+1 Recommend
0 collections
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      Women's Access and Provider Practices for the Case Management of Malaria during Pregnancy: A Systematic Review and Meta-Analysis

      Read this article at

          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.


          Jenny Hill and colleagues conduct a systematic review and meta-analysis of women’s access and healthcare provider adherence to WHO case-management policy of malaria during pregnancy.

          Please see later in the article for the Editors' Summary



          WHO recommends prompt diagnosis and quinine plus clindamycin for treatment of uncomplicated malaria in the first trimester and artemisinin-based combination therapies in subsequent trimesters. We undertook a systematic review of women's access to and healthcare provider adherence to WHO case management policy for malaria in pregnant women.

          Methods and Findings

          We searched the Malaria in Pregnancy Library, the Global Health Database, and the International Network for the Rational Use of Drugs Bibliography from 1 January 2006 to 3 April 2014, without language restriction. Data were appraised for quality and content. Frequencies of women's and healthcare providers' practices were explored using narrative synthesis and random effect meta-analysis. Barriers to women's access and providers' adherence to policy were explored by content analysis using NVivo. Determinants of women's access and providers' case management practices were extracted and compared across studies. We did not perform a meta-ethnography. Thirty-seven studies were included, conducted in Africa (30), Asia (4), Yemen (1), and Brazil (2). One- to three-quarters of women reported malaria episodes during pregnancy, of whom treatment was sought by >85%. Barriers to access among women included poor knowledge of drug safety, prohibitive costs, and self-treatment practices, used by 5%–40% of women. Determinants of women's treatment-seeking behaviour were education and previous experience of miscarriage and antenatal care. Healthcare provider reliance on clinical diagnosis and poor adherence to treatment policy, especially in first versus other trimesters (28%, 95% CI 14%–47%, versus 72%, 95% CI 39%–91%, p = 0.02), was consistently reported. Prescribing practices were driven by concerns over side effects and drug safety, patient preference, drug availability, and cost. Determinants of provider practices were access to training and facility type (public versus private). Findings were limited by the availability, quality, scope, and methodological inconsistencies of the included studies.


          A systematic assessment of the extent of substandard case management practices of malaria in pregnancy is required, as well as quality improvement interventions that reach all providers administering antimalarial drugs in the community. Pregnant women need access to information on which anti-malarial drugs are safe to use at different stages of pregnancy.

          Please see later in the article for the Editors' Summary

          Editors' Summary


          Malaria, a mosquito-borne parasite, kills about 600,000 people every year. Most of these deaths occur among young children in sub-Saharan Africa, but pregnant women and their unborn babies are also vulnerable to malaria. Infection with malaria during pregnancy can cause severe maternal anemia, miscarriages, and preterm births, and kills about 10,000 women and 100,000 children each year. Since 2006, the World Health Organization (WHO) has recommended that uncomplicated malaria (an infection that causes a fever but does not involve organ damage or severe anemia) should be treated with quinine and clindamycin if it occurs during the first trimester (first three months) of pregnancy and with an artemisinin-based combination therapy (ACT) if it occurs during the second or third trimester; ACTs should be used during the first trimester only if no other treatment is immediately available because their safety during early pregnancy has not been established. Since 2010, WHO has also recommended that clinical diagnosis of malaria should be confirmed before treatment by looking for parasites in patients' blood (parasitology).

          Why Was This Study Done?

          Prompt diagnosis and treatment of malaria in pregnancy in regions where malaria is always present (endemic regions) is extremely important, yet little is known about women's access to the recommended interventions for malaria in pregnancy or about healthcare providers' adherence to the WHO case management guidelines. In this systematic review and meta-analysis of qualitative, quantitative, and mixed methods studies, the researchers explore the factors that affect women's access to treatment and healthcare provider practices for case management of malaria during pregnancy. A systematic review uses predefined criteria to identify all the research on a given topic. Meta-analysis is a statistical method for combining the results of several studies. A qualitative study collects non-quantitative data such as reasons for refusing an intervention, whereas a qualitative study collects numerical data such as the proportion of a population receiving an intervention.

          What Did the Researchers Do and Find?

          The researchers identified 37 studies (mostly conducted in Africa) that provided data on the range of healthcare providers visited, antimalarials used, and the factors influencing the choice of healthcare provider and medicines among pregnant women seeking treatment for malaria and/or the type and quality of diagnostic and case management services offered to them by healthcare providers. The researchers explored the data in these studies using narrative synthesis (which summarizes the results from several qualitative studies) and content analysis (which identifies key themes within texts). Among the studies that provided relevant data, one-quarter to three-quarters of women reported malaria episodes during pregnancy. More than 85% of the women who reported a malaria episode during pregnancy sought some form of treatment. Barriers to access to WHO-recommended treatment among women included poor knowledge about drug safety, and the use of self-treatment practices such as taking herbal remedies. Factors that affected the treatment-seeking behavior of pregnant women (“determinants”) included prior use of antenatal care, education, and previous experience of a miscarriage. Among healthcare providers, reliance on clinical diagnosis of malaria was consistently reported, as was poor adherence to the treatment policy. Specifically, 28% and 72% of healthcare providers followed the treatment guidelines for malaria during the first and second/third trimesters of pregnancy, respectively. Finally, the researchers report that concerns over side effects and drug safety, patient preference, drug availability, and cost drove the prescribing practices of the healthcare providers, and that the determinants of provider practices included the type (cadre) of heathcare worker, access to training, and whether they were based in a public or private facility.

          What Do These Findings Mean?

          These findings reveal important limitations in the implementation of the WHO policy on the treatment of malaria in pregnancy across many parts of Africa and in several other malaria endemic regions. Notably, they show that women do not uniformly seek care within the formal healthcare system and suggest that, when they do seek care, they may not be given the appropriate treatment because healthcare providers frequently fail to adhere to the WHO diagnostic and treatment guidelines. Although limited by the sparseness of data and by inconsistencies in study methodologies, these findings nevertheless highlight the need for further systematic assessments of the extent of substandard case management of malaria in pregnancy in malaria endemic countries, and the need to develop interventions to improve access to and delivery of quality case management of malaria among pregnant women.

          Additional Information

          Please access these websites via the online version of this summary at

          Related collections

          Most cited references 36

          • Record: found
          • Abstract: found
          • Article: found
          Is Open Access

          Why Do Women Not Use Antenatal Services in Low- and Middle-Income Countries? A Meta-Synthesis of Qualitative Studies

