Factors influencing implementation of interventions to promote birth preparedness and complication readiness

Background The recent WHO report on health promotion interventions for maternal and newborn health recommends birth preparedness and complications readiness interventions to increase the use of skilled care at birth and to increase timely use of facility care for obstetric and newborn complications. However, these interventions are complex and relate strongly to the context in which they are implemented. In this article we explore factors to consider when implementing these interventions. Methods This paper reports a secondary analysis of 64 studies on birth preparedness and complication readiness interventions identified through a systematic review and updated searches. Analysis was performed using the Supporting the Use of Research Evidence (SURE) framework to guide thematic analysis of barriers and facilitators for implementation. Results Differences in definitions, indicators and evaluation strategies of birth preparedness and complication readiness interventions complicate the analysis. Although most studies focus on women as the main target group, multi-stakeholder participation with interventions occurring simultaneously at both community and facility level facilitated the impact on seeking skilled care at birth. Increase in formal education for women most likely contributed positively to results. Women and their families adhering to traditional beliefs, (human) resource scarcities, financial constraints of women and families and mismatches between offered and desired maternity care services were identified as key barriers for implementation. Conclusions Implementation of birth preparedness and complication readiness to improve the use of skilled care at birth can be facilitated by contextualizing interventions through multi-stakeholder involvement, targeting interventions at multiple levels of the health system and ensuring interventions and program messages are consistent with local knowledge and practices and the capabilities of the health system. Electronic supplementary material The online version of this article (doi:10.1186/s12884-017-1448-8) contains supplementary material, which is available to authorized users.


Background
Our systematic review on the impact of birth preparedness and complication readiness (BPCR) interventions on birth with a skilled attendant revealed that BPCR is a complex intervention, highly dependent on the context in which it is implemented [1]. We also found that BPCR interventions vary in terms of approaches, actors involved, in definitions applied, in outcomes measured and in the strategies used to evaluate them.
The concept of BPCR emerged almost 20 years ago and is described as a process of planning for birth and anticipating actions in case of obstetric emergencies in order to reduce delays in seeking skilled care [2]. In 2005 BPCR was included in the World Health Organization (WHO) antenatal care package [3,4], with emphasis on the following elements: deciding on desired place of birth; preferred birth attendant; location of the closest facility for birth and in case of complications: funds for expenses related to birth and/or complications; supplies necessary to bring to the facility; an identified labour and birth companion; an identified support to look after home and other children while the woman is away; transport to a facility for birth or when complications arise; and identification of compatible blood donors when needed. At around the same time, Johns Hopkins Program for International Education in Gynecology and Obstetrics (JHPIEGO) developed a BPCR matrix acknowledging the important role of coordinated efforts of all 'safe motherhood stakeholders' for implementing BPCR. The matrix delineates roles and activities of policymakers, facility managers, providers, communities, families, and women in ensuring that women and newborns reach accessible, appropriate, acceptable and good quality care during pregnancy, childbirth and postpartum [2].
Despite widespread promotion and inclusion of BPCR in Safe Motherhood interventions, evidence on the effect of BPCR interventions remains limited. Our recent systematic review of the available evidence found that BPCR, as part of a package of interventions, has the potential to increase skilled care at birth and timely use of facility care for obstetric and newborn complications [1]. The results of the review have been included in recently published WHO guidelines on health promotion interventions for maternal and newborn health, where WHO recommends implementation of BPCR interventions [5].
To support those who plan to implement BPCR interventions, we conducted a secondary analysis of the papers included in our systematic review [1] and additional studies identified, in order to identify factors influencing implementation. We explore stakeholder perceptions and experiences of BPCR interventions, identify barriers and facilitators to BPCR implementation, and discuss how these relate to improvements in use of skilled care at birth.

