Levels, Patterns and Determinants of Reversible Contraceptive Use to Limit Family Planning in India: Data from the National Family Health Survey, 2015-2016 | BMC Women’s Health
The study used unit-level data from the fourth cycle of the National Family Health Survey (NFHS 4), conducted in 2015-2016. This survey adopts a two-stage stratified systematic random sampling. At the first stage, the primary sampling unit was selected and the household was selected at the second stage. This survey collected information for representation of the districts of India. The main objectives of the survey are to provide reliable estimates on fertility, maternal and child mortality, family planning, reproductive and child health, nutritional status of children, use of maternal and child health services and the autonomy of women. The survey adopted the multi-stage sampling design to collect the samples . The NFHS collected data using different interview schedules – household schedule and eligible female, male, and biomarker schedule. The NFHS 4 survey interviewed a total of 699,686 women aged 15-49 in India. The paper limits the sample to those who were currently married women during the survey period. After excluding single, divorced, separated and widowed women, the total sample used for the analysis is 511,377 currently married women. Among the total currently married women, the total demand for family planning includes 339,537 women and the demand for limitation includes 282,795 women. The details of using the sample distribution are shown in Fig. 1. Of the women who requested methods of limitation, the satisfied request for limitation includes 245,941 women (185,498 by modern permanent method, 39,021 by modern reversible method and 21,422 by traditional methods) .
NFHS 4 collected information on demand and use of the contraceptive method for family planning. Outcome variables were prepared based on the following questions. “Do you want to limit your pregnancy or childbearing permanently?” If so, this is called limiting demand. If women have a limited demand, the following question for them: “Are you using contraceptives to limit your pregnancy permanently?” If the answer is yes, then it is “satisfied demand for limitation”, otherwise “unmet need for limitation”.
The selected outcome variables are (a) demand satisfied to limit childbearing (SLD) (b) limited demand satisfied by modern reversible contraceptive methods (mrSLD) (c) limited demand satisfied by traditional contraceptive methods ( tSLD). Here, SLD refers to when women used contraceptive methods to satisfy their limiting demand. mrSLD means when women were using modern reversible methods of contraception (pill, IUD, injectable, male condom, female condom, Standard Days Method (SDM), diaphragm, foam/jelly, breastfeeding method and amenorrhea (LAM) and other modern methods) to meet their limited demand. And, tSLD refers to when women were using traditional methods of contraception (rhythm, withdrawal, and other traditional methods) to meet limited demand.
The explanatory variable used in the analyzes includes a range of socioeconomic and demographic variables, which have been significantly associated with unmet and met need for family planning in India and elsewhere (8-11). These independent variables are respondent’s age (15–19, 20–24, 25–29, 30–34, 35+), parity (0, 1, 2, 3, 4+), having at least a son (had at least one son [Yes]had no son [No]), female years of schooling (no schooling, 1–5, 6–10, 11+), wealth status (poorest, poorest, middle, richest, and wealthiest ), religion (Hindu, Muslim, Christian and other) and social group (General, other backward class [OBC]Scheduled caste [SC]Scheduled Tribe [ST] and don’t know), exposure to mass media (had not listened to family planning program through mass media [No]had listened to the family planning program through the media [Yes]), place of residence (urban, rural) and geographic region (North, Center, East, Northeast, West and South). The division of geographic regions is taken from the NFHS-4 report .
Bivariate and multivariate analysis including cross-tabulation and binary logistic regression was used to achieve the study objectives. Bivariate analysis with a chi-square test of significance was applied to examine patterns of limiting demand, satisfied limitation, and use of a contraceptive method by those who satisfied the limiting demand. Three separate binary logistic regression analyzes were performed to determine the socioeconomic correlates of SLD, mrSLD, and tSLD. These variables are binary in nature like SLD (unmet need ‘0’, demand met ‘1’), mrSLD (no ‘0’; yes ‘1’) and tSLD (no ‘0’; yes ‘1’). The study used revised definitions of unmet need to limit . The results of the binary logistic regression analysis were presented as adjusted odds ratios (AOR) with a 95% confidence interval (CI). All analyzes were performed using statistical STATA (version 14.1 SE).