**1. Introduction**

Contraceptive use is considered as essential for protecting women's health and rights, influencing fertility and population growth, and helping to promote economic development, especially in sub-Saharan Africa. Contraceptive use helps to avoid maternal deaths by preventing unwanted pregnancies and abortions. It also helps to determine the number of children in the family and enhances adequate child spacing [1,2]. Some contraceptive methods such as condoms help, not only for family planning purposes, but also to prevent sexually transmitted infections, for instance, HIV, among others.

The various studies in the literature point out that satisfying a woman's need for contraception, particularly in family planning, may considerably reduce fertility [3,4]. However, young women confront many problems in contraceptive method use, mainly in family planning services, for instance, fear of side effects, cost, and lack of enough information [5]. In general, contraception use and family planning are essential for improving the health of the population. Although many United Nations member countries, especially those in the developed world, have strong family planning programmes, this is not the case in some sub-Saharan African countries, some sub-regions of Asia and Latin America and in the Caribbean, where, regardless of a rise in the contraceptive-prevalence rate, a large number of mainly poor, uneducated women and those with limited access to family planning services, continue to have unmet needs for contraceptives [6,7].

The prevalence of any type of contraception-use method among married women in Rwanda has improved in the last fifteen years. It was 17.4% among married women in 2005, tripled (52%) between 2005 and 2010, and was 53% in 2014 [8]. This may be attributable to the Government of Rwanda's commitment to strengthen the health sector, especially family planning services. The maternal and child mortality rate in Rwanda has also tremendously declined over a period of 20 years. As shown by the Rwanda Demographic and Health Survey reports, in 2000, 2005, 2010 and 2015 there were 1 071, 750, 476 and 210 deaths per 100,000 live births respectively for maternal and for child mortality [8]. In neighboring countries, especially in East Africa, in the last 15 years contraceptive use has generally increased. The prevalence of current contraceptive use among married women in Uganda was 26% in 2004, 34% in 2010 and 38% in 2016; in Tanzania it was 26.4% in 2005, 34.4% in 2010 and 38.4% in 2015 and 2016; in Kenya, it was 39% in 2003, 46% in 2008 and 2009 and 58% in 2014; and in Burundi it was 27.7% in 2010 and 29% in 2016 and 2017 [9–12]. Identifying determinants of contraceptive use among women of reproductive age is essential in order to formulate adequate health programmes, policies and strategies and possible interventions that can promote the well-being of children and maternal health in general. We now consider the problem of contraceptive use.

Various studies in the literature assessed the determinants of modern contraceptive (pill, IUD, injections, condoms, female sterilization, male sterilization, implants/norplant, lactational amenorrhea, standards days methods) use only [13–15] and did not include traditional methods (withdrawal, periodic abstinence) which may sometimes be the most accessible. The current study addresses this problem by combining all types of contraceptive-use methods among married women of childbearing age, nationwide. In addition, the RDHS data set has inherent nonlinear, spatial and random effects that needed to be accounted for. We consider these aspects to capture the heterogeneity of the data. Consequently, the main objectives of the current study were to address these problems in their entirety, by analyzing the factors associated with any type of contraceptive use and mapping the spatial distribution of use at district level in Rwanda. These objectives were achieved by applying structured spatial modelling that accounted for fixed, nonlinear, spatial and random effects. To the best of our knowledge, there is no study in the literature that assessed the determinants of contraceptive use among women in Rwanda, using a structured spatial model. It was expected that the findings from the present study would help policy makers and other public health institutions in Rwanda to visualize the spatial distribution of the use of any type of contraceptive method among women of reproductive age at district level. Therefore, it would help them to improve the current strategies when targeting the districts of low prevalence of contraceptive use.
