Despite the decline in fertility reported in sub-Sahara Africa, fertility level remains high. For instance, sub-Sahara Africa has witnessed decline in fertility level from 6.3 to 4.6 between 1990 and 2019 relative to global fertility decline of 3.2 in 1990 to 2.5 in 2019 (UNDESA, 2020) Similarly, within the same period, use of modern contraceptives among women of reproductive age has increased from 13% to 29% (UNDESA, 2020) Notwithstanding the improvement witnessed between 1990 and 2019 in sub-Sahara Africa, Total Fertility Rate (TFR) in the continent remains high, thereby contributing significantly to the high population in the region. In Nigeria, fertility level remains high (TFR=5.0), with a corresponding low modern contraceptive prevalence (mCPR) among currently married women 15-49 years, estimated at 18% (NBS and UNICEF, 2022) This is against Federal Government of Nigeria target of 27% by the year 2020 (NPC and ICF International, 2019). Two years down the line, the goal of achieving modern contraceptive prevalence of 27% is far from being achieved.

Despite the decline in fertility reported in sub-Sahara Africa, fertility level remains high. For instance, sub-Sahara Africa has witnessed decline in fertility level from 6.3 to 4.6 between 1990 and 2019 relative to global fertility decline of 3.2 in 1990 to 2.5 in 2019 (UNDESA, 2020) Similarly, within the same period, use of modern contraceptives among women of reproductive age has increased from 13% to 29% (UNDESA, 2020) Notwithstanding the improvement witnessed between 1990 and 2019 in sub-Sahara Africa, Total Fertility Rate (TFR) in the continent remains high, thereby contributing significantly to the high population in the region. In Nigeria, fertility level remains high (TFR=5.0), with a corresponding low modern contraceptive prevalence (mCPR) among currently married women 15-49 years, estimated at 18% (NBS and UNICEF, 2022) This is against Federal Government of Nigeria target of 27% by the year 2020 (NPC and ICF International, 2019). Two years down the line, the goal of achieving modern contraceptive prevalence of 27% is far from being achieved. 

Vast number of studies (Adebowale et al., 2013; Ahinkorah et al., 2021; Ahmed & Seid, 2020; Asiimwe et al., 2014; Geremew & Gelagay, 2018; Jacobs et al., 2017; Lasong et al., 2020) have identified individual and contextual level factors as predictors of modern contraceptive use among women. Individual level factors such as maternal age, parity, level of education and contextual factors such as place of residence, exposure to media and household wealth index predominate in the majority of these studies. However, there seems to be paucity of studies that have systematically explore the influence of joint decision making on modern contraceptive use overtime despite studies (Eshete & Adissu, 2017; Mutombo & Bakibinga, 2014; Nketiah-Amponsah et al., 2012, 2022) confirming positive association between joint decision and use of modern contraceptive. According to these studies, in settings where both partners jointly decide on contraceptive use resulted in higher contraceptive prevalence rate relative to areas where contraceptive use is not jointly decided. These studies also found knowledge of contraceptives, younger age group, wealth index, level of education, level of woman’s autonomy and number of living children were mainly associated with joint decision on contraceptive use. 

Moreso, even though women have the right to take decision regarding their reproductive health independently, in most societies, especially low-income countries, the decision of the woman is often set aside by their partner (Abu Aragaw, 2015; Delbiso, 2013; Hameed et al., 2014) This will obviously affect the fertility behaviour of the woman including fertility preference and decision to control number of children. Other African studies (Asresie et al., 2020; Bogale et al., 2011; Eshete & Adissu, 2017) have found a link between joint contraception decision-making and place of residence. Some authors (Eshete & Adissu, 2017) in Southern Ethiopia for instance, found more than half of the women jointly decide contraceptive use with their partner, although with differentials according to place of residence. More women residing in urban areas were more likely to jointly decide the use of contraceptive with the partners relative to their rural counterpart. The urban-rural differentials revealed better contraceptive knowledge (which might be attributed to better exposure to media and other information sources) and age difference among women residing in the rural area predicted joint decision on contraceptive use, while among women from rural area, high fertility level and support from the parent of the women significantly predicted joint decision on contraceptive among partners (Eshete & Adissu, 2017).   

Notwithstanding studies(Blackstone et al., 2017; Bogale et al., 2011; Prata et al., 2017; Upadhyay & Karasek, 2012) that examined factors associated with joint decision on contraceptive use, emphasizing the role of partners, this study, examines the predictors of joint decision on modern contraceptive use over a ten year period 2008-2018, taking into account the low modern contraceptive prevalence that has persist over time and paucity of studies examining trends using data from different data point.   