Methods
This article reports a secondary analysis of studies identified in a systematic review conducted in 2013 [1,6] and additional articles identified through a subsequent search. The systematic review included articles published in English between 2000-2012, identified from PubMed, Embase and CINAHL plus a manual search of the grey literature and a database that included results of systematic mapping of maternal health research in low-and middle-income countries [7]. The original review was concerned with effects on care seeking including use of a skilled attendant at birth (SBA) or facility birth, use of antenatal care (ANC) as well as effects on knowledge and preparations made for BPCR.
For the findings presented in this paper, we conducted a narrative synthesis of qualitative information on implementation factors from the 64 papers. We used an adapted SURE (Supporting the Use of Research Evidence) framework to guide the extraction of relevant information from studies and to structure the synthesis [39]. The framework comprises a comprehensive list of barriers and facilitators to implementing health systems interventions including stakeholder knowledge and attitudes, health service delivery factors, and social and political considerations; the framework has been used in other systematic reviews of qualitative evidence [40,41].

Description of included studies
Characteristics of the 64 included studies are presented in Table 1; some studies report on the same BPCR programme or intervention and are listed together. Most of the studies of BPCR interventions were conducted in South Asia (Nepal n = 7; India n = 6; Bangladesh n = 4; Pakistan n = 1; and Tibet n = 1), followed by East Africa (Tanzania n = 7; Ethiopia n = 6; Uganda n = 4; Eritrea n = 1; and Kenya n = 1), West Africa (Burkina Faso n = 3; Nigeria n = 3; and Benin n = 1), South East Asia (Cambodia n = 1; and Indonesia n = 1), and Latin America (Guatemala n = 1); one study included multiple countries, and two articles were literature reviews. BPCR implementation strategies varied and often included multiple interventions, which are summarized in Table 2. These included house visits by volunteers who provided education on BPCR, training of health workers in facilities to provide BPCR as part of ANC, provision of education materials or other visual aids with BPCR information, community mobilization activities to increase awareness on BPCR and mass media campaigns with BPCR messages.
Studies define BPCR and its main components variously which complicates interpretation of results, context and policy advice. For example, the most commonly described components include: identifying funds for birth and emergency expenses; deciding on a preferred birth attendant; identifying transport to the health facility for birth or complications; choosing the place of birth and location of nearest facility; and knowledge and identification of danger signs in pregnancy. Other definitions include identifying compatible blood donors, preparing supplies, identifying a birth companion, and discussing   Interventions included facility upgrades, quality of care and BPCR and community mobilization. SBAs, fieldworkers and village doctors were trained to disseminate BPCR messages that were also incorporated into a variety of visual aids during home visits, group discussions at clinics and village meetings. Comparison district received facility upgrade but no community intervention; control district received no intervention.

Iliyasu 2010
Cross-sectional study NIGERIA, Kano State Study aim: to assess men's perception of high risk pregnancy and danger signs; birth preparedness and complication readiness, and participation in maternity care.
Main findings: Half of the men considered bleeding a danger sign. One third mentioned convulsion as danger sign. Les than a third of the men made arrangements for mother's health care, transportation and delivery or made savings for obstetric emergencies. One third of the men accompanied women to maternity care. Higher participation was observed in younger educated men.

Jennings 2010
Quasi-experimental pre and post comparative study with a control group

BENIN, Zou/Collines region
Intervention: Introduction of the job aids: a set of pictorial counseling cards designed to support communication to women about care during and after pregnancy according to national guidelines. Intervention components: training, organizational changes, and field support. All health care personnel at the intervention sites were trained for three days in the content and use of the counseling cards, interpersonal communication, and quality improvement.
Main findings: The study measured three outcomes: (1) quality of counseling provided to pregnant women; (2) provider perceptions regarding use of the job aids; and (3) women's knowledge of messages relating to maternal and newborn care. Women in the intervention arm received more recommended messages than in the control arm. Increased communication skills regarding use of visual aids and verification of understanding was seen in the intervention arm. Improvements in knowledge among pregnant women were observed in the area of birth preparedness, recognition of danger signs, and clean delivery.