          Introduction Recent estimates of global maternal mortality ratios (MMRs) suggest a substantial decline in recent years [1],[2]. However, current rates of decline will still fall well short of meeting Millennium Development Goal 5 (MDG 5): reducing maternal mortality by 75% by 2015 [3]. Data from the World Health Organization (WHO) indicate that in many low- and middle-income countries (LMICs), especially in sub-Saharan Africa, the rate of decline in MMR is less than 1% per year, and in some countries (e.g., South Africa, Nigeria, Mozambique, and Swaziland) rates even appear to be increasing [1],[4]. This slow rate of progress is starkly highlighted in the most recent “Countdown to 2015” report, which found that only nine of the 74 countries with the highest MMRs in the world were on target to achieve MDG 5 [5]. WHO reports and experts in the field consistently highlight the lack of access to local, adequately resourced health care facilities as an important reason for the relatively slow rate of progress towards achieving MDG 5 [6],[7]. Access includes ensuring comprehensive antenatal care coverage for all pregnant women. Recent estimates indicate that the number of women in LMICs attending at least one antenatal appointment increased from 64% in 1990 to 81% in 2009, and those attending four or more times rose from 35% to 51% over the same period [2]. However, major disparities exist within and between continents, between countries, and between urban and rural populations [8]. As with the MMR figures, the rate of progress is slowest in sub-Saharan Africa, where antenatal coverage rates have improved slightly during the last two decades, but the number of women visiting four or more times has remained static, at about 44% [2]. Although the correlation between “inadequate” antenatal care and high maternal mortality is complicated and contentious, it is widely accepted that antenatal care presents opportunities to identify pregnancy risks and, in a broader sense, to monitor and support the general health care of women who may be susceptible to a range of potentially fatal pathologies including HIV, anaemia, malnutrition, tuberculosis, and malaria [5]–[9]. Global implementation of strategies designed to encourage antenatal attendance tend to be based on the assumption that if high-quality services are provided, people will come to them. However, data from quantitative population-level studies suggest that this is not necessarily the case for some groups of pregnant women. Well-documented socio-demographic data indicate that women from relatively poor backgrounds, living in rural areas, and/or with low levels of education are less likely to access antenatal services, even if they are provided [10]–[12]. Other factors, including having a husband with a low level of education, living a long distance from a clinic, and having high parity, have also been identified as barriers [13]–[17]. Similar factors emerge in reviews of barriers to antenatal care in developed countries [18]–[21], which suggests that the issues for women who remain marginalised at local, national, and global levels are much the same. Based on the results of a WHO antenatal care randomised trial [22], the standard measure of adequate antenatal care delivery is a minimum of four antenatal visits (with the first occurring during the first trimester) for a woman and her foetus, if they are judged to be healthy following a standard risk assessment [23]. Although some authorities, e.g., the US Agency for International Development, have noted the need for woman-centred, individualised, culturally specific programmes [24], the recent BMJ Best Practice guidance on routine antenatal care provision lists a wide range of routine screening, testing, and health education topics, with little emphasis on individual concerns and circumstances [25]. Evidence equating risk-focused, low-frequency antenatal care with clinical outcomes in LMICs is limited, but a recent Cochrane review found that population groups in LMICs receiving fewer antenatal visits (4–6) had an increased risk of perinatal mortality and, in particular, stillbirth [26]. The author of a WHO commentary on this review hypothesizes that the excess perinatal loss for women in LMIC settings may be due to inadequate local tailoring of risk assessment, low numbers of staff, and inadequate training [27]. The WHO manual on antenatal care [23] does not specify how antenatal care should be funded, the nature and relevance of staff attitude and training, or what resources should be available at which level of care provision. However, tacit assumptions are likely to include that staff are available and have high levels of communication and interpersonal skills, and that the programme is affordable, otherwise it would be unlikely to function. Despite the findings of the review, and speculation about the components and the effectiveness of the WHO programme, it remains the standard for adequate antenatal care provision. Given the potential significance of context in mediating whether women access antenatal care, qualitative studies may provide fresh insights into pertinent issues in specific settings. In terms of LMICs, such studies suggest that some women do not attend antenatal facilities because of deeply held cultural beliefs and/or tribal traditions surrounding the nature of pregnancy and childbirth [28],[29]. Qualitative studies can also illuminate the effect of local policies and incentives, such as the use of antenatal clinic cards to guarantee intra-partum hospital access—a controversial practice in a number of African countries because of the potential for discrimination against women who don't have any record of antenatal clinic attendance [30]. However, because of the highly contextualised nature of individual qualitative studies, policy makers often overlook them, and their findings remain outside of global, national, and local health care strategies [31]. Systematic review and synthesis of qualitative studies can generate hypotheses about how successful programmes work, and why unsuccessful programmes fail certain individuals and groups [32]. To address the latter question with regards to inadequate accessing of antenatal care, we planned to locate, analyse, and synthesise qualitative studies exploring the views, beliefs, and experiences of women from LMICs who did not access antenatal care at all, or accessed it inadequately, according to the WHO definition given above. The intention was to develop hypotheses about lack of attendance that could inform policy development, based on a new understanding of why some women still don't access antenatal care, even when it is made available. Qualitative Meta-Synthesis Methodology The emphasis in meta-synthesis is on rigorous study selection and the careful interpretation of data across studies, contexts, and populations. This combination and interpretation of findings from a number of systematically selected studies in a particular subject area shares methodological similarities with its quantitative equivalent, meta-analysis. When meta-synthesis is used to explain or interpret existing knowledge, e.g., alongside meta-analysis, it can be aggregative and deductive [32]. However, when it is exploring fields where there is little prior information, it is undertaken as an inductive method, designed to generate theoretical insights and hypotheses that can be tested in future research [32]. In the latter case, the classic approach is meta-ethnography [33]. As with qualitative research, the direct findings of meta-synthesis are not usually generalisable, but the theoretical insights or hypotheses arising from the synthesis of the included studies should be transferable to other similar settings and contexts [34]–[36]. In meta-synthesis, as in grounded theory, the comprehensiveness of the analysis is determined by the concept of theoretical saturation. Theoretical saturation is reached when new studies do not change the emerging theory or hypothesis, and when a systematic search for disconfirming cases in all the included studies reinforces the theoretical insights. Given the scope and rigour of meta-synthesis reviews, there is greater potential for them to inform practice, influence policy, and underpin strategy than for individual qualitative studies [37],[38]. Methods Search Strategy and Selection Criteria The search strategy was designed to locate qualitative studies exploring the antenatal care experiences, attitudes, and/or beliefs of women from LMICs who had chosen to access antenatal care late (after 12 wk gestation), infrequently (less than four times), or not at all [7]. We searched for any studies that might include qualitative data, including survey-based studies with open-ended written responses, mixed methods studies, focus groups, and one-to-one interviews. No language restriction was imposed. All electronic searches used keywords covering the main search domains including “antenatal”, “prenatal”, “maternity”, “pregnancy”, “care”, “service”, “provision”, “access”, and “attendance”. The searches were conducted across a range of medical, sociological, and psychological databases (MEDLINE, Embase, PubMed, AMED [Allied and Complementary Medicine Database], BNI [British Nursing Index], CINAHL [Cumulative Index to Nursing and Allied Health Literature], PsycINFO, Wilson Social Science Abstracts), as well as continent-specific databases such as Latindex (Literatura Latino-Americana e do Caribe em Ciências da Saúde) for South American publications and AJOL (African Journals Online) for articles published in Africa. Where possible, we sought to narrow the search to LMICs by incorporating the World Bank's list of low- and middle-income economies in the search terms [39]. Some specific papers were recommended by colleagues, and we hand-searched relevant journals in the departmental and university libraries. Other articles were obtained from reference lists published in identified studies. The initial search included papers published between 1 January 1980 and 31 March 2011. An updated search was completed on 14–15 February 2012, after which the contents pages of relevant journals were reviewed (via Zetoc) as they were published. These updated searches have provided a means to check that the thematic structure and synthesis developed in the primary analysis continue to hold true as new studies are published (“theoretical saturation”). Both authors reviewed all of the included papers independently, and then reached a final agreement on inclusion by consensus. All of the papers meeting our eligibility criteria were assessed for quality using an appropriate published tool [40]. This tool incorporates a pragmatic grading system [41] and uses an A–D scoring system. The authors determined grades by consensus, and studies scoring C+ or higher were included in the final review (see Table S1). Analysis and Synthesis Our intention was to generate new theoretical insights that could form the basis for hypothesis testing in the future, so we used the meta-ethnographic approach developed by Noblit and Hare [33]. This approach has been used successfully in meta-synthesis studies related to several different health care settings [19],[35],[42]–[44]. It is not restricted to ethnographic studies, as the approach can incorporate the full range of qualitative methods. We began by identifying the findings from one paper and comparing them with the findings from another, to generate a “long list” of emerging concepts. These early concepts were then examined to identify similarities, in a process that is termed “reciprocal translation”. During this process, some concepts were collapsed together to create a parsimonious thematic structure. Each author then reviewed the themes independently to ensure there were no data that were at odds with our analysis and that no data remained unexplained. This stage of the process is analogous to searching for discomfirming data and is termed “refutational translation” in meta-ethnographic studies [33]. The themes were then synthesised into a “line of argument” synthesis—a phrase or statement that summarises the main findings of the study and the theoretical insights that they generate. This synthesis was then used to create a hypothetical model to explain why women fail to make adequate use of antenatal services in LMICs. Reflexive Accounting In qualitative research, the researcher is the instrument of measurement, and the final analysis is a product of the interaction between the researcher and the data. Reflexive accounting allows the reader of the final research product to assess the degree to which the prior views and experiences of the researchers may have influenced design, data collection, and data interpretation in any specific study. In this case, S. D. believed that interpersonal relationships were likely to be critical in mediating antenatal care use, and K. F. believed that whether women accessed care was most likely to be influenced by personal and/or localised socio-economic circumstances. To minimise the effect of these beliefs, both authors were particularly rigorous in looking for refutational data in these specific areas as the analysis progressed. Results Our search to 31 March 2011 generated a total of 3,622 hits, including 625 duplicates, which were removed at this stage. Of the remaining 2,997 articles, 2,892 were excluded by title and abstract because they failed to address the initial selection criteria. Most of the studies removed at this stage were excluded for one of three reasons: (1) they were conducted in high-income countries, (2) they were obviously quantitative, or (3) they were not about access to antenatal care. Of the remaining 105 papers, a further 75 were removed after independent full text review by the authors, largely because they lacked sufficient qualitative data (n = 36), were based on the experiences of women who attended antenatal services regularly rather those who didn't (n = 25), reflected the views of service providers rather than the women attending care (n = 8), or were concerned with access to health care generally as opposed to antenatal care specifically (n = 6). This left 30 papers that were taken forward for quality assessment. Following independent review, the authors agreed that nine studies failed to meet the quality requirements, leaving 21 that were taken forward for analysis and synthesis (see Figure 1 for details of the selection process). Of the nine studies excluded, three were mixed methods studies with very limited qualitative data, two reported on the views of health care providers with little emphasis on the responses of service users, two presented qualitative information in a quantitative format (frequency of responses), and two failed to meet the quality criteria for design, methodology, and/or analysis. Only one study meeting the inclusion and quality criteria was identified by the updated searches since 31 March 2011 [45], and this was used to check the explanatory power of the final thematic structure, synthesis, and interpretation. 10.1371/journal.pmed.1001373.g001 Figure 1 Flow chart summarising search strategy. Description of the Studies The 21 papers in the final full synthesis represent the views of women from 15 countries (Bangladesh [×2], Benin, Cambodia, Gambia, India, Indonesia, Kenya, Lebanon, Mexico, Mozambique, Nepal, Pakistan, South Africa [×4], Tanzania [×2], and Uganda [×2]) and include data from more than 1,239 participants (minimum 10, maximum 240) who were either interviewed directly or gave their opinion as part of a focus group (see Table S1 for full details of the included studies). Two of the studies utilised a mixed methods approach, and although these studies contained limited qualitative information, the narrative data were pertinent and reasonably well reported. Ten of the 21 studies were conducted in a rural setting, three took place in an exclusively urban environment, and the remaining eight involved both urban and rural settings. The earliest paper was published in 1992 and the most recent in 2011, with the majority (n = 17) being published within the last ten years. More than half of the included papers (n = 12) were published within the last three years, which suggests an upswing in interest in this area of research (see Table 1 for a summary of included papers). 10.1371/journal.pmed.1001373.t001 Table 1 Summary of included studies. Authors [Reference] Year of Publication Country Location—Type of Region Number of Participants Method Used Quality Grading Abrahams et al. [46] 2001 South Africa Cape Town—semi-urban 32 Interviews C+ Myer and Harrison [47] 2003 South Africa Hlabisa district—rural 29 Interviews B Pretorius and Greeff [48] 2004 South Africa Mafikeng-Mmbatho districts—rural 18 Interviews C+ Mrisho et al. [49] 2009 Tanzania Lindi and Tandahimba districts—rural 58 Focus groups B Matsuoka et al. [50] 2010 Cambodia Kampong and Cham provinces—rural 66 Interviews and focus groups B Choudhury and Ahmed [51] 2011 Bangladesh Rangpur and Kurigram districts—rural 20 Interviews C+ Chapman [52] 2003 Mozambique Vila-Gondola—semi-urban 83 Interviews A Grossmann-Kendall et al. [53] 2001 Benin Cotonou and Ouidah districts—urban and rural 19 Interviews C+ Ndyomugyenyi et al. [54] 1998 Uganda Kigorobya sub-country—rural 80–120a Focus groups C+ Gcaba and Brookes [55] 1992 South Africa Durban—urban 10 Interviews B Atuyambe et al. [56] 2009 Uganda Wakiso district—rural 92 Focus groups B Stokes et al. [57] 2008 Gambia Kiang West district—rural 83 Interviews and focus groups C+ Griffiths and Stephenson [58] 2001 India Pune and Mumbai—mix of urban and rural at each location 45 Interviews B Simkhada et al. [59] 2010 Nepal Kathmandu area—semi-urban and rural 30 Interviews B Titaley et al. [60] 2010 Indonesia Garut, Sukabumi, and Ciamis districts, West Java—semi-urban and rural 119 Interviews and focus groups B Family Care International [61] 2003 Kenya Homabay and Migori districts—mix of urban and rural in each 27–47a Interviews and focus groups B Tinoco-Ojanguren et al. [62] 2008 Mexico Chiapas—mix of urban and rural 16 Interviews C+ Mumtaz and Salway [63] 2007 Pakistan Punjab—rural 39–55a Interviews and focus groups B Chowdhury et al. [64] 2003 Bangladesh Dhaka and Upazila—urban and rural 16 Interviews B Mubyazi et al. [65] 2010 Tanzania Mkuranaga and Mufinidi districts—both rural 240 Interviews and focus groups B+ Kabakian-Khasholian et al. [66] 2000 Lebanon Bekaa, Akkar, and Beirut—rural, semi-rural, and urban 117 Interviews C+ a A range is given for these studies, as the authors list the number of focus groups conducted, with a minimum and maximum number of participants; e.g., ten focus groups with 8–12 participants. Description of the Themes The emerging concepts and themes are summarised in Table 2. We identified a total of seven emerging themes and three final themes (summarised below), two of which relate specifically to initial attendance at antenatal facilities, and a further, service-oriented, theme relating to maintaining attendance. 10.1371/journal.pmed.1001373.t002 Table 2 Summary of themes. Initial Concepts (Findings from Primary Papers) Relevant Papers (References) Emerging Themes Final Themes Awareness of signs/symptoms of pregnancy 46–51 Pregnancy awareness and disclosure—awareness of signs and symptoms of pregnancy; cultural reasons for keeping pregnancy secret Pregnancy as socially contingent and physiologically healthy—pregnancy as a normal life event—only attend antenatal care when sick; lack of awareness of pregnancy indicators; lack of understanding of antenatal care benefits; embarrassment; cultural and supernatural implications of pregnancy disclosure; preference for traditional healers and medicines (including cost savings) Cultural reasons for keeping pregnancy secret 46,48,49,52–57 Don't recognise/understand Western approaches to health care 46,47,54,56,58,60 Resistance to risk-averse care models—don't recognise/understand Western approaches to health care; lack of perceived benefits; pregnancy as a normal life event; reliance on traditional/alternative antenatal practices; influence of family members Lack of perceived benefits of attendance 46–48,51,56,58,59,62–64 Pregnancy as a normal life event 46,50–54,56,58–61,66 Reliance on traditional/alternative antenatal practices 50,52,54,58,60–62 Influence of family members 50,51,62–64 Costs (direct and indirect) 46,49,50–56,58–66 Prioritising limited resources for basic survival—costs (direct and indirect); laziness Resource use and survival in conditions of extreme poverty—costs (direct and indirect), transport, and distance; time off work and child care—may be made to wait several hours; inadequate infrastructure (especially in rural areas); potential for accident/attack en route Laziness 46,47,49,62 Lack of transport and distance to clinic 46,48,50,54–56,58,60,65 Difficult and dangerous travel —lack of transport and distance to clinic; inadequate infrastructure Inadequate infrastructure 48,49,55,58,60–62,64 Lack of staff/medicine/care at clinic 49,50,54,56,58,65 Attending clinics is not worth the effort—lack of staff/medicine/care at clinic; waiting times at clinic Not getting it right the first time—poor staff attitude; inflexibility of antenatal care services; issuing of cards for delivery at a hospital (women don't return) and staff giving card holders preferential treatment; few, poorly trained staff; lack of facilities; lack of medicines Waiting times at clinic 46,48–50,52, Attendance only to get a card (for hospital delivery) 46,47,50,52,61 Locally determined rules of access—attendance only to get a card; inflexible booking systems Inflexible booking systems 46,63 Poor staff attitude 46,48–51,53,55–57,62,65,66 Insensitivity, disrespect, and abuse —poor staff attitude; embarrassment Embarrassment (about examination or inability to pay) 46,49,56,65, Theme One: Pregnancy as Socially Contingent and Physiologically Healthy This theme incorporates two concepts (highlighted below) that emphasise some of the cultural and contextual nuances associated with pregnancy. Many women in these studies described pregnancy as a healthy physical state and saw little reason to visit health professionals when there was no perceived threat to their well-being. In some cultures this reluctance to engage with antenatal services was further compounded by a belief that pregnancy disclosure could lead to unwanted religious or spiritual complications. Pregnancy