Sociodemographic Factors on Contraceptive Use among Ever-Married Women of Reproductive Age: Evidence from Three Demographic and Health Surveys in Bangladesh

Contraceptive use among married women of reproductive age has showed a substantial progress over the last few decades in Bangladesh. This study explores the sociodemographic factors associated with contraceptive use among ever-married women of reproductive age in Bangladesh by utilizing the information extracted from three of the Bangladesh Demographic and Health Surveys (BDHSs) in 1993–1994, 2004 and 2014. Bivariate analysis was conducted by performing chi-squared test of independence to explore the relationship between selected sociodemographic factors and dependent variables. To know the adjusted effects of covariates, a popular binary logistic regression model was considered. Respondents’ current age, place residence, division religion, education, age at first marriage, family planning (FP) media exposure, ideal number of children and fertility preferences are the significant determinants according to the most recent survey, BDHS 2014. However, wealth index and a respondent’s current working status were found to be significant factors only in BDHS 2004. The results of the study strongly recommend efforts to increase the education level among poor people, particularly among women in Bangladesh. Program interventions, including health behavior education and family planning services and counselling, are especially needed for some categories of the population, including those living in rural areas, Sylhet, Chittagong and Dhaka divisions, as well as illiterate and Muslim ever-married women.


Introduction
A high population growth rate is one of the leading social problems experienced by the developing world. It is responsible not only for increasing the level of poverty but also for decreasing life expectancy [1][2][3]. Contraception proves to be an effective medical intervention for controlling the fertility rate and also very helpful in developing maternal and child health [4].
Over the last 20 years, contraceptive use has shown a remarkable increase in all over the world, particularly in developing countries. The growing use of contraceptive methods in developing countries has reduced unintended pregnancies and maternal mortality by 40% [5]. Moreover, a study conducted in Bangladesh [6] provides evidence that the unintended pregnancy rate was higher (33%) among women who used contraceptives before their last pregnancy than women (23%) who did not use any contraceptives.
According to the reaffirmed commitments of the United Nations General Assembly, the new sustainable development agenda includes two targets (Target 3.7 and Target 5.6) relevant for family planning [7]. Target 3.7 states that, "by 2030, ensure universal access to sexual and reproductive health-care services, including for family planning, information and education, and the integration of reproductive health into national strategies and programmes". On the other hand, the statement of Target 5.6 is to "ensure universal access to sexual and reproductive health and reproductive rights as agreed in accordance with the Programme of Action of the International Conference on Population and Development and the Beijing Platform for Action and the outcome documents of their review conferences".
In the mid-1970s, the total fertility rate (TFR) of Bangladesh was more than six births per woman and the contraceptive prevalence rate (CPR) was less than 10% [8]. According to the 1993-1994 Bangladesh Demographic and Health Survey (BDHS), the CPR had risen to 45% and besides the TFR had fallen to 3.4 [9]. The 2014 BDHS showed that 54% of currently married women of reproductive age in Bangladesh are using modern contraception methods, and the TFR had fallen to 2.3 [10].
Despite the improvement of declining TFR in Bangladesh, women of reproductive age of this country need to pay special attention to contraceptive use, since Bangladesh still has a long way to go to reach the targeted CPR level of over 70% [11]. To increase CPR, an explainable and evidence-based strategy is needed. In this context, identification of possible factors for contraceptives use is one of the major component to formulate this strategy.
In the past, several studies have been conducted in Bangladesh to identify the factors that have significant impact on contraceptive use [12][13][14][15][16]. Most of this research was based on the nationally representative sample surveys. This indicates a comparative study can be performed to identify the factors having significant effects on contraceptive use in Bangladesh by using national surveys conducted in different years. Therefore, the main objective of our study is to determine the major sociodemographic determinants of contraceptive use in Bangladesh, using the data extracted from three of the BDHSs conducted in 1993-1994, 2004 and 2014.

Data Sources
The analysis is based on secondary data obtained from the 1993-1994, 2004 and 2014 BDHSs (data freely available at: http://dhsprogram.com/data/available-datasets.cfm). The samples for 1993-1994 to 2014 BDHSs were nationally representative and cover the entire population residing in noninstitutionalized dwelling units in the country. The survey used the sampling frame from the list of enumeration areas (EAs) from the 1991, 2001 and 2011 Population and Housing Censuses of the People's Republic of Bangladesh, provided by Bangladesh Bureau of Statistics (BBS) [10].
The BDHSs covered 9681, 10,811 and 17,989 residential households from 1993-1994, 2004 and 2014, respectively. From these sampled households, 9640, 11,440 and 17,886 ever-married women were interviewed, respectively. In this study, data are restricted to ever-married women aged 15-49. Based on these criteria, sample sizes for this study from the several BDHSs were 9495 ever-married women in 1993-1994, 11,290 ever-married women in 2004 and 17,863 ever-married women in 2014. Data were weighted to represent the structure of the Bangladeshi population using weighting factors provided with the BDHS [9,10,17].

