | 0.4 | 0.6 | 0.6 | 0.5 |
| Prepregnancy overweight–obesity# | 6.5 | 5.6 | 4.3 | 6.8 |
| Periconceptional opioid use** | 1.5 | 1.5 | 1.5 | 1.4 |
| Periconceptional trimethoprim–sulfamethoxazole or nitrofurantoin use** | 0.8 | 0.6 | 0.6 | 0.6 |
| Periconceptional fever†† | 1.4 | 1.5 | 0.6 | 1.9 |
| First-trimester oral contraceptive use‡‡ | 0.8 | 0.8 | 0.8 | 0.8 |
| TOF | ||||
| Male sex | 10.8 | 11.5 | 11.6 | 11.4 |
| First-degree family history of CHD | 3.1 | 3.1 | 3.2 | 3.1 |
| Nulliparous§§ | 4.6 | 4.6 | 4.5 | 4.7 |
| Maternal age 35 y or older | 4.3 | 4.2 | 4.2 | 4.2 |
| Pregestational diabetes¶ | 1.9 | 1.9 | 1.9 | 1.8 |
| Prepregnancy overweight–obesity# | 8.3 | 7.7 | 6.4 | 8.8 |
| Use of assisted reproductive technology or clomiphene citrate for help becoming pregnant|||| | 1.2 | 1.0 | 1.1 | 1.0 |
| Periconceptional opioid use** | 1.3 | 1.0 | 1.1 | 1.0 |
| Periconceptional selective serotonin reuptake inhibitors** | 1.5 | 1.5 | 1.5 | 1.5 |
Botto LD, Lin AE, Riehle-Colarusso T, Malik S, Correa A, National Birth Defects Prevention Study. Seeking causes: Classifying and evaluating congenital heart defects in etiologic studies. Birth Defects Res A Clin Mol Teratol. 2007 Oct; 79(10):714–27.
Cases and controls with complete information on all risk factors.
Missing data imputed assuming no association between unobserved missing covariates and case and control status.
Missing data imputed assuming that cases are 50% less likely than controls to have a positives value for the missing covariate.
Missing data imputed assuming that cases are 50% more likely than controls to have a positives value for the missing covariate.
Type I diabetes diagnosed any time or Type II diabetes diagnosed before the pregnancy.
BMI ≥25 kg/m2.
Any report of use between the month before through the third month of pregnancy.
Maternal report of fever between the month before through the third month of pregnancy.
Any reported use in between the first through the third month of pregnancy.
No previous pregnancies.
Use of intracytoplasmic sperm injection, in vitro fertilization, gamete intrafallopian transfer, zygote intrafallopian transfer, or clomiphene citrate.