Implementation strategies for infection prevention and control promotion for nurses in Sub-Saharan Africa: a systematic review

Background Despite impressive reductions in infectious disease burden within Sub-Saharan Africa (SSA), half of the top ten causes of poor health or death in SSA are communicable illnesses. With emerging and re-emerging infections affecting the region, the possibility of healthcare-acquired infections (HAIs) being transmitted to patients and healthcare workers, especially nurses, is a critical concern. Despite infection prevention and control (IPC) evidence-based practices (EBP) to minimize the transmission of HAIs, many healthcare systems in SSA are challenged to implement them. The purpose of this review is to synthesize and critique what is known about implementation strategies to promote IPC for nurses in SSA. Methods The databases, PubMed, Ovid/Medline, Embase, Cochrane, and CINHAL, were searched for articles with the following criteria: English language, peer-reviewed, published between 1998 and 2018, implemented in SSA, targeted nurses, and promoted IPC EBPs. Further, 6241 search results were produced and screened for eligibility to identify implementation strategies used to promote IPC for nurses in SSA. A total of 61 articles met the inclusion criteria for the final review. The articles were evaluated using the Joanna Briggs Institute’s (JBI) quality appraisal tools. Results were reported using PRISMA guidelines. Results Most studies were conducted in South Africa (n = 18, 30%), within the last 18 years (n = 41, 67%), and utilized a quasi-experimental design (n = 22, 36%). Few studies (n = 14, 23%) had sample populations comprising nurses only. The majority of studies focused on administrative precautions (n = 36, 59%). The most frequent implementation strategies reported were education (n = 59, 97%), quality management (n = 39, 64%), planning (n = 33, 54%), and restructure (n = 32, 53%). Penetration and feasibility were the most common outcomes measured for both EBPs and implementation strategies used to implement the EBPs. The most common MAStARI and MMAT scores were 5 (n = 19, 31%) and 50% (n = 3, 4.9%) respectively. Conclusions As infectious diseases, especially emerging and re-emerging infectious diseases, continue to challenge healthcare systems in SSA, nurses, the keystones to IPC practice, need to have a better understanding of which, in what combination, and in what context implementation strategies should be best utilized to ensure their safety and that of their patients. Based on the results of this review, it is clear that implementation of IPC EBPs in SSA requires additional research from an implementation science-specific perspective to promote IPC protocols for nurses in SSA.