awareness and disclosure: “It's better to wait, to see if you really are pregnant” For many respondents, traditional or cultural beliefs dictated that they should wait until they had missed several periods before confirming a pregnancy [46]–[51]. Sometimes it's difficult to tell that you are pregnant. Some people have irregular periods, they miss periods for months only to find they are not pregnant, so it is better to wait, to see if you are really pregnant. [Pregnant woman, rural South Africa] [47] This belief limited early accessing of care. Even when women suspected they were pregnant, the motivation to visit an antenatal clinic was often superseded by cultural and superstitious beliefs about pregnancy disclosure [52]–[63]. In rural Pakistan, the shame (sharam) associated with pregnancy, because of the obvious relationship with sexual activity, meant women were less willing to be seen in public places [63]. The shame of being pregnant and the subsequent reluctance to be seen in public was also a factor for pregnant teenagers in Uganda [56]. In other parts of Africa and South East Asia, the potential to be “cursed” by evil spirits or jealous or vindictive contemporaries had a detrimental effect on pregnancy disclosure [52],[53],[55],[64]. One South African woman who had recently experienced a neonatal death explained her loss in the following manner: I think my boyfriend's previous girlfriends were jealous of my pregnancy and they bewitched me. [55] These kinds of beliefs appeared to be relatively common in rural communities and discouraged women from visiting public places, especially antenatal clinics, where a visit would be perceived as a public declaration of pregnancy. Resistance to risk-averse care models: “What is the point in going for a check-up in a healthy condition?” In many of the studies, women reported that they didn't feel the need to seek professional care when there was nothing wrong with their pregnancy [46]–[49],[52]–[54],[58],[63],[64]. As no-one expects to be sick during pregnancy, visiting the centre for a check-up is not necessary. What is the point in going for a check-up in a healthy condition? [Non-user of antenatal care services, rural Bangladesh] [64] Pregnancy was viewed as a normal life event rather than a medical condition requiring professional monitoring and supervision. This was especially true for multiparous women who had experienced one or more healthy pregnancies [50],[54],[58]. If a woman has always delivered well, she does not see the need for antenatal care attendance or visiting the health unit to deliver. [Pregnant woman, rural Uganda] [54] In some hierarchical cultures the decision to engage with antenatal services was made by tribal elders, husbands, mothers-in-law, or senior family members rather than the women themselves [50],[56],[59],[62]–[64]. Findings from a Nepalese study highlight the central role played by the mother-in-law when it came to making decisions about whether to go for antenatal care. My mother-in-law doesn't help. It might be due to her past experiences. She used to do all the work by herself during her time of pregnancy, so she wants me to do the same. I have lots of work here at home so I don't go for [antenatal care] check-ups. [Non-user of antenatal care services, rural Nepal] [59] Theme Two: Resource Use and Survival in Conditions of Extreme Poverty All of the studies were conducted in populations affected by poverty, and our findings suggest that, in such circumstances, limited personal resources were often directed towards immediate survival needs rather than specific pregnancy-related concerns. Even when antenatal care was offered free of charge, the cost of transport (sometimes across difficult or dangerous terrain), the loss of women's labour to the family, and the possibility of having to pay for additional medicines rendered attendance impossible. Using resources for health care or basic survival: “If there is no money, we can't go” In virtually all of the identified studies [46],[49],[50]–[56],[58]–[66], the costs (both direct and indirect) of visiting antenatal facilities were viewed as a significant factor in restricting or inhibiting access to antenatal care: It is good to go to the doctor during pregnancy, but if there is no money we can't go. I wanted to go but I didn't have the money to pay. [Limited user of antenatal care services, Mumbai, India] [58] Even in countries offering free access to antenatal care, the unanticipated costs of paying for drugs, tests, and medical cards placed an additional strain on limited family finances. The reason I did not go back there [to the antenatal clinic] is because my husband only buys what he wants when he is given the prescription. For example, when there are three things prescribed he buys only two. So, why should I take the trouble to go there for nothing? If I go to the clinic every month, he will just ask how much I think he earns to be able to buy so many medications for me. [Limited user of antenatal care services, Benin] [53] The indirect costs of getting to and from antenatal facilities were highlighted consistently in the included studies, especially those conducted in rural areas [46],[48],[50],[54]–[56],[58],[60]. The prohibitive costs of taxi and bus fares prevented some women from visiting antenatal clinics, and, in cases of extreme poverty, even the most basic forms of transport came at an unaffordable price. When I was pregnant what prevented me from seeking healthcare was lack of transport money because my legs were a problem. I used to live far away in the hills and I could not ask anyone to take me on a bicycle because I would be asked for money. [Adolescent limited user of antenatal care services, rural Uganda] [56] Some of the respondents' accounts indicated that the need for women to contribute to relatively meagre household resources was more than simply a useful contribution. It was perceived to be crucial for survival, especially at significant times in the farming cycle: When I had a third pregnancy, it was harvest season. So I wanted to help my husband, even during the pregnancy. [Non-user of antenatal care services, rural Cambodia] [50] Difficult and dangerous travel: “It is so far and the road condition is too bad” Many of the studies included in this synthesis were conducted in predominantly rural areas with relatively basic transport networks. For pregnant women living in towns and villages without community health care facilities, the need to journey to distant locations to receive antenatal care presented travelling difficulties, which they were unwilling or unable to overcome [46],[47],[50],[52],[58],[60],[62]. I never visited the health centre to check my pregnancy because it is so far and the road condition is too bad. [Non-user of antenatal care services, rural Cambodia] [50] Even in situations where women were prepared to make lengthy journeys on foot, sometimes walking for three to four hours, the associated risks and challenges occasionally prevented them from doing so. In parts of Africa, the prospect of being attacked by wild animals or aggressive men deterred women from making these journeys, whilst in South East Asia, the deterioration of the roads during the rainy season had a similar detrimental effect. This suggests that the barriers were not just nonexistent or expensive transportation, or inadequate roads, but also the fear of physical harm, which outweighed any benefits that might be gained from antenatal care: It is really hard when it is raining. We are afraid we will fall over because the road is so slippery and we are pregnant. The health centre is far and you can see that the road is poor. [Limited user of antenatal care services, West Java, Indonesia] [60] Theme Three: Not Getting It Right the First Time Given the very real and critical issues of how women perceive pregnancy, and of the economic and physical sacrifice needed from the woman and her family to get her to a central antenatal clinic, it is crucial that the services she receives when she gets there are “fit for purpose”, and that the benefits are perceived to outweigh potential harms. Unfortunately, for many of the women included in this review, this was not the case. Attending clinics is not worth the effort: “It's better to go to the TBA [traditional birth attendant]” The relatively poor economic circumstances of the countries included in this study meant that clinics were often severely under-resourced. Pregnant women who initially recognised the benefits of antenatal care and who made the often significant financial and personal sacrifices to visit health care facilities were often disappointed by the lack of resources they found when they finally got there. As a consequence, they made the decision not to return [48]–[50],[54],[56],[58],[65]. I don't visit the health centre for antenatal care because the health centre doesn't have enough medical equipment. When we have a problem, all they will probably do is refer us to the referral hospital…. [Non-user of antenatal care services, rural Cambodia] [50] The amount of time women had to wait to be seen by health professionals, especially after long and difficult journeys, was a common cause of complaint and discouraged some of them from attending again [46],[48],[49],[52],[54],[65]. Pregnant women also complained about the cursory nature of consultations in understaffed clinics and made the decision to revert to more traditional forms of antenatal care. They just touch your abdomen, it's better to go to the TBA [traditional birth attendant] because the TBA examines the mother and tells her how the baby is lying in her stomach. [Pregnant woman, rural Uganda] [54] Locally determined rules of access: “If you do not have a card, they will not accept you” Our findings suggest that in a number of cases, particularly in sub-Saharan Africa, the practice of giving antenatal cards to women attending the clinic is being poorly managed and is having a detrimental effect on continued access. Some health care providers use the clinic card as “a passport” and refuse to admit labouring women to a clinic or hospital if they do not have one [46],[49],[52],[53],[65]. This kind of negative reinforcement has created a situation in which pregnant women visit an antenatal facility only once—to get a “clinic card”. I am just afraid of being denied services when I need them, so one must just go [to antenatal care] to get the [clinic] card. If you do not have a card, they will not accept you when there is a problem…otherwise we could just rest at home. [Woman in ninth month of pregnancy, rural Tanzania] [65] Disrespect and abuse: “They don't care for patients” One of the most common reasons given for delaying or restricting antenatal visits was the poor attitude of staff at health care facilities [46],[48],[50],[51],[53],[56],[62],[63],[65],[66]. Findings from countries in Africa, Asia, and South America highlight insensitivity, rudeness, humiliation, neglect, abuse, and even physical violence by health centre staff as key factors in limiting women's accessing of antenatal care. Sometimes the poor attitude of health care providers was described by what they failed to do, as recounted by a young woman in Uganda: They [health care workers at an antenatal clinic] don't care for patients, for example when you go in the morning they will ask you “at your home don't you sleep”. When you go at lunch time they will ask you whether at your place you don't take lunch. And when you go in the evening they will tell you they have closed up. [56] Authors also reported that women felt intimidated because of the potential for abuse: When you see the health agent yelling at women who are not feeling well, you are afraid of telling them what is wrong with you too…. [Pregnant woman, Benin] [53] In other contexts, women recounted being punished or humiliated because of some perceived minor misdemeanour: If you arrive late at the clinic, the staff rebukes and punishes you with a fine or they order you to clean the floor or sweep the surroundings. [Limited user of antenatal care services, rural Tanzania] [65] In all of these examples, women reported feeling reluctant to return for another appointment, and some reverted to more traditional forms of antenatal care as a consequence. Line-of-Argument Synthesis Antenatal care provision that is based on a concept of pregnancy as a potentially risky biomedical state, and that, as a consequence, provides mechanisms focused mainly on surveillance in more or less centralised locations, is contextually at odds with the theories, beliefs, and socio-economic situations of pregnant women and their families in a range of LMICs. This situation is compounded when accessing services presents additional risks to women and their families, in terms of lost labour or income, or physical danger; when the promised care is not delivered because of resource constraints; and when women experience covert or overt abuse in care settings. Hypothesis Based on the Findings Following the claim by Pawson [67] that “programmes are theory incarnate”, our data can illuminate the potential inconsistencies between theories underpinning antenatal care programmes based on the WHO antenatal care model [23] and the themes that underpin the beliefs, actions, and experiences (the local context) of those to whom these programmes are targeted (see Figure 2). We hypothesize that the dissonance between these two frames of reference might explain the lack of initial attendance at antenatal clinics, as described in the first row of Figure 2. The second row of the figure illustrates a second misalignment, this time between the principles assumed to underpin antenatal care provision, and the experiences of women who use them. We hypothesize that this misalignment may explain the lack of return visits for antenatal care after the first encounter. 10.1371/journal.pmed.1001373.g002 Figure 2 Hypothetical model of inadequate access to antenatal care in low and middle income countries. Testing for Theoretical Saturation The data from the one paper [45] we identified after the end of our formal search phase in March 2011 can be incorporated into our explanatory model, suggesting theoretical saturation. We would argue that future studies should therefore focus on testing our hypothesis, and designing specific solutions to the problem of inadequate attendance building on this summary of all the relevant qualitative data to date. This approach would avoid the problem of “analytic interruptus” described by Statham in relation to the continual reproduction of single-site qualitative studies with no attempt to translate the emerging theoretical insights into action [68]. Discussion Some of the issues identified by this meta-synthesis are common to other areas of maternity care and health care in general. At the family level, these include lack of household resources, especially when faced with the problem of formal and informal payment or services [69],[70], and the problems inherent in travel to centralised health care services [46],[51]–[54],[71]. Restricted autonomy for women has been identified as a factor underpinning inability to make personal decisions about health service use [72], and this factor is one of the underlying elements of the “three delays” hypothesis relating to lack of accessing of emergency care in labour [73]. There is also an increasing recognition of the problem of human rights abuses in health care in general [74]–[76]. From a theoretical perspective our findings suggest the hypothesis that, in certain contexts, there may be a misalignment between the theories that underpin the provision of antenatal care and the beliefs and socio-economic contexts of women who access services irregularly or not at all. The dissonance between these two frames of reference might explain the lack of initial access to antenatal care. A second disparity, this time between the nature of antenatal provision and the expectations of the women who use the services, may explain the lack of continued engagement. We are not aware of previous studies that have integrated these factors into an analysis of antenatal care use based on women's views and experiences, or that have identified pregnancy as a socially risky but physically healthy state. Minimising social stigma and risk requires care provision that is discrete and certainly not provided in public places subject to long queues for services. Strategies incorporating culturally appropriate understandings of maternity care tailored to individual communities are rare, but participatory programmes where local women and community leaders are actively engaged in the planning of local antenatal services have been shown to be effective in increasing antenatal coverage and reducing maternal and infant mortality [77]. This approach is well illustrated by a recent Cambodian study, which showed a 22% increase in antenatal attendance when service users and influential stakeholders became involved in the planning of community maternity services [78]. From a socio-economic perspective our findings suggest that, even in situations where women recognise the benefits of antenatal care and are willing, in principle, to attend, the physical barriers and even physical risks of getting to and from the clinic, coupled with the potential loss of crucial family resources, can be prohibitive. Even if services are free (with no covert point-of-care costs) and safe transport systems are provided, taking women from essential farming duties on long trips to and from central clinics might, at the extreme, still present a risk to family food supplies. In this case, the benefits of antenatal care must weigh strongly in the balance for service users before uptake will increase. Also, even where women do have a degree of personal autonomy, those who see pregnancy as a healthy state, but as socially risky, are still unlikely to value and attend programmes that expose their pregnant state, and that are largely focused on identifying and averting risk. This is especially pertinent when both the direct and opportunity costs are high, travel to central locations is difficult and dangerous, and the services they receive are of poor quality and overtly or covertly abusive. Projects designed to incentivise pregnant women to attend antenatal care have been implemented successfully in some LMICs. The Janani Suraksha Yojana cash transfer programme in India, where women are paid a small amount to attend antenatal care and give birth in a recognised health care facility, has had a significant effect on antenatal attendance and subsequent levels of neonatal and perinatal mortality [79]. An alternative, transport-based project in eastern Uganda, where local motorcycle riders were contracted to take women to and from antenatal clinics throughout their pregnancy, also showed a significant increase in antenatal attendance [80]. However, doubts remain about the practicality and sustainability of these kinds of initiatives, and, as our findings illustrate, many pregnant women remain unconvinced by the focus and quality of antenatal programmes, and seem unlikely to make full use of antenatal facilities unless care quality is improved. Given that data were not available from every region of every LMIC, it is possible that our line-of-argument synthesis, and our model, do not apply to all contexts in which antenatal care is underused. However, the comprehensiveness of our analysis is reinforced by evidence of theoretical saturation, both from our refutational analysis, and from the paper [45] published after the end of our formal search phase in March 2011. Our hypothetical model can explain the findings of this study, including the influence of cultural beliefs and lack of respect from health care professionals. In addition, the findings of the meta-synthesis are similar to those of a parallel review of women's accounts of non-accessing or limited accessing of antenatal care in resource-rich countries [19]. Given the range of countries that were represented in the meta-synthesis, and the date range of the publications (over nearly two decades), the issues seem to be universal and persistent. We hope that our synthesis illustrates the need to move from small repeated studies of the same problem in local contexts towards a more comprehensive understanding of the potential dissonance between context and service delivery mechanism across all of these settings. A more thorough evaluation using the realist review approach could test this hypothesis, and contribute towards a more detailed understanding of this issue [81]. This could provide the basis for a new approach to the design and delivery of antenatal care, founded on a careful analysis of distinctive local beliefs, values, and resource availability. Such an approach could identify a way of moving services away from broad population-based solutions, towards new service designs based on what works, for whom, in what circumstances [81]. Conclusion Despite moderate success in reducing MMRs and increasing antenatal care coverage, the global targets associated with MDG 5 seem unlikely to be attained by 2015, especially in many LMICs. So far, practical initiatives to address these issues have tended to adopt centralised, public provision of antenatal care based on utilitarian strategies designed to minimise clinical risk. This approach benefits some women and infants, but it marginalises others, as the programme design does not take into account necessary survival decisions, beliefs, attitudes, or cultural theories that may be intrinsic to the local context. Measures designed to sustain and maintain access in LMICs are likely to be more effective when policy makers and service providers align their programmes with the theoretical and philosophical constructs and the severe practical constraints that underpin the local community context. Such programmes must ensure that, once they access services, all pregnant women are treated with dignity, respect, and compassion. If programme delivery is not aligned with local contexts in this way, the findings from this meta-synthesis suggest, even the best and most physically accessible services may remain underused by some local pregnant women. Supporting Information Table S1 Assessment of quality of included studies. (RTF) Click here for additional data file.
            • Record: found
            • Abstract: found
            • Article: found
            Is Open Access

