Retention in a Study of Prenatal Care: Implications of attrition for program planning

1
Retention in a Study ofPrenatal
CareImplications of attrition forprogram
planning
Suzanne Tough, PhD11th Annual Maternal and
Child Health Epidemiology ConferenceMiami,
FloridaDecember 2005
2
Community Perinatal Care Study

• Background
• Study Question
• Methods
• Results
• Conclusions
• Public Health Implications

3
Background

• Prenatal care is publicly funded in Canada and
  widely accepted as the standard of care for
  pregnant women
• Prenatal care provides an opportunity to address
  medical and psychosocial concerns that impact
  health
• Challenges exist with retaining high-risk women
  in prenatal care (i.e. young, low income,
  non-Caucasian)

4
Study Question

• Did socio-demographic characteristics, lifestyle
  choices, psychosocial factors, and life events
  differ between women who completed and women who
  did not complete a community-based randomized
  controlled trial of supplementary prenatal care?

5
Community Perinatal Care (CPC) Study

• BASICS
• prospective randomized controlled trial
• pregnant women attending low risk maternity
  clinics in Calgary, Alberta were recruited prior
  to first prenatal appointment
• study groups
• 1. control (standard of care at prenatal
  clinic)
• 2. control Nurse
• 3. control Nurse Home Visitor

6
Just to let you know

• Overall CPC results indicated that women who had
  the intervention increased their use of
  pregnancy-related resources
• Early prenatal class
• Parenting classes
• Nutrition counseling
• Written material
• Agencies who provide lists of child care
  facilities

7
Methods
DATA COLLECTION

• 3 telephone questionnaires
• first trimester
• mid-pregnancy
• 6-8 weeks post-partum
• Questionnaires gathered information on
• demographics
• obstetrical history
• lifestyle choices
• psychological functioning
• social support and network orientation
• life events

8
Methods
CLASSIFICATION

• CPC participants were classified as
• 1. RESPONDER (completed all 3 questionnaires)
• 2. DROP-OUT (withdrew from program prior to
  completion of the 3rd questionnaire)
• 3. UNREACHABLE (could not be contacted by
  telephone at some point after
• the initial questionnaire)

OR
OR
9
Recruitment and Retention Rates

• From a total of 1737 women who agreed to
  participate
• 1352 / 1737 77.8 completed the study
• 177 / 1737 10.2 dropped out
• 92 / 1737 5.3 became unreachable
• NOTE Non-completion rates were the same across
• intervention groups

(There were also 116 / 1737 6.7 who did not
complete the study for other reasons, such as
miscarriage)
10
Results Summary

• COMPARED TO RESPONDERS
• A significantly greater proportion of women who
  dropped out of the study had characteristics that
  included
• young maternal age (lt25 years)
• non-Caucasian ethnicity
• low income (lt40,000 / year)
• low education (lt high school)
• daily smoker
• use of food bank
• low social support
• negative network orientation
• history of depression
• came from a family with separated / divorced
  parents
• during 12 months prior to pregnancy

11
Results Summary

• COMPARED TO RESPONDERS
• A significantly greater proportion of women who
  became unreachable had characteristics that
  included
• young maternal age (lt25 years)
• single
• low income (lt40,000 / year)
• low education (lt high school)
• daily smoker, use of street drugs, and/or use
  of food bank
• low social support
• negative network orientation
• history of alcohol, drug, unemployment, suicidal
  thoughts/attempts or depression problems
• came from a family with separated / divorced
  parents
• during 12 months prior to pregnancy

12
Common Characteristics

• young maternal age (lt25 years)
• low income (lt40,000 / year)
• low education (lt high school)
• daily smoker
• use of food bank
• low social support
• negative network orientation
• history of depression
• came from a family with separated / divorced
  parents
• during 12 months prior to pregnancy

13
Logistic Regression Results

• The key predictors of women who dropped out of
  the study included

14
Logistic Regression Results

• The key predictors of women who became
  unreachable during the study included

15
Subtle CPC Findings
AMONG THOSE WHO STAYED IN THE STUDY

• Characteristics of women who reported need of
  additional support
• First parity
• Low income
• Non-Caucasian
• Abuse during pregnancy
• Young maternal age
• Lower network orientation
• Low self-esteem

16
Conclusions

• There are missed opportunities to engage women
  with the most complex needs in prenatal care
  programs
• If women can be engaged in the system, there
  are opportunities to improve resource utilization
  and informational support, potentially improving
  the birth experience

17
Public Health Implications

• Healthcare providers may wish to improve their
  ability to engage and retain high risk women
• Programs designed to engage and retain high risk
  women should consider different delivery models
  that facilitate access and use of services
• extended office hours, group prenatal care, easy
  to access venues (i.e. grocery stores)

18
Who We Worked With

• Calgary Childrens Initiative
• Calgary Immigrant Womens Association
• University of Calgary
• Pediatrics
• Community Health Sciences
• CHAPS

• Low Risk Prenatal Clinics (NE/NW)
• Calgary Home Visitation Collaborative
• Calgary Health Region
• Decision Support Research Team Healthy
  Communities
• Health Survey and Evaluation Unit
• Physician Partnership Steering Committee

19
The CPC Study Team