Background
Infectious diseases generate significant morbidity and mortality worldwide [1,2]. Despite reductions in global prevalence over the last 20 years, a disproportionate amount of the disease burden related to infectious diseases remains in Sub-Saharan Africa (SSA) [3,4]. Despite impressive reductions in infectious disease burden within SSA [3,4], communicable illnesses (along with maternal, neonatal, and nutritional causes) comprised 61% of the disability-adjusted life year (DALY) burden in the region as of 2015 [5]. Half of the top ten causes of poor health or death in SSA are infectious diseases [2,3]. With rapid economic, social, and geographical shifts occurring in the region [6][7][8], emerging and re-emerging infectious diseases are playing an increasingly important role in the burden of disease. Recent outbreaks like the Ebola viral disease (EVD) epidemics of 2014 and 2018 have highlighted the impact of infectious diseases on healthcare systems and the communities they serve. In the context of already resource-challenged health care systems, the EVD epidemic of 2014 devastated the healthcare infrastructure of three Western African countries [9], and impacted four more [10]. By April 2016, 28, 616 confirmed EVD cases and 11,310 deaths were reported for the region [9]; the economic cost of the epidemic was estimated at $3.6 billion, with $2.2 billion in gross domestic product lost in Guinea, Liberia, and Sierra Leone in 2015 [9]. Similar expenditures can be observed for HIV services and care within the region: fiscal requirements for HIV therapy up to 2050 are projected to be as high as 21% and 80% of the GDPs of South Africa and Malawi respectively [11]. Currently, an EVD epidemic in the Democratic Republic of the Congo (DRC) is evolving into a significant public health endeavor, leading to the second largest EVD epidemic in history, with over 500 confirmed cases [12]. Infectious diseases and the damage they can cause to patients, healthcare workers, and health systems remain among the most pressing priorities to be addressed in SSA.
Within the broader category of infectious diseases, healthcare-acquired infections (HAIs) are a major challenge. HAI rates are generally higher in low-income compared to high-income countries [13], with substantial variation across and within countries of all income levels: the cumulative incidence of HAIs ranges from 5.7 to 48.5% within African countries [14]. Traditionally defined, HAIs are infections patients acquire while receiving care in a healthcare facility [15,16]. Yet, HAIs that impact healthcare workers providing patient care are equally important, especially nurses [17,18]. While many different types of health care workers (i.e., laboratory technicians, physicians, water and sanitation staff) are at increased risk of acquiring infectious diseases in the healthcare setting, this study focused on nurses for the following reasons: (a) nurses have unique needs (they spend the most amount of time with patients than any other health worker [19] and operate in highly unstandardized and variable circumstances); (b) nurses are by far the largest cadre of healthcare workers in SSA (even though their needs often take second place or are lumped to those of physicians or other healthcare workers).
Two diseases, EVD and tuberculosis (TB), provide excellent exemplars of infectious diseases that disproportionately affect nurses while caring for patients. A total of 718 healthcare worker EVD infections occurred in West Africa, with 396 (55%) confirmed cases among nurses [20]. A combined cumulative incidence rate of EVD among nurses was 43.7 per 1000 in the region, compared to 29.5 and 40.4 per 1000 among physicians and laboratory technicians respectively [20]. Similarly, high rates of TB infections are observed in healthcare workers [21]. The median incidence rate of TB among healthcare workers in SSA was 3871 per 100,000 [21], making the risk of contracting TB among healthcare workers, including nurses, in SSA greater than the risk in the general population in SSA [22]. In terms of HAIs, nurses are often unduly infected, leaving significant burdens on the health system.
Inadequate adherence to infection prevention and control (IPC) standards place millions of patients and healthcare workers, especially nurses, at risk of infectious diseases worldwide, including HAIs. "IPC is a scientific approach and practical solutions to prevent harm caused by infections to patients and health workers" [23]. Effective IPC knowledge and practice are the keystones of a strong healthcare system. The causes of high HAI rates include poor environmental hygiene, inappropriate medical waste disposal, inadequate infrastructure, insufficient equipment, and poor knowledge of infection control protocols all contribute to high HAIs [13].
For example, nurses in SSA may not have enough resources, like biohazard bins or waste bags, to adequately dispose of infectious medical materials [18]. Additionally, nurses may not have access to the following: an adequate healthcare infrastructure to provide safe patient care, familiarity with IPC policies or regulations within their healthcare facilities, and knowledge of effective screening and triage practices to minimize transmission of infectious diseases entering the health facility [18]. The causes of poor IPC places nurses at increased risk of acquiring an infectious disease while serving patients; however, HAIs among healthcare workers and patients are preventable.
