Impact of Family Case Management and WIC on Birth Outcomes of Medicaid Recipients in Illinois 2000-2002

1
Impact of Family Case Management and WIC on Birth
Outcomes of Medicaid Recipients in
Illinois2000-2002

• Stephen E. Saunders, M.D., M.P.H.
• Associate Director for Family Health
• Illinois Department of Human Services

2
Co-authors

• Mary Ellen Simpson, R.N., Ph.D.
• William Sappenfield, M.D., M.P.H.
• Ralph Schubert, M.A., M.Sc.
• Arden Handler, Dr.PH
• Deb Rosenberg, Ph.D.

3
Welcome to Illinois

• 12.4 million people
• 2/3 of population in Chicago and Collar
  Counties
• 70 of counties are rural (lt50,000 pop.)
• gt99 of population served by a local health
  department
• 180,000 live births
• 44 of live births covered by Medicaid

4
Family Case Management

• Outreach and Case Finding
• Assessment and Care Plan Development
• Referral and Follow-up
• Advocacy

5
Family Case Management

• Statewide
• 116 Local Agencies
• Local Health Departments
• Federally Qualified Health Centers
• Community Based Organizations
• 44 Million Budget
• 281,000 Pregnant Women and Infants Served Annually

6
Special Supplemental Nutrition Program for Women,
Infants and Children (WIC)

• Nutrition education
• Breastfeeding education and support
• Supplemental nutritious foods
• Access to health care services

7
WIC Program Characteristics

• Statewide
• 100 Local Agencies
• Local Health Departments
• Community Based Agencies
• 34 Million Budget ( Food)
• 508,300 Women, Infants and Children Served
  Annually

8
Theoretical Model of FCM
Positive Resulton PregnancyTest
Prenatal Care
FCMAssessment
WICFood Supplement
Domestic Violence screening
HealthEducation
Care Plan Provides continued follow up and
Linage to the following servicesHOME VISIT
Education
Housing
UltimateGoal
Mentoring Support
Mental Health
Substance Abuse Treatment

• Lower Prematurity
• Raise Birthweight
• Lower Intrauterine Growth Restriction

Smoking Cessation
9
Study Background

• Results of previous research examining
  programmatic effects of perinatal case management
  services using linked program and vital records
  computer files has been questioned because of
  preterm delivery bias and selection bias

10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
Research Question

• Does enrollment in Family Case Management and/or
  WIC reduce the likelihood of early preterm, late
  preterm or term SGA births?

14
Selection Bias

• Participants were not randomly assigned to the
  intervention and comparison groups, so they may
  not be equivalent with regard to race, age,
  parity, medical risk factors use of prenatal care
  or other variables.

15
Program Entry (Prematurity) Bias

• Some of the women in the comparison group may
  have given birth before they would typically have
  enrolled in WIC or FCM
• Some of the women in the intervention group may
  have enrolled in WIC or FCM after it was possible
  to have a preterm delivery.

16
Controlling For Selection Bias

• Women with no PNC or late PNC were not included
  in the study
• All have PNC started before the 5th month of
  pregnancy
• Logistic regression

17
Controlling For Program Entry Bias

• Women who enrolled after possible preterm
  delivery were included as non-program
  participants

18
Controlling For Medical Risk Factor Bias

• Models adjusted for medical risk factors
• There were more adverse medical risk factors in
  women who participated in the programs
• This bias is against the program effect

19
Controlling for the Black Box of FCM

• We were unable to examine referrals and content
  of FCM because of data limitations.

20
Program Participation by Year
21
Cornerstone MIS

• Supports FCM, WIC, Immunization Others
• Common Enrollment
• Case Management Service Delivery
• 117 Screens
• 300 Locations
• 3,000 Workstations

22
Methods

• Link Birth Certificate, Medicaid and Cornerstone
  Management Information System files
• Concatenated 2000-2002 files
• Analyze singleton births on Medicaid

23
Methods

• Measure completed months of pregnancy
• Entered programs prior to gestational age of
  outcome measured
• Entered prenatal care before 5th month of
  pregnancy
• Enrolled in programs at least 1 month

24
Logistic Regression Models

• Outcomes early preterm birth (5-6th month), late
  preterm birth (7th month) and term SGA births
  (8th month or more)
• Adjusted for socio-demographic factors and
  health-related factors

25
Results Early Preterm Birth (5-6th mo.)

• Odds Ratio 0.76
• 95 Confidence Interval 0.69 -- 0.83
• Adjusted for age, education, race, ethnicity,
  material status, smoking, alcohol, parity risk
  for age, medical risks, parity
• N203,450
• 50th Percentile birth weight (g) for early
  preterm birth 496g-1637g

26
Results Early Preterm Birth
N203,450 Adjusted for age, education, race,
ethnicity, marital status, smoking, alcohol,
medical risk factors parity
27
Results Late Preterm Birth (7th mo.)
Adjusted for medical risks, race, smoking,
parity risk for age, marital status N199,413 50
th Percentile birth weight (g) for late preterm
birth 1918g-2667g
28
Results SGA Term Birth (8th month or more)
Adjusted for medical risks, race, marital
status, smoking, alcohol, and parity risk for
age N184,224
29
Conservative Estimates

• Excluded women without prenatal care
• Moved women who entered WIC or FCM late in
  pregnancy into the comparison group
• Adjusted for a wide array of demographic and
  health status variables
• Addressed 4 of 5 sources of bias reported in
  previous studies

30
Limitations

• No randomization (Observational study design)
• Did not account for variations in service
  delivery after enrollment

31
Conclusions

• FCM WIC showed a protective effect for both
  early and late preterm delivery
• 24 reduction in extreme prematurity
• Neither program impacts term SGA births
• Program benefits are in addition to those of PNC
• More rigorous control for bias

32
Public Health Implications

• Substantiate previous studies that WIC or FCM
  reduces the risk of early and late preterm
  delivery
• FCM effects may be due to nutrition, health
  education or social service programs