            Factors Affecting the Delivery, Access, and Use of Interventions to Prevent Malaria in Pregnancy in Sub-Saharan Africa: A Systematic Review and Meta-Analysis

            Introduction Malaria in pregnancy can have important consequences for the mother, foetus, and newborn child, yet the harmful effects are preventable [1]. The adverse outcomes of malaria in pregnancy can be substantially reduced by interventions that have been available for over two decades [2]–[4] and that are inexpensive and cost-effective [5]. Access to and use of these interventions by pregnant women is, however, extremely low, representing a failure of the public health community. In areas of stable malaria transmission in Africa the World Health Organization (WHO) recommends a package of intermittent preventive treatment in pregnancy (IPTp) with sulphadoxine–pyrimethamine (SP) and use of insecticide-treated nets (ITNs), together with effective case management of clinical malaria and anaemia [6]. IPTp consists of two doses of SP taken 1 mo apart commencing in the second trimester [7],[8]. Both IPTp and ITNs are commonly delivered through antenatal clinics (ANCs) through collaboration between malaria and reproductive health programmes. The Roll Back Malaria Partnership aims to ensure that all pregnant women receive IPTp and at least 80% of people at risk from malaria in areas of high-intensity transmission use ITNs by 2010 [9], with even more ambitious targets of 100% for both interventions by 2015 [10]. Achievement of high coverage of these preventive interventions among pregnant women remains elusive for many countries in sub-Saharan Africa [11],[12]. A recent review of national survey data shows that in 27 countries with survey data between the years 2009 and 2011, the median coverage of two doses of SP was 24.5% (range 7.3%–69.4%), even though the median coverage for at least two ANC visits was 84.6% (range 49.7%–96.9%, 22 countries, 2003–2011) (A. M. van Eijk, personal communication), representing substantial missed opportunities at ANCs. Despite the call for universal ITN coverage [13] and all 45 malaria-endemic countries having a policy of providing ITNs to pregnant women, the median use of an ITN the previous night among pregnant women in 37 countries from survey data for the years 2009–2011 was 35.3% (range 5.2%–75.5%) (A. M. van Eijk, personal communication). According to a Countdown to 2015 report, in 20 countries with data, IPTp and ITNs, together with case management of malaria during pregnancy, have the lowest coverage among all the interventions delivered to pregnant women at ANCs [14]. Evidence on the determinants of coverage and reasons for the failure in delivery and uptake of IPTp and ITNs from qualitative [15] and quantitative studies is currently disparate, in addition to which, many relevant reviews are now outdated [5],[16]–[18]. We therefore undertook a systematic review to update the evidence and to integrate findings from three separate syntheses of studies on (1) barriers to achieving high coverage, (2) determinants of uptake, and (3) interventions to increase coverage. We then explored the extent to which the intervention studies have addressed known barriers and determinants, and identified critical gaps in the knowledge required for the formulation of effective strategies. The review was restricted to sub-Saharan Africa as the only malaria-endemic region with a specific WHO strategy for the prevention of malaria in pregnancy, which includes both IPTp with SP and ITNs. Methods Search Strategy We performed a systematic and comprehensive literature search of electronic databases on 23 April 2013, including the Malaria in Pregnancy Library (; updated 20 April 2013) and the Global Health Database [19], and a search of bibliographies of retrieved articles. The Malaria in Pregnancy Library contains peer-reviewed published and unpublished literature compiled from 40 sources including PubMed, the Global Health Library, Google Scholar, Lilacs (Latin American and Caribbean Health Sciences Literature), Popline, the ProQuest Digital Dissertations and Theses database, Web of Knowledge, WorldCat, and registers of trials and studies [20]. A full account of the search terms used is presented in Table S1. Study Inclusion Criteria and Analysis Strategy Titles and abstracts were reviewed independently by two authors (J. Hill and J. Hoyt/A. M. van Eijk). Studies were eligible for inclusion if they met the following criteria: (1) reported an original research study; (2) addressed barriers to, facilitators of, or determinants of the delivery or uptake of IPTp and/or ITNs in pregnancy, or evaluated the impact of an intervention to increase the coverage of IPTp and/or ITNs in pregnancy; (3) were published between 1 January 1990 and 23 April 2013; and (4) were conducted in sub-Saharan Africa. No restrictions were placed on publication language or study design, i.e., quantitative, qualitative, and mixed methods studies were included, and both peer-reviewed papers and grey literature were included. Studies meeting the inclusion criteria were grouped according to whether their content addressed (1) barriers or facilitators, (2) determinants, and/or (3) evaluation of intervention(s); some studies contributed to more than one of these content groups (Figure 2). Studies with content on barriers or facilitators and/or determinants were then further categorised into studies exploring factors among pregnant women, healthcare providers, or both. Studies with content on delivery interventions were categorised by intervention, i.e., IPTp, ITNs, or both. The kappa statistic was used to measure the chance-adjusted inter-rater agreement for eligibility. Data Extraction Two authors extracted data and appraised the quality and content of included studies. J. Hill and J. Hoyt/A. M. van Eijk extracted quantitative and qualitative data on barriers and facilitators from quantitative, qualitative, and mixed methods studies using pre-existing themes used by the authors of the included studies, which were stratified according to whether the views or perspectives were those of pregnant women or healthcare providers; the views or perspectives mainly comprised self-reported information but also observed data. The barrier and facilitator themes were then divided into four predetermined categories adapted from the literature [21],[22] for pregnant or postpartum women (Box 1) and for healthcare providers (Box 2). Because facilitators uniformly reflected the converse of the barriers, we report only the barriers (Table S4). A. M. van Eijk and J. Hoyt/L. D'Mello-Guyett extracted quantitative data from quantitative and mixed methods studies that explored the determinants of receipt of one or two doses of IPTp and ITN ownership and use, henceforth referred to as “determinants”. J. Hill and J. Hoyt/L. D'Mello-Guyett extracted quantitative, qualitative, and descriptive data from the studies evaluating delivery strategies for IPTp and/or ITNs according to the type of delivery intervention, e.g., promotion, training, or type of delivery mechanism. Box 1. Barriers from the Women's Perspective by Level Individual level: factors related to a woman's knowledge, thoughts, beliefs, actions and behaviour, pregnancy, and health status Social/cultural/household level: factors related to a woman's economic and social position, household factors including gender roles, societal and cultural norms and traditions, and religious practices Environmental level: factors related to seasonality of malaria, weather conditions, physical access, and transportation Healthcare system level: factors related to the various components and quality of the healthcare system, such as staff attitudes or performance, medication, service provision, and user fees Box 2. Barriers from the Healthcare Provider Perspective by Level Individual level: factors related to the knowledge, attitudes, and performance of individual healthcare providers Organisational level: factors related to the operation of the health facility unit, such as management, staff rosters/rotation, and services Healthcare system level: factors that are dependent on higher levels of the healthcare system related to the various components and quality of services, such as supply of drugs or ITNs, development and dissemination of policy guidelines, training and supervision of staff, and imposition of user fees Non-Healthcare system: macro-level factors external to the healthcare system such as media, water supply, side effects of medications, and women's practices Two authors (J. Hill and J. Hoyt/L. D'Mello-Guyett) assessed the quality of reporting of individual studies using a checklist of criteria developed a priori based on criteria and methods described in the literature. For observational quantitative studies the criteria of reporting were randomised sample selection, multivariate analysis, and minimising bias through study design and analysis [23],[24]. For qualitative studies the criteria were the extent to which the authors described the sampling strategy, the effects of reflexivity, and methods to ensure reliability and validity [25],[26]. For mixed methods studies, the following reporting criteria were used: justification of mixed methods, clearly described sampling strategy, clear reporting of methods for the qualitative component, analysis strategy, multivariate analyses, minimising bias, and integration of qualitative and quantitative findings [27],[28]. For intervention studies, reporting criteria were presence/type of control, steps to reduce bias, and the extent to which authors described confounding, loss to follow-up, and external validity [29]. No studies were excluded on the basis of quality. Data Synthesis Barriers and facilitators were described and explored using content analysis and narrative synthesis of qualitative and quantitative data. Data from the pregnant women's perspective were synthesised across four levels (individual, household/social/cultural, healthcare system, and environmental) and assessed in relation to receipt of IPTp, ITN ownership, and ITN use. Similarly, data from the healthcare provider perspective were synthesised across four levels (individual, organisational, healthcare system and non-healthcare system) and assessed in relation to the delivery of IPTp and ITNs in the ANC setting. The intervention studies were grouped into common strategies and explored using a narrative synthesis to summarise each intervention and to compare and contrast findings between studies evaluating similar strategies for scaling up one or both malaria interventions. Statistical Analysis We conducted a meta-analysis of data on determinants using Stata version 12 (StataCorp) and Comprehensive Meta-Analysis (Biostat, Summary odds ratios (ORs) were calculated using random effects models based on the approach of DerSimonian and Laird [30]. Data were extracted from studies using the following hierarchy based on availability: raw data (numerators and denominators); computed unadjusted ORs, computed adjusted ORs. The use of adjusted (by multivariate analysis) or cluster-adjusted ORs as provided by the studies is indicated in the meta-analysis forest plots. If studies presented results for both “1+ doses” and “2+ doses” of IPTp, only the data for “2+ doses” was used. We conducted sub-group analysis and considered the following factors for IPTp: number of SP doses (1+ or 2+), location of enrolment (community or clinic), study population (postpartum women, a mixed population of postpartum and pregnant women, or pregnant women only), and study country. The subgroup analysis for ITNs considered location of enrolment (community or health facility), study population (postpartum women, a mixed population of postpartum and pregnant women, or pregnant women only), study country, and—for ITNs—type of net (ITN or untreated net) and definition of net use (last night or during pregnancy). Sensitivity analysis was conducted to assess the potential effect of study quality on the examined associations. We assigned studies a score based on the quality assessment, and studies that failed to report on three or more quality criteria scored as low-to-moderate quality. The I 2 and 95% CI were used to quantify heterogeneity [31]. Synthesis across the Barriers, Determinants, and Intervention Studies We compared identified barriers with the determinants identified in the meta-analysis and aligned them with the intervention studies. The barriers were first collapsed into a limited number of key categories using a coding template, and the implications for intervention for each category of barriers were described. We then matched the proposed interventions derived from the barrier studies against the intervention studies included in the review to assess the extent to which the intervention studies addressed the barriers identified in the observational studies. Results Study Selection and Characteristics The primary search identified 1,780 citations (1,240 from the Malaria in Pregnancy Library, 540 from the Global Health Database, and two from bibliographies and authors), from which 271 duplicates were removed (Figure 1). From the remaining 1,511, 1,280 articles were excluded on the basis of abstracts. Of 231 full-text articles reviewed, 133 were excluded as they did not meet the inclusion criteria, the full text was not available, or they contained duplicate data, leaving 98 included articles. There was close agreement between reviewers on the included studies (kappa score of 0.86). 10.1371/journal.pmed.1001488.g001 Figure 1 Flowchart of studies included in the review. 10.1371/journal.pmed.1001488.g002 Figure 2 Analysis strategy. MiP, malaria in pregnancy. Of the 98 included studies, 81 contributed data on barriers and determinants (Table 1), and 20 studies contributed data on interventions that aimed to increase coverage and uptake of IPTp (Table 2) or ITNs (Table 3). One study did not contain data in a usable format for the meta-analysis [32]. The key characteristics of the barrier and determinant studies and of the intervention studies are provided in Table S2. 10.1371/journal.pmed.1001488.t001 Table 1 Data extracted for barriers and determinants by study. Study IPTp ITN Facility-based surveys Barriers Determinants Barriers Determinants Akaba 2013 [34] √ √ √ √ De Allegri 2013 [82] √ √ Aluko 2012 [71] √ √ Amoran 2012a [35] √ √ Amoran 2012b [72] √ √ Arulogun 2012 [55] √ Bouyou-Akotet 2013 [113] √ Diala 2012 [40] √ Iliyasu 2012 [36] √ √ Mubyazi 2012 [63] √ Mutagonda 2012 [43] √ Namusoke 2012 [59] √ Onoka 2012a [37] √ √ Onoka 2012b [114] √ Onwujekwe 2012 [61] √ d'Almeida 2011 [115] √ Donkor 2011 [48] √ Manirakiza 2011 [116] √ √ Napoleon 2011 [117] √ √ Nduka 2011 [118] √ Okonta 2011 [73] √ √ Olajide 2011 [74] √ √ Tutu 2011 [119] √ Smith Paintain 2011 [64] √ Gross 2011 [33] √ Ambrose 2011 [77] √ Sande 2010 [45] √ √ Antwi 2010 [53] √ √ Mubyazi 2010 [52] √ √ Smith 2010 [47] √ Karunamoorthi 2010 [67] √ √ Wagbatsoma 2010 [120] √ Akinleye 2009 [121] √ Takem 2009 [122] √ Klebi 2009 [123] √ Musa 2009 [69] √ Njoroge 2009 [65] √ √ Adjei 2009 [49] √ √ Mubyazi 2008 [18] √ Pettifor 2008 [76] √ √ Anders 2008 [56] √ Onyeaso 2007 [60] √ √ Mnyika 2006 [124] √ Launiala 2007 [44] √ Brentlinger 2007 [62] √ √ Kweku 2007 [83] √ Van Geertruyden 2005 [78] √ √ Gates Malaria Partnership 2005 [39] √ Mubyazi 2005 [41] √ Nganda 2004 [125] √ √ √ Ashwood-Smith 2002 [54] √ Community-based surveys Hill 2013 [126] √ √ Ankomah 2012 [80] √ √ Ansah-Ofei 2011 [46] √ Auta 2012 [81] √ √ Zere 2012 [127] √ √ √ Faye 2011 [128] √ O'Meara 2011 [85] √ Ndyomugyenyi 2010 [50] √ √ Grietens 2010 [51] √ Sangare 2010a [57] √ √ Mbonye 2010 [129] √ Sangare 2010b [70] √ √ Beiersmann 2010 [84] √ Acquah 2009 [130] √ Brabin 2009 [42] √ Gies 2009 [90] √ Gikandi 2008 [131] √ √ Marchant 2008 [58] √ √ Belay 2008 [79] √ √ Hassan 2008 [132] √ Kiwuwa 2008 [133] √ √ Ouma 2007 [91] √ PSI Burundi 2006 [134] √ PSI Rwanda 2006 [135] √ PSI Zambia 2006 [136] √ Mbonye 2006a [38] √ Mbonye 2006b [66] √ van Eijk 2005 [68] √ √ √ Guyatt 2004 [137] √ √ √ Marchant 2002 [75] √ √ Summary total 38 31 28 27 10.1371/journal.pmed.1001488.t002 Table 2 Evaluation of interventions aimed at increasing coverage of IPTp (six studies). Study/Measure Description (Country) Baseline Point of Evaluation Intervention (Percent) Control (Percent) p-Value Intervention (Percent) Control (Percent) p-Valuea p-Valueb Msyamboza 2009 [86] IPTp delivered by community health workers (Malawi) IPTp 2+ 36/87 (41.4) 47/107 (43.9) 0.77 663/912 (72.7) 412/897 (45.9) 19 y) and married women were the most likely to use an ITN. Women with higher education or greater knowledge of malaria or ITNs were more likely to use ITNs than women with lower education or less knowledge, and women who were employed in a wage-paying job were also more likely to use ITNs during pregnancy than farmers or housewives. Women who had received IPTp were more likely to use ITNs. The effect of education on ITN use showed significant variation by country (p = 0.028; Text S2), and the effect of marital status on ITN use varied significantly by location of enrolment (p = 0.001; Text S2). Sensitivity analysis indicated a stronger association between ITN use and marital status in the low-to-moderate quality studies compared to the better quality studies (Text S2). 