The World Health Organization (WHO) [24] has identified a set of evidence-based recommendations on the key components of an IPC program for a national and facility level. These IPC core components include dedicated programs with teams of specialty trained IPC professionals, guidelines, training and education, surveillance, implementation of multi-modal IPC strategies, monitoring/auditing and providing feedback, establishing requirements for workload, staffing, and bed occupancy, and ensuring that the built environment, equipment, and materials are available for IPC practices [24]. These core components are the foundation for the two different branches of IPC precautions: standard precautions and transmission-based precautions [25]. Standard precautions are the basics of the IPC precautions. Used for all patients, this branch of precautions includes hand hygiene practices, appropriate use of personal protective equipment (PPE), respiratory hygiene, appropriate patient placement, injection safety, disinfection, and medical waste disposal [26]. When implemented correctly by health workers, these precautions keep the health worker protected from infection and keep infections from spreading among patients [26]. In addition to standard precautions, the second branch of IPC precautions are transmission-based precautions [26]. The three transmission-based precautions are contact, droplet, and airborne [26]. Contact precautions are used when patients are colonized with an infectious agent and the risk of further transmission is high [26]. For some infectious agents, specialized precautions called administrative precautions are used to further control the spread of infection. Administrative precautions focus on reducing the risk of exposure to patients infected with specific infectious diseases [27]. Administrative control activities include screening, diagnosing, and treating infectious agents [27]. For example, TB is an infectious disease that requires both administrative precautions and transmission-based precautions [27]. Rapidly screening TB suspects via intensive case finding expedites patient diagnosis, which allows for therapy to be expedited as well. Once TB patients are placed on effective therapy, they are no longer infectious to others [28][29][30].
With sufficient resources to advance health system strengthening initiatives, prevention of infectious diseases, including HAIs, is achievable. Within the global health context, many initiatives have used strategies to incorporate IPC evidence-based practices (EBPs), like standard, transmission-based, and administrative precautions, into standard healthcare practice. Many EBPs are known for many healthcare challenges [31]. EBPs, like hand hygiene, are effective interventions known to reduce infectious agents among patients and healthcare workers [32][33][34]. Unfortunately, EBPs, including those for IPC, are not effectively implemented in many low-and middle-income countries (LMICs) in SSA [35]. Within implementation science, a variety of implementation strategies have been used to integrate EBPs into clinical practice in LMICs. Proctor et al. [36] defines implementation strategies as "methods or techniques used to enhance the adoption, implementation, and sustainability of a clinical program or practice". In SSA, many implementation strategies have been used to promote IPC protocols [32,[37][38][39][40]. All of these strategies have produced outcomes associated with the original EBP or the strategy utilized. Additionally, implementation outcomes measure the degree in which implementation strategies have been successfully utilized. Implementation outcomes are "the effects of deliberate and purposive action to implement new treatments, practices, and services [41]. In SSA, implementation outcomes have been measured to assess if IPC EBPs have been successfully implemented.
The Conceptual Model of Implementation Research is a framework that outlines the relationships between an EBP, implementation strategies utilized to promote the EBP, and outcomes of the implementation strategies [42]. This framework provides the conceptual underpinnings for the primary research question of this review: how are implementation strategies used to support IPC promotion for nurses in SSA. Using this model, hand hygiene, waste disposal, and correct PPE use are examples of EBPs. Trainings and stakeholder buy-in sessions are examples of implementation strategies, and number of nurses trained and number of trainings conducted are examples of implementation outcomes. Given the burden of infectious diseases in SSA, the promotion of IPC protocols for healthcare workers is critically needed. Yet, limited literature exists on how implementation strategies have been used to advance IPC, for nurses, a commonly overlooked healthcare worker cohort. To address this gap, the purpose of this review is to synthesize and critique what is known about implementation strategies to promote IPC for nurses in SSA.