Kabakyenga 2011
Cross-sectional survey UGANDA, Mbarara district Study aim: To explore the association between knowledge of obstetric danger signs and birth preparedness among recently delivered women: Main findings: More than half of the women knew at least one danger signs during pregnancy, childbirth than during the post-partum period. Few women had knowledge of 3 or more key danger signs during the three periods. Of the four birth preparedness practices; 91% had saved money, 71% had bought birth materials, 61% identified a health professional and 61% identified means of transport.
Overall one third of the respondents were birth prepared (saved money, bought materials, identified health professional and identified transport). Young age and high levels of education had synergistic effect on the relationship between knowledge and birth preparedness.

Kabakyenga 2012
Cross-sectional survey UGANDA, Mbarara district Study aim: to assess the influence of birth preparedness practices and decision-making on location of birth and assistance by SBAs. Main findings: One third of the women had been prepared for childbirth and the prevalence of assistance by SBAs in the sample was two thirds. Decision making on location of birth was the husband in the majority of cases. When women made the final decision on location of birth in consultation with either the spouse or other people, the likelihood of giving birth assisted by a skilled birth attendant was very high and low when they made the decision alone.

Kakaire 2011
Cross-sectional facility based study UGANDA, Kabale district Study aim: to assess factors associated with birth preparedness and complication readiness as well as the level of male participation in the birth plan and healthcare Table 1 Characteristics of included studies ( a studies included in the systematic review) (Continued) seeking for emergency obstetric referrals. Main findings: Nearly half of the women had saved money in the event of complications and were joined by their men to ANC and during labour.

Karkee 2013
Prospective cohort study NEPAL, Kaski district Study aim: to assess birth preparedness level in expectant mothers and to evaluate its association with skilled attendance at birth. Main findings: The majority of women were birth prepared, 72% prepared the five activities (identification of delivery place, identification of transport, identification of blood donor, money saving, and antenatal care check-up). Of the cohort 85% SBA and it appeared that the more arrangements made, the more likely were the women to have skilled attendance at birth.

Kaso 2014
Cross-sectional study ETHIOPIA, Robe Woreda, Oromia Region Study aim: to assess knowledge and practices with respect to birth preparedness and complication readiness and factors associated in rural community among women of reproductive age. Main findings: Few respondents were prepared for birth and its complications and was higher amongst educated women.  One group before and after evaluation BURKINA FASO, Koupela Community and facility-based HCWs and SBAs provided one-on-one counselling with pregnant women and families on key messages focused on BPCR using a flip chart. These messages were reinforced through district-based radio messages and theatre plays. Facilities  Study aim: to examine whether BPCR counseling provided to during routine prenatal visits increased the probability of delivering in a health facility. Main findings: Exposure to information varied and not all BPCR messages were received equally. The four messages together (information on danger signs; promotion of facility-based delivery; information on the cost of delivery and advice on transportation during labour and in cases of obstetric emergencies) did not significantly influence the use of SBA.

Soubeiga, 2014
Systematic Review N/A Study aim: to evaluate the impact BPCR interventions in reducing maternal and neonatal mortality in low-income countries. Main findings: Meta-analysis showed no significant reduction on maternal mortality but identified an 18% reduction in neonatal mortality risk. There was a slight increase in the probability of facility-delivery. Note: seven out of the twelve included studies implemented action-learning cycles with women's groups.

