Measuring unintended pregnancy: Are women who report mistimed and unwanted pregnancies different

1
Measuring unintended pregnancy Are women who
report mistimed and unwanted pregnancies
different?

• Denise DAngelo, MPH, Brenda Colley Gilbert, PhD,
  MSPH, Roger Rochat, MD, Joan Herold, PhD, John
  Santelli, MD, MPH

2
Background

• Unintended pregnancy is common in the US
• 1994 NSFG found 49 of all pregnancies 31 of
  pregnancies that ended in a live birth were
  unintended
• Many surveys measure UP in terms of mistimed
  unwanted pregnancies, but few studies report or
  analyze groups separately

3
Study Objective

• Determine if it is reasonable to combine women
  with mistimed and unwanted pregnancies into the
  single category of unintended pregnancy by
• Describing the characteristics of women who
  report mistimed and unwanted pregnancies and
  examining how they compare to women with intended
  pregnancies
• Measuring the association between maternal
  demographic and psychosocial factors and
  pregnancy intention

4
Data Source

• 1998 Pregnancy Risk Assessment Monitoring System
  (PRAMS) data covering 15 states
• States included AL, AK, AR, CO, FL, IL, LA, ME,
  NM, NY, NC, OK, SC, WA, WV
• Data from NY do not include NYC NM data includes
  July 1997- December 1998 births
• Overall weighted response rates ranged from 70
  in NM to 83 in ME
• 25,027 respondents representing the 986,510 women
  who gave birth in those states in 1998

5
Overview of PRAMS

• PRAMS is an on-going population based
  surveillance system developed, technically
  supported, and funded in part by the CDC
• PRAMS uses a mixed-mode survey (mail telephone)
  to collect self-reported information on perinatal
  maternal health indicators from mothers who
  recently gave birth to a live infant

6
Methods

• The prevalence of selected maternal demographic
  psychosocial variables and infant outcome
  variables was determined
• Unadjusted relative risks 95 confidence
  intervals were calculated to identify
  associations between the variables and pregnancy
  intention
• Data analysis was performed using SUDAAN software
  to accommodate PRAMS complex sampling design

7
Intention definition

• Intended pregnancy wanted pregnancy then or
  sooner
• Mistimed pregnancy wanted pregnancy later
• Unwanted pregnancy didnt want pregnancy then or
  at any time in the future

8
Prevalence of Pregnancy by Intention, 15 PRAMS
States, 1998
9
Summary of Prevalence

• Women reporting mistimed unwanted pregnancies
  were more likely than women with intended
  pregnancies to
• be unmarried
• have a HS education or less
• be black or NA/AN
• be Medicaid recipients
• smoke
• enter PNC after the 1st trimester
• not breastfeed
• report that partner did not want the pregnancy
• report being physically abused during pregnancy
• have a LBW infant

10
Associations-Introduction

• Did 3 analyses of association---each comparing 2
  pregnancy intention groups (3rd group excluded)
• Mistimed Intended
• Unwanted Intended
• Unwanted Mistimed

11
Associations Results

• See attached tables

12
Limitations

• Survey responses are self-reported
• Concept of pregnancy intention may not be
  meaningful to some women
• Pregnancies that do not result in a live birth
  are excluded
• Data do not provide a national estimate of
  unintended pregnancy
• Descriptive analysis only

13
Conclusions

• Women with mistimed and unwanted pregnancies were
  different from women who reported intended
  pregnancies
• Women who reported mistimed unwanted
  pregnancies had similar characteristics for 11/13
  attributes examined
• For each of the 11 attributes, risk was greater
  for women who reported unwanted pregnancies than
  mistimed.
• Specific differences between women who report
  mistimed unwanted pregnancies were age and
  parity

14
Interpretation Recommendation

• There is justification for combining women with
  mistimed unwanted pregnancies into the single
  category of unintended pregnancy in that both
  groups differ from those with intended
  pregnancies
• The use of the single category of unintended
  masks the increased risk among women with
  unwanted pregnancies, and the differences between
  women with mistimed unwanted pregnancies
• It would be beneficial to separate groups when
  possible for analysis and especially for policy
  or program planning

15
Next Steps

• Additional analysis (multivariate) to determine
  variables on which to focus intervention
  strategies
• Examination of mistimed unwanted pregnancies
  that did not end in a live birth