Search strategy
A systematic approach was used to identify articles from the following databases: PubMed, Ovid/MED-LINE, Embase, Cochrane, and CINHAL. PubMed and Ovid/MEDLINE were selected for their referencing of the biomedical literature. Embase was selected for its focus on international scholarship and a global audience. Cochrane and CINHAL were selected for their reporting on systematic reviews and nursing literature respectively. Two reviewers (AEBC, KAR) independently searched the databases using the search terms for nurses/ nursing; IPC, standard precautions or transmission-based precautions; and Sub-Saharan Africa or individual countries in the region. The complete search syntax for each database are included in Additional file 1. As per the recommendations of Whittemore and Knafl [43], the reference sections of each article were reviewed for additional studies meeting eligibility criteria: a methodology known as citation indexing. Figure 1 depicts the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram used to report study findings [44]. The protocol for this review was not registered.

Eligibility criteria
Inclusion and exclusion criteria were developed to identify research that empirically evaluated or tested implementation strategies to promote IPC protocols for nurses in SSA. Studies were eligible for inclusion if written in English, peer-reviewed, published between 1998 and 2018, implemented in SSA, targeted nurses, and promoted IPC EBPs. All study designs (RCT, crosssectional, cohort, qualitative) were eligible. Studies were excluded if they did not meet the aforementioned criteria. For example, non-empirical studies (reviews, commentaries, briefs, editorials, reports, guidelines) and studies that did not specifically evaluate or test an implementation strategy (e.g., prevalence studies, modeling studies) were excluded from this review.

Study selection and data extraction
Titles and abstracts identified in the initial search were de-duplicated and then screened using inclusion and exclusion criteria for full-text review. Citation-index searching was conducted on articles that met inclusion criteria following full-text review, and eligible articles were similarly screened. The following implementationrelated data elements were extracted from studies included in the review: author, year of publication, purpose, country, study design, IPC EBPs, implementation strategies, and implementation outcomes.
For each article, implementation strategies and outcomes were categorized based on the definitions produced by Powell et al. [45] and Proctor et al. [41] respectively. Further categorization of implementation outcomes based on their association with the EBPs or implementation strategies to promote the EBPs was conducted. For example, hand hygiene is an EBP. An educational training (implementation strategy) may be provided to nurses to promote the uptake of hand hygiene practices. The outcomes of this initiative may be related to the EBP and/or to the training. An example of a possible outcome related to the above scenario for the EBP is penetration (i.e., increased number of nurses practicing hand hygiene out of total number of nurses). An example of a possible outcome associated with the implementation strategy (training) is acceptability (i.e., 95% of nurses agreed the training material on hand hygiene was good, informative, and useful). Two reviewers (AEBC, KAR) screened all articles for inclusion/ exclusion. After all articles were screened and deemed to meet the inclusion criteria of the review, data extraction was then conducted by the primary reviewer (AEBC).

Quality screening
Each included article was assessed for quality using tools developed by the Johanna Briggs Institute (JBI). JBI focuses on the promotion of evidence-based practices through effective healthcare measures to a global community [46]. Implementation and translational science are key methodologies embraced and promoted by JBI [46], making these appraisal criteria particularly wellsuited to the purpose and scope of this review. JBI uses individual assessment tools for each type of study (randomized control trial, cohort, case-study, qualitative investigation, etc.) to assess quality. Study designs that had global cut-off criteria were assessed using that criteria. For studies where no global criteria existed, the reviewers established pre-determined cut-off scores before the appraisal process was initiated. Each JBI tool has specific questions addressing bias, confounding variables, statistical analyses, methodological validity, and outcome reliability [47]. Table 1 contains the quality appraisal and scoring criteria utilized in this review.
For quantitative studies, the JBI-Meta Analysis for Statistics Assessment and Review Instrument (MAStARI) was used. For qualitative studies, the JBI-Qualitative Assessment and Review Instrument (QARI) was used. For mixed methods studies, JBI does not have an appraisal tool. The mixed methods appraisal tool (MMAT) produced by Pluye et al. [49] at McGill University was used. This tool has specific quantitative, qualitative, and integration (mixed methods) questions to assess study quality. Two reviewers (AEBC, KAR) appraised all studies for quality, and an additional reviewer (JES) appraised a random sample of 20% (n = 20) of the articles. Any discrepancy in the appraisal process was discussed by all reviewers and consensus was reached.

Database and citation indexing search results
The initial database search produced 8400 results. Further, 2159 duplicates were removed. Of the 6241 studies that remained, 511 met the criteria for fulltext review. Forty-six studies met the inclusion criteria. Citation indexing yielded an additional 15 studies that met the inclusion criteria. A total of 61 studies have been included in this review (see Fig. 1 for PRISMA flow diagram) [32-34, 38-40, 50-104]. Table 2 provides a summary of the individual study characteristics of the studies included in this review. Table 3 provides a table of evidence of the studies in the review. The majority of the studies were conducted in South Africa (n = 18, 30%), within the last 18 years (n = 41, 67%), and utilized a quasi-experimental design (n = 22, 36%). After South Africa, the other countries where the majority of studies were conducted were Nigeria (n = 5, 8%), Kenya (n = 4, 7%), and Zambia (n = 4, 7%). The majority of studies in this review focused on HIV (n = 24, 39%), TB (n = 6, 10%), Ebola (n = 6, 10%), or did not focus on a specific disease or infection (n = 14, 23%). The nondisease studies focused on standard precautions.