Taleb 2015
Qualitative program evaluation

BANGLADESH, Netrokona district
Intervention: Implementation of Individuals, Families and Communities (IFC) program which focused on BPCR and working with TBAs to serve a new role in MNH which prioritized education, referral and social support of women rather than birth attendance with the aim to influence the social and cultural norms and practices surrounding care seeking in order to increase the utilization of skilled care. In order to promote BPCR, community-and facility-based health workers were trained to assist pregnant women and their families in creating a plan and to build community awareness of the importance of BPCR. A Birth and Emergency Preparedness Plan (BEPP) card was produced illustrating the following preparations: selecting a birth attendant; choosing a birth place and transportation to reach the birthplace; organizing with a birth companion; identifying a potential blood donor; developing a strategy to save money for costs related to pregnancy; and identifying where to seek care in the case of complications. Women receive the card either from health care providers in facilities during ANC visits or from CHWs through home visits. Main findings: Qualitative assessment revealed a more general trend towards planning for birth and complications and increase in knowledge of danger signs. Additionally a shift was identified in choosing for skilled care, although this was plans with husband and family. In the systematic review, it was impossible to determine which strategy and which components, or which combination of strategies and components, was most effective in improving health seeking outcomes [1]. Despite these differences, this analysis of contextual and implementation factors provides an understanding of some common barriers and facilitators to implementing interventions that promote BPCR. The SURE framework is added as Additional file 1.

Perspectives of women and their families
Some studies reported women's perceptions of birth as a normal and 'natural event' which could be successful at home, and that this often reduced the urgency to plan for facility birth. Other studies similarly reported beliefs that pregnancy outcomes are predetermined and 'in God's hands' , therefore there was no perceived need to be prepared for birth [13,14,45,63]. In two studies in Tanzania and Nepal 'modern' health services were regarded as the 'last resort' to be used only after complications arose [14,55]. Improving knowledge of danger signs is an essential element of most BPCR interventions. However, promoting this knowledge without ensuring awareness of the need for planning for normal birth might unintentionally results in the perception that no actions need to be taken if all signs are 'absent' and promote the notion that uncomplicated births indeed are best at home [14]. Quasi-experimental studies identified strong correlations between education level and BPCR, and concluded that BPCR interventions were more successful, and facility delivery more likely, among women with higher levels of education [11,13,15,17,21,56,57]. We found evidence in the studies that despite being able to recognise danger signs during pregnancy sometimes women remain silent and do not seek care because of cultural beliefs about the underlying causes. For example, in Tanzania obstructed labour, retained placenta and eclampsia were associated with adultery [47,58]. In some countries BPCR actions are limited due to fear of unfavourable outcomes and the belief that 'preparing' could bring bad luck [20,45]. In Tanzania and Kenya, although families reportedly discussed pregnancy and childbirth together (including husbands and wives), the studies indicate that taboos still exist and that this can restrict BPCR discussions. For example, announcing pregnancy and informing the husband when labour starts, is believed to bring misfortune [53], limiting husbands' ability to make timely preparations [47]. Cultural beliefs and norms also hindered transport preparations in some contexts, as women refrained from crossing a river since this was believed to cause abortion/preterm birth [47], and travel at night was considered dangerous due to active evil spirits [20]. In Bangladesh and Kenya, purchasing relevant items in preparation for birth was reportedly discouraged, especially items for the baby. Financial preparations were perceived as wasteful as it is unknown if the child would survive [20,53,58]. Identifying a SBA beforehand was a key BPCR message in all studies, and this inevitably involved making a choice about where to give birth and with whom. Some studies emphasized the importance of providing women with clear information during ANC on who is considered a SBA [50,55]. In some contexts, traditional birth attendants (TBAs) are the preferred attendant as the first point of care in pregnancy and when complications arise. They are considered 'skilled' because of their years of experience, ability to perform important rituals and willingness to attend women at home [48,59,60]. Similarly, women's understanding of the expected date of birth could limit timely preparations for facility birth or birth with a SBA, as some women perceived the estimated date of birth as exact date of birth, thus awaiting this exact time to make further plans [14,20,45,47].
Most interventions took place in contexts where men or other family members are the main decision makers and gatekeepers to women's timely access to care. Despite this, men are often excluded from maternal health interventions, and this can impact on the likelihood that BPCR actions are taken. In studies in Tanzania women reported that men may cause delay in seeking transport for women in labour or with complications for several reasons: they are unavailable at the time, they may not be aware of the emergency, or they feel ashamed to be seen supporting their wives [14,47]. A study in India reported that although men appear willing to perform certain tasks, primarily related to financial contributions, they often do not take on more proactive or supportive roles and are reluctant to get fully involved [61]. In a Nepali programme, women requested volunteers to increase awareness of husbands and mothers in-law, to help them in childbirth preparations [55]. Descriptive studies indicate that when men are involved in making plans for birth, they more often accompany their wives to the antenatal clinic and labour ward [16,18,45]. Some studies specifically involved men, by providing information to men (68), or training men as maternal health promoters (56,58). In the latter intervention, implemented in Eritrea and Tanzania, men delivered BPCR  messages to households and communities, which was well received and contributed to men's understanding of the importance of timely care seeking during pregnancy and for childbirth.