Dependent Variable
The dependent variable was current contraceptive use, categorized dichotomously as a "Yes/No" variable. Respondents who were currently using a contraceptive method were categorized as "Yes", otherwise as "No".

Statistical Methods
Frequency distribution was used to describe the background characteristics of the respondents. Two-way contingency tables along with Pearson's chi-squared test were used to examine the relationships between sociodemographic factors and a dependent variable. To get the adjusted effects of selected factors, we consider a statistical model appropriate for binary response, namely binary logistic regression [18] model in multivariate setup. The result was presented as odds ratio (OR) and 95% confidence interval (CI). The Statistical Package for Social Science (SPSS v20.0, IBM Corporation, Armonk, New York, NY, USA) software was used for data analysis.

Results
The percentage of women in each category of the selected variables for each survey point are displayed in Table 1. With regard to the respondent's current age, about half of the women in all three surveys were under age 30. About 7% respondents in all surveys were residing in Barisal division, while slightly over one-third of the respondents live in Dhaka division. Considering the educational status, in the first two surveys, about half of the respondents (58.1% in 1993-1994 and 41.2% in 2004) had no education, while in BDHS 2014, this rate was only one-quarter (24.9%). With respect to the highest educational level, there is a clear indication that the percentage of women attaining secondary and higher education has been increasing in Bangladesh since 1993-1994. The percentage of the respondents who resides in urban areas have increased from 11.5% in 1993-1994 to 28.3% in 2014. Across the three BDHSs, the proportion of Muslims showed a slight increase from 87.8% to 90.1% between 1993-1994 and 2014. The proportion of women with exposure to family planning information in the media declined from 45.5% in 1993-1994 to 19.9% in 2014. Regarding age at first marriage of the respondents, the percentage of marriage took place before 18 years of age has declined from 88.1% in 1993-1994 to 76.5% in 2014 BDHS. In terms of household wealth quintiles, the proportion of women in middle income households has increased since 1993-1994. The proportion of respondent's working status rose from 16% in 1993-1994 to 33.1% in 2014. The percentage of women reporting three as an ideal family size have increased in recent times. Regarding women's fertility preferences, the percentage of women wanting no more children has increased slightly from 48.4% in 1993-1994 to 56.7% in 2014. Concerning contraception use, the percentage of women practicing contraception increased from 42% in 1993-1994 to 58.9% in 2014.
With regards to education in all three surveys period, women who attained primary education and again those having secondary and higher levels of education were more likely to use contraceptives than women having no education. In 2014, the rural women were 16% (OR = 0.84, 95% CI = (0.77, 0.91)) less likely to prefer contraceptives than their urban counterparts. For religion in all three surveys, women belonging to the non-Muslim religious groups had more likely used contraceptives compared with Muslim women.
In 1993-1994, the odds of women who were exposed to family planning information in the media were 16% (OR = 1.16, 95% CI = (1.02, 1.31)) more likely to use modern contraception as compared to married women that were not exposed family planning information in the media. Among the wealth index, the results showed that for the 2004 BDHS, women from middle and rich households was (12% and 19%, respectively) more likely to use contraceptives than women in the poor households. Moreover, for the last two surveys (BDHSs 2004 and 2014), the women who thought that four or more as their ideal number of children were (37% and 27%, respectively) less likely to use modern contraception than women whose ideal number of children was 0 or 1. In all three surveys, women who wanted to have another child and as well as women who were uncertain about having another child were less likely to use modern contraception than women who wanted no more children.

