Do free caesarean section policies increase inequalities in Benin and Mali?

The number of maternal deaths has decreased worldwide since 1990; however, 275,000 women still died giving birth in 2015, and almost half of these deaths occurred in sub-Saharan Africa [1]. A systematic review concluded that poorer women or women living in rural areas have less access to skilled delivery than richer women or those living in urban areas [2]. Inequalities also exist in terms of access to caesarean sections (C-sections), a major life-saving intervention needed by 3.6 to 6.5% of pregnant women [3]. Although C-section rates have increased in most low- and middle-income countries, these rates have typically increased faster for women in the richest quintile than for those in the poorest one [4]. A recent study of 2003–2013 showed that C-sections were extremely rare (less than 1%) among rural poorer women in seven out of 11 studied West and Central African countries, and only one country had a C-section rate higher than 2%. In contrast, eight of these countries had C-section rates higher than 4% among richer urban women [5].

In the 1980s, many African countries introduced user fees at the point of service to improve the quality of health services and access to primary health care. This change was supported by the Bamako Initiative (BI), which aimed to promote community financing of health services [6, 7]. User fees at the point of health service became a major barrier to health care access, especially for vulnerable populations [8]. This barrier is especially relevant for maternal health [9]. The fear of having to pay for excessive expenses may even lead some women to deter or delay their decision to seek care when they are dealing with obstetric complications [10]. Because C-sections are an expensive intervention, their access is directly influenced by household wealth and other non-financial factors [11], and the procedure can lead to catastrophic expenses [12, 13].

Some African countries have implemented user fee exemption policies for maternal health care services to improve access to maternal health services. In the 2000s, Benin and Mali, two western African countries with very high maternal mortality rates (respectively 405 and 587 per 100,000 live births), a low human development index ranking (167 and 175 out of 188,) and a high fertility rate (4.9 and 6.4 births per women) [14], decided to remove user fees only for women who receive a C-section.

In April 2009, the government of Benin introduced a national user fee exemption policy concerning all C-sections in selected public and private hospitals that offer emergency obstetric care. Hospitals receive 100,000 CFA (US$166) per C-section, which covers pre-operative laboratory tests, medications, surgery kits, blood, hospitalization for 7 days and transportation to a hospital if the woman is referred [15, 16]. The state is the principal financer of the reform [16].

The government of Mali introduced a user fee removal reform on January 1, 2005; this reform concerns all C-sections in the public sector and covers the surgical procedure and pre-operative examinations, the surgical kit and postoperative treatment (a standardized set of products and medications), and hospitalization. Structures receive 30,000 FCFA (US$50) for a simple C-section and 42,000 FCFA (US$70) for a complicated C-section in addition to a surgical kit [17, 18].

In contrast to other financing mechanisms, such as conditional cash transfers or targeted vouchers, the fee exemption policy for C-sections in Mali and Benin concerns all women and does not target poorer, rural or non-educated women. This policy has had a positive impact on access to C-sections and facility based delivery (FBD) among favoured and unfavoured women [19]. The impact of user fee exemptions on socioeconomic inequality is still highly uncertain [20]. In Mali, a patient survey study conducted 5 years after implementation of the free C-section policy to estimate the distribution of C-sections across socioeconomic groups showed that wealthier women clearly had greater access to C-sections than poorer women [21]. Some studies showed that user fee removal was far from sufficient to ensure equity in access to maternal health care, and in some cases, it increased existing inequality in access [22‐26].

Two recent studies used a robust approach via the difference-in-differences method to assess the effects of pregnancy-related fee removal on inequalities in access to maternal health services in Ghana, Senegal, Kenya and Burkina Faso [22, 27]. Only one of these studies evaluated the impact of the fee exemption policy on inequalities in access to C-section and found that the reform had its greatest impact on rural and less educated women [27]. However, this study did not determine whether inequalities were reduced after the policy was implemented in the studied countries compared to countries without a fee exemption policy.

The aim of this study was to observe whether there was an increase or a decrease in urban/rural and socioeconomic inequalities in access to C-sections and facility based deliveries after the free C-section policy was introduced in Mali and Benin.

We observed the evolution of inequalities in access to C-sections and facility based deliveries in the two countries through an observational study using repeated cross-sectional surveys.

We used the three DHSs for each country to analyse 47,302 women who delivered a live-born child in the 5 years prior to the interview (23,266 women in Benin and 24,036 women in Mali).

Figure 2 illustrates the evolution of C-section rates between study periods in Benin and Mali. In these two countries, regardless of the period, the C-section rates were higher for women in the most advantaged categories (urban, educated or richest) than for those in the other categories.

Table 1 presents the characteristics of women by period and country.

The number of pregnant women differs greatly by period and country because the births from the study periods do not correspond to the births from the DHSs. Indeed, the policy introduction date is shifted in time: 2005 in Mali and 2009 in Benin. This difference in timing explains why there are more studied births in period 2 than in period 3 in Benin and why these results contrast with the results for Mali.

The mode of delivery was available for 99.8% of the included women (47,209 women), and the place of delivery was available for 99.8% of the included women (47,217 women).

In both countries, we found no significant change in socioeconomic inequalities in access to C-section and FBD after the introduction of the free C-section policy, except for a reduction in education-related inequalities in access to C-section in Mali.

Otherwise, all significant changes that we observed began prior to the introduction of the policy.