Study Area 

Nigeria has one of the highest fertility rates in the world with a TFR of 5.0 and low modern contraceptive prevalence of 18% among married women 15-49 years (NBS and UNICEF, 2022). The study performed a secondary analysis of the three most recent repeated cross sectional nationally representative survey data (2008, 2013 and 2018).   

Data Source, population, sampling, and sample size 

Data were pooled from three successive Demographic and Health Surveys (individual recode) datasets for the year 2008, 2013 and 2018. The DHS is a nationally representative survey which collects data from women of reproductive age 15-49 years who had given birth to at least one live birth for the five-year period preceding the survey and have at least three surviving children. The sampling design consisted of stratified three-stage cluster approach. By grouping each state into urban and rural areas, stratification was achieved. Within each cluster, a complete list of households was generated, resulting in a sampling frame for selecting households. Analysis of data was restricted to women currently married and living with their partner, having at least three surviving children in the age group 20-40 years. Weighted sample size 2018, 2013 and 2008 was (14,410, 16,804 and 18,009) making a total sample of 49,223 women.   

Measurement of Variables 

The outcome variable is decision to jointly use contraceptive between the woman and her partner. A question was asked from the woman – who makes decision when using contraception? The response to the question was categorized into – mainly respondent; mainly husband/partner; joint decision; others. However, in this study, the responses were categorized into two namely: joint decision coded as “1”, otherwise “0”. Explanatory variables used in this study were based on their significant association with the outcome variable at the bivariate analysis and literature (Blackstone et al., 2017; Nketiah-Amponsah et al., 2012) which include: age, age at birth, place of residence (rural or urban), region of residence, educational attainment, occupation (working; not working), household wealth index (based on NDHS classification), antenatal attendance, place of delivery, decision on large household purchases and exposure to mass media. Media exposure was measured from a composite of three variables namely: frequency of listening to radio, television and reading newspaper. It was then dichotomized as either exposed to at least one media source, coded as 1, and not exposed to any media source, coded as 0. Knowledge of contraceptives was defined as knows at least one method of contraception, coded as 1, while knows no method was coded as 0.  The variable Children Ever Born (CEB) was recoded into high fertility (TFR>4) and low fertility (TFR less than 4).

In performing the analysis, due to the complex nature of the DHS data, Stata “svy” command was applied to handle the complex DHS design. Both univariate and multivariate analyses were performed.  At the univariate level, descriptive analysis namely (frequency and percentage) was performed. At the bivariate level, association were tested using chi square and unadjusted binary logistic regression (tables not shown). At the multivariate level, binary logistic regression was performed. The binary logistics regression provides logit and odds effect of each of the explanatory variables on joint decision on contraceptive use among women and their partners. The results were presented as Adjusted Odds Ratio (AOR) and confidence intervals at 95% level of significance.

Ethical Approval

Approval to use the dataset was gotten from Measure DHS website upon registration and making request for the use of the data. The protocol used for the survey was reviewed and approved by the National Health Research Ethics Committee of Nigeria (NHREC) and the ICF Institutional Review Board. All the questionnaires were finalized in English and translated into the three major languages of Hausa, Yoruba and Igbo.

Socio-demographics

Table 1 presents the socio-demographic characteristics of the women. Women age group 20-30 years account for 64% in 2008 and 2013 and 61% in 2018. Nearly two thirds were in the low fertility regime (63% vs. 63% vs. 64%) for the year 2008, 2013 and 2018 respectively. Higher proportion (45%) of the women had no formal education, (28.8%) had secondary education, while (8%) had tertiary education. The proportion of women residing in rural area was 69% in 2008, 64% in 2013 and 60% in 2018. Also, more than two thirds of the women had between 1-4 children (75% vs. 73% vs. 72%) for the year 2008, 2013 and 2018 respectively. The level of autonomy among the women revealed little or no change over the ten-year period; 51% had autonomy in 2008, 46% in 2013 and 51% in 2018. Joint decision making on contraceptive use between the women and their partner was very low. Only 9% of the women in 2008, 10% in 2013 and 12% in 2018 jointly decide on contraceptive use with their partner.

Table 1: Socio-demographics Characteristics

The pattern of association between socio-demographic characteristics and joint decision on contraceptive use was similar for the ten-year period (table 2). For the year 2008, higher proportion of respondents who jointly decide on contraceptives were women in the younger age group 20-30 years (52%) low fertility regime (72%), whose age at first birth is between 20-29 years (62%), affiliated to Christianity (75.5%), had secondary education (48%), from the richest household wealth (50%), residence in urban area (59.4%), from the South-west region (43.3%), working (82.5%), with between 1-4 surviving children (79.1%), exposed to media (68,2%) and had autonomy (72.5%).