10.1371/journal.pmed.1001488.g004 Figure 4 Summary odds ratios of determinants of ITN use assessed in 17 studies with quantitative data. SES, Socio-economic status. Intervention Studies Interventions to increase coverage of IPTp The evidence from four studies that evaluated community-based distribution of IPTp suggests that community resources have the potential to complement the delivery of IPTp through ANCs to increase access to and uptake of IPTp among pregnant women [86]–[89] (Table 2). However, there was evidence that community-based distribution may concurrently reduce women's attendance at ANCs, though this was not consistent across the four studies: two studies showed reduced ANC attendance in the intervention sites [86],[87], and two showed increased ANC attendance [88],[89]. An alternative to delivering IPTp through community-based programmes is to employ community-based resource persons to promote IPTp, while referring women to ANCs to be given SP. This approach had substantial success in Burkina Faso, and resulted not only in higher IPTp coverage (71.8% versus 49.2% in intervention and control groups, respectively; p<0.001) but also in women attending the ANC earlier, in their first or second trimester (81.3% versus 70.4% in intervention and control groups, respectively; p<0.001), and in more women making two or more visits (89.3% versus 75.3% in intervention and control groups, respectively; p<0.001) [90]. One intervention study evaluated strategies to improve healthcare provider knowledge and performance on how to deliver IPTp. The study was undertaken in Kenya, 4 y after the national IPTp policy was adopted, and suggests that retraining of healthcare providers on the delivery, timing, and dosing of IPTp significantly increased coverage of IPTp (36.9% versus 10.9% in intervention and control groups, respectively; p<0.001) [91]. Interventions to increase coverage of ITNs The included intervention studies evaluated two main channels for delivering ITNs to pregnant women: campaign delivery (non-targeted) [89],[92]–[94] and routine delivery to pregnant women through ANC services (targeted), with three alternative mechanisms evaluated at ANCs: distribution of free nets with [95]–[97] or without social marketing [98],[99], and distribution of subsidised vouchers [83],[84],[100]–[102]. One study compared the impact of ANC delivery alone versus ANC delivery plus community-based distribution of subsidised nets in Niger (Table 3). Campaign delivery of ITNs to households with pregnant women [89], households with children under 5 y [94], or poor households [93] had limited impact on increasing coverage among pregnant women with one exception, which was a campaign in Senegal that delivered ITN vouchers to all households with children under 5 y, alongside vitamin A and mebendazole (an anthelmintic) [92] (49.2% versus 28.5% ITN coverage in intervention versus control groups, respectively; no statistical analysis reported). In a comparison study in Tanzania, the Tanzania National Voucher Scheme, which provides a voucher subsidy to pregnant women at ANCs, which is then used to purchase an ITN from a contracted retailer, achieved greater coverage than a 3-d mass campaign targeting ITNs to households with infants, based on the assumption that infants sleep with their mothers, a common practice in this setting, or ITNs sourced from retailers [94]. The voucher scheme was, however, inequitable, with fewer poorer women receiving nets [100]. In a comparison study of routine ANC delivery of ITNs alone and ANC delivery plus community-based distribution, there was no significant difference in ITN use among pregnant women between groups [103]. Routine delivery of ITNs through ANCs, by comparison, appeared to be more successful in reaching pregnant women, with four studies demonstrating an increase in ITN coverage among pregnant women compared to baseline [98],[99],[101],[102]. Programmes that delivered vouchers, as opposed to free nets, to women at ANCs experienced more operational challenges [83], and were dependent on retailers having ITN stock available [84]. Social marketing campaigns have been effective in promoting the use of ITNs in some settings through extensive media and educational campaigns that increase awareness about the benefits and importance of ITN use (especially for pregnant women), coupled with provision of readily available ITNs at low cost. They are, however, comparatively expensive to implement and sustain [104]. Implications for Interventions to Address Barriers We aligned the barriers to uptake of IPTp and ITNs against the findings from the intervention studies to determine the extent to which these interventions addressed known barriers (Tables 4 and 5). There were four key categories of barriers to women receiving IPTp: pregnant women's knowledge of IPTp, access to an ANC, affordability of ANC services, and quality of ANC services. Women's lack of knowledge of IPTp was very common and yet may be improved through relatively simple promotional activities delivered through all available channels, such as community-based resource persons, facility-based counselling and education, and messaging via the media and local leaders. We identified only one relevant intervention study, which evaluated community-based promotion of IPTp in Burkina Faso [90]. Women's access to an ANC was a barrier in remote settings, where community-based distribution or outreach services may be required to supplement ANC services. Four studies evaluating community-based distribution of IPTp were identified in the review, using a combination of existing [87],[88] or new community resource persons [86],[89]. 10.1371/journal.pmed.1001488.t004 Table 4 Synthesis matrix comparing findings from observational studies with those of intervention studies for IPTp. Type of Factor Findings from Observational Studies Findings from Intervention Studies Categories Derived from Barriers Implications for Interventions to Increase Uptake Type of Intervention Evaluated Number of Intervention Studies Pregnant women factors Category 1—pregnant women's knowledge Example barriers• Lack of knowledge of the preventive benefits of IPTp• Belief that use of drugs or SP in pregnancy is unsafe, e.g., could cause abortion• Fear of perceived side effects of SP• Unaware of the dangers of malaria in pregnancy Promotion of IPTp strategy and safety of SP for IPTp through a variety of channels, e.g., community-based, clinic-based, media, local leaders Community-based promotion of IPTp and referral of women to ANC 1 study in Burkina Faso (Gies 2009 [90]) Category 2—access to ANC Example barriers• Poor access to ANC• Direct and indirect costs of accessing ANC• Commitments to farming, employment, or childcare• Unwillingness to reveal pregnancy• Lack of awareness of importance of ANC services Community-based distribution of IPTp in hard-to-reach populations with limited access to ANC, e.g., through community-based volunteers and/or community-based referral systems to increase use of ANC Community-based distribution in settings with poor access to ANC, or community-based distribution in settings with existing drug distribution programmes, e.g., onchocerciasis, or community-based referral of women to ANC 3 studies evaluating community-based distribution of IPTp (Okeibunor 2011 [89], Msyamboza 2009 [86], Mbonye 2007 [87]); 1 study in Uganda (Ndyomugyenyi 2009 [88]); 1 study in Burkina Faso (Gies 2009 [90]) Category 3 –affordability of ANC services Example barriers• ANC registration fees• Laboratory fees• Cost of SP• Unofficial penalties charged by healthcare providers for late ANC attendance See healthcare provider factors Category 4—quality of ANC services Example barriers• Providers do not offer IPTp• SP unavailable• Lack of water or cups for DOT• Poor attitudes of healthcare providers• Lack of information or instructions given by healthcare providers regarding IPTp See healthcare provider factors Healthcare provider factors Category 1—provider knowledge Example barriers• Poor knowledge of IPT strategy, timing and dosage of SP• Imprecise estimation of gestational age• Confusion about when to give IPTp in relation to treatment of malaria, HIV, or other• Perception that women will or should not take SP on empty stomach Training and supervision of healthcare providers Training of healthcare providers 1 study in Kenya (Ouma 2007 [91]) Category 2—provider attitudes Example barriers• Health education not given in local language• Information and instructions on IPTp not given to pregnant women• Providers do not offer IPTp• Providers treat women with lack of respect Training and supervision of healthcare providers on provider–client interactions None None Category 3—health facility organisation Example barriers• Restrictive ANC hours• Lack of cups or drinking water• Frequent provider absence from work• Ineffective staff rosters Reorganisation of staff rosters, opening hours, etc., and better management, supervision, and accountability of staff None None Category 4—inadequate guidance on IPTp Example barriers• Variation in information given to healthcare providers on IPTp• No guidelines available at facility• Lack of supervision and monitoring of IPTp• Lack of recent training on IPTp• Private facilities following different practices• Incompatibilities between delivery of IPTp and other health interventions Provision of consistent, simple guidelines to all health facilities, both public and private sectors, together with training and supervision Modelling the effect of simple guidelines on coverage with IPTp 1 study in Tanzania (Gross 2011 [33]) Category 5—fees for ANC services Example barriers• ANC registration fees• Cost of SP• Unofficial penalties charged by healthcare providers for late ANC attendance Modification or removal of user fees and regulation against imposition of penalties None None Category 6—supply of SP Example barriers• SP unavailable• Poor stock control Timely procurement and distribution systems for SP, and system to prioritise use of funds for SP at health facilities None None 10.1371/journal.pmed.1001488.t005 Table 5 Synthesis matrix comparing findings from observational studies with those of intervention studies for ITNs. Type of Factor Findings from Observational Studies Findings from Intervention Studies Categories Derived from Barriers Implications for Interventions to Increase Uptake Type of Intervention Evaluated Number of Intervention Studies Pregnant women factors Category 1—pregnant women's knowledge Example barriers• Lack of knowledge of benefits of ITNs for mother and child• Discomfort of using ITNs• Lack of habit of using ITNs• Fear of chemicals used on ITNs• Perception that there are no mosquitoes Promotion of ITN strategy and safety of insecticides used to treat nets through a variety of channels, e.g., community-based, clinic-based, media, local leaders Promotional campaigns using a variety of channels, e.g., social marketing, clinic-based, media 3 social marketing studies by PSI in Burundi (2007 [97]), Kenya (2008 [95]), and Madagascar (2009 [96]) Category 2—household or cultural constraints Example barriers• Lack of support from husband and/or community• Lack of cultural habit of using ITNs• Cultural beliefs, e.g., resemblance of ITNs to burial shrouds Promotion of ITN strategy and safety of insecticides used to treat nets through a variety of channels, e.g., community-based, clinic-based, media, local leaders As above As above Category 3—access to ITNs Example barriers• Lack of retailers• Cost of ITNs• Inability to pay top-up fees on vouchers• Direct and indirect costs of accessing ITN distribution points Delivery of free ITNs to pregnant women through ANC or campaigns, or delivery of voucher subsidies through ANC or campaigns, or community-based distribution of subsidised ITNs Delivery of free ITNs to pregnant women through ANC or campaigns, or delivery of voucher subsidies through ANC or campaigns, or community-based distribution of subsidised ITNs 3 studies evaluated free ITNs: 2 studies through ANC (Pettifor 2009 [98], Guyatt 2003 [99]) and 1 study through campaign delivery (Thwing 2011 [92]); 7 studies evaluated voucher subsidies: 2 studies via campaign delivery (Ahmed 2010 [93], Khatib 2008 [94]), 5 studies via ANC (Beiersmann 2010 [84], Marchant 2010 [100], Hanson 2009 [101], Muller 2008 [102], Kweku 2007 [83]); 1 study community-based: Nonaka 2012 [103] Healthcare provider factors Category 1—provider knowledge Example barrier• Lack of knowledge of ITN benefits for mother and child Training and supervision of healthcare providers on ITNs None None Category 2—provider attitudes Example barriers• Providers refuse to offer ITNs to pregnant women• Providers impose eligibility criteria for ITNs or vouchers Better training, management, supervision, and accountability of staff None None Category 3—health facility organisation Example barriers• Vouchers not available at facility• As for IPTp Reorganisation of staff rosters, hours, etc., and better management, supervision, and accountability of staff None None Category 4—fees for ANC services Example barriers• ANC registration fees• Cost of ITNs Removal of user fees and regulation against imposition of penalties None None Category 5—supply of ITNs/vouchers Example barriers• Poor stock control• Stockouts of ITNs• Vouchers not available Timely procurement and distribution systems for ITNs or vouchers None None Six key categories of barriers to healthcare providers delivering IPTp were identified: provider knowledge of IPTp, provider attitudes, health facility organisation, policy and guidance, fees for services, and supply of SP. Poor knowledge and poor administration of IPTp guidelines by healthcare providers appear to be substantial barriers to achieving high coverage, as highlighted in several studies included in this review. Provider knowledge of the IPTp strategy could be improved through retraining and closer supervision by district staff; however, only one study was identified that evaluated the impact of retraining of healthcare providers in Kenya on the delivery, timing, and dosing of IPTp [91]. Simplified policy and guidance on IPTp would be a relatively simple intervention to improve healthcare provider practice in delivering IPTp, and while no relevant intervention study was identified, one study in Tanzania modelled the effect of simplified guidelines on coverage with IPTp, demonstrating that coverage could be increased with simplified guidance [33]. No intervention studies were identified that addressed supply of SP, even though this was one of the commonest barriers identified in the observational studies. Poor healthcare provider attitude is a generic problem often entrenched in resource-constrained healthcare system and public sector settings, and may be difficult to address; no relevant intervention studies were identified. Similarly, user fees at ANCs are a generic barrier to ANC services, and no intervention studies were identified that addressed this. Three key categories of barriers to women receiving and using ITNs were identified: pregnant women's knowledge of ITNs, household or cultural constraints, and access to ITNs. As for IPTp, pregnant women's knowledge of ITNs as well as certain household and cultural constraints could be addressed through promotion of ITNs through a variety of channels. Social marketing using extensive media and educational campaigns has been used in a large number of countries, and three evaluation studies were identified in this review [95]–[97]. Access to ITNs has been a problem for women in terms of direct and indirect costs, ITN availability, and access to distribution points. Three studies evaluated the delivery of free ITNs to pregnant women through ANCs [98],[99] or campaigns [92], one study evaluated community-based delivery of subsidised ITNs [103], and seven studies evaluated voucher subsidies delivered through ANCs [83],[84],[100]–[102] or campaigns [93],[94]. Categories of barriers to healthcare providers delivering ITNs were similar to those for the delivery of IPTp: provider knowledge, provider attitudes, health facility organisation, fees for services, and supply of ITNs. We did not find any relevant studies that evaluated interventions that directly addressed these provider barriers. Discussion To our knowledge this is the first systematic review of the factors affecting the delivery, access, and use of interventions to prevent malaria in pregnant women that uses research findings from quantitative, qualitative, and mixed methods studies, that assesses both user and provider perspectives, and that integrates these findings with intervention studies. This analysis provides a comprehensive basis for identifying key bottlenecks in the delivery and uptake of IPTp and ITNs among pregnant women, and for understanding which scale-up interventions have been effective, in order to prioritise which interventions are most likely to have the greatest impact in the short or medium term. Barriers to the delivery of IPTp and ITNs were found at different levels of implementation, and broadly fall into policy and guidance, healthcare system issues, health facility issues, and healthcare provider performance. Whilst many of the barriers reflected broader weaknesses in the healthcare system, some were specific to the intervention. With regard to IPTp, a key identified barrier to effective delivery was healthcare provider confusion about the