Perspectives of community stakeholders
Pregnancy and childbirth are usually regarded as family events and the wider community rarely plays a major role in preparation or readiness activities [47,50,61]. However, other community stakeholders who were not always targeted by BPCR interventions, such as community leaders, responded positively to interventions and implementers [57,61,62]. Interventions that did include community BPCR components resulted in increased awareness of maternal deaths in the community [43] and increased feeling of responsibility for pregnant women in the community. The latter through interventions which were specifically applicable to communities at large [50], such as developing transport or financial support systems, or through linking interventions to existing community structures [43,47]. In one study in Nepal, women requested more detailed information on where to go for birth and how to arrange transport [55]. In one study in India, community leadership was particularly supportive where youth groups held activities to increase awareness of maternal health problems, which subsequently increased direct interaction with government officials on problems faced by health providers and the women themselves [61]. Community transport and financial schemes for maternal emergencies were successful if supervised by transparent, trustworthy and stable leadership [57,63] and reversely inefficient when corrupted or insufficiently managed [43,57].
Implementers of BPCR messages, health workers or volunteers, were generally well accepted by individuals, families and communities. In one Tanzanian study, home visits by volunteers were especially appreciated for the time spent on discussions and questions [45]. BPCR messages were easily understood by women respondents in Nepal and Burkina Faso [23,52]. Studies that used visual aids such as cards, posters or booklets were positively received and understood [32,42,43,45,50].
Implementers generally reported they were satisfied with their activities and job aids [22,23,42,45,47] and felt supported by combinations of job aids with training, field support, and organizational change [22,23]. Some village volunteers felt appreciated by their communities and were committed to activities [47,55], others felt overburdened by study tasks [23]. Implementers in Nepal and Benin reported that job aids included too many or a repetition of messages or lacked concrete activities or examples for preparations to effectively help women and their families [22,23]. Facility interventions of BPCR during ANC in Benin and Tanzania increased workload by requiring more time with pregnant women [14,22].

Health service delivery factors
In some studies hospital staff and researchers noted that shortage of staff at facilities, in particular during nights and weekends, and high staff turnover limited intervention effectiveness. For example high turnover of facility personnel in Tanzania and Bangladesh [43,56], limited government training or supervision of staff in Burkina Faso and India [64,70] and limiting staff payments, causing health staff to run private business and as a consequence SBA absence in the clinics in Cambodia [42]. Giving staff additional tasks while at the same time maintaining morale and commitment was not easy, despite provision of additional training [43,47,49,51,53,56,64,66]. Several BPCR interventions reported in the included studies incorporated service delivery improvements including training of facility-based [43-45, 48, 49, 52, 54, 68] and community-based [46,47,64,67] health workers on (emergency) obstetric care, improving facility infrastructure and supplies [43,47,53,56,64] as well as strengthening the transport and referral system [46,51,53,56,64]. A study in Tanzania made additional effort to improve the supply of drugs, provided essential obstetric equipment, and facilitated strengthening of the logistics system at facility level [56]. However, no improvements were seen as ordered supplies were frequently not available mainly due to logistics problems at higher levels in the health system [56].
Women's and family member's views about the accessibility of care seemed to influence their willingness to prepare for facility birth. Lack of availability of care 24 h a day [42] or lack of awareness that lower-level health facilities also provide childbirth services [53] could result in women choosing home birth. Even if women considered facility birth, large distances to nearest facilities, poor road conditions and lack of transport options [47,[52][53][54] made facility birth unrealistic. Cost for transportation, formal and informal costs for facility-based services and general out-of-pocket payments limited affordability of care [14,47,53,55,56,61] and made TBA care often the cheaper option [42]. Perceived lack of quality of care at facilities including accounts of staff being described as inattentive and unconcerned or needing bribes before treatment was another barrier for SBA [53,58].

