Pregnancy Factors Associated with Low Birth Weight and Infant Birth Defects, Maltreatment, and Morta

1
Pregnancy Factors Associated with Low Birth
Weight and Infant Birth Defects, Maltreatment,
and Mortality

• Paper Presented to 26th International Congress of
  Applied Psychology
• Athens, Greece
• July 20, 2006
• Michael B. Resnick, Ed.D
• Currently Adjunct Professor of Pediatrics
• Founding Director of Maternal Child Health and
  Education Research and Data Center
• University of Florida, College of Medicine,
  Department of Pediatrics
• Gainesville, Florida 32610-0296
• USA
• Phone Number 352-334-1360
• Website https//www.cms.peds.ufl.edu
• Email MResnick_at_ufl.edu

2
Co-Authors Collaborators
Pregnancy Factors and Low Birth Weight and Infant
Outcomes 26th International Congress of Applied
Psychology

• Mario Ariet, PhD Professor of Medicine -
  Computer Science
• Chang Xing Ma, PhD Assistant Professor of
  Statistics Biostatistician, Maltreatment
• Yiwei Tang, MD, MS Assistant Professor of
  Pediatrics Data Manager, Biostatistics, Birth
  Defects
• Vivian Chang, MD Coord. Computer Applications
  Data Manager, Birth Defects
• Terry Wei, MD, MS Assistant Professor of
  Pediatrics Biostatistics
• Rajeeb Das, MPH Data Manager, Birth Vital
  Statistics Medicaid Eligibility
• Jeff Roth, PhD Associate Professor of
  Pediatrics Program Evaluation
• Randy Carter, PhD Professor of Statistics
  Biostatistician
• Li Yan, PhD Assistant Professor of Statistics
  System Administrator
• Sam Wu, PhD Associate Professor of Statistics
  Data Manager
• PV Rao Adjunct Professor of Statistics
• Julia Ackerman, MPH Research Assistant
  Technical Assistant
• Floridas Departments of Health, Children and
  Families, Education, and Agency for Health Care
  Administration

MCHERDC, University of Florida, College of
Medicine, Department of Pediatrics, Gainesville
Florida, USA
3
Description of Study
Pregnancy Factors and Low Birth Weight and Infant
Outcomes 26th International Congress of Applied
Psychology

• Approx. 1.2 million births to Florida residents
  between 1996-2001
• 535,045 (46) received prenatal care, delivery,
  and postpartum services financed by Medicaid
  (income level lt 185 of Federal Poverty Level)
• Merged pregnancy and delivery records for all
  mothers who gave birth in Florida between
  1996-2001
• Identified sociodemographic, behavioral, health,
  and educational characteristics of these mothers,
  from information obtained from Florida Birth
  Vital Statistics
• Analyzed 6 birth cohorts between 1996-2001to
  ascertain the effect of maternal, birth and
  infant characteristics on 4 birth outcomes
• Low Birth Weight (LBW)
• Birth Defects (BD)
• Maltreatment (M)
• Mortality (MO)

MCHERDC, University of Florida, College of
Medicine, Department of Pediatrics, Gainesville
Florida, USA
4
Predictive Factors
Pregnancy Factors and Low Birth Weight and Infant
Outcomes 26th International Congress of Applied
Psychology

• Pregnancy Health Status variable is an index
  of maternal medical condition number of
• health problems (diagnoses and procedures)
  associated with pregnancy and/or delivery
• and whether delivery was by Cesarean Section
  as reported on Floridas Birth Certificate
• - Functions as a single measure
  indicative of a womans health at time of delivery

MCHERDC, University of Florida, College of
Medicine, Department of Pediatrics, Gainesville
Florida, USA
5
Outcomes
Pregnancy Factors and Low Birth Weight and Infant
Outcomes 26th International Congress of Applied
Psychology

• Definitions
• Very Low Birth Weight equals a birth weight less
  than 1500 grams
• Low Birth Weight equals a birth weight less than
  2500 grams
• Infant Mortality all live births who then die
  within the first year of life
• Birth Defect one of the 45 ICD-9 (740.0-759.9)
  classifications that US states with a Birth
  Defects Registry report annually to the US
  Centers for Disease Control and Prevention. Data
  Reported here is from Floridas Birth Defects
  Registry
• Verified infant maltreatment being reported to
  the Department of Children and Families,
  investigated, and determined by being supported
  by evidence that can be substantiated (such as a
  pediatricians testimony/evaluation) and meets
  the criteria described by law
• Maltreatment includes the following 3 categories
• Abuse any willful act that results in physical,
  mental, or sexual injury that causes or is likely
  to cause the childs physical, mental, or
  emotional health to be significantly impaired
• Neglect any failure or omission by a caretaker
  to provide the care, supervision, services or
  protection necessary to maintain the childs
  physical and mental health
• Threatened Harm a willful act that is
  intrinsically harmful or dangerous which could
  clearly and immediately result in injury or harm

MCHERDC, University of Florida, College of
Medicine, Department of Pediatrics, Gainesville
Florida, USA
6
Statistical Methods
Pregnancy Factors and Low Birth Weight and Infant
Outcomes 26th International Congress of Applied
Psychology

• Analyzed the relationship between 13 pregnancy
  related sociodemographic, behavioral, and health
  factors and 4 birth and infant outcomes for the
  combined 1996-2001 birth cohort by using
  categorical generalized linear models
• Response variable and log linear function is the
  number of adverse outcomes assuming Poisson error
  distribution (rare events)
• Using the log of the total number of children in
  each cell as an offset
• Fitted using the GENMOD (General Linear Model)
  Procedures of SAS (Statistical Analysis System)

