Background: A community-based intervention known as Safe Motherhood Action Groups (SMAGs) was implemented to increase coverage of maternal and neonatal health (MNH) services among the poorest and most remote populations in Zambia. While the outcome evaluation demonstrated statistically significant improvement in the MNH indicators, targets for key indicators were not achieved, and reasons for this shortfall were not known. This study was aimed at understanding why the targeted key indicators for MNH services were not achieved. Methods: A process evaluation, in accordance with the Medical Research Council (MRC) framework, was conducted in two selected rural districts of Zambia using qualitative approaches. Focus group discussions were conducted with SMAGs, volunteer community health workers, and mothers and in-depth interviews with healthcare providers. Content analysis was done. Results: We found that SMAGs implemented much of the intervention as was intended, particularly in the area of women’s education and referral to health facilities for skilled MNH services. The SMAGs went beyond their prescribed roles to assist women with household chores and personal problems and used their own resources to enhance the success of the intervention. Deficiencies in the intervention were reported and included poor ongoing support, inadequate supplies and lack of effective transportation such as bicycles needed for the SMAGs to facilitate their work. Factors external to the intervention, such as inadequacy of health services and skilled healthcare providers in facilities where SMAGs referred mothers and poor geographical access, may have led SMAGs to engage in the unintended role of conducting deliveries, thus compromising the outcome of the intervention. Conclusion: We found evidence suggesting that although SMAGs continue to play pivotal roles in contribution towards accelerated coverage of MNH services among hard-to-reach populations, they are unable to meet some of the critical sets of MNH service-targeted indicators. The complexities of the implementation mechanisms coupled with the presence of setting specific socio-cultural and geographical contextual factors could partially explain this failure. This suggests a need for innovating existing implementation strategies so as to help SMAGs and any other community health system champions to effectively respond to MNH needs of most-at-risk women and promote universal health coverage targeting hard-to-reach groups. Keywords: Access to healthcare, Process evaluation, Maternal health, Neonatal health, Remote and rural areas, Lay health workers, Zambia * Correspondence: email@example.com School of Nursing and Public Health, University of KwaZulu-Natal, Durban, South Africa School of Public Health, Department of Epidemiology and Biostatistics, University of Zambia, Lusaka, Zambia Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Jacobs et al. Implementation Science (2018) 13:74 Page 2 of 10 Background The SMAGs are groups of women and men working Zambia is one of the countries in the Sub-Saharan region as CHWs, traditional birth attendants (TBAs), child with poor maternal outcomes [1, 2]. Despite reductions in health promoters or growth monitoring promoters, mal- maternal and neonatal mortality rates in Zambia, approxi- aria agents and lay counsellors. The SMAGs were re- mately 398 deaths per 100,000 live births and 24 per 1000 cruited and trained in safe motherhood skills. The live births were reported in the recent Zambia Demographic choice and recruitment of SMAGs were guided by the Health Survey (ZDHS) . Arguably, most of these deaths Zambia Ministry of Health standards that entail one could be averted if all women utilised the maternal and CHW for every 500 of the population. A standard train- neonatal health (MNH) services . Unfortunately, in many ing programme of 5 days was used to empower SMAGs developing countries similar to Zambia, poor women living with safe motherhood knowledge and skills, specifically in the most remote and rural areas are least likely to receive for promoting antenatal care, delivery in a health facility adequate MNH care, including antenatal care (ANC), skilled with a skilled provider, postnatal home visits and essen- birth attendance (SBA) and postnatal Care (PNC), due to tial neonatal care. SMAGs were specifically trained in fo- service utilisation challenges such as geographical and cused antenatal care to identify danger signs, encourage socio-cultural barriers [4–8]. In the most remote and rural women to start ANC early, attend ANC at least four areas of Zambia, only a third of women receive SBA and times and receive skilled deliveries. In addition, SMAGs PNC . There is a need to improve the utilisation of and were trained in essential newborn care, including the access to MNH care services for women, particularly those provision of effective cord care, early initiation of exclu- in rural areas. Increasingly, community-based interventions sive breastfeeding and reporting maternal and neonatal through community health workers (CHWs) have been deaths that in the community. receiving recognition as an effective strategy to improve The key roles SMAGs played in the intervention were utilisation and access related to health facility-based services to refer women for ANC, delivery and complications [9–11]. By serving as a linkage between the community and during pregnancy, delivery and the postnatal period; ac- the formal health system, CHWs are well-placed to bridge tively following up women to close the gap in the MNH the service delivery gap in poor-resource settings , often continuum of care; and providing facility-based birth through effective referral to health services [12, 13]. Zambia preparedness messages to pregnant women and their has also been responding to the challenges of poor access spouses in the community. The work by the SMAGs and utilisation of MNH services through community-based was voluntary, and they could be called upon anytime a interventions, and among them are the