Use of reproductive healthcare services is critical for the health and well-being of mothers and their babies. Utilization of antenatal care, in particular, has been linked to lower rates of both maternal and infant morbidity and mortality [1]-[3]. However, strategies for increasing service utilization, including improving health education, providing financing for maternity services, and increasing the supply of trained birth attendants, have had limited impact [4]-[6]. The use of reproductive care, therefore, may be influenced by additional parameters. Lack of empowerment among women in the context of romantic relationships may be particularly important to the utilization of reproductive health services [7], because the existence of partnerships is inextricably linked to the need for reproductive healthcare.

Previous research examining the link between empowerment and use of antenatal care is limited and inconsistent. First, results are derived primarily from a handful of studies [8]-[13], many of which demonstrate strong associations between certain aspects of empowerment, such as freedom of movement, but not others, such as control over finances. Additionally, studies often neglect to consider experiences of partner abuse alongside other aspects of empowerment. Partner abuse is the attempt to control one’s partner by through emotional, physical, or sexual violence [14]. Evidence suggests that experiences of partner abuse may reduce the woman’s ability to use antenatal care [15]-[17]. Furthermore, these studies are primarily derived from populations of women in Asia and East Africa and thus have limited generalizability to other low-income contexts. Last, none of these studies examine how the association between empowerment and antenatal care may be moderated by education, a potentially modifiable factor that is increasingly promoted by international development agencies.

Accordingly, we aimed to examine the relationship between several measures of empowerment and use of inadequate antenatal care within a nationally representative cohort in Ghana. Our study focused on Ghanaian women, a population in which the reported maternal mortality ratio is 450 deaths per 100,000 live births, one of the highest in the world [18]. Furthermore, we included several socio-demographic characteristics to adjust for their associations with utilization of inadequate antenatal care and to determine if the association between empowerment and antenatal care was moderated by education.

We found that one in four women who had been pregnant and in a relationship in the last 12 months received inadequate antenatal care in Ghana. We also found that women who are less empowered, as measured by their experience of physical abuse, are far more likely to be in this at-risk population. Our results are consistent with other literature suggesting that partner abuse is associated with lower likelihoods of receiving any antenatal care and sufficient antenatal care [15]-[17],[24],[25]. They are also consistent with analyses conducted among women in Bangladesh and Egypt demonstrating that physical abuse is significantly associated with antenatal care even after accounting for other indicators of empowerment [15],[25]. We found that the experience of physical abuse was a strong marker for low empowerment as the majority of women who reported physical abuse also reported emotional abuse and limited freedom of movement. This study is the first of which we know to demonstrate the empirical importance of empowerment in reproductive health care services use in West Africa.

Our study also finds a strong protective effect of education among women who experience physical abuse. In fact, among women with at least some formal education, having experienced physical abuse is not associated with receiving inadequate antenatal care, while for those with no formal education, the association between physical abuse and inadequate care is pronounced. These findings likely support the importance of achieving the Millennium Development Goal to promote gender equality and empower women by 2015 by increasing literacy and formal education among girls [26]. Future research, however, should aim to replicate these findings as our sample size among women with no formal education was small and our confidence interval was wide.

This analysis has important implications for future research and practice. Women who experience physical abuse not only experience negative physical and mental health consequences related to the abuse [27] but may also experience sub-optimal levels of reproductive health care, potentially also affecting their own and their babies’ health. Hence, antenatal care providers who care for women who are not receiving adequate care may want to consider screening for or making provisions for additional care or services women may need due to consequences of physical abuse and low empowerment, particularly because physical abuse co-occurs alongside many other instances of partner control. Community health care or social workers may be important links for these patients during the antenatal and postnatal periods. Furthermore, although current literature tends to consider empowerment in isolation of experiences of partner abuse or violence, our findings suggest that future research should consider including the measurement of physical abuse as an indicator of empowerment. Including this measure may more accurately identify women who experience low levels of empowerment and thus better target this population for interventions. Last, although improving access to education and ensuring quality educational programs may be challenging, our research suggests education may be a powerful tool for reducing the effects of low empowerment on women’s health.

Despite the strengths of our study, including the use of a nationally representative sample as well as several measures of empowerment and socio-demographic factors, we recognize the limitations for our analysis. First, data were self-reported and thus are subject to misclassification. Social desirability bias may have led to under-reporting certain aspects of empowerment, particularly the measures of abuse, and over-reporting participation in prenatal care. Recall bias may have also affected results as women were asked about the use of antenatal care up to 12 months earlier. We expect these potential biases, however, to non-differentially affect responses, thus potentially biasing results towards the null. Second, we could not exclude women with pathological or high-risk pregnancies; the existence of such could be associated with both her number of antenatal care visits and empowerment, thus potentially confounding our results. We believe, however, that this confounding would potentially bias our results towards the null as pathological pregnancies are likely associated with higher numbers of antenatal care visits and lower levels of empowerment. Third, our analysis is cross-sectional and thus can make no assertions about causality but may be useful as a foundation for future work examining temporal effects. Additionally, our empowerment measures were asked with respect to the last 12 months and although women in our sample were pregnant within the last 12 months, use of antenatal care may have preceded this period. We do not expect the timing of these measurements to differentially affect women in our sample, thus dampening any associations found in our analysis. Fourth, because women were pregnant during this period about which the empowerment variables were asked, the frequencies of physical abuse, in particular, may be lower than they would be if the women were not pregnant, as rates of physical abuse have been shown to be lower during pregnancy [28]. Last, although we attempted to use several constructs to measure empowerment based on prior literature taken primarily from Asia and East Africa, it is not clear to what extent these measures are generalizable to Ghana. Future efforts to validate measures of empowerment for Ghana or West Africa generally are warranted. Additional concepts of decision-making and economic dependence were unavailable in our dataset and thus could not be included in our analysis.

Empowerment appears to have substantial effects on the use reproductive health care within this national sample of women in Ghana. Our findings suggest that improving reproductive health will require reducing partner abuse and enhancing empowerment among women in this and other low-income countries, particularly among those with no formal education. Furthermore, as reproductive health and empowerment within the context of intimate partner relationships are both predicated on partnerships, the involvement of male partners is critical for improving reproductive health outcomes. Last, increasing education among girls in these settings is likely a strong approach for improving reproductive health and buffering the effects of low empowerment among women.

CU conceptualized and helped conduct the national study and provided comments on the final version of the manuscript. HS, AOA, MC, and EB conceptualized the analysis and interpreted results. HS conducted the analysis and drafted the paper. All authors critically reviewed and edited the paper and approve its final version.