Marriage, Fertility and Births

in a Northern District of Bangladesh

M. Rashid1, A.B. Labrique2, A.B. Massie2, A.A.Shamim1, A.Huq1, Parul Christian2, Rolf W. Klemm2, K.P West, Jr.2
1 JiVitA Bangladesh (NIPHP), 2 Center for Human Nutrition Johns Hopkins University Bloomberg School of Public Health
Background Results Figure 1. Seasonal Variation in Pregnancy Events

Several studies in Bangladesh have documented the seasonal The mean age for all pregnant women was 25 years 5 months, 0.1
Conceptions (N=32,214)

variations in marriages, conceptions, births and other pregnancy while the mean for newly married pregnant women was exactly 16 Births (N=17,629)
MC (N=3,419)
outcomes. The reasons for the seasonality of some of these events years. Around 43% of all women had some schooling and only 10% 0.09 MR (N=5,688)
SB (N=781)
are not well understood and many hypotheses have been postulated, were involved in any income-generating activities. In contrast, 76.6% Marriage (N=10,289)
0.08
ranging from agricultural demands on the workforce to contraceptive of the cohort of newly married women had some schooling, and 5%
availability and access. were generating income for their families. Husbands of newly married 0.07
women were also better educated (59.6% with some schooling) than
Bangladesh, one of the most successful examples of social the husbands of the general population of married women (43.2%). 0.06

Proportion
marketing of contraceptives, boasts a 54% contraceptive prevalence
(UNFPA, 2002). However, more research is warranted into the A peak in marriage was seen in July in both years. Among the 0.05

behaviors of newly married women and couples and the contraceptive entire population of eligible women of reproductive age, we observed
choices available to prevent unwanted pregnancies and empower an increase in conceptions in the months of March/April in both years, 0.04

couples to delay pregnancy, if so desired. followed by a spike in miscarriages shortly after. MRs also closely 0.03
follow the onset of pregnancies, as expected. A peak in stillbirths is
Since JiVitA actively enrolls newly married women who move observed shortly before the peak birth season in October/ November/ 0.02
into the study area within 4 months of their marriage date, we have a December. (Figure 1)
unique opportunity not only to characterize seasonal patterns of 0.01

marriage, pregnancy and associated outcomes, but also to Time to Pregnancy

0
characterize the reproductive health patterns of these newly married Among 9,708 newly married women on whom data were Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct
women. available, a life table analysis (Table 1) revealed that 25% of the 2001 2002 2003

cohort became pregnant by the end of their 6th month after marriage,
and 50% by the 13 th post-nuptial month. At the end of two years, only
30% had not become pregnant (Figure 2).
Objectives Table 2. Contraceptive Use Patterns Among Conclusions
At the time of their first trimester enrollment interview, about
• To describe the seasonality of 44% of newly married women and 37% of their husbands indicated Newly Married and Multigravid The patterns of seasonality of marriage, conceptions and
marriage, conception and pregnancy they had not wanted the pregnancy. In 35% of couples, both husband Women by Wantedness of Pregnancy births seen in Gaibandha are comparable to findings from Matlab in
outcomes in the northern rural District and wife expressed a desire to have delayed the current pregnancy. southern Bangladesh (Becker et al, Pop Stud 1986) and Haryana,
of Gaibandha over a period of two Despite this, within families where neither partner wanted pregnancy, India (Anand et al, Ind Ped 2000). The study community is
years of population surveillance, 88.4% were not practicing any method of contraception (Table 2). predominantly agrarian, so seasonality of marriage, conception and
• To characterize the time between Most commonly used in this group were oral contraceptive pills birth may be explained in part by the sowing and harvesting
marriage and pregnancy among (OCPs) (9.1%), with condoms being used by only 2.2% of couples. seasons. A marriage peak in July occurs soon after the ‘Boro’ rice
newly married women and This contrasts sharply with practices among multigravid women harvest (during April-May), when farmers have money to arrange
among whom 34.3% reported use of OCPs and 7.4% injectables. marriage ceremonies.
• To explore contraceptive choices by
these newly married women in light of As the cohort of newly married women has not been followed
wantedness of pregnancy. for sufficient time, pregnancy event seasonality was looked at using
data from all enrolled women of reproductive age. The peak of births
Table 1. Life Table of Pregnancy Events in “Newly in October/ November/ December corresponds with a peak in
conceptions in March/April. The drop in marriages and conceptions
Married Women”, Between 9/2001 and 9/2003
Methods in winter may reflect the ‘Aman’ harvest and ‘Boro’ sowing seasons,
when women’s participation in fieldwork increases.
The JiVitA study is following 113,060 married women of
reproductive age for incident pregnancies in 19 Unions of rural The lifetable analysis shows that newly married women
Gaibandha and Rangpur Districts. Since the study’s onset in July become pregnant very soon after marriage; half the cohort were
2001, 35701 newly pregnant women have been enrolled and visited pregnant within the first year of their marriage. However, within this
weekly by locally hired and trained female workers called Field group, more than one third of couples reported not wanting their
Distributors (FD). Every 5 weeks, the FDs conduct house-to-house pregnancy. Contraceptive use, however, among these newlyweds
surveillance of all enumerated married women of reproductive age to does not reflect well the expressed desire for delayed conception
identify incident pregnancies and to enroll newly married women Figure 2. Cumulative pregnancies by Time (Months) (Table 2, Column 1). This contrasts sharply with high contraceptive
entering the study area. since Marriage use documented in the general population of multigravid married
women (Table 2, Column 3).
To date, 10278 newly married women have been enrolled into
the JiVitA-1 trial. Data from enrolled women was used to look at There may be several reasons that newly married women are
seasonality of marriage, conception and pregnancy outcomes, not better empowered to control their first pregnancy, including family
including livebirths, stillbirths, miscarriages and premature and social pressures, restricted access to information, or even non-
terminations of pregnancy. Proportionate distributions of these events availability of methods. This phenomenon may also represent a
over a 25-month period were plotted to determine seasonality of these missed opportunity for family planning interventions to target birth
events. Approximate dates of conception have been extrapolated by control counseling to newly married couples in this population.
shifting the dates of pregnancy enrollment back by eight weeks, which
is the average delay between conception and enrollment.
A lifetable analysis was used to explore the distribution of time Acknowledgements
to pregnancy after marriage among 9708 newly married women on
whom data was available. In an effort to ascertain whether newly The financial support of the United States Agency for International
married couples delay their first pregnancy through contraception, we Development (USAID), the Bill and Melinda Gates Foundation, the
explored the contraceptive use of pregnant women who reported that Sight and Life Research Institute, Micronutrient Initiative and the
neither spouse wanted the pregnancy. Government of Bangladesh is acknowledged.