Similarly, in the year 2013, higher proportion of respondents who jointly decides contraceptive use were evenly distributed among the age group 20-30 years (50.0%) and 31-40 years (50.0%). However, the proportion was higher among women in the low fertility regime (73.4%), whose age at first birth is between 20-29 years (59.6%), affiliated to Christianity (75.6%), had secondary education (55%), from the richest household wealth (52%), residence in urban area (65.0%), from the South-west region (38.1%), working (83.0%), with between 1-4 surviving children (78.1%), exposed to media (92.1%) and had autonomy (80.4%).

In the year 2018, the proportion of respondents who jointly decides contraceptive use was higher among the age group 20-30 years (51.4%), women in the low fertility regime (72.0%), whose age at first birth is between 20-29 years (58.1%), affiliated to Christianity (69.6%), had secondary education (53.3%), from the richest household wealth (42.0%), residence in urban area (65.7%), from the South-west region (31.8%), with between 1-4 surviving children (76.1%), exposed to media (87.4%) and had autonomy (75.4%).

Overall, for the ten years period 2008-2018, younger age group 20-30 years, women in the low fertility regime, age at first birth, religion, level of education, household wealth, place and region of residence, work status, number of surviving children, exposure to mass media and autonomy were all significantly associated with joint decision on contraceptives use among the women and their partners.

Table 2: Socio-demographics Characteristics and joint decision on contraceptive use

**variable work status was missing in the 2018 dataset

Table 3: Socio-demographics Characteristics and joint decision on contraceptive use

**variable work status was missing in the 2018 dataset

Table 4 presents the result of the binary logistic regression. In the year 2008, older age group 31-40 years (OR=1.26; 95% C.I=1.0-1.5), place of residence (OR=0.72; 95% C.I=0.6-0.9), region of residence: North-east (OR=0.23; 95% C.I=0.2-0.3), North-west (OR=0.17; 95% C.I=0.1-0.3), level of education: Primary (OR=1.73; 95% C.I=1.2-2.3), secondary (OR=2.17; 95% C.I=1.6-2.9) and tertiary education (OR=3.46; 95% C.I=2.4-5.1), religion: Islam (OR=0.71; 95% C.I=0.6-0.9), Household wealth: rich (OR=2.41; 95% C.I=1.7-3.5) and richest (OR=3.33; 95% C.I=2.2-5.0), Work status: working (OR=1.50; 95% C.I=1.2-1.9) and autonomy: has autonomy (OR=1.50; 95% C.I=1.3-1.8) all significantly predicted joint decision on contraceptive use among women and their partner.

For the year 2013, factors that were significantly associated with joint decision on contraceptive use include: older age group 31-40 years (OR=1.35; 95% C.I=1.1-1.6), region of residence: North-east (OR=0.41; 95% C.I=0.3-0.6), North-west (OR=0.42; C.I=0.3-0.6), level of education: primary (OR=2.30; 95% C.I=1.7-3.0), secondary (OR=3.48; 95% C.I=2.6-4.7) and tertiary education (OR=4.90; 95% C.I=3.4-7.0), religion: Islam (OR=0.67; 95% C.I=0.5-0.8), household wealth: poorer (OR=1.59; 95% C.I=1.0-2.5), middle (OR=2.44; 95% C.I=1.5-3.8), rich (OR=2.93; 95% C.I=1.8-4.7) and richest (OR=5.22; 95% C.I=3.2-8.5), media exposure: yes (OR=1.27; 95% C.I=1.0-1.6) and autonomy: has autonomy (OR=2.08, 95% C.I=1.8-2.5).

For the year 2018, significant predictors of joint decision on contraceptive use are:  rural residence  (OR=0.79; 95% C.I=0.7-0.9), region of residence: North-west (OR=0.76; 95% C.I=0.6-1.0), South-east (OR=0.67; C.I=0.5-0.8), and South-south (OR=0.56; 95% C.I=0.4-0.7), level of education: primary (OR=2.36; 95% C.I=1.9-2.9), secondary (OR=2.65; 95% C.I=2.1-3.1) and tertiary education (OR=3.54; 95% C.I=2.7-4.7), religion: Islam (OR=0.41; 95% C.I=0.3-0.5) and traditional (OR=0.23; 95% C.I=0.1-0.7), household wealth: middle (OR=1.63; 95% C.I=1.2-2.1), rich (OR=2.02; 95% C.I=1.5-2.7) and richest (OR=2.78; 95% C.I=2.0-3.8), media exposure: yes (OR=1.50; 95% C.I=1.2-1.8), autonomy: has autonomy (OR=1.82, 95% C.I=1.6-2.1) and fertility regime: high fertility (OR=1.08; 95% C.I=1.6-2.1). Factors associated with joint decision on contraceptive use in 2008, 2013 and 2018 were similar except for media exposure and fertility in 2008, residence in 2013 and age and fertility regime in 2018 that were not significant.