Social and political factors
Three studies in Nepal show implementation was affected by political instability and civil war, hindering implementation location or intensity [50,55,68]. The situation in Nepal reduced accessibility to health facilities due to security concerns in some areas [68] and in one study changed the implementation location from rural to urban although the intervention was thought to be most effective in rural geographical locations [50].
BPCR interventions in the studies included in this review were primarily implemented in patriarchal societies where gender inequality pervades household decision making [54,57,71]. Several interventions attempted to address this specific barrier to BPCR, by involving locally influential people. For example, TBAs, traditional healers, church leaders, community or clan elders and political leaders were consulted prior to the intervention or were encouraged to become involved as active promoters [62,63,66]. In Indonesia, promotion material featured a popular singer which helped to give the campaign a brand name which spoke to the majority of the target population [49].
One study in Tanzania showed benefits of collaboration with key-stakeholders and government leadership [47], in other countries government policies or changes in for example payment regulations could preclude BPCR implementation [42]. Although national policy changes in favour of MNHC priorities assisted the implementation process in Guatemala [51], BPCR implementation was limited by other national policy changes, such as ending cost-sharing policy in Kenya, increasing overall costs of care [53], and consequently reports of corruption [53,56].
Good rapport between the donor agency and government workers together with communities facilitated BPCR implementation, especially when engaging communities in problem solving in relation to BPCR [57,68,71]. Translating Non Governmental Organization (NGO) concepts into locally meaningful capacities or strategies, not only required time, but also adaptive skills from NGO workers from headquarter to local NGO level in Kenya and Tanzania [48,58]. Ability of national and local policy makers to continuously connect with communities and hospitals, through newly established village health worker committees or supportive supervision of health services, seem important factors for sustainability of interventions and finances [51,63]. Scale-up of the BPCR intervention was facilitated in Burkina Faso by connecting to existing health system structures [71]. In Cambodia, short-term BPCR interventions implemented independently of the health system structure, led to fear among programme planners and implementers, that project staff would become unemployed or move to the private sector once the intervention ended [42].

Discussion
Linking implementation factors with improvements in care seeking outcomes For studies included in the original systematic review that reported improvements in use of SBA or birth in a facility, the important factors that seemed to make a difference were positive perceptions of the intervention among women, families and community members and readily understood BPCR messages which motivated behaviour change [43,45,52]. Positive behaviour change, which could be making more preparations for childbirth or choosing to give birth with a skilled attendant, seemed to be more likely in women with higher levels of formal education [11,13,15,17,21,56,57], when husbands (as well as women) were targeted with BPCR messages on the need for SBA [43,45,52,61], or when the intervention helped to lessen the influence of traditional or cultural barriers [51]. In studies where BPCR interventions led to no or marginal improvements in number of women giving birth with a skilled attendant or at a facility, preference for home birth and underlying cultural beliefs mitigated the need for facility birth [47,53,55].
Political instability [50,55,68] and short duration of interventions [42,50] were reported to be among factors which prevented BPCR interventions impacting on care seeking outcomes. On the contrary, in studies reporting improvements in SBA or facility birth, BPCR was generally implemented within a package of interventions designed to address both demand and supply strengthening [43,46,49,51,56,64]. It is well recognized that community sensitization to BPCR without concurrent improvements in access to facilities and the quality of care provided will have little impact on care seeking or other health outcomes [43,55]. When interventions were integrated into existing government health service delivery systems [51,70], or were delivered in partnership with relevant safe motherhood stakeholders to ensure close links between the community and facilities [43,45], this seemed to increase impact on care-seeking outcomes. According to study evaluations the inclusion of perceptions of hospital management and sub-district policy makers on BPCR should be part of and could facilitate NGO implementation strategies [65]. These stakeholders could be offered support to make quality improvements in health policy making [72].

