Maternal and Infant Risk Factors Associated with Adverse Kindergarten Outcomes Haihong Lia, Changxin

1
Maternal and Infant Risk Factors Associated
with Adverse Kindergarten Outcomes Haihong Lia,
Changxing Maa, Samuel Wua,, PV Raoa ,Wei Cuib,
Yiwei Tangb, Li Yana, Mario Arietc, Ed Feaverd,
Michael Resnickb, Jeffrey Rothb aDepartment of
Statistics, bPediatrics, cMedicine, University of
Florida, Gainesville, Florida, dChiles Center,
University of South Florida
RESULTS
SUMMARY
BACKGROUND
Relative Risks of Adverse Kindergarten Outcomes
There is a paucity of population-based,
longitudinal studies that examine the
relationship between mothers and infants health
conditions during pregnancy, delivery, and birth,
and childrens subsequent educational functioning
in kindergarten

• A mother whose pregnancy or delivery was funded
  by Medicaid (based on family income less than or
  equal to 185 of the federal poverty level) was
  more likely to have a child with three of four
  adverse kindergarten outcomes
• Sixteen socio-demographic, behavioral and
  pregnancy/newborn health factors were also
  significantly related to four adverse
  kindergarten outcomes

OBJECTIVE
To ascertain differences in childrens
functioning in kindergarten, adjusting for
differences in family socio-demographic
characteristics, mothers health status at
delivery, and childrens health status at birth
CONCLUSIONS
STUDY SAMPLE

• The association between poverty during pregnancy
  and birth and childrens poor academic
  functioning at kindergarten was confirmed after
  taking into account initial differences in
  mothers and infants health status
• Mothers health status at delivery, infants
  health status at birth, and familys
  socio-demographic characteristics need to be
  considered when assessing childrens school
  performance

Data from Floridas Vital Statistics (1996 birth
cohort) and Agency for Health Care Administration
(Medicaid beneficiaries) were linked to Floridas
Department of Education kindergarten data
(2001-2002)
OUTCOMES

• Four kindergarten outcomes were considered,
  controlling for18 explanatory factors
• readiness to start school
• exceptional student education
• suspension in and out of school
• retention

POLICY IMPLICATIONS

• Programs to improve kindergarten outcomes would
  require
• moving families out of poverty
• reducing smoking during pregnancy
• Increasing high school graduation rates
• increasing inter-pregnancy interval
• improving pregnant womens pregnancy health
  status
• improving newborns health status

METHODS
Blank cells indicate that risk factor was not
statistically significant and was not involved in
any significant two-way interaction with other
main effects at the 0.01 level.

• Generalized linear models, assuming Poisson error
  distribution and using the log of the total
  number of children in each cell as an offset,
  were fitted using the GENMOD procedure of SAS
• Method modeled the log of the probability of
  kindergarten outcome as a linear function of
  socio-demographic and biomedical variables
• Stepwise model building with backward selection
  and forward selection was employed
• Adjusted relative risks (RR) and 95 confidence
  interval were estimated for each control
  variable, adjusting for all other confounding
  factors

ACKNOWLEDGEMENT
The authors thank Debby Walters, Florida Agency
for Health Care Administration, Dan Thompson and
Carol Graham, Florida Department of Health for
their assistance, cooperation, and advice.