Regardless of the study period, older age group 31-40 years, age at first birth 30 years and older, rural residence, region of residence: north-east, north-west, south-south and south-west region, level of education: primary, secondary and tertiary, religion: Islam and traditional, household wealth: poorer, middle, rich and richest, media exposure: women exposed to mass media, work status: working women and autonomy: women with autonomy were all significantly associated with joint decision on contraceptives among women and their partner.

Table 4: Binary logistic regression showing predictorsof joint decision on contraceptive use.

*=statistically significant at 0.05, **=statistically significant at 0.001

This study has examined factors associated with decision to jointly decide contraceptive use among women and their partners using evidence from a repeated cross-sectional nationally representative data for the period 2008-2018. Our study found factors that consistently predicted joint decision on contraceptive use include place of residence, region of residence, level of education, religion, household wealth and autonomy. Women residing in rural area were less likely to jointly decide contraceptive use with their partner. This is likely because women in rural areas are more likely to attain low economic status such as level of education, thereby impacting on their level of autonomy and their decision-making power. In another study (Eshete & Adissu, 2017), the authors were of the opinion that higher contraceptive use among urban women might be attributed to the fact that they are better exposed to information about family planning. Also, in our present study, increasing level of education and being affiliated to Christian religion was associated with higher likelihood of joint decision on contraceptive use. This corroborated the findings of previous studies (Eshete & Adissu, 2017; Mutombo & Bakibinga, 2014; Nketiah-Amponsah et al., 2012). The positive effect of education can be explained in several ways. First, women who are well educated will no doubt be better informed and have accurate knowledge about contraceptives in addition to the benefits associated with its use relative to the uneducated women. Also, women who are still schooling are more likely to use contraceptive for the purpose of avoiding getting pregnant which can cause interference in their education. These findings, also aligns with an Ethiopian study (Eshete & Adissu, 2017). In the study, women with no formal education and affiliated with the Islam religion had lower odds of making joint decision on contraceptive use with their partners. The influence on religion on contraceptive use might be due to religious factors that tend to influence the acceptability of contraceptives among couples from different religious settings (Tiruneh et al., 2016).

In this our current study, higher household wealth was associated with greater odds of joint decision-making process on contraceptive use. This importance of this finding can be attributed to greater capacity to obtain modern contraceptives without the need to rely on their partner. Also, it is important to note that wealth index according to the DHS measures is an aggregate of family assets including television, radio, phone, cars among others all of which can positively contribute to the ability of the woman to access information among others (Currie, 2009). In this present study, where mass media and work status were significant, women exposed to mass media and working had higher odds of making joint decision on contraceptive use relative to their counterpart not exposed to mass media and not working. Similarly, women with autonomy also demonstrated higher likelihood of joint decision on contraceptive use. Previous studies (Eshete & Adissu, 2017; Mutombo & Bakibinga, 2014) have also reported positive association between joint decision and use of modern contraceptives. This suggests an increasing need for better spousal communication among couples on reproductive health and use of contraceptives.

The study concluded by recognizing the importance of socio-economic characteristic of women and autonomy as factors to be considered when designing policies and programmes targeting the promotion of joint decision on contraceptives among women and their partner, with the overall goal of promoting increased contraceptive prevalence rate. This implication of our findings is that any intervention seeking to promote joint contraceptive use among couples must take into consideration the socioeconomic, cultural, and religious barriers that adversely affect the ability of women to make informed decision on their reproductive health. Also, future family planning program should make concerted effort at ensuring that male partners are well considered when developing their programs and the need to promote spousal communication on issues related to use of contraceptives. 

Notwithstanding the importance of this study on policy, some limitations are worth noting. First, because this study made use of secondary data, the reliability of the findings will to a large extent be dependent on the quality of reporting during the survey. Also, the outcome variable talks about joint decision making. However, response for the male partner was obtained from the woman’s perspective and might be subject to bias. Nevertheless, our study represents one of the few studies that have examined predictors of joint decision making using repeated cross sectional nationally representative data across three data points, thus making important contribution to literature.

the authors declared no conflicting interest.

No funding was received for the study.