Limitations
Firstly, the variety of definitions and topics used to describe BPCR complicate interpretation of results, context and policy advice. Seven authors did not specify any BPCR definition [42,46,53,56,61,64,73]. Six interventions were focused on complication readiness only [43,46,47,59,67,70]. Five authors did explain that BPCR should also include preparedness for routine birth, [48,[51][52][53][54] although in one study this was not part of the intervention [43]. In areas with extremely low SBA use, ensuring improved conditions at home [preparing birth kits, a clean confinement room] were considered improvements in BPCR [43,46,47,59,67]. Secondly, skilled birth attendants are variously described in the included studies, making extrapolation of results from one context to another difficult. Ten studies defined skilled birth attendants as [47, 51-53, 55, 56, 59, 64, 67, 70] doctors, midwives or nurses. In some studies -as per WHO definition -unqualified staff including nurse aides were considered SBAs [47,53,56,64]. Other studies reported on health facility births [43,44,46,49,51,61,64,67,68] which does not necessarily mean the presence of a SBA. Two interventions included promotion of skilled care irrespective of location which consequently resulted in inclusion of home birth with SBAs [45,50]. As study contexts varied vastly, comparing studies on intervention and outcomes remains difficult. In our view defining interventions and outcome measures properly is crucial in BPCR, as improper definitions complicate interpretation of outcomes.
Further research of robust design is needed to: agree on key definitions and priority BPCR actions; assess the effect of including men and other key-stakeholders on care-seeking outcomes; and to understand how cultural factors influence BPCR implementation [5]. Although we aimed to include additional qualitative studies, only few were identified. Few studies reported on barriers or facilitators related to the intervention or program itself, such as resource implications, intervention integrity, leadership, and only some reported on donor policies or legislations. Often studies did not elaborate how and why such factors lead to successful or non-successful outcomes. These aspects also require further consideration in future research on BPCR interventions or packages.
The SURE framework proved useful in assessment of factors influencing implementation although we would argue that especially in low-resource settings 'cultural factors' should be added as a separate category. Cultural factors are generally under researched in maternal health studies [74] and such adaptation of existing models could form a way to swiftly create insight into complexities of implementing health interventions locally [75].

Conclusion
Implementation of BPCR interventions to improve the use of skilled care at birth requires careful consideration of contextual factors influencing implementation. When developing programmes and interventions, BPCR messages and strategies should match and respect the target audience and the different decision makers in maternal health and their values, as well as the organisation and capacities of the local health system. When mismatch occurs, such as when increased demand for facility births meets unprepared facilities in contexts where essential and comprehensive emergency obstetric care services are not available, this could cause considerable damage to the often already fragile trust the community has in the formal health system and increase complications or mistreatment of women giving birth at facilities. For this reason, it is important that BPCR is implemented alongside other interventions and activities to strengthen the supply and improve the quality of maternity care services. Implementation of BPCR should always include preparedness for both routine childbirth care and for complications, ensuring women and families have discussed the plans. Local socio-economic realities and determinants however remain a heavy burden for effective implementation of BPCR, and therefore require actions with the community and other stakeholders. BPCR messages should therefore be adapted to the local context in terms of availability, accessibility and affordability of health facilities and services. Inclusive and active involvement of all levels of stakeholders, including health officials and policymakers, appears to be a crucial step for securing linkages between the actions of all respective stakeholders that optimizes chances for women and newborns to reach needed care and contributes to the success of BPCR.