Our results challenge ideas about free healthcare policies for all, which mainly benefit the most favoured social groups, such as the richest people or those who live in urban areas [36]. This study provides evidence that a user fee exemption policy does not necessary lead to an increase in existing inequalities and can benefit the least advantaged population categories. Similar results have already been shown not only in Africa but also in Asia and Latin America [7, 37]. However, even though a free C-section policy does not increase inequalities, these inequalities are still present, and increased health care access is insufficient for the poorest, non-educated and rural women. Some prior studies have already reached the same conclusions [24, 38, 39]. Similar to other reports [40‐42], we recommend switching to mechanisms that combine both a universal approach (health care for all) and targeted action for vulnerable populations to address this issue. The goal is to ensure equal health care access across individuals.

Our results on wealth-related inequalities are consistent with the results found by McKinnon et al. [22] concerning user fee removal policies focused on pregnant women (not specifically focused on C-sections) in other sub-Saharan African countries. They did not find robust evidence that this reform was associated with a reduction in wealth-related inequality in access to FBD. We found no difference in the evolution of wealth-related inequalities in access to C-section after implementation of the policy. The free policy benefitted the richest and poorest women in the same way. This conclusion is very similar to findings from another study, which concluded that the user fee policy in Burkina Faso benefited all categories of women, including the poorest women [43].

We found that education-related inequalities related to maternal education in access to C-section decreased in Mali after implementation of the free policy. Education allows women to evaluate whether they require treatment [44]. Prior to implementation of the free policy (when C-sections were more expensive), we hypothesize that educated women tended to pay for this intervention because they were aware of its benefits for themselves and their newborn infant when faced with obstetric complications. In contrast, non-educated women (often the poorest) could not afford to pay for an expensive intervention, regardless of its necessity. After implementation of the reform, even if non-educated women were not sensitive to the role of C-sections, they were more likely to follow their doctor’s decision when the fear of having a financial burden disappeared.

Furthermore, user fee exemptions can contribute to improving the decision-making power of women in health matters [45]. This policy might thus encourage pregnant women to request a C-section.

In Benin, unlike in Mali, the results revealed no significant differences in access to C-sections between urban and rural women or between educated and non-educated women, before or after the implementation of the policy. Benin and its population of 11 million is approximately 10 times smaller in terms of surface area than Mali (population 18 million) (Word Bank data). This difference implies that women have a better access to health care during pregnancy, even if they live in rural areas, as the distance to a health centre is less important.

Furthermore, according to the DHS 2011–12 in Benin, women who gave birth within the last 5 years prior to the interview were more urban (40% vs 19% in Mali) and more educated (28% received a primary education or more vs 16% in Mali) [46, 47]. The only related inequality was that the richest women had greater access to C-sections than the poorest women. Policies must therefore focus on measures that truly eliminate this inequality.

We did not find any studies that confirmed our findings on urban-rural access to C-sections in Africa. This result is an original finding showing that a user fee exemption at the point of service is not enough to reduce urban-rural disparities in access to C-section, even if rural women are often the poorest. Among the approximately 32,469 women living in rural areas who were included in our analysis, 27% belonged to the poorest quintile of wealth, while 49% of urban women were in the richest quintile.

Urban-rural inequalities in Mali decreased between period 1 (the end of the 1990s) and period 2 (the early 2000s). The Reference Evacuation System (RES) launched in 2002 in Mali can explain this decrease. The RES relies on improvements in communication, transportation, community cost-sharing, training and equipment in referral hospitals. A study showed that the RES had a positive impact on C-section rates among rural populations [48].

While the free C-section policy in Mali helped reduce education-related inequalities, some measures taken before the exemption had an effect on the inequalities between urban and rural areas. Mali still needs to focus more on the poorest women to achieve wealth-related equity.

This study has limitations. First, we used DHS data to investigate access to C-sections and FBDs. The reliability of this kind of survey can be questioned as every variable is self-reported and thus potentially subject to possible misclassification and recall bias. Moreover, this kind of survey collects information only on live births, leading to the omission of all stillbirths in these analyses. Stillbirth rates are high in western African countries [49], and there is evidence that C-section rates and intrapartum stillbirth rates are correlated [50]. However, we could not study these cases with our data. Second, this study did not consider the level of implementation of the free C-section policy in both countries. Previous studies showed that the cost of C-sections was still high in many hospitals in Benin after the policy was introduced [15] and that 91% of women still paid for their C-sections in a rural area of Mali during the period 2008–2011 [51]. The impact of this policy on socioeconomic inequalities might have been more important than the observed impact if C-sections were truly free for every woman. Third, DHSs provide information only on whether a child was born by C-section, but we could not verify whether the procedure was required. A recent study showed that C-section rates in Mali and Benin are very high for low-risk women and for women with a previous C-section [52]. These findings suggest that some of these C-sections would not be medically justified. The World Health Organization (WHO) stated in 2015 that “C-sections should be undertaken when medically necessary” [53], and there is no evidence showing a benefit of C-sections for women or infants who do not require this procedure. Fourth, we have only 3 years of data in Benin after the implementation of the policy. This limitation might partly explain why we found no significant evolution of inequalities in Benin after introduction of the policy. The use of the next DHS is necessary to study the sustainability of these impacts.

This study provides evidence that urban/rural and socioeconomic inequalities in C-section access did not change substantially in Benin or Mali after the countries implemented free C-section policies.

To achieve universal health coverage (UHC) and allow every woman to have access to C-sections without suffering from financial issues, the removal of financial barriers such as user fees must be a priority.

However, to improve equity in access to C-sections, user fee exemptions should not be enacted alone, and a voluntary governmental policy must be established to target vulnerable women without leaving anyone behind.