Huaide Ye, PhD

1
The Impact of WIC on Birth Outcomes in Low And
Medium Income Neighborhoods
Huaide Ye, PhD Ruth Shrock, MS, MPH Corey Y.
Hamilton, MS, RD, LD David Schor, MPH, MD Ohio
Department of Health
AMCHP 2005 Annual Conference February 19 23,
2005 Washington, DC
2
Objective

• The purpose of this presentation is to explore
  the impact of WIC (the Women, Infants, and
  Childrens Supplemental Food Program) on birth
  outcomes in different income neighborhoods.

3
Method of Linking 2001 2002 WIC Prenatal Data,
2002 Birth Data, and 2000 US Census
Individual Level
Aggregated Level
Relationship many to one
Relationship one to one
4
n 2,338
n 4,678
Per Capita Income in 1999 gt 23,132 (gt75
percentile)
n 2,338
Ohio Neighborhood Types by Census Block Group in
2000 Total Number of Block Group N 9,354
5
Neighborhood Types in Franklin County
6
Total Births of Ohio Residents in 2002 147,832
Excluded in the Analyses
7
Number and Percent of Births by WIC Status and
Neighborhood Type
WIC Pregnant Women
All Births
8
Birth Outcome Variables

• Very Low Birth Weight (lt 1,500 g)
• Moderate Low Birth Weight (1,500 g 2,499 g)
• Preterm LBW (lt 2,500 g lt 37 Wks Gestation )
• Small-for-Gestation-Age (SGA) (lt 2,500 g 37
  Wks Gestation)
• Hoffman's Below 10th Birth Weight for
  Gestational Age

9
Very Low Birth Weight (lt 1,500 g)
(n136,065 after excluding missing birth weight
block group cases)
10
Moderate Low Birth Weight (1,500 g 2,499 g)
(n136,065 after excluding missing birth weight
block group cases)
11
Preterm LBW (lt 2,500 g lt 37 Wks Gestation)
(n136,065 after excluding missing birth
weight, gestational age, block group cases)
12
SGA (lt 2,500 g 37 Wks Gestation)
(n136,065 after excluding missing birth
weight, gestational age, block group cases)
13
Hoffman's Below 10th Birth Weight for
Gestational Age
(n135,897 after excluding missing Hoffmans
Index block group cases)
14
Demographic Social Economic Variables

• Plurality (Singleton/Multiple Births)
• Mothers Race Ethnicity
• Mothers Age at Birth
• Mothers Marital Status
• Mothers Education
• Mothers Parity
• Mothers Weight Gain during Pregnancy

15
Multiple Births (Twin, Triplet, Quadruplet, etc.)
(n136,095 after excluding missing block group
cases)
16
Mothers Black Race
(n136,095 after excluding missing block group
cases)
17
Mothers Hispanic Ethnicity
(n135,745 after excluding unknown mothers
ethnicity missing block group cases)
18
Mothers Age
(n136,095 after excluding missing block group
cases)
19
Mother Who is Unmarried
(n136,095 after excluding missing block group
cases)
20
Mothers Education lt 12 Years
(n136,095 after excluding missing block group
cases)
21
Mothers High Parity ( 4 or More Parities)
(n135,847 after excluding missing number of
living or dead previous live births block group
cases)
22
Mothers Weight Gain during Pregnancy
(n131,553 after excluding missing weight gain
cases block group cases)
23
Maternal Infant Medical Variables

• Mothers Previous Preterm Birth
• Mothers Pregnancy Anemia
• Mothers Pregnancy Induced Hypertension
• Mothers Diabetes
• Mothers Complications of Labor/Delivery
• Mothers C-section Status
• Infants Congenital Anomalies
• Infants Delivered in Level III Hospital
• Mothers Transferring Hospital Status
• Infants Transferring Hospital Status

24
Mothers Previous Preterm Birth
(n136,095 after excluding missing block group
cases)
25
Mothers Pregnancy Anemia
(n136,095 after excluding missing block group
cases)
26
Mothers Pregnancy Induced Hypertension
(n136,095 after excluding missing block group
cases)
27
Mothers Diabetes
(n136,095 after excluding missing block group
cases)
28
Mothers Complications of Labor/Delivery
(n136,095 after excluding missing block group
cases)
29
Mothers C-section Status
(n136,095 after excluding missing block group
cases)
30
Infants Congenital Anomalies
(n136,095 after excluding missing block group
cases)
31
Infant Delivered in Level III Hospital
(n135,831 after excluding missing hospital code
block group cases)
32
Mothers Transferring Hospital
(n136,095 after excluding missing block group
cases)
33
Infants Transferring Hospital
(n136,095 after excluding missing block group
cases)
34
Maternal Behavioral Variables

• Maternal Smoking during Pregnancy
• Maternal Drinking Alcohol during Pregnancy
• Inadequate Prenatal Care (Kotelchuck Index)
• Prenatal Care Began at 1st Trimester

35
Maternal Smoking During Pregnancy
(n135,607 after excluding missing cigarette
smoking block group cases)
36
Maternal Drinking Alcohol During Pregnancy
(n135,576 after excluding missing drinking
alcohol block group cases)
37
Maternal Prenatal Care (Kotelchuck Index)
(n131,531 after excluding missing Kotelchuck
Index block group cases)
38
Prenatal Care Began at 1st Trimester
(n134,178 after excluding missing month prenatal
care began block group cases)
39
Results from Multivariate Analyses
40
Logistic Regression of LBW (lt 2,500 g) in Each
Type of Neighborhood

• In low income neighborhoods, non-WIC mothers are
  20 more likely to have LBW babies than WIC
  mothers without control of other variables.
• In medium and high income neighborhoods, non-WIC
  mothers are significantly less likely to have LBW
  babies than WIC mothers without control of other
  variables.

41
List of Controlled Variables

• Complications of Labor
• C-section
• Maternal Infant Transferring Hospital
• Infant Congenital Anomalies
• Infant Delivered in Level III Hospital
• Maternal Smoking Drinking Alcohol
• Prenatal Care (Kotelchuck Index)

• Infant Gender
• Plurality
• Mothers Race Ethnicity
• Mothers Age Education
• Maternal Parity
• Mothers Weight Gain during Pregnancy
• Mothers Previous Preterm Birth
• Mothers Anemia, Hypertension, Diabetes

42
Logistic Regression of LBW (lt 2,500 g) in Each
Type of Neighborhood (to be Continued)

• In low income neighborhoods, after controlling
  all listed variables, non-WIC mothers are 25
  more likely to have LBW babies than WIC mothers
  (a 5 increase in comparison with no control
  model).
• In medium income neighborhoods, with controls,
  non-WIC mothers are 10 more likely to have LBW
  babies than WIC mothers (the direction is totally
  reversed).
• In high income neighborhoods, with controls,
  there is no difference in LBW between WIC
  non-WIC mothers.

43
Conclusion

• Participation in the WIC program has a strong
  association with better birth outcomes in low and
  medium income neighborhoods in Ohio.
• The strongest association with birth outcomes is
  seen in women who participated in WIC during
  pregnancy in low income neighborhoods.
• This mostly descriptive study prompts additional
  questions about outreach dose effect of WIC
  investigation of systematic biases and what
  could be learned by linkage to PRAMS.

44
Contact Information
Huaide Ye, Ph.D. Epidemiology Investigator 3 Ohio
Department of Health Bureau of Health Services
Information and Operational Support 246 North
High Street Columbus, Ohio 43216-0118 Phone
[phone] E-Mail hye_at_gw.odh.state.oh.us