Healthcare worker cadres
This review was conducted to investigate what is known about implementation strategies utilized to promote IPC protocols for nurses. Forty-seven (77%) studies included healthcare worker samples, including physicians, pharmacists, laboratory technicians, nurse aids, trained birth attendants, residents, and nurses. Fewer studies (n = 14, 23%) had sample populations comprising nurses only.

IPC evidence-based practices
A variety of EBPs were represented in this review. The majority of studies focused on administrative precautions (n = 36, 59%). Standard precautions and transmissionbased precautions represented forty-one (34%) and four (7%) studies respectively. Among the administrative precautions, treatment was the most frequently reported EBP (n = 21, 34%), followed by screening (n = 11, 18%) and diagnosis (n = 4, 7%). Studies focusing on HIV treatment or TB screening comprised the majority of studies in this section. Only four (6%) studies focused on transmission-based precautions. Each study reported on a unique precaution. Transmission-based studies focused on correct PPE use, appropriate patient placement, immunization, and post-exposure prophylaxis.           Studies that addressed standard precautions did so by incorporating multiple precautions (n = 11, 18%) or focused on hand hygiene only (n = 8, 13%). Other studies that focused on standard precautions addressed medical equipment disinfection or appropriate waste management.

Implementation strategies
The most frequent implementation strategies used to promote IPC protocols in included studies were education (n = 59, 97%), quality management (n = 39, 64%), planning (n = 33, 54%), and restructure (n = 32, 53%). A variety of educational strategies were used to promote IPC protocols for nurses. Strategies used included didactic lectures, simulations, on-site mentorship, visual reminders, and demonstrations. Quality management strategies generally consisted of audit and feedback sessions provided to nurses in real time to promote the uptake of an IPC EBP. Planning strategies consisted of collaborations, partnerships, or buy-in sessions that were established at a higher administrative level for nurse IPC involvement. Planning strategies were mostly utilized in conjunction with other strategies. For example, a partnership between the government of a country and an academic institution may be formalized to provide nurses training or mentorship on IPC protocols. Planning strategies were also used to inquire about nurses' experiences with IPC EBPs. A variety of restructure strategies were used to promote IPC. Nurses may be task-shifted to include IPC activities within their scope of work, provided additional resources (i.e., alcohol-based hand gel) to make adhering to IPC protocols easier, or promoted to a higher nursing position whereby IPC became the focus of the new role. Only eight (n = 8, 13%) studies used financing incentives to promote IPC protocols among nurses. When used, a financial strategy was generally associated with external funders providing resources to initiate an EBP intervention or providing over-time compensation to healthcare workers. For example, the Elizabeth Glaser Pediatric AIDS Foundation, the Axios Foundation, and the Boehringer Ingleheim Pharmaceutical Company provided funds to initiate an antiretroviral therapy (ART) initiation program for HIV patients in Cameroon [88]. Zero (0%) policy strategies were used.
An exhaustive list of implementation strategies is provided for each study in Table 4. Bolded strategies targeted nurses. In order to be included in this review, studies needed to report the use of implementation strategies for nurses. Many studies used implementation strategies that targeted patients, non-nurse healthcare workers, or aspects of the healthcare system. Few studies used implementation strategies solely for nurses. A summary of implementation strategies used for nurses and non-nurses (i.e., patients or a healthcare system-specific matter) is provided in Table 5. Most studies used education (n = 58, 95%) for nurses; a drastic contrast to the number of studies that used education (n = 1, 2%) for non-nurses. Other discordant results between strategies used for nurses compared to strategies used for non-nurses included planning, restructure, and finance. These strategies were used more for nonnurses than nurses. Twenty-one (34%), 25 (41%), and six (10%) studies used planning, restructure, and finance strategies respectively for non-nurses. Planning, restructure, and finance were used in 12 (20%), seven (12%), and two (3%) studies respectively for nurses.