Safe Motherhood woman in the community needed their services. The Action Groups (SMAGs). Neighbourhood Health Committees and health facility staff, including district coordinators, supervised the im- Description of the community-based intervention plementation of the SMAG activities. In 2013, through the Health for the Poorest Population (HPP) programme, a community-based intervention was Outcomes of the community-based intervention designed to strengthen MNH services. Through this pro- An outcome evaluation was conducted based on house- ject, intensified efforts were made to reduce disparities hold survey data collected at three time points during the in MNH services through SMAGs. The goal was to implementation of the intervention: baseline, mid-point make pregnancy safer through accelerated delivery of (mid-line) and end-point (end-line). priority MNH interventions (ANC, SBA and PNC). The Observations from the outcome evaluation showed a SMAGs, although not scaled up in all the districts, have statistically significant effect of the deployment of the been in existence in Zambia since 2003. The aim of the SMAG intervention on most of the MNH outcomes of SMAGs programme was to raise awareness about preg- interest, ANC, SBA and PNC, with an increasing trend nancy and birth-related complications and to reduce over time. Although statistically significant, the increase critical delays in decision-making at a household level in coverage over time was programmatically marginal, about seeking life-saving maternal healthcare in health and the programme’s targets for coverage were not met facilities . The programme was also aimed at im- (42 versus 80% for focused ANC, and 49 versus 60% for proving access to MNH services through linkages be- SBA) except for PNC for within 48 h by SMAGs (22 ver- tween the community and the healthcare facility . sus 20%). This study was therefore aimed at understand- Specifically, in the HPP project, the aim of the SMAG ing why the targeted key indicators for MNH services intervention was to reduce disparities in intervention were not achieved. coverage for MNH services and help meet the national There were also variations at the district level in cer- targets (80 and 60% for ANC at least four times and tain outcome indicators of interest, such as postnatal SBA, respectively) for coverage among the remote and care, the details of which were reported by Jacobs et al. poorest populations in rural areas.  on the outcome evaluation of the intervention. Jacobs et al. Implementation Science (2018) 13:74 Page 3 of 10 Similar to the findings in other community-based Key constructs of the process evaluation intervention studies [16–18], the observed results of The UK MRC framework  was adopted to guide the the outcome evaluation needed further explanation, identification of relevant key constructs and to generate taking into account interactions between contextual evaluation questions in this study. According to Moore factors and the SMAG intervention. Therefore, the et al. , despite a need to understand casual assump- purpose of this paper was to understand why the tar- tions that underpin an intervention in complex interven- geted key indicators for MNH services were not tions such as the HPP project, there is also a need to achieved. Specifically, the paper explored the follow- understand how the intervention worked by scrutinising ing questions: (1) Was the intervention (the SMAGs) its plausibility and the relations between implementa- implemented as it was intended? (2) What are the tion, mechanisms of impact and context. The SMAG factors external to the intervention that may have in- intervention was regarded as complex because it com- fluenced implementation of the intervention? and (3) prised multiple interacting components and a number of What are the possible mechanisms that likely explain targets to be met. According to the MRC framework, an the gap between achieved and targeted outcomes? intervention may have limited effects or positive out- This study will provide an understanding of possible comes due to its implementation processes such as fidel- explanations for the partial success of the SMAG ity, whether the intervention was implemented as intervention, taking into account the contextual fac- intended or the degree to which an intervention is deliv- tors under implementations occurred, so as to inform ered as intended; the dose; the quantity of the interven- similar future programmatic decision-making. tion implemented; and its reach, whether the intended audience comes into contact with the intervention or not . While the implementation context includes Methods anything external to the intervention that may act as a Study setting barrier or facilitator to its implementation . Further, The study was conducted in two remote districts, lo- the mechanism of impact guides an understanding of cated in Luapula and the northern provinces of Zambia. how an intervention was delivered and how the effects The districts are among the four districts for the HPP of the intervention occurred. An illustration of these key where the SMAGs programme was implemented . constructs and the assumptions on their interaction with To select the study districts, we first stratified the dis- the intervention is provided in Fig. 1. tricts into two provinces. Within each of the two prov- Existing evidence shows that the outcome of a complex inces, we randomly selected one of the two intervention community-based intervention can be influenced by the in- districts by flipping a coin. From each of the selected teractions between the stated three key constructs, namely districts, two intervention health facilities were randomly implementation, context and mechanisms [19, 21]. selected using a lottery method, where all the facilities were assigned numbers, after which two numbers were Causal assumptions for the intervention selected at random. The logic ‘inputs-processes-outputs-outcomes-impact model’ was used as a theory of change to guide the Design implementation