Association between timing and number of antenatal care visits on uptake of intermittent preventive treatment for malaria during pregnancy among Malawian women

Malaria remains a critical public health issue, with particular burden of disease on children aged younger than 5 years and pregnant women [1]. Malaria in pregnancy (MIP) poses a serious threat to mothers, fetuses and newborns, leading to infant and maternal mortality, low birth weight, and maternal anaemia [1]. Globally, MIP is responsible for 10,000 maternal deaths per year [2]. Furthermore, in sub-Saharan Africa, it accounts for 20 and 11% of stillborn and neonatal deaths, respectively [3, 4].

Malawi is a malaria-endemic country. Between 2009 and 2014, an infection rate of 19.6% was reported among pregnant women attending their first antenatal care (ANC) service in a peri-urban area of Ndirande, in Blantyre [5]. In addition to the aforementioned impact of MIP, the economic burden of malaria in Malawi is high. An average Malawian family spends approximately US$17.48 per malaria episode, which is higher than a week’s income for most Malawian families [6]. Considering the adverse effects of MIP, the use of intermittent preventive treatment for malaria during pregnancy (IPTp) is one of the most important strategies adopted by the Malawi Ministry of Health (MoH) to reduce the MIP burden [7].

In IPTp, pregnant women are prescribed sulfadoxine–pyrimethamine (SP) [8]. The World Health Organization (WHO) recommends that women in malaria-endemic areas should be given the IPTp dose during ANC visits as early as possible in the second trimester and that subsequent doses should be given at 4-week intervals [9]. An appropriate dose of IPTp reduces maternal mortality along with the risks of maternal anaemia and low birth weight [10]. A Malawian study, conducted in Blantyre from July 1997 to April 1999, reported a significant association between SP use and reduced placental malaria parasite prevalence among IPTp non-users (31.9%) and users (22.8%) [11].

In 1993, Malawi became the first country worldwide to implement an IPTp policy, recommending two doses of SP to all pregnant women [12‐14], 5 years before the WHO released its IPTp recommendations [15]. Thereafter, the coverage with two doses of IPTp (formerly, the recommended dose) has been relatively high in Malawi compared to the rest of sub-Saharan Africa [16‐18]. In 2015, 63% of pregnant women received at least two doses; the percentage of women receiving one IPTp dose increased from 80.7% in 2006 to 89% in 2015 [19, 20]. In Malawi, monitoring of the uptake of 3 IPTp doses started in 2014, after IPTp policy was updated by the WHO in 2012 [9, 19]; however, the uptake of 3 doses was as low as 12% in 2014. Although the 3-dose recommendation was new in 2014 and despite it being a regional leader in IPTp uptake, Malawi still failed to meet its own 2011–2015 Malaria Strategic Plan target of having 80% coverage of the uptake of 2 IPTp doses among pregnant women [18, 19, 21]. Nevertheless, increasing its uptake is crucial to reducing MIP burden. The effects of socio-demographics (e.g., education and parity) on IPTp uptake have been well documented [22‐27]. For instance, a 2009 study conducted in 8 Malawian districts revealed that women who had secondary- or tertiary-level education were more likely to take the recommended number of IPTp doses than those without formal education [24]. However, studies have also reported contrasting findings regarding the influence of socio-demographics on IPTp uptake [22‐25, 27]. Although a Tanzanian study found parity to be a crucial predictor of IPTp uptake [23], no association was reported in western Nigeria [28].

Although several studies have examined the effects of ANC attendance on IPTp uptake, the association has been inconsistent [26, 29‐33]. For example, a Kenyan study found an association between the timing of first ANC visit and uptake of IPTp [29], whereas studies in Uganda, Tanzania, and Nigeria have reported no such association [34‐37]. In Cameroon, less frequent clinic visits during pregnancy prevented the uptake of the recommended IPTp dose [38], whereas no such association was observed in Kenya [29]. The observed differences might be due to the various assessment durations; the Kenyan study [29] included not only currently pregnant women but also mothers with children aged under 1 year, whereas the Cameroonian study [38] was restricted to pregnant women. In addition, this inconsistency might be explained by the lack of consideration of the potential interactive effects of socio-demographics and ANC attendance on IPTp uptake.