Implementation outcomes
For this review, the outcomes of the studies have been analyzed into two parts: outcomes associated with the EBPs and outcomes related to the implementation strategies used to promote an IPC EBP.

Implementation outcomes for EBPs
All implementation outcomes, except cost, were measured in the studies of this review. Most studies reported penetration (n = 22, 37%) and feasibility (n = 13, 21%) as the most common outcomes produced (see Table 5). Four (7%) and 18 (30%) studies measured penetration for nurses and non-nurses respectively (see Table 5). For the non-nurse category, penetration was measured as patient agreement to initiate an EBP (i.e., screening for TB or antiretroviral therapy (ART) uptake). Feasibility was frequently measured as nurse report that EBPs could be implemented within their scope of work. Additionally, feasibility was used to report barriers to EBP implementation. For example, limited hospital infrastructure (i.e., no running water) was identified as a limitation to hand hygiene adherence among nurses.
Few studies measured adoption (n = 5, 8%) and sustainability (n = 5, 8%). Adoption was discussed as patient willingness to accept an EBP with nurse support, facilitylevel uptake of an EBP, or nurse report of healthcare improvements as a result of an implemented EBP. Sustainability was measured as nurses' ability to maintain an EBP for a few weeks or longer. Four (7%) studies discussed acceptability. Fidelity was also discussed in three (5%) studies. Fidelity was reported as adherence to an EBP by nurses. One (2%) study addressed appropriateness, and zero (0%) studies reported cost. Lastly, 21 (35%) studies did not measure any EBP outcome. In these studies, the outcomes of interest were related to the implementation strategies. Table 4 provides a complete list of implementation outcomes associated with EBPs for each study. Nurse-specific outcomes associated with EBPs are bolded in Table 4.      patients stated clinic-based methods were generally acceptable and not out of the ordinary, however, they want to ensure patients were not being stigmatized. Thus, they suggested that ALL patients (not just some patients, like HIV+) be screened; HCWs stated they favored incentive-based system best 2. FEASIBILITY: Clinic costs and transportation times were also listed as issues that might inhibit this strategy's successful implementation; Some TB patients stated they liked the home-based method better, as it allowed patients to feel more comfortable, less costs, but it did have the potential to create stigma; In terms of the material-based method, many participants stated food and money were good incentives, yet they questioned if the government would be able to implement this form of incentive system. Additional concerns related to this method were around if this method was sustainable. 21

Implementation outcomes for implementation strategies
The most common outcomes measured for implementation strategies used in this review were penetration and feasibility; a similar finding to the outcomes associated with EBPs. Forty-four (72%) studies reported penetration. Thirty-four (56%) and seven (12%) studies reported penetration outcomes for nurses and non-nurses respectively (see Table 5). For nurses, penetration was most frequently used to measure the number of nurses experiencing or undergoing a given implementation strategy (e.g., number of nurses trained in hand hygiene). For the non-nurse category, penetration was used to scale-up patient services or measure loss to follow-up (see Table 5). Feasibility was measured in 28 (46%) studies, with six (10%) and 22 (36%) studies measuring feasibility for nurses and non-nurses respectively. Feasibility measured if an IPC protocol could be successfully implemented for nurses. Four studies successfully implemented the WHO hand hygiene campaign to promote hand washing for healthcare workers [32,58,80,81], demonstrating that the intervention could be implemented. Feasibility also described barriers to the successful implementation of an implementation strategy or to the procurement of resources for IPC protocol implementation. Soap, alcohol-based hand gels, and towels were procured for nurses to eliminate barriers related to hand disinfection, making it feasible for nurses to implement the EBP. The third most common outcome reported was acceptability (n = 18, 30%). Acceptability was measured in 13 (21%) and five (8%) studies for nurses and non-nurses respectively (see Table