of the intervention. Figure 2 describes A process evaluation was conducted retrospectively, in the inputs, outputs (activities, participation) and their accordance with the Medical Research Council (MRC) links to outcomes. Based on the model, the inputs framework [19, 20], between November 2016 and Janu- included implementation plans, human resources, ary 2017, using focus group discussions (FDGs) and funding and working with district health teams. The in-depth interviews (IDIs). Process evaluations have been processes included training of SMAGs and procure- reported as an essential part of community-based inter- ment of supplies, including bicycles and medicines, ventions [19, 21], needed to provide insight on how well and the creation of data collection tools/systems that programme activities are implemented, and performing would facilitate the development of the community within the context in which implementation occurs . Health Management Information Systems (HMIS). According to Moore and colleagues , effect sizes These processes were expected to lead to short-term alone may not inform policy and implementers on how results that were expressed as output indicators, such such community-based interventions may be replicated as numbers of CHWs trained and referrals conducted. or reproduced in specific contexts. Moore et al. further It was also assumed that the processes of the inter- argued that process evaluations are needed to assess fi- vention would ultimately lead to medium-term out- delity and quality of implementation, as well as to iden- comes of the intervention based on baseline coverage tify causal mechanisms and contextual factors associated data, such as the proportion of mothers receiving at with the variations in the outcomes of interventions. least four ANC visits during pregnancy. Finally, the Jacobs et al. Implementation Science (2018) 13:74 Page 4 of 10 Fig. 1 Key constructs of the process evaluation and the relations among the constructs  impact was the long-term goal of the project that women and SMAGs who were purposively selected. The would include a reduction in neonatal, infant and ma- inclusion criteria for focus groups with women were ternal morbidity and mortality. However, it was also women of reproductive age, with children less than 1 year noted that there would be external factors likely to old and living within the study community during their interact with this theory of change. most recent pregnancy. The SMAGs were included in the study with the help of healthcare workers at the facility Study participants and sampling level if they were above the age of 18 years, both male and Participants engaged in the intervention were purposively females, working within the communities under study on sampled for in-depth interviews and focus group discus- the implementation of the intervention and living either sions. Healthcare providers were purposively selected for within or beyond 5 km radius from the health facility. A in-depth interviews based on their active involvement in total of 78 participants were interviewed, 34 SMAGs, 36 maternal and neonatal health as well as in the interven- mothers and 8 healthcare providers from Samfya and tion. Focus group discussions were conducted with Luwingu districts. Fig. 2 Logical model for the Health for the Poorest Populations project Jacobs et al. Implementation Science (2018) 13:74 Page 5 of 10 Data collection and tools was done by cross-examining the data [25, 26]. Triangu- Eight in-depth interviews were conducted with health- lation is a recognised method to increase the credibility care providers, using qualitative research techniques of data analysis . This was achieved through data tri- to explore issues related to the implementation of the angulation whereby the perspectives of the different re- intervention, such as referral practices and supervi- spondent groups were explored. We also maintained a sion at the community level. In addition, eight FGDs detailed audit trail of all decisions through a codebook, were conducted, two from each facility. At each of coding discussions and meetings. the four facilities, one FGD was with SMAGs and an- other with mothers. Trained research assistants with Results experience in qualitative studies collected the data, A total of 78 participants were interviewed, 34 SMAGs, 36 1 year after the intervention. Focus group discussions mothers and 8 healthcare providers from two rural districts. were conducted by a pair of research assistants, who The majority of the SMAGs were female (59%) with an age were of thesamegenderand fluent in thelocal lan- range of 24 to 71 years. All the SMAGs (100%) received guage (Bemba). One research assistant facilitated the training on safe motherhood skills before the intervention sessions while the other one managed audio record- and were working in the community. Three out the eight ings and took field notes. The research assistants healthcare providers (38%) were female; one of them was a underwenta2-daytrainingprior to thedatacollec- community health assistant and another a classified daily tion and were supervised by one of the co-authors employee. The mothers’ age range was 19 to 46 years; none (CJ). The data collection tools were piloted in a simi- of them reported that they have never been to school, and lar facility not included in the study (Additional file 1). the majority (64%) had not completed primary education. The average duration of FGDs and KIIs was 45 min. The interviews were delivered on a face-to-face basis, at the Thematic areas health facilities. Informed consent was obtained from all On the whole, SMAGs were able to implement the the participants, and digital voice recorders were used to intended intervention, by identifying pregnant women document the interviews and discussions. and encouraging them to attend ANC visits, assisting with birth preparedness plans and clinic attendance for Data management and analysis skilled birth assistance as well as postnatal care. In Recorded data were