timing of the two doses of IPTp and whether IPTp can be given on an empty stomach. This confusion stemmed from a combination of unclear policy and guidance, inadequate training, and lack of information and job aids on IPTp. Several studies reported conflicting national policies with regards to provision of IPTp in relation to management of HIV and other diseases or conditions, and when to give IPTp if women have been treated for malaria, a problem also identified in another review [105]. Also, some studies reported that healthcare providers expressed uncertainty over the effectiveness of SP for IPTp. Clearly there is an urgent need for countries to update national IPTp policy and guidance, and to ensure that this information reaches frontline providers at ANCs and outpatient departments providing treatment to pregnant women for illness, e.g., through directives or memos from the Director of Medical Services, as done in Kenya (M. J. Hamel, personal communication). The recent WHO IPTp policy update recommendation with simplified guidance on IPTp dosing, which also restates the continued effectiveness of IPTp with SP, serves as an important opportunity for national programmes to update and reinvigorate their IPTp strategy [106]. Organisational problems at the facility level were also common, such as lack of privacy and confidentiality in the health encounter [51] and the restriction of hours of ANC services, resulting in high client-to-staff ratios, long waiting times [49],[52], and reduced consultation times, all of which contribute to poor quality of care at ANCs. Absenteeism and high staff rotation at the facility leading to lack of continuity of care and high workload among staff on duty was also reported [62]. Most of these organisational problems present another area for improvement in the short term that does not require additional resources, though it will require better management and accountability by the heads of health facilities. Other barriers were, however, dependent on higher levels of the healthcare system, such as high staff turnover [62], understaffing (particularly in remote areas), poor infrastructure [41], poor supervision, and poor use of data to identify problems and inform decision-making. These problems are inherent in the healthcare systems in some areas in some countries, and will require longer term strategies and increased investment in healthcare system strengthening. Also persistently reported across the studies and dependent on action taken at higher levels were stockouts of both SP for IPTp and ITNs, and lack of water or cups for providing IPTp by DOT. The reviewed studies did not explore the reasons for the stockouts, but they are likely to be a combination of lack of funding at the national level for procurement of commodities (i.e., specific to IPTp and ITNs) and problems in supply chain management. Barrier studies among women highlighted additional healthcare system barriers leading to poor uptake of IPTp and/or ITNs. Having to pay user fees or pay for SP, drinking water for DOT, or ITNs was a common barrier, as were the indirect costs associated with visiting ANCs, such as transport, food, and opportunity costs. This finding was supported by the meta-analysis of determinants of coverage among pregnant women, which showed that socio-economic status and employment status are important predictors of IPTp and ITN coverage, respectively. These inequities may to some extent reflect the determinants of women's access to ANCs, where user fees are routinely applied to registration, consultations, laboratory tests, and drugs, as identified in a review of factors affecting utilisation of antenatal care in developing countries [107]. However, in some instances user fees are also applied to SP (e.g., where women have to purchase SP or water to take IPTp by DOT) and to ITNs [108]. This situation calls for a review of charging policies for IPTp and ITNs across national programmes, and of user fees and charges at ANCs in general. Another common barrier to ANC utilisation was the poor quality of interactions between healthcare providers and pregnant women [38],[41]. Women were generally perceived as passive recipients and were provided with little or no information about the services provided [44], and women with a low social position, such as adolescents [51], and less educated women are most vulnerable. This issue appears to be a problem in some resource-poor settings and is more difficult to tackle. However, educating women about their rights and about the ANC services available to them may go some way to empowering women to be able to demand better services. This finding is supported by the fact that pregnant women's lack of knowledge and understanding of IPTp and ITNs was consistently reported in both the barrier and determinant data as an important factor preventing the uptake and use of IPTp and ITNs. Women who understand the benefits of IPTp and the safety of SP, and how and when to take it, are more likely to take it. However, many women do not receive adequate information about IPTp, and this can result in fears that the drug causes harm, even abortion [15], or women showing preference for an alternative drug. Whilst there are some reports that women experience side effects from IPTp, the severity and extent of these events are not clearly described. There were also reports of women fearing that the chemical used on ITNs would harm the foetus [15]. Whilst knowledge is also an important facilitator of ITN use, barrier studies reveal important deterrents to ITN use such as the inconvenience and discomfort of use [109], especially in the dry season, and the lack of a culture or habit of net use. These findings were consistent with the meta-analysis of determinants in that coverage of both IPTp and ITNs was lower among women with no education and, in some countries, women living in rural areas; these women were less likely to access ANC and/or health education services. The meta-analysis was useful in identifying other important risk groups. Younger or adolescent women, unmarried women, and less educated women were significantly less likely to use ITNs. The barrier studies show that this may be related to lower affordability and in-household access among these women. Adolescents, unmarried women, and less educated women therefore constitute high-risk groups for targeting ITNs. This suggests that ministries of health need to pay more attention to IPTp and ITN promotion and health education, with additional targeting of risk groups, as well as using new innovations for communication of messages, since traditional health education is not offered at all facilities or is not always effective. Women seeking care at ANCs often have to overcome barriers at the household or societal level, and these barriers are more challenging to address. Women have commitments to farming or employers and the responsibility of childcare, and often have to defer to their husbands or in-laws in decision-making over accessing ITNs or use of household income to pay for ANC services. In a review of ANC access, use of ANCs was shown to increase with husband's educational level and was an even stronger predictor than women's education in some settings [107]. Local cultural norms and practices present a considerable barrier to women accessing ANC services in some but not all study countries, with wide variation within countries and between countries, a finding also reported in the review by Pell et al. [15]. In comparison to the observational studies, the review identified comparatively few studies that evaluated interventions to promote scale-up of these interventions, particularly for IPTp. Whilst many of the barriers to IPTp and ITN coverage identified in the observational studies related to healthcare providers and service delivery, very few studies that evaluated interventions to improve service delivery were found. Similarly, very few studies explored the determinants of delivery of either IPTp or ITNs among healthcare providers, or supply-side interventions designed to improve the quality of delivery of IPTp or of ITNs with a chosen strategy, whether it be campaigns or routine delivery through ANCs. Of the six studies that evaluated interventions to increase coverage of IPTp, all but one targeted women's knowledge or access, the last being a healthcare provider intervention. Consideration of the context for employing community-based distribution of IPTp is important; this distribution strategy appears to be an effective additional strategy to boost coverage in areas where there is already a successful community-based distribution programme, as seen in the onchocerciasis control programme in Uganda [88], but may serve to undermine women's attendance at ANCs in areas where ANC attendance is fragile. Community-based promotion, on the other hand, has the potential benefit in some settings of increasing access and uptake of IPTp by providing women with information about the importance and benefits of IPTp, and at the same time reinforcing the message that women should obtain antenatal care from ANCs, where they benefit from the full range of focussed ANC services [90]. While 13 studies were identified that evaluated the effectiveness of alternative delivery strategies to increase ITN coverage among pregnant women, the study objectives and designs were heterogeneous; hence, it was not possible to draw generalisable conclusions. Nevertheless, ANC services appear to be an important source of free ITNs for pregnant women in rural areas, a finding supported in a review of best practices of ITN programmes in sub-Saharan eastern Africa [108]. Strengths and Limitations of the Review The review triangulates data from quantitative, qualitative, and mixed methods studies to increase the content validity and comprehensiveness of the review; it does not, however, attempt a full meta-ethnography of qualitative data, which has been undertaken recently by others [15],[110]. The meta-analysis of determinants was used to explore the range of effects between studies and to provide a pooled analysis to support the findings of the narrative (interpretive) synthesis. Although the use of cluster-unadjusted ORs may have overestimated precision, these were limited to four out of 36 studies. There was considerable heterogeneity among studies included in the meta-analysis, and we explored only a limited number of variables in the subgroup analysis to assess whether these could explain the differences between studies (Text S2). The lack of adjustment for ANC attendance in studies using community-based surveys means that the determinants of IPTp use may be partly driven by determinants of ANC access. However, the differences in the results between studies that enrolled women in the community and those that enrolled women in clinics in the subgroup analysis were not significant (Text S2). Whilst distinguishing between use of SP for treatment versus use for prevention poses an important challenge in interpreting community surveys, this limitation was not measured in the studies included in the meta-analysis. Whilst no restrictions were placed on the language of publication, and no studies were excluded on the basis of language, the focus the Malaria in Pregnancy Library (the primary source of studies) to date has been on the European family of languages and predominantly English. Reviewer bias was limited by the use of two independent reviewers to assess inclusion criteria. Reporting of included studies was assessed for quality, and reporting quality for the majority of studies was assessed to be fair. There were three quantitative studies that met no reporting quality criteria and 13 studies that met only one criterion (10 quantitative and three intervention studies). Findings from the studies with data on barriers were found to be entirely consistent with findings from other studies, and provided no new or surprising themes, and inclusion of these studies did not alter the study findings. Our review includes 98 studies from across sub-Saharan Africa, with 77 of these specifically containing data on barriers and determinants of delivery, access, and use of IPTp and ITNs among healthcare providers and pregnant women; this is a sizeable body of evidence. In summary, the delivery and uptake of IPTp and ITNs by pregnant women is impeded by a wide range of factors among both pregnant women and the healthcare system, each influenced by an array of social, cultural, economic, and institutional factors, with each factor influenced by the others in a complex interchange. There are also geographic variations, with some barriers more prominent in some countries than in others. Notwithstanding this complexity, many of the barriers highlighted in this review are relatively consistent across countries and are surmountable: barriers that programmes can address in the near term with limited additional investment. Delivery of ITNs through ANCs presents a narrower range of problems than delivery of IPTp. Actions to increase coverage of IPTp and ITNs in the short term would be (1) to simplify country policies and guidance to align the updated WHO IPTp policy [106] with the new WHO policy for focused antenatal care, consisting of four visits in the second and third trimesters, and ensure effective dissemination to frontline healthcare providers through training and job aids; (2) to earmark funding for procurement of SP and ITNs; (3) to review ANC fee structures; and (4) to launch targeted promotional campaigns to reach high-risk populations of pregnant women, according to local settings, e.g., rural, poor, or adolescent women. Promotional campaigns will need to reflect the needs of women and offer services they will accept at a price they can afford. Other barriers are more entrenched within the overall healthcare system and will require medium- to long-term strategies to improve the overall quality of antenatal services and encourage the habit of ANC use among women. New multifaceted interventions should be explored, such as quality improvement initiatives that link improvements in delivery of IPTp and ITNs to other core ANC services, management tools for facility-level decision-making, and innovations, such as use of mobile phones for defaulter tracing, supply chain/stock control, reporting of health management information systems data on coverage, and surveillance. Increasing drug resistance means that IPTp with SP will most likely be replaced by more complicated and expensive drug regimens [4],[111], or new strategies, such as intermittent screening and treatment [112]. Intermittent screening and treatment will require adjustments to be made in the ANC setting [47],[64], and will not have the added benefit of IPTp in controlling infections that cannot be detected by rapid diagnostic tests or microscopy. Malaria prevention estimates have increased only modestly between 2007 and 2010 (from 13.6% to 21.5% coverage for IPTp and from 17.0% to 38.8% coverage for ITN use) [138]. Conclusion Our synthesis shows that the key barriers to access, delivery, and use of IPTp and ITNs are relatively consistent across countries. These barriers may be helpful as a checklist for use by country malaria programmes and/or policy-makers to identify factors influencing uptake of these interventions in their specific location or context. The review also highlights the need for multi-country studies that evaluate targeted or multifaceted interventions aimed to improve the delivery and uptake of IPTp and ITNs. More research is also needed to understand and improve the policy change process to facilitate future replacement of SP with alternative drug regimens for IPTp or alternative strategies such as screening and treatment that will present even greater challenges for delivery. Supporting Information Table S1 Search terms and databases used in the review. (DOCX) Click here for additional data file. Table S2 Study characteristics. Table S2.1. Characteristics of studies on determinants, barriers, and facilitators. Table S2.2. Characteristics of intervention studies. (DOCX) Click here for additional data file. Table S3 Checklist for quality of reporting. Table S3.1. Checklist for quality of reporting: quantitative studies. Table S3.2. Checklist for quality of reporting: qualitative studies. Table S3.3. Checklist for quality of reporting: mixed methods studies. Table S3.4. Checklist for quality of reporting: intervention studies. (DOCX) Click here for additional data file. Table S4 Barriers and facilitators to delivery, access, and use of IPTp and ITNs. Table S4.1. Barriers and facilitators to receipt of IPTp from the perspective of pregnant and recently delivered women. Table S4.2. Barriers and facilitators to ITN ownership and use from the perspective of pregnant and recently delivered women. Table S4.3. Barriers and facilitators to the delivery of IPTp from the healthcare provider perspective. Table S4.4. Barriers and facilitators to the delivery and use of ITNs from the healthcare provider perspective. (DOCX) Click here for additional data file. Text S1 PRISMA statement. (DOC) Click here for additional data file. Text S2 Meta-analysis of determinants of IPTp and ITN use in pregnancy. (PDF) Click here for additional data file.
              • Record: found
              • Abstract: found
              • Article: not found

