Unintended pregnancy and its association with shorter breastfeeding duration: an assessment of Bolivia and Paraguay

1
Unintended pregnancy and its association with
shorter breastfeeding duration an assessment of
Bolivia and Paraguay

• Carrie Shapiro-Mendoza, MPH, PhD1
• Beatrice J. Selwyn, ScD1
• David P. Smith, PhD1
• Maureen Sanderson, PhD2
• 1 School of Public Health,
• University of Texas Health Science Center
  at Houston
• 2 School of Public Health,
• University of Texas Health Science Center
  at Brownsville

2
 We are guilty of many errors and faultsBut our
worst crime is abandoning the children,Neglecting
the fountain of life.Many things we need can
wait.The child cannot.Right now is the time
bones are being formed,Blood is being made,
senses are being developed.To the child we
cannot answer "Tomorrow."The child's name is
"Today." -Gabriela Mistral, (1889-1957)
1945Nobel Laureate in Literature, Chile
3
International health priority

• To decide when and how many children to bear
• 1995 Cairo Convention
• basic human right
• maternal health linked to child morbidity
• Breastfeeding (BF)
• exclusive BF for 1st six months
• continued BF until 2 years

4
Objectives

• To explore the association between pregnancy
  intention and breastfeeding duration
• To compare associations among Bolivian and
  Paraguayan children
• To differentiate between mistimed and unwanted
  pregnancies

5
Covariates available from the data

• maternal age
• maternal education
• marital status
• parity
• rural versus urban residence
• socioeconomic status
• current pregnancy status
• contraceptive use

effect measure modifier
6
Methods

• Demographic and Health Survey (DHS)
• Bolivia 1998
• Paraguay 1990
• Cross-sectional survey
• nationally representative sample
• childbearing age women (15-49 years)
• stratified, 2-stage multi-cluster sample design
• Response rate 93

7
Data collection activities

• Face-to-face household interviews
• Mothers asked about
• reproductive history
• child morbidity and mortality
• breastfeeding status
• fertility preferences
• sociodemographic characteristics

8
To measure pregnancy intention

• Mothers asked
• At the time you became pregnant with name of
  last born child
• did you want to become pregnant then - intended
• did you want to wait until later - mistimed
• or did you want no more children at all -
  unwanted

9
Breastfeeding duration

• Maternal report
• Number of months a child was breastfed
• If child was still breastfeeding at the time of
  the interview
• childs age at that time recorded
• censored observation

10
Study population for analysis

• Children whose mothers report that their
  pregnancy was or was not intended at the time of
  conception.
• Inclusion criteria for children
• Last born
• Surviving
• Singleton births
• Born in the 36 months prior to interview
• Sample size
• Bolivia (n 3445)
• Paraguay (n 1837)

11
Analysis of the data

• Calculated incidence rates and hazards ratios
• Survival analysis
• Kaplan-Meier estimates and graphs
• Cox proportional hazards (PH) regression
• Tested PH assumption
• Complex sampling design weights

12
Selected maternal sociodemographic
characteristics of unwanted pregnancies in
Paraguay and Bolivia.

• Older
• Less educated
• Multiparous
• No flush toilet or latrine
• Rural (Bolivia only)
• Not currently using a modern contraceptive method
  (Bolivia only)
• Currently working

13
Intention status at conception of children in
Paraguay and Bolivia in percentages
14
Comparison of breastfeeding survival curves for
last born surviving children, Bolivia DHS 1998
and Paraguay DHS 1990.
Bolivia
Paraguay
15
Proportion of children continuing to breastfeed
up until 36 months by pregnancy intention status.
16
Adjusted survival curves for breastfeeding
duration derived from baseline hazards of
multivariate Cox regression analysis, stratified
by parity and pregnancy intention status for
Bolivia DHS 1998.
Adjusted for maternal education, region of
residence, and current pregnancy status
17
Survival curves for breastfeeding duration
derived from baseline hazards of multivariate Cox
regression analysis, stratified by parity and
pregnancy intention status for Paraguay DHS 1990.
Adjusted for maternal education, region of
residence, and current pregnancy status
18
Strengths

• Distinguished between mistimed and unwanted
  pregnancies
• Used survival analysis
• Measured BF duration vs never/ever
• Measured risk
• Compared different populations

19
Limitations

• Retrospective report
• Recall bias
• Post-hoc rationalization
• Validity reliability unintended pregnancy
  questions
• Selective survival bias
• Other confounders

20
Comparison of breastfeeding survival curves for
last born surviving children, by current and
retrospective maternal report, Bolivia DHS 1998
and Paraguay DHS 1990.
21
Conclusions

• Primiparae with unwanted pregnancies have shorter
  breastfeeding durations in Bolivia, but not in
  Paraguay, taking into account child age,
  education, and socioeconomic status, and
  residence.

22
(No Transcript)
23
Comparison of maternal age
24
Comparison of maternal education
25
Comparison of maternal education
26
Comparison of current contraceptive use
27
Comparison of current modern contraceptive use by
intention status
28
(No Transcript)
29
(No Transcript)
30
Unintended Pregnancy and
Contraceptive use in Paraguay and
Bolivia
Paraguay,
23
58
1990
Country and Year
Bolivia,
54
84
19931994
0
20
40
60
80
100
Percent
of births unplanned
If want no more births, not using effective
method
31
Causes of unintended pregnancy

• Lack of access to effective contraception or
  abortion
• Insufficient knowledge about family planning
• Individual feelings, motivations, attitudes, and
  beliefs
• Non-supportive sexual partners
• Social forces
• political or economic environments
• religious authority

32
Causes of unintended pregnancy

• Failure to use contraception
• Contraceptive failures
• only accounted for 68 of the pregnancies that
  were reported as unintended.
• 1995 NSFG, Trussell et al. 1999

33
Hazards ratios for breastfeeding termination, by
pregnancy intention status, stratified by parity,
Bolivia DHS, 1998.
Primipara
Multipara
Adjusted for maternal education, urban versus
rural residence, and current pregnancy status.
34
Hazards ratios for breastfeeding termination, by
pregnancy intention status, stratified by parity,
Paraguay DHS, 1990.
Primipara
Multipara
Adjusted for maternal education, urban versus
rural residence, and current pregnancy status.