transcribed verbatim, and translated addition, SMAGs were able to refer women with compli- from Bemba to English, supplemented with field notes. cations in pregnancy and following delivery, as well as All transcripts were assigned a unique identifier and doing follow-up visits for those under their care and imported into NVivo 13 for data management and ana- submitting the necessary written reports. However, the lysis. Data was coded by two individuals, a trained re- following themes under the three major constructs of search assistant and one of the co-authors (CJ). An implementation process were found to potentially influ- iterative inductive thematic approach  was used ence the implementation of the intervention and likely through repeated rounds of reading and re-reading to to explain the failure to meet outcome targets: service clarify coding differences and to ensure consistency for shortages, geographic barriers, socio-cultural factors and subsequent analyses. Coders first independently listened implications of personalising care (see Fig. 3). to some recordings, reviewed a sample of the transcripts and began to formulate draft codes and themes. The re- searchers then met after coding the first six interviews to discuss the coding. Discrepancies were discussed until consensus was reached. Coding meetings with the re- search team and an experienced research assistant were held every week to create a mutual understanding of codes and refine the coding framework. The two coders examined and assigned sections of text to codes, repre- senting themes or subthemes. Extracts of data were coded and memos were written to record emerging im- pressions of the data. Coded data extracts were further discussed among all the authors and merged into cat- egories before refining them into themes. To further ver- ify our results, we returned to the raw data. To enhance Fig. 3 Summary of the findings based on the key constructs of the study validity, triangulation of different data sources MRC framework (FGDs and IDIs) between different respondent groups Jacobs et al. Implementation Science (2018) 13:74 Page 6 of 10 Implementation processes SMAGs were seen to have taken over the roles of Service shortages TBAs, which made it harder for healthcare providers and community members to shift their expectations, and Lack of supervision Most SMAGs expressed concern in fact, expected SMAGs to do the work of the TBAs. over lack of supervision and refresher courses from the health facility and district health staff. They indi- It is hard to differentiate the care given by SMAGs cated that supportive supervision was only provided from that of TBAs because it is like their roles have in the initial phases of the SMAG programme, which been mixed… yes uh, the SMAGs are doing the roles was demotivating and a cause of dropout for some of the TBAs. (HCW: IDI2:1) SMAGs. When the programme started in 2013, they used to Implementation context come and check on us. But nowadays we just work Geographic barriers with the staff. Otherwise, there is no support provided to us, and others end up withdrawing. Referral to clinics Distance to the health facility, poor (Mother: FGD3:1) road conditions and inadequate transport to the health facility emerged as prominent challenges that hindered SMAGs’ ability to refer women to the health facilities for Lack of equipment SMAGS also reported that lack of skilled care. protective clothing such as gumboots, raincoats and torches made movements at night or during the rainy The points that my sisters have said are true, we have season challenging. These additional supplies would a challenge here because some of our areas are very have assisted SMAGs to work more efficiently through far......... others have to cross two streams, they live the nights and rains. very far where there is no SMAG. So the point here is that we have to sacrifice to go and see them, and if Yes ......these are things that we lack and without you don’t have a bicycle to use it is a problem. these, our work is very challenging, things like (SMAG: FGD2:3:3) raincoats, torches because they [patients] come to wake us up in the night to escort them. (SMAG: FGD2: 2:7) Means of transport The need for SMAGs to transport pregnant women to the health facility, and the difficulty thereof, hampered their ability to promote skilled deliv- Shortage of staff The respondents reported that lack of eries in health facilities. A female SMAG respondent skilled health personnel in some facilities made the work confirmed that some communities were more than 4 h of SMAGs more difficult in that some women in the walk to the health facility. community did not see any benefit of being referred to a health centre where there was no skilled personnel. Some of them are within the clinic catchment area of Ndoki, but others take about 4 hours walking to reach We have a problem because some women do not see here so if there is no bicycle you walk with a pregnant any need of going to the facility only to be attended woman, imagine the challenge. (SMAG: FDG2: 1:1) by someone who is not a nurse, and especially a young man. (SMAG: FGD2:2:2) All the SMAG respondents stated that bicycles, a major means of transport that was used in the commu- nities, were either not available or were inadequate since Conduct of deliveries Discussion with respondents re- these were shared among all the SMAGs in the vealed that SMAGs conducted deliveries in the health community. facilities. According to most healthcare providers, the SMAGs were a relief in facilities when healthcare pro- Socio-cultural factors viders were overwhelmed with work or not available. Gender roles An intended gender-neutral role for Ah for delivery, usually they [SMAGs] refer clients SMAGs, which entails similar roles between male and fe- to clinic, but when I am not at the facility since I male SMAGs, was reported as a challenge for effective am alone, they [SMAGs] do conduct deliveries. provision of referral services by the SMAGs. Male SMAGs (HCW: IDI1:1) expressed