Menu
Menu
Menu
Menu
Postpartum months provide a challenging period for poor women. This study examined patterns of menstrual resumption, sexual behaviors and contraceptive use among urban poor postpartum women. Women were eligible for this study if they had a birth after the period September 2006 and were residents of two Nairobi slums of Korogocho and Viwandani. The two communities are under continuous demographic surveillance. A monthly calendar type questionnaire was administered retrospectively to cover the period since birth to the interview date and data on sexual behavior, menstrual resumption, breastfeeding patterns, and contraception were collected. The results show that sexual resumption occurs earlier than menses and postpartum contraceptive use. Out of all postpartum months where women were exposed to the risk of another pregnancy, about 28% were months where no contraceptive method was used. Menstrual resumption acts as a trigger for initiating contraceptive use with a peak of contraceptive initiation occurring shortly after the first month when menses are reported. There was no variation in contraceptive method choice between women who initiate use before and after menstrual resumption. Overall, poor postpartum women in marginalized areas such as slums experience an appreciable risk of unintended pregnancy. Postnatal visits and other subsequent health system contacts provide opportunities for reaching postpartum women with a need for family planning services. © 2011 The New York Academy of Medicine.
The study was conducted in two Nairobi slum settlements namely: Korogocho and Viwandani. The Nairobi Urban Health and Demographic Surveillance System (NUHDSS) has prospectively monitored about 60,000 individuals living in the two slums since 2002, with routine updates conducted every 4 months. The two informal settlements share common slum characteristics such as poor sanitation, high school dropout, congestion, crime, unemployment, high disease burden, and limited access to proper health facilities. The slums are served by some government health centers, together with several private for-profit outlets, faith-based organizations, nongovernmental organizations, not-for-profit health care providers, and retail outlets selling over-the-counter medicines including contraceptives. The current study uses data from the Maternal and Child Health (MCH) component of a 5-year Urbanization, Poverty, and Health Dynamics longitudinal study. This is an ongoing open cohort, where women are recruited into the study if they had a birth from September 2006 onwards and they were living in the Nairobi Urban Health and Demographic Surveillance area. The MCH study is nested into the NUHDSS and relies on previously collected rich sociodemographic data from all women resident in the study area. The first baseline round of the MCH component was conducted between February and April 2007 and since then additional waves of data collection have been conducted. A total of 2,994 women had been recruited into the study by the end of August 2008 (Table 1). Interviews were conducted in Swahili, the commonly spoken national language in the settlements. During every visit, trained fieldworkers recruit new mothers who form a new cohort and updates are conducted for those mothers previously recruited. Number of women recruited/interviewed for the study during the period 2007–2008 aThese two cohorts recorded relatively higher loss to follow-up/attrition rates mostly due to the political instability resulting from national elections that were conducted around this period and the resulting higher rate of changes of residences and outmigrations Details of reproductive events such as breastfeeding, postpartum abstinence, postpartum amenorrhea, sex, contraceptive use, and condom use are documented in a month-to-month calendar format since the birth of the index child. For the current analysis, data from four cohorts of women collected between February 2007 to August 2008 are utilized (Table 1). The fieldwork duration for the third wave was relatively long because data collection was disrupted by the presidential election campaigns that covered most of December 2007 and later resulted into post-election violence in the first few months of 2008. To assess the effect of loss to follow-up, we compared characteristics for women who were recruited initially and women who were successfully followed up during the subsequent updates/surveys (results not shown). The distributions of selected indicators such as parity, marital status, age of woman, location, etc. were generally comparable between women present at recruitment and those who were successfully re-interviewed during the first update which allowed on average a maximum of 12 months. The entire analysis is restricted to the first 12 postpartum months. Much of the analysis is performed in woman-months. A woman followed up for 1 year or more contributes 12 woman-months. Nearly all women in the early recruitment cohorts contributed a full 12 months of data. Women in cohort 4 contribute an average of 9 months of data. This censoring for a small minority of the sample is addressed by life-table methods and by data presentation in terms of ordinal months since birth. This approach allowed us to measure exposure to pregnancy and contraceptive use on a month-to-month basis and allowed us to assess the contribution of several event-states such as abstinence, amenorrhea, and contraception in the first 12 months of postpartum. Five women with missing information in one of the months of (retrospective) follow-up were excluded. In order to jointly assess the timing and interactions of menstrual and sexual resumption, the time since birth of the child was classified into ordinal months of mutually exclusive categories of protection and risk periods in relation to contraceptive use. For a specific woman-month: Descriptive statistics was used to assess patterns of postpartum infecundity (amenorrhea), contraceptive use, and sexual resumption as well as for summarizing study cohorts by selected characteristics such as age of women at recruitment, marital status, education, ethnicity, parity, fertility desires, and source of contraception. Survival analysis techniques were used to assess the time to event (resumption of menstrual flow and resumption of sexual activities and time to first contraceptive use).
N/A