Discussion and Conclusions
Our study was designed to identify the major factors contributing to the changes in contraceptive use in the last two decades, by using the information extracted from the 1993-1994, 2004, and 2014 BDHSs.
The findings of the study provide evidence that contraceptive use decreases with age, indicating older women were more reluctant to use contraceptives than their younger counterparts. This type of relationship between age and contraceptive use has been observed similar to another existing study [19]. Consistent increase of contraceptive use rate from age of 25 to 44 years indicated that in the reproductive period, the better part of women reached their desired number of children and then were willing to terminate a pregnancy by using modern contraceptive methods [20].
Division also mattered in the pattern of contraceptive use. Our analysis strongly indicated that, over time, contraceptive use rates are higher in all divisions except Chittagong and Sylhet divisions. Previous studies also showed the conservativeness of the residents of Sylhet and Chittagong [21,22]. One of the possible reasons for this result is that the lack of concentration of policy makers on certain divisions in Bangladesh. Therefore, we strongly recommend that policymakers implement more programs associated with family planning in every division of Bangladesh.
The results of the study confirmed that women who attained primary education and again those with secondary and higher levels of education were more likely to use contraceptives than women with no education. This is because higher educational attainment can provide opportunity for better information on contraceptive methods and better access to services. Education can generate an openness among women to new ideas, such as family planning methods [23]. Moreover, women with more education are more likely to be engaged in professional and other employment activities, and thus may be more likely to want to limit their number of children. Several authors have also found a positive relationship between female education and contraceptive use in Bangladesh [22][23][24]. Our findings are consistent with most of the literature of South Asia and elsewhere [20,[25][26][27][28][29][30]. Women with more education were more likely to use contraceptives than uneducated women in India as well as Bangladesh [31].
According to previous studies, contraceptive use rates are expected to be higher in urban areas [20,22,23,31]. Our analysis also suggested that urban women were more likely to use contraception than rural women. This may be due to have better access to contraception of urban women than rural women. Urban women are usually keen to accept any method of contraception [32].
Religion has been detected to play an important role in using contraception. With regards to religion in all three surveys, Muslim women have lower odds of contraceptive use compared with non-Muslim women. This indicates that the less favorable attitude of Muslims toward family planning is gradually changing. Other studies also observed similar patterns [23,33]. Muslim women are likely to have a lower approval rate for contraceptive use [32].
Our analysis reveals a significant association between contraceptive practice and women with exposure to family planning information in the media. As for exposure to family planning messages via various forms of mass media, the results were significant in all three surveys. This finding is consistent with those earlier conducted in Bangladesh [22,34]. This finding is similar to that found by a study conducted in South Asia [20,35]. Of the respondents, those women who had no exposure to mass media were less likely to use modern methods than those who did have exposure to media in India [31]. Mass media exposure (radio/TV) has an important effect on reproductive behavior [32]. Contraceptive use and notions of ideal family size were changing at the same time as the use contraceptives [35]. One of the explanations for this result is that media can both apprize and motivate couples, even about such complex subjects as their reproductive means and goals.
In all three survey periods, age at first marriage was also found a significant factor. However, we did not find too much progress regarding to child marriage. Therefore, we strongly suggest the policy makers to take initiatives for reducing child marriage and increasing the rate of contraceptive use among married women in Bangladesh.
In terms of household wealth quintiles, the results showed that women from middle and rich households were more likely to use contraceptives than women in the poor quintile in 2004. A similar finding is observed in studies conducted in different African countries [36,37]. However, no significant effect of wealth quintiles was observed in other two survey periods.
Regarding ideal family size, the trend showed that for the last two surveys, only women who reported four or more as their ideal number of children were less likely to use contraception than women whose ideal number of children was 0 or 1 in 2004 and 2014. The relationship between ideal family size and contraceptive use has been found to be significant, which corresponds with many of the previous studies [19,38].
Women's fertility preference is also an important determinant of contraceptive use. In all three surveys, women who wanted to have another child and as well as women who were uncertain about having another child were less likely to use contraception than women who wanted no more children. This finding is similar to the earlier study [39]. One possible explanation for this could be that women using contraceptives have more control over their reproductive health [40].
In summary, we conclude that a respondent's current age, place residence, division, religion, level of education, age at first marriage, FP media exposure, ideal number of children and fertility preferences are the significant determinants according to the most recent survey, BDHS 2014. However, wealth index and a respondent's current working status were found to be significant factors only in BDHS 2004. The results of the study strongly recommend efforts to increase the education level among poor people, particularly within the women in Bangladesh. Program interventions, including health behavior education and family planning services and counselling, are especially needed for some categories of the population, including those living in rural areas, Sylhet, Chittagong and Dhaka divisions, illiterate women, and Muslim ever-married women. According to the FP 2020 of Bangladesh, the Sylhet and Chittagong divisions demand special focus to increase the use of contraceptives to 60% by 2021. This will be necessary as we move forward to address the goals of sustainable development goals.