concern that some mothers, particularly those Jacobs et al. Implementation Science (2018) 13:74 Page 7 of 10 unmarried, found it difficult to seek care from a male Addressing personal problems SMAGs were regarded SMAG on pregnancy or childbirth-related issues and needs. as individuals who could not only handle some of their Male SMAGs, therefore, preferred to send female SMAGs health needs but also mothers’ personal problems. Some to young unmarried mothers to identify problems. mothers indicated that the trust they had in SMAGs motivated them to seek advice from the SMAGs. One of Just as someone said, most women who get pregnant the mothers had this to say: and are not married feel shy to come and tell a male SMAG, so we send a female SMAG to see her. Then We trust them, which is why we go to them. Even that female SMAG will now come and tell us what if you are quarrelling at home, we get up and go to the problem is. (SMAG: FGD1:2:8) the SMAG. When you are pregnant, and tell them listen, my husband has failed to prepare. Come and teach him so that he knows how to prepare the Gender preferences Healthcare providers confirmed the clothes for the baby, because others are drunkards. SMAGs’ assertions over gender preferences by mothers (Mother: FGD3:9) and that male referrals are few and far in between, which may indicate the compromised effectiveness of male SMAGs in their role compared to female SMAGs. Use of personal resources Sometimes, the SMAGs had to use their personal resources to meet some costs such Yes, in fact even referrals that I usually receive most as transport costs to get the mothers to the health facil- of them are those that are referred by female SMAGs, ities. Most women and their families could hardly afford as for male SMAGS, I think it takes a year ah...ah it decent clothing for their newborn baby because of can happen once a year. (HCW: 1DI2:4) poverty. Sometimes, a woman and her husband may not even Bearers of ‘life and death’ reports When SMAGs have things for the baby, so you the SMAG have to escorted a woman for facility deliveries, they had to wait for give them a chitenge or a nappy. So that money has the women to deliver. The long hours and sometimes days to be paid for by us SMAGs just to help our friends to of waiting were a challenge to the roles of SMAGs. When deliver on time. (SMAG: FGD1:1:6) asked why they waited, most SMAGs explained that com- munities viewed them as bearers of ‘life and death’ news. In such circumstances, the SMAGs would take up the responsibility of securing clothing for the newborn baby, Concerning waiting for the woman until she is and related requirements, just to overcome the barrier discharged, you get worried because pregnancy is a and get women to the health facility. These expenses matter of life and death. So this makes us wait until a made it difficult for the SMAGs, as they themselves were woman delivers because once the family members also poor. hear that the woman has delivered, they become relieved. That is why she is saying we wait. Discussion (SMAG: FGD1:2:6) The current study, a process evaluation approach, was designed to provide insights into explanatory mecha- nisms for the variations observed between actual and Mechanism of impact targeted outcomes for MNH services, ANC and SBA, Personalising care following a SMAG community-based intervention in Zambia. We found, firstly, that SMAGs implemented Going beyond the call of duty SMAGs provided perso- much of the intervention as was intended, particularly nalised care beyond their role in the intervention. The education of women and referral to health facilities for personalised care included among others house chores ANC, SBA, PNC and complications. Secondly, SMAGs and marital counselling. These extended roles were out- went above and beyond their prescribed roles to assist side their prescribed scope of work but appreciated by women with household chores and personal problems the pregnant women whose physical conditions necessi- and even used their own resources to support needy tated support. One of the mothers said: households, so as to enhance the success of their work. Thirdly, SMAGs reported what could be regarded as de- When they come to visit us they bring us food ficiencies in the intervention itself, due to poor ongoing and even draw for [us] water just like that. support, inadequate supplies and lack of effective trans- (Mother: FGD2:1:3) portation needed to enable success in their work, leading Jacobs et al. Implementation Science (2018) 13:74 Page 8 of 10 to attrition of SMAGs from the intervention. Fourthly, CHWs has been reported in other rural studies that also the beneficiaries did not perceive the intervention to be involved referral or community members to health facil- gender-neutral as was assumed during intervention de- ities . For the SMAG intervention to function and be sign; in that, mothers preferred to disclose their health sustainable, empowering CHWs with financial incentives issues to female SMAGs rather than male SMAGs, appears to be critical in order to enable them to effect- thereby compromising the intensity of the intervention. ively provide their services and meet needs of commu- Lastly, factors external to the intervention may have nity members in rural and poor communities [37, 38]. compromised the results, particularly the inadequacy of The remoteness of rural communities has been reported health services in facilities where SMAGs referred in other studies as a barrier to mothers’ ability to reach mothers, made worse by the long distances and difficult the healthcare facilities [5, 39–41]. These findings sug- terrain women had to travel in the company of SMAGs. gested the need for effective and sustainable means of The health facility-related service deficiencies led transport in community-based interventions designed to SMAGs to engage in the unintended role of conducting improve access by linking communities to the health deliveries. Some SMAGs were also known to be TBAs, system through SMAGs . whose