Racial Differences in Risk Status and in Preterm Delivery among Very Low Risk Women

1
Racial Differences in Risk Status and in Preterm
Delivery among Very Low Risk Women
Kristen Helms, MSPH Nedra S. Whitehead, PhD
MS Division of Reproductive Health, National
Center for Chronic Disease Prevention and Health
Promotion, Centers for Disease Control and
Prevention
2
AcknowledgementPRAMS Working Group

• Alabama Albert Woolbright, PhD, MPH
  Alaska - Kathy Perham-Hester, MS, MPH Arkansas -
  Gina Redford, MAP Colorado - Alyson Shupe, Ph.D
  Florida Helen Marshall Georgia Carol Hoban,
  MS, MPH Hawaii Limin Song, MPH, CHES Illinois
  - Theresa Sandidge, MA Louisiana - Joan
  Wightkin Maine - Martha Henson Maryland - Diana
  Cheng, MD Michigan - Yasmina Bouraoui, MPH
  Minnesota Jan Jernell Mississippi - Linda
  Pendleton, LMSW Montana JoAnn Dotson Nebraska
  - Debbi Barnes-Josiah, PhD New Jersey - Lakota
  Kruse, MD New Mexico - Ssu Weng, MD, MPH New
  York State - Anne Radigan-Garcia New York City -
  Fabienne Laraque, MD North Carolina - Paul
  Buescher, PhD North Dakota Sandra Anseth, RN
  Ohio Amy Davis Oklahoma - Dick Lorenz, MS
  Oregon - Ken Rosenberg, MD, MPH Rhode Island -
  Sam Viner-Brown South Carolina Sylvia Sievers,
  PhD Texas - Ramdas Menon, Ph.D Utah - Lois
  Bloebaum Vermont - Peggy Brozicevic Washington
  - Linda Lohdefinck West Virginia - Melissa
  Baker, MA

3
Background

• Demographic and behavioral factors are associated
  with preterm delivery
• Black women are twice as likely to deliver
  preterm as white women
• Very young and older women at higher risk
• Women who smoke at increased risk
• Women from lower socioeconomic class at higher
  risk
• Racial and ethnic differences in the prevalence
  of risk factors
• More black women than white women give birth when
  they are teenagers or unmarried
• Fewer black and Hispanic women have higher
  education and receive prenatal care in the first
  trimester
• More white and Native American women smoke during
  pregnancy
• Little information on overall risk status of
  women giving birth and how it differs by race

4
Research Questions

• 1. Who is Low Risk?
• Examine racial and ethnic differences in womens
  pre-pregnancy risk status based on demographic
  and behavioral factors
• 2. Are there racial differences in preterm
  delivery among low risk women?
• Determine if racial and ethnic differences in
  preterm delivery exist among women who are at
  very low risk for preterm delivery before
  pregnancy based on demographic and behavioral
  factors

5
Sampling and Data Collection MethodsPregnancy
Risk Assessment Monitoring System (PRAMS)

• Sampling
• Population - based
• Frame state birth certificate files
• 2-6 months after delivery
• High-risk women are over sampled
• States annual sample size 1600-3000
• Data Collection
• Dillmans Tailored Design Method
• Questionnaire mailed 2-3 times
• Mail non-responders interviewed by telephone
• Representativeness
• Weighted for sampling design, nonresponse, and
  non- coverage of the frame
• Generalizable to state residents with recent live
  
• births

6
States Included in the Analysis
WA
ME
NY
MI
NE
OH
IL
IN
UT
WV
CO
NC
OK
NM
AR
SC
GA
LA
AK
FL
7
Study and Analysis Methods

• Study population
• Singleton births
• Year 1988 2000 births
• States Included
• Response rates 70
• Data collected on maternal income
• Contributed 1 12 years of data
• Analysis methods
• SUDAAN to calculate population prevalence

8
Definition of Very Low Risk WomenIn Order
Assessed

• Birth Certificate Information
• 18 34 years at sampled birth
• 12 years of education
• Married
• Began prenatal care in first trimester
• Inter-pregnancy interval 6 months
• PRAMS Questionnaire Information
• Nonsmoker in 3 months before pregnancy
• No income from public assistance
• Income at or above median for state and year
• Pre-pregnancy body mass index 19.8 26.0

• Source U.S. Census Bureau, Current Population
  Survey

9
Racial and Ethnic Distribution
Native American includes Alaskan Natives.
Hispanic ethnicity was not available for all
states and years. Analysis of ethnicity was
limited to white women because there were few
Hispanic women of other races.
10
Response Rates and Sample Size

• Response Rates
• Overall 75
• Low risk by birth certificate information 85
• Sample size Very low risk
• Total 13,502 Total
  7,485
• White 12,001 (89) Non-Hispanic
  7,144 (95)
• Black 733 ( 5) Hispanic
  341 ( 5)
• Native American 230 ( 2)
• Asian Pac Islander 467 ( 4)

Available ethnicity only.
11
Distribution of Risk Status among PRAMS States
12
Risk Status by Race and Ethnicity
Very Low Risk (Birth Certificate and
Questionnaire Variables)
Low Risk (Birth Certificate Variables Only)
13
Population Percentage of Reason for Exclusion
from Low Risk Status, In Order Evaluated
14
Who Is Low Risk?