Understanding the factors affecting IPTp uptake is critical for promoting and designing targeted interventions to improve the coverage of IPTp. Studies on the influence of both ANC attendance and socio-demographics on IPTp have focused only on their independent effects; thus far, limited empirical evidence is available on their interactive effects. Examining the effect modifiers of the relationship between ANC attendance and IPTp uptake is critical for identifying groups with a high risk of low IPTp uptake. Furthermore, although both the number of ANC visits and timing of first ANC visit are prominent factors, studies on ANC attendance and IPTp uptake have not elucidated how these two factors (i.e., number and timing) independently affect IPTp uptake, considering that the timing of first ANC visit may also influence the number of ANC visits, as reported in Tanzania [39].

Therefore, this study examined the relationship between ANC attendance, socio-demographics, and IPTp uptake among Malawian women by using a nationally representative sample. In particular, the study aimed to: (1) examine the factors (i.e., ANC attendance and socio-demographics) associated with the uptake of the recommended IPTp dose; (2) evaluate the independent association between the number of ANC visits and timing of first ANC visit on IPTp uptake; and, (3) investigate the moderating effects of socio-demographics on the relationship between ANC attendance and IPTp uptake.

According to literature, this is the first study attempting to identify the association between timing of the first ANC visit and number of ANC visits, and also to evaluate the interactive effects of ANC attendance and socio-demographics on IPTp uptake. Having more ANC visits was expected to increase the likelihood of taking the recommended IPTp dose. However, the study revealed that although inadequate ANC visits (< 4) were noted, women who had their first ANC visit earlier in the first and second trimesters respectively had 2.24- and 2.19-fold increased odds of receiving the recommended IPTp dose. The effects of adequate ANC visits on taking the recommended IPTp dose were significant in married women, those aged ≤ 34 years, and those in the middle or poor wealth categories, whereas early ANC initiation was associated with an increased likelihood of taking the recommended IPTp dose among rural women only.

The rate of recommended IPTp uptake was 30.2%. A substantially lower rate (11%) was reported in Simuyu, Tanzania [44]. The change in the WHO IPTp policy in 2014 (minimum number of doses from 2 to 3) may have been partially responsible for this low uptake because most countries were still in the process of adopting the new guidelines. In the current study, only 48.1% of women had adequate ANC visits, similar to the 46% reported in an earlier Malawian study [45]. These results suggest that IPTp coverage and ANC attendance remain a challenge. The results increase the understanding of the factors affecting the uptake of such services, particularly emphasizing the associated effects of ANC attendance; further relevant research is warranted.

This study determined that women with adequate ANC visits were more likely to take the recommended IPTp dose, consistent with studies conducted in Cameroon and Ghana [31, 33]. This may be because IPTp policies in these countries recommend that IPTp should be administered at ANC visits, preferably as direct observation therapy (DOT). Consequently, women with higher numbers of ANC visits are more likely to complete the required IPTp doses. By contrast, the Kenyan study found no such association [29]. Other factors, such as availability of drugs at ANC clinics and effective DOT implementation, might also contribute to the observed differences [37, 46]. Additional studies are required to understand the health facility-based factors that could influence IPTp uptake, such as accessibility of drugs at clinics and availability of clean water for DOT.

The current study’s results support the findings observed in Tanzania and Ghana [23, 33], where women with ANC initiation in the first and second trimesters reported a higher IPTp uptake. However, other authors [34‐37, 47] have reported no such association, potentially because of the different designs employed in these studies; for instance, Kibusi et al. [23] conducted a quantitative study, whereas Rass et al. [37] conducted a qualitative study. If a woman reports for ANC in the third trimester, she may not receive the full IPTp dose because of the delay in initiation of IPTp uptake.