              Social and Cultural Factors Affecting Uptake of Interventions for Malaria in Pregnancy in Africa: A Systematic Review of the Qualitative Research

              Background Malaria during pregnancy (MiP) results in adverse birth outcomes and poor maternal health. MiP-related morbidity and mortality is most pronounced in sub-Saharan Africa, where recommended MiP interventions include intermittent preventive treatment, insecticide-treated bednets and appropriate case management. Besides their clinical efficacy, the effectiveness of these interventions depends on the attitudes and behaviours of pregnant women and the wider community, which are shaped by social and cultural factors. Although these factors have been studied largely using quantitative methods, qualitative research also offers important insights. This article provides a comprehensive overview of qualitative research on social and cultural factors relevant to uptake of MiP interventions in sub-Saharan Africa. Methods and Findings A systematic search strategy was employed: literature searches were undertaken in several databases (OVID SP, IS Web of Knowledge, MiP Consortium library). MiP-related original research, on social/cultural factors relevant to MiP interventions, in Africa, with findings derived from qualitative methods was included. Non-English language articles were excluded. A meta-ethnographic approach was taken to analysing and synthesizing findings. Thirty-seven studies were identified. Fourteen concentrated on MiP. Others focused on malaria treatment and prevention, antenatal care (ANC), anaemia during pregnancy or reproductive loss. Themes identified included concepts of malaria and risk in pregnancy, attitudes towards interventions, structural factors affecting delivery and uptake, and perceptions of ANC. Conclusions Although malaria risk is associated with pregnancy, women's vulnerability is often considered less disease-specific and MiP interpreted in locally defined categories. Furthermore, local discourses and health workers' ideas and comments influence concerns about MiP interventions. Understandings of ANC, health worker-client interactions, household decision-making, gender relations, cost and distance to health facilities affect pregnant women's access to MiP interventions and lack of healthcare infrastructure limits provision of interventions. Further qualitative research is however required: many studies were principally descriptive and an in-depth comparative approach is recommended.