• 9.6 of women who had a live birth met the
  classification for very low risk
• Proportion of women who were very low risk
  differed by race and ethnicity
• 2 for Native Americans to 12 for Asian Pacific
  Islanders
• Most frequent reason women were not classified as
  very low risk
• Marital status (17)
• Maternal age (17)
• Education (16)
• Income level (12)
• Primary reason for exclusion varied by race and
  ethnicity
• White non-Hispanic women maternal age (16)
• White Hispanic women education (38)
• Black and Native American women marital status
  (39 and 25, respectively)
• Asian Pacific Islanders income level (20)

15
Prevalence of Preterm Delivery Among All Women by
Race and Ethnicity
16
Prevalence of Preterm Delivery by Risk Status
17
Distribution of Gestational Age among Very Low
Risk Women
18
Gestational Age Among Preterm Births of
Very Low Risk Women
19
Are there racial differences in preterm delivery
among very low risk women?

• Racial differences in the prevalence of preterm
  delivery exist even among very low risk women
• The pattern of racial differences is different
  among very low risk women and higher risk women
• Very low risk women are less likely to have a
  preterm birth than higher risk women among
  all racial and ethnic groups

20
Limitations

• Data were not available on some important risk
  factors
• Pre-pregnancy health problems such as diabetes or
  hypertension
• Maternal drug use
• Income level and maternal BMI were missing for 8
  to 16 of respondents
• Amount of missing data varied by race
• May have resulted in
• Underestimation of the proportion of women who
  were very low risk
• Overestimation of preterm delivery for some
  racial groups
• Questions on income
• Varied considerably
• Different time periods used
• Some used household income and some family income
• Data were categorical
• Estimation of median is imprecise and some women
  whose income was close to the median may be
  misclassified

21
Conclusions

• Few women who give birth can be classified as
  very low risk at the time of conception
• Risk status and risk factors differ greatly by
  race and ethnicity
• Demographic, behavioral or economic risk factors
• do not explain the increased risk of preterm
  delivery among black women
• may explain the increased risk of preterm
  delivery among Native American women

22
Conclusions

• Differences should be considered in research on
  racial differences in pregnancy outcome and in
  planning interventions
• Implication for Preconceptional Care
• Identify those with less than ideal risk profile
• Further research
• Explore differences among racial and ethnic
  groups in the distribution of gestational age
• Explore relationship between race and preterm
  delivery among low risk primiparous women
• Explore relationship between race and preterm
  delivery among very low risk women whose prior
  birth was preterm

23
Thank you!

• Contact information
• Kristen Helms, MSPH
• khelms_at_cdc.gov
• 803-749-5647

24
Additional Information on Bias for Nonresponse
Missing Information
25
Bias Assessment Methods

• Assessed bias due to nonresponse
• Examined prevalence of preterm delivery
• Among women low risk by birth certificate
  information
• Between
• Respondents
• Nonresponsdents
• By race and ethnicity
• Assessed bias due to missing information
• Data missing 10 of respondents
• Level of income
• BMI
• Compared prevalence of preterm delivery
• Among women who were low risk by previous
  variables
• Between
• Women within range for low risk for variable of
  interest
• Women out of range for low risk for variable of
  interest
• Women with missing information for variable of
  interest

26
Bias Assessment

• Nonresponse
• Among white women
• Preterm birth was less prevalent among
  respondents than nonrespondents
• Among black women
• No difference
• Among Native American and Asian Pacific Islander
  women
• Preterm birth more common among respondents than
  nonrespondents
• May cause overestimation of relative differences
  between white and other races
• Missing information
• Among white women
• Those with missing information had higher
  prevalence of preterm birth than those classified
  as very low risk
• Among Black and Asian Pacific Islander women
• Those with missing information had lower
  prevalence of preterm birth than very low risk or
  higher risk women
• Among Native American women
• No consistent pattern
• May cause overestimation of relative differences
  between black or Asian women and white
  women