birth attendance activities have been terminated In these remote areas, women found it easier to disclose despite their many years of experience in assisting births pregnancy-related issues to female SMAGs than male in the communities. Previous studies have indicated that SMAGs, as healthcare occurs within cultural-bound SMAGs were chosen from the existing TBAs and other norms and sensitive socio-cultural factors, often more community health volunteer [27, 28]. However, health common in remote rural communities [43, 44]. The find- workers allowed SMAGs known to have worked as ings thus suggest a need for gendered SMAG roles  TBAs to conduct deliveries in the health facility when and male-female task sharing functions if MNH services facilities experienced absenteeism, shortages or work are to be optimised for marginalised women in remote pressure. The unintended consequence of hiring previ- and hard-to-reach areas. Feldhaus et al. argued that ous TBAs meant that SMAGs could conduct deliveries pairing male and female SMAGs may potentially address in health facilities, but these were not recognised as and accommodate gender preferences among SMAGs for skilled attendance, even though some of these deliveries pregnancy-related issues . All in all, these findings may have been performed under the supervision of highlight the interrelationships between the contextual nurses. Such implementation mechanisms could explain factors and the implementation processes of the the failure of the SMAG intervention to meet the set intervention. targets for SBA. The role confusion between activities of However, this study has potential limitations that SMAGs and TBAs should be considered in future inter- should be noted. Firstly, the collection of data for this ventions to avoid unintended consequences . Enhan- process evaluation was conducted after the outcome cing strategies that provide awareness of the specific analysis, which may have been compromised by recall roles of CHWs is critical to guide expectations of com- bias in retrospectively reconstructing events, and a pro- munities and healthcare providers towards the specific spective process evaluation design would have been services in community-based interventions . We also more preferable. Secondly, self-reported accounts of suggest the critical need of ensuring availability of skilled intervention fidelity in this study are prone to bias and personnel in health facilities to avoid similar unintended may have lower reliability than observational measures consequences of SMAGs . . However, the adoption of a data triangulation ap- Further, the findings in this study are consistent with proach was intended to allow for validation of findings the studies in Zambia and elsewhere [28, 29], which have from different sets of participants. Thirdly, data for this highlighted the importance of recruiting SMAGs within study was collected from respondents who were still ac- their communities where they are trusted and preferred. tive as SMAGs, and therefore not representative views The SMAGs’ unique position and moral authority within from inactive SMAGs. these communities had the potential to influence out- comes to the intervention positively [28, 29, 32–34] Conclusion given their existing social networks [28, 29] and inclin- In conclusion, we found evidence suggesting that although ation to be natural helpers through personalised care SMAGs continue to play pivotal roles in contribution to- provision . However, provision of personalised care wards accelerated coverage of maternal and neonatal by SMAGs also meant going beyond their scope and health services among hard-to-reach populations, they are using their own personal resources, an indication of the unable to meet some of the critical sets of MNH service inadequacy of the health system and a threat to optimal targeted indicators. Complexities of the implementation provision of referral services, a key component of the mechanisms coupled with the presence of differential set- SMAGs intervention. Use of personal resources by ting specific social, cultural and geographical contextual Jacobs et al. Implementation Science (2018) 13:74 Page 9 of 10 factors could partly explain this failure. More specifically, Availability of data and materials The datasets generated and/or analysed during the current study are not we think that deficiencies in the implementation mecha- publicly available in keeping with the anonymity of study participants but nisms and the factors external to the SMAG intervention are available from the corresponding author on reasonable request. such as the health services inadequacy of skilled health- Authors’ contributions care providers in facilities where SMAGs referred mothers CJ and MM contributed to the conceptualisation of the study design. CJ as well as poor geographical access may have led SMAGs collected the data and analysed it. MM and CM were doctoral supervisors to engage in the unintended role of conducting deliveries. who supported CJ in writing the manuscript. MM and CM contributed substantially in the revision and interpretation of the study findings. All This could have compromised the performance of the authors read and approved the final version of the manuscript. SMAGs and led to the failure to meet the set targets for key MNH service indicators among women in the most Ethics approval and consent to participate rural and remote communities. These observations point This study is part of a larger doctoral research project aiming to determine the utilisation of maternal and neonatal health services among the remote to limitations in past and existing efforts to improve and poorest populations of Zambia. This study was approved by the Tropical MNCH health service delivery. There is thus need to Disease Research Centre in Zambia (TDRC; Ref No: TRC/C4/07/2015) and the innovate and re-package existing implementation strat- University of KwaZulu-Natal in South Africa (Ref No: BE363/15). Formal per- mission was also obtained from the Zambia Ministry