Table 1.
Participant characteristics for the VCSIP study training and testing dataset and the ELGAN and RICHS validation datasets.
| Characteristic | VCSIP (training and testing) | ELGAN (EPIC validation) | RICHS (450K validation) |
|---|---|---|---|
| Maternal smoking—self-report (%) [ (%)] | |||
| Not exposed | 24 (25) | 356 (89) | 202 (85) |
| Exposed | 72 (75) | 43 (11) | 35 (15) |
| Maternal plasma cotinine (ng/mL)—smokers only () | |||
| Randomization | NA | NA | |
| Mid-gestation | NA | NA | |
| Late-gestation | NA | NA | |
| Maternal race [ (%)] | |||
| American Indian or Alaska Native | 1 (1) | NA | |
| Asian | 0 (0) | NA | |
| Black | 5 (5) | 111 (28) | NA |
| Multiple race | 7 (7) | 14 (4) | NA |
| Other race | 1 (1) | 19 (5) | NA |
| White | 82 (85) | 245 (61) | NA |
| Missing | 0 | NA | |
| Maternal ethnicity [ (%)] | |||
| Not Hispanic or Latino | 91 (96) | 364 (91) | NA |
| Hispanic or Latino | 4 (4) | 34 (9) | NA |
| Missing | 1 | 1 | NA |
| Maternal education [ (%)] | |||
| Master’s degree (MA, MS) and above (PhD, MD) | 8 (8) | 63 (16) | NA |
| Bachelor’s degree (BA, BS) | 14 (15) | 81 (21) | NA |
| Some college, no degree; associate’s degree (AA, AS); trade school | 16 (17) | 92 (24) | NA |
| High school degree, GED or equivalent | 21 (22) | 102 (26) | NA |
| Less than high school | 37 (39) | 51 (13) | NA |
| Missing | 0 | 10 | NA |
| Marital status [ (%)] | |||
| Married or living with a partner | 25 (26) | 310 (78) | NA |
| Single, never married; partnered (boyfriend or girlfriend), not living together | 29 (30) | 74 (19) | NA |
| Widowed; separated; divorced | 42 (44) | 15 (4) | NA |
| Private insurance [ (%)] | |||
| No | 62 (65) | 137 (34) | NA |
| Yes | 34 (35) | 262 (66) | NA |
| Gestational diabetes [ (%)] | |||
| No | 84 (88) | 362 (93) | NA |
| Yes | 12 (13) | 28 (7) | NA |
| Missing | 0 | 9 | NA |
| Gestational hypertension [ (%)]a | |||
| No | 96 (100) | 370 (93) | NA |
| Yes | 0 (0) | 29 (7) | NA |
| Missing | 0 | 0 | NA |
| Preeclampsia [ (%)]a | |||
| No | 96 (100) | 326 (82) | NA |
| Yes | 0 (0) | 73 (18) | NA |
| Missing | 0 | 0 | NA |
| Child sex [ (%)] | |||
| Female | 51 (53) | 189 (47) | NA |
| Male | 45 (47) | 209 (53) | NA |
| Unknown | 0 | 1 | NA |
| Gestational age [ (%)] | |||
| Mean | 39.21 | 25.68 | 39.36 |
| Median (min, max) | 39 (37, 42) | 26 (23, 27) | 39 (39, 40) |
| Preterm [ (%)]a | |||
| Preterm ( or CBI indicates preterm at birth) | 0 (0) | 399 (100) | 0 (0) |
| Term ( or CBI indicates not preterm at birth) | 96 (100) | 0 (0) | 237 (100) |
| Birth weight (g) [ (%)] | |||
| Mean | 3,328 | 832.61 | NA |
| Median (min, max) | 3,359 (2,376, 4,241) | 830 (420, 1,418) | NA |
| Missing | 1 | 0 | NA |
Note: For categorical variables we present the total per cohort above the columns and the and percentage for each category. When data is missing, we do not include the missing observations in calculation of percentages. For the ELGAN dataset (), we were restricted in reporting summary statistics and counts . For the RICHS dataset (), we were limited to metadata available in the GEO repository. AA, associate of arts; AS, associate of science; BA, bachelor of arts; BS, bachelor of science; CBI, child birth information; DNAm, DNA methylation; ELGAN, Extremely Low Gestational Age Newborn; EPIC, Infinium MethylationEPIC array; GA, gestational age; GED, General Educational Development; GEO, Gene Expression Omnibus; IQR, interquartile range; MA, master of arts; max, maximum; MD, doctor of medicine; min, minimum; MS, master of science; NA, not available; PhD, doctor of philosophy; RICHS, Rhode Island Children’s Health Study; VCSIP, Vitamin C to Decrease the Effects of Smoking in Pregnancy on Infant Lung Function.
In VCSIP DNAm analysis (), we excluded placentas from patients with preeclampsia, preterm delivery, gestational hypertension, or collection h after delivery owing to concern of sample degradation.