Deepali Godha, David Hotchkiss, and Anastasia Gage Tulane University

Understanding the role of child marriage on reproductive health

outcomes: evidence from a multi-country study in South Asia

Deepali Godha, David Hotchkiss, and Anastasia Gage

Tulane University

Background Child marriage remains a pervasive problem in South Asia and

sub-Saharan Africa

Increasingly recognized as a violation of human rights

Previous research has associated child marriage with a number of adverse health and social outcomes

But most studies to date have focused on India – very little research in other countries

Purpose of study To assess the association of child marriage and

reproductive health outcomes in four South Asian countries after controlling for individual-, and household- level factors

Prevalence of child marriage in South Asia is high Among currently married women 20 to 24 years of age, percent

married prior to age 15 ranged from 10 percent in Nepal to 38 percent in Bangladesh.

Data Most recent Demographic and Health Surveys in four

countries India (2005/6) Bangladesh (2007) Nepal (2006) Pakistan (2006/7)

Sample: Ever-married women 20-24 years of age Sample size ranged from 1,546 in Pakistan to 22,807 in India

Methods (1) Dependent variables

Fertility Early fertility – childbirth within the first year of marriage Multiple unwanted pregnancies Having had at least one pregnancy termination Having had at least one unwanted pregnancy

Fertility control Lack of fertility control prior to the first birth Low lifetime fertility control – history of rapid repeat childbirth

Maternal health care utilization

Methods (2) Independent variables

Age at marriage dummy variables Married at 14 years of age or younger Married at 15 to 17 years of age

Other individual- and household-characteristics

Statistical methods Logistic regression models to examine the association between

various outcome variables and age of marriage Unit of analysis is individual woman

Descriptive results

Characteristic India Bangladesh Nepal Pakistan

Total married 17 years of age or younger

58.9 77.2 62.5 50.3

- Married 15-17 years of age

41.6 39.3 50.0 36.3

- Married 14 years of age or younger

17.3 37.9 12.5 14.0

Husband older by 10 years of age or more

14.3 41.1 8.8 19.4

Primary or no education 54.5 46.7 64.6 78.4

Percent of women married as children ranges from 50% to 77%

Descriptive results

Characteristic India Bangladesh Nepal Pakistan

Fertility

Early fertility 19.9 19.2 16.5 18.7

Pregnancy termination 15.3 14.1 13.3 17.6

Unwanted pregnancy 15.2 19.8 24.0 13.9

Multiple unwanted pregnancy 3.1 1.7 3.1 4.0

Fertility control

Lack of fertility control prior to 1st birth

90.8 67.6 82.8 NA

Low lifetime fertility control 23.0 14.8 21.5 30.8

Percent of women with various fertility and fertility control outcomes

Logistic regression results

Characteristic India Nepal

Married ≤ 14 Married 15-17 Married ≤ 14 Married 15-17

Fertility

Early fertility 0.9 1.0 0.6 0.8

Pregnancy termination 1.6 1.4 2.3 2.3

Unwanted pregnancy 1.6 1.7 1.9 1.5

Multiple unwanted pregnancy 2.5 2.3 3.7 1.6

Fertility control

Lack of fertility control prior to 1st birth 1.7 1.3 2.2 2.5

Low lifetime fertility control 3.8 2.6 4.5 3.4

Odds ratios: comparison group is women married at 18 years and older

Blue font indicates statistical

significance (p<0.05)

Logistic regression results

Characteristic Bangladesh Pakistan

Married ≤ 14 Married 15-17 Married ≤ 14 Married 15-17

Fertility

Early fertility 1.0 0.9

Pregnancy termination 4.5 3.4 2.2 1.9

Unwanted pregnancy 1.1 1.2 2.9 3.8

Multiple unwanted pregnancy 4.7 4.1 24.2 18.2

Fertility control

Lack of fertility control prior to 1st birth

1.5 1.1 NA NA

Low lifetime fertility control 5.9 3.7 8.3 5.7

Odds ratios: comparison group is women married at 18 years and older

Blue font indicates statistical

significance (p<0.05)

Conclusions In South Asia, child marriage is significantly associated with many

negative fertility and fertility control outcomes (and maternal health care utilization)

Women married in early adolescence show a higher propensity towards most negative outcomes than women married in middle adolescence

Child marriage adds a layer of vulnerability to women that leads to poor fertility control and fertility related outcomes

Limitations Outcomes are self-reported: may be prone to bias

Data are cross-sectional: difficult to attribute causality

High odds ratios and wide confidence intervals may be an indication of small cell sizes or small probabilities

Cannot distinguish between natural and induced pregnancy termination

MEASURE Evaluation PRH is a MEASURE project funded by the United States Agency for International Development (USAID) through Cooperative Agreement GHA-A-00-08-00003-00 and is implemented by the Carolina Population Center at the University of North Carolina at Chapel Hill in partnership with Futures Group International, Management Sciences for Health, and Tulane University. Views expressed in this presentation do not necessarily reflect the views of USAID or the U.S. Government. MEASURE Evaluation PRH supports improvements in monitoring and evaluation in population, health and nutrition worldwide.