MCHERDC, University of Florida, College of
Medicine, Department of Pediatrics, Gainesville
Florida, USA
7
Statistical Methods
Pregnancy Factors and Low Birth Weight and Infant
Outcomes 26th International Congress of Applied
Psychology

• Approach modeled the log of the mean rate of each
  outcome as a linear function of pregnant womens
  socio-demographic, behavioral, and health related
  characteristics
• Stepwise model building with backward selection
  was employed
• Starting with a model including all main effects
  and two-way interactions of these factors

MCHERDC, University of Florida, College of
Medicine, Department of Pediatrics, Gainesville
Florida, USA
8
Statistical Methods
Pregnancy Factors and Low Birth Weight and Infant
Outcomes 26th International Congress of Applied
Psychology

• After controlling for all other variables in the
  model, the relative risk (RR) for each outcome
  was analyzed,
• RR is the risk of an adverse pregnancy/birth/infan
  t outcome for a factor relative to the risk of
  the same outcome
• RRs that were found to be significantly different
  from 1, at the p.01 level, were identified as
  indicators of significant differences in risks
• RRs significantly different from 1 implies that
  the risk in the category in question differed
  from the risk in the corresponding reference
  category

MCHERDC, University of Florida, College of
Medicine, Department of Pediatrics, Gainesville
Florida, USA
9
Results Descriptive Statistics
Pregnancy Factors and Low Birth Weight and Infant
Outcomes 26th International Congress of Applied
Psychology
MCHERDC, University of Florida, College of
Medicine, Department of Pediatrics, Gainesville
Florida, USA
10
Overall Statistically Significant Relative Risk
Results
Pregnancy Factors and Low Birth Weight and Infant
Outcomes 26th International Congress of Applied
Psychology
MCHERDC, University of Florida, College of
Medicine, Department of Pediatrics, Gainesville
Florida, USA
11
Example of Application
Pregnancy Factors and Low Birth Weight and Infant
Outcomes 26th International Congress of Applied
Psychology
MCHERDC, University of Florida, College of
Medicine, Department of Pediatrics, Gainesville
Florida, USA
12
Conclusions
Pregnancy Factors and Low Birth Weight and Infant
Outcomes 26th International Congress of Applied
Psychology

• There is a significant relationship between
  mothers Pregnancy Health Status,
  sociodemographic, behavioral, and educational
  characteristics, to 4 adverse birth and infant
  outcomes (LBW, BD, M, MO)
• These 13 pregnancy factors, associated
  characteristics must be accounted for or
  controlled for when using these outcomes as
  Health Status Indicators
• There is a similar pattern of association between
  the pregnancy factors and the 4 birth/infant
  outcomes
• Current and improved statistical modeling
  strategies, improved software operating systems,
  hardware and storage capability, improved
  deterministic merging algorithm enables
  population based studies to be undertaken which
  accounts for and controls for the independent
  relationship of multiple variables on independent
  outcomes, rather than just reporting raw rates

MCHERDC, University of Florida, College of
Medicine, Department of Pediatrics, Gainesville
Florida, USA
13
Conclusions
Pregnancy Factors and Low Birth Weight and Infant
Outcomes 26th International Congress of Applied
Psychology

• The following pregnancy risk factors are
  associated with either infant LBW, BD, M, or MO.
  The more of these risk factors a pregnant women
  has, the greater the risk of an adverse birth or
  infant outcomes
• Mothers with low incomes
• Mothers who have a health, behavioral, (or both)
  problems during pregnancy
• Mothers with less than a high school education
• Mothers without a permanent partner
• Mothers who smoke during pregnancy
• Mothers who drank during pregnancy
• Mothers who do not have adequate prenatal care
• Infants whose father is not identified on the
  birth certificate
• Mothers with multiple births
• Mother over 35 and under 19

MCHERDC, University of Florida, College of
Medicine, Department of Pediatrics, Gainesville
Florida, USA
14
Discussion
Pregnancy Factors and Low Birth Weight and Infant
Outcomes 26th International Congress of Applied
Psychology

• Identification of these factors during pregnancy,
  delivery/birth or the immediate postpartum period
  could be used as an early screening method for
  referral for assessment of infants for early
  developmental delay or disability
• Since neglect and threatened harm are the
  majority of maltreatment cases reported in
  infants, early intervention and/or case
  management services for these families may be
  very beneficial
• Interpregnancy medical health (Family Planning),
  social, behavioral, educational, and financial
  intervention may be very beneficial in reducing
  adverse pregnancy outcomes in subsequent
  pregnancies
• Nations, States, and/or Cities, birth and infant
  outcomes must be presented based on comparable
  sub-population data in order to make meaningful
  comparisons and policy decisions
• Improving the health status of women prior to and
  during the interconceptual period must be a
  priority if we want to improve birth and infant
  outcomes

MCHERDC, University of Florida, College of
Medicine, Department of Pediatrics, Gainesville
Florida, USA
15
Current Work
Pregnancy Factors and Low Birth Weight and Infant
Outcomes 26th International Congress of Applied
Psychology

• A pregnancy screening tool is being developed and
  field tested for expectant and new parents to
  identify high risk pregnancies, so
  preconceptional and interconceptional
  interventions may be implemented in order to
  reduce adverse birth and infant outcomes
• Toddler, preschool, and school age longitudinal
  follow-up is currently being undertaken to
  ascertain the consequences of these adverse
  outcomes and associated pregnancy risk factors
• Statistical methods are being developed to
  ascertain direct and indirect effects of
  intermediate or intervening factors on all
  outcome measures

MCHERDC, University of Florida, College of
Medicine, Department of Pediatrics, Gainesville
Florida, USA