                Author and article information

                Role: Academic Editor
                PLoS Med
                PLoS Med
                PLoS Medicine
                Public Library of Science (San Francisco, USA )
                August 2014
                5 August 2014
                : 11
                : 8
                [1 ]Department of Clinical Sciences, Liverpool School of Tropical Medicine, Liverpool, United Kingdom
                [2 ]Disease Control Department, London School of Hygiene & Tropical Medicine, London, United Kingdom
                Hospital Clinic Barcelona, Spain
                Author notes

                The authors have declared that no competing interests exist.

                Conceived and designed the experiments: JH JW. Performed the experiments: JH LDM JHo AvE. Analyzed the data: JH LDM JHo AvE JW. Wrote the first draft of the manuscript: JH. Contributed to the writing of the manuscript: JH LDM JHo AvE FtK JW. ICMJE criteria for authorship read and met: JH LDM JHo AvE FtK JW. Agree with manuscript results and conclusions: JH LDM JHo AvE FtK JW.


                This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

                Page count
                Pages: 28
                This work was supported by a Master Service Agreement (contract # 20762) from the Bill & Melinda Gates Foundation to the Liverpool School of Tropical Medicine. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
                Research Article
                Medicine and Health Sciences
                Health Care
                Primary Care
                Infectious Diseases
                Parasitic Diseases
                Public and Occupational Health
                Women's Health
                Custom metadata
                The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the paper and its Supporting Information files, and the published studies are available on the the Malaria in Pregnancy (MiP) Library (, the Global Health Database (, and the International Network for Rational Use of Drugs (INRUD) Bibliography (



                Comment on this article