Compared with other studies [23, 25, 26, 33, 38], this study further demonstrated the independent association between timing of the first ANC visit, in addition to the number of ANC visits, and IPTp uptake among pregnant women. In particular, the results suggest that reporting for the first ANC earlier during pregnancy, despite having an inadequate number of ANC visits, was associated with recommended IPTp uptake. ANC clinics are the main sources of disseminating crucial health information to women during pregnancy. Therefore, women may be advised on the importance of IPTp when they visit the ANC clinic early, and therefore, they may be able to source the drug from elsewhere if they fail to attend subsequent ANC visits [48].

Increased odds of recommended IPTp uptake in those with adequate visits despite having late initiation of ANC were also observed, but with a wider confidence interval, possibly because of the smaller sample number in this category. Two Ghanaian studies reported that 4 or more ANC visits were associated with higher IPTp uptake [26, 33]. The current findings further revealed the significance of this factor, independent of the timing of the first ANC visit, suggesting the need for healthcare workers to encourage late ANC initiators to make frequent ANC visits. However, women who report late for ANC with complications, such as malaria infection, may make frequent clinic visits and subsequently receive more SP as remedial treatment. In addition, the observed increased odds should be interpreted with caution because women who reported for ANC early in the third trimester may have completed the recommended IPTp dose, but those who reported late in the third trimester may not have.

Studies have reported age and marital status [23, 49], wealth [24, 30], education [23, 24, 30], and parity [23, 26, 30] to be significant predictors for IPTp uptake. However, this study did not observe a significant association between socio-demographics and IPTp uptake. Instead, moderating effects of socio-demographics on the association between ANC attendance and IPTp uptake were found in the sub-group analyses. Specifically, the effect of adequate ANC visits on taking the recommended IPTp dose was significant among married women, those in the poor and middle wealth groups, and women aged ≤ 34 years (p < 0.05). Notably, this relationship was significant at all levels of parity, with the highest odds observed for the multigravida group. In addition, the effect of desirable ANC timing on IPTp uptake was significant only among rural residents. Thus, tailor-made behavioural-change messages on IPTp uptake for different vulnerable groups of women are required to ensure effective scale-up of IPTp uptake.

The use of a nationally representative sample could strengthen the generalization of these results to Malawian women; however, because information on IPTp uptake and ANC attendance relied on recall, recall bias may have occurred. Recall ability is considered to be associated with the vividness, meaningfulness and significance of the event [50]. Recall may be better for less frequent and more significant events, such as the ones assessed in this study (IPTp uptake and ANC attendance), however, recall bias might still have compromised the findings. The MDHS data used for the analysis have not been verified with participants’ recalled information for either IPTp uptake or ANC attendance from the mothers on the ANC cards, which might have helped in reducing recall bias. Future studies should consider verifying recalled information by using the mothers’ ANC or health cards. Limiting the participants to those with a live birth in the 2 years preceding the survey also may have helped to reduce recall bias. Crucial factors, such as previous malaria infection, infection of family members, and health facility-based factors (i.e., drug stock-outs and ANC health worker attitude), which might affect pregnant mothers’ intention to return to ANC and thus their IPTp uptake, could not be evaluated in this study. Finally, because this was a cross-sectional study, causality could not be inferred.

ANC attendance, namely the number of ANC visits and timing of the first visit, strongly influences IPTp uptake among Malawian women. The results demonstrated the independent association of these two ANC attendance assessments on IPTp uptake. In addition to frequent ANC visits, earlier ANC initiation should also be emphasized, which may simultaneously facilitate recommended IPTp uptake to appropriately protect pregnant women from the risk of malaria infection. The significant moderating effects of socio-demographics can facilitate the identification and targeting of vulnerable groups to promote the recommended IPTp uptake rate in Malawi. Further research is required to examine the effects of health facility-based factors on IPTp use and the effects of IPTp dose on birth outcomes. Qualitative studies interviewing both women and health personnel are also required for further elucidating barriers to recommended IPTp uptake.