of Community Develop- egies, such as recruiting adequate health care providers ment, Mother and Child and from the traditional leaders through the District and ensure adequate provision of logistics and equipment, Medical Officers of the respective selected districts. Informed consent was so as to help SMAGs and any other community health obtained from each study participant after explaining the objectives of the study and procedure. systems champions to effectively respond to MNH needs of most-at-risk women and promote universal health Competing interests coverage targeting hard-to-reach groups. If this is done The authors declare that they have no competing interests. correctly, policy and programme managers will be armed with systems and strategies that adequately and appropri- Publisher’sNote ately match the complex nature of the implementation Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. mechanisms and external contextual factors that hinder successful implementation of community-based interven- Author details tions such as SMAGs in hard-to-reach areas. School of Nursing and Public Health, University of KwaZulu-Natal, Durban, South Africa. School of Public Health, Department of Epidemiology and Biostatistics, University of Zambia, Lusaka, Zambia. Strategic Centre for Health Systems Metrics and Evaluations (SCHEME), School of Public Health, Additional file University of Zambia, Lusaka, Zambia. Africa Health Research Institute, KwaZulu Natal, South Africa. Additional file 1: Discussion and interview guides. (DOCX 45 kb) Received: 3 January 2018 Accepted: 21 May 2018 Abbreviations ANC: Antenatal care; CHW: Community health worker; FGD: Focus group References discussion; HPP: Health for the poorest populations; IDI: In-depth interview; 1. Central Statistical Office (CSO) [Zambia], Ministry of Health (MOH) [Zambia], MNH: Maternal and neonatal health; MRC: Medical research council; and ICF International. Zambia Demographic and Health Survey 2013-14. PNC: Postnatal care; SBA: Skilled birth attendants; SMAGs: Safe motherhood Rockville: Central Statistical Office, Ministry of Health, and ICF International; action groups 2014. 2. World Health Organization UNICEF. Trends in maternal mortality: 1990-2015: estimates from WHO, UNICEF, 503 UNFPA, World Bank Group and the Acknowledgements United Nations Population Division: executive 504 summary. 2015. The authors thank all the study participants for their time, officials at the 3. Tunçalp Ӧ, Were W, MacLennan C, et al. Quality of care for pregnant Ministry of Health, the district health management teams and technical women and newborns—the WHO vision. BJOG Int J Obstet Gynaecol. 2015; advisors that supported the implementation of the SMAG intervention. 122(8):1045–9. 4. Jacobs C, Moshabela M, Maswenyeho S, Lambo N, Michelo C. Predictors of antenatal care, skilled Birth attendance, and Postnatal care Utilization Funding among the remote and Poorest rural communities of Zambia: a Multilevel analysis. Front Public Health. 2017;5:11. This doctoral research was partly funded by the African Union 5. Sialubanje C, Massar K, Hamer DH, et al. Personal and environmental factors Commission Scholarship for female students in Africa awarded to associated with the utilisation of maternity waiting homes in rural Zambia. the corresponding author. BMC pregnancy childbirth. 2017;17(1):136. Data collection for this study was partially supported by the African 6. Chama-Chiliba CM, Koch SF. Utilization of focused antenatal care in Zambia: Doctoral Dissertation Fellowship (ADDRF) award offered by the examining individual-and community-level factors using a multilevel African Population and Health Research Centre in partnership with analysis. Health Policy Plan. 2013;30(1):78–87. the International Development Research Centre (IDRC) and by the 7. Ameh S, Adeleye OA, Kabiru CW, et al. Predictors of poor pregnancy Norwegian Agency for Development (NORAD) through the outcomes among antenatal care attendees in primary health care facilities NORHED QZA-0848 QZA-MW-13/00032 small grant awarded to in Cross River State, Nigeria: a multilevel model. Matern Child Health J. 2016; the corresponding author. 20(8):1662–72. https://doi.org/10.1007/s10995-016-1965-5. 8. Bourke L, Humphreys JS, Wakerman J, et al. Understanding drivers of rural The above-stated funders had no role in the design of the study and collection, and remote health outcomes: a conceptual framework in action. Aust J analysis, interpretation of data or in writing the manuscript. Rural Health. 2012;20(6):318–23. Jacobs et al. Implementation Science (2018) 13:74 Page 10 of 10 9. August F, Pembe AB, Mpembeni R, et al. Effectiveness of the home based sector: a comparative analysis of factors shaping relationships in four life saving skills training by community health workers on knowledge countries. Glob Public Health. 2017;12(11):1404–32. of danger signs, birth preparedness, complication readiness and facility 34. Puett C, Coates J, Alderman H, et al. Quality of care for severe acute delivery, among women in rural Tanzania. BMC pregnancy childbirth. malnutrition delivered by community health workers in southern 2016;16(1):129. Bangladesh. Matern Child Nutr. 2013;9(1):130–42. 10. Adam MB, Dillmann M, M-k C, et al. Improving maternal and newborn 35. Terpstra J, Coleman KJ, Simon G, et al. The role of community health health: effectiveness of a community health worker program in rural Kenya. workers (CHWs) in health promotion research: ethical challenges and PLoS One. 2014;9(8):e104027. practical solutions. Health Promot Pract. 2011;12(1):86–93. 11. Chopra M, Sharkey A, Dalmiya N, et al. Strategies to improve health 36. Sips I, Mazanderani AH, Schneider H, et al. Community care workers, poor coverage and narrow the equity gap in child survival, health, and nutrition. referral networks and consumption of personal resources in rural South Lancet. 2012;380(9850):1331–40. Africa. PLoS One. 2014;9(4):e95324. 37. Becker J, Kovach AC, Gronseth DL. Individual empowerment: how 12. Haines A, Sanders D, Lehmann U, et al. Achieving child survival goals: potential community health workers operationalize self-determination, self-sufficiency, contribution of community health workers. Lancet. 2007;369(9579):2121–31. and decision-making abilities of low-income mothers. J Community Psychol. 13. Perez F, Ba H, Dastagire SG, et al. The role of community health workers in 2004;32(3):327–42. improving child health programmes in Mali. BMC Int Health and Hum 38. Roman LA, Lindsay JK, Moore JS, et al. Community health workers: examining Rights. 2009;9(1):28. the helper therapy principle. Public Health Nurs. 1999;16(2):87–95. 14. Ensor T, Green C, Quigley P, et al. Mobilizing communities to improve 39. Sacks E, Masvawure TB, Atuyambe LM, et al. Postnatal care experiences and maternal health: results of an intervention in rural Zambia. Bull World Health barriers to care utilization for home-and facility-delivered newborns in Organ. 2014;92(1):51–9. Uganda and Zambia. Matern Child Health J. 2017;21(3):599–606. 15. Jacobs C, Michelo C, Chola M, et al. Evaluation of a community-based 40. SNa P, Fylkesnes K, Moland KM, et al. Rural-urban inequity in unmet intervention to improve maternal and neonatal health service coverage in obstetric needs and functionality of emergency obstetric care services in a the most rural and remote districts of Zambia. PLoS One. 2018;13(1): Zambian district. PLoS One. 2016;11(1):e0145196. e0190145. 41. Onah HE, Ikeako LC, Iloabachie GC. Factors associated with the use of 16. Tesfaye S, Barry D, Gobezayehu AG, et al. Improving coverage of postnatal maternity services in Enugu, Southeastern Nigeria. Soc Sci Med. 2006;63(7): care in rural Ethiopia using a community-based, collaborative quality 1870-8. https://doi.org/10.1016/j.socscimed.2006.04.019. improvement approach. J Midwifery Womens Health. 2014;59(s1):S55–64. 42. Turinawe EB, Rwemisisi JT, Musinguzi LK, et al. Towards promotion of 17. Wilunda C, Tanaka S, Putoto G, et al. Evaluation of a maternal health care community rewards to volunteer community health workers? Lessons from project in south west Shoa Zone, Ethiopia: before-and-after comparison. experiences of village health teams in Luwero, Uganda. Res Health Sci. Reprod Health. 2016;13(1):95. 2016;1(2):85. 18. Sharma S, Van Teijlingen E, Belizán JM, et al. Measuring what works: an 43. Akeju DO, Oladapo OT, Vidler M, et al. Determinants of health care impact evaluation of women’s groups on maternal health uptake in rural seeking behaviour during pregnancy in Ogun State, Nigeria. Reprod Nepal. PLoS One. 2016;11(5):e0155144. Health. 2016;13(1):32. 19. Moore GF, Audrey S, Barker M, et al. Process evaluation of complex 44. Worthington RP, Gogne A. Cultural aspects of primary healthcare in India: a interventions: Medical Research Council guidance. BMJ. 2015;350:h1258. case-based analysis. Asia Pac Fam Med. 2011;10(1):8. 20. Rosecrans A, Gittelsohn J, Ho L, et al. Process evaluation of a multi- 45. Elazan SJ, Higgins-Steele AE, Fotso JC, et al. Reproductive, maternal, institutional community-based program for diabetes prevention among first newborn, and child health in the community: task-sharing between male nations. Health Educ Res. 2007;23(2):272–86. and female health workers in an Indian rural context. Indian J Community 21. Craig P, Dieppe P, Macintyre S, et al. Developing and evaluating Med: official publication of Indian Assoc Pre Soc Med. 2016;41(1):34. complex interventions: the new Medical Research Council guidance. 46. Feldhaus I, Silverman M, LeFevre AE, et al. Equally able, but unequally BMJ. 2008;337:a1655. accepted: gender differentials and experiences of community health 22. Harachi TW, Abbott RD, Catalano RF, et al. Opening the black box: using volunteers promoting maternal, newborn, and child health in Morogoro process evaluation measures to assess implementation and theory building. region, Tanzania. Int J Equity Health. 2015;14(1):70. Am J Community Psychol. 1999;27(5):711–31. 47. Mabe PA, West SG. Validity of self-evaluation of ability: a review and meta- 23. Carroll C, Patterson M, Wood S, et al. A conceptual framework for analysis. J Appl Psychol. 1982;67(3):280. implementation fidelity. Implement Sci. 2007;2(1):40. 24. Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol. 2006;3(2):77–101. 25. Miles MB, Huberman AM, Saldana J. Qualitative data analysis: a method sourcebook. CA, US: Sage publications; 2014. 26. Boeije H. A purposeful approach to the constant comparative method in the analysis of qualitative interviews. Qual Quant. 2002;36(4):391–409. 27. Sialubanje C, Massar K, Horstkotte L, et al. Increasing utilisation of skilled facility-based maternal healthcare services in rural Zambia: the role of safe motherhood action groups. Reprod Health. 2017;14(1):81. 28. Sialubanje C, Massar K, Hamer DH, et al. Reasons for home delivery and use of traditional birth attendants in rural Zambia: a qualitative study. BMC Pregnancy Childbirth. 2015;15(1):216. 29. Titaley CR, Hunter CL, Dibley MJ, Heywood P. Why do some women still prefer traditional birth attendants and home delivery?: a qualitative study on delivery care services in west java province, Indonesia. BMC Pregnancy Childbirth. 2010;10(1):43. https://doi.org/10.1186/1471-2393-10-43. 30. Swider SM. Outcome effectiveness of community health workers: an integrative literature review. Public Health Nurs. 2002;19(1):11–20. 31. Byrne A, Morgan A. How the integration of traditional birth attendants with formal health systems can increase skilled birth attendance. Int J Gynecol Obstet. 2011;115(2):127–34. 32. Simwaka BN, Theobald S, Willets A, et al. Acceptability and effectiveness of the storekeeper-based TB referral system for TB suspects in sub-districts of Lilongwe in Malawi. PLoS One. 2012;7(9):e39746. 33. Kok MC, Ormel H, Broerse JE, et al. Optimising the benefits of community health workers’ unique position between communities and the health
Implementation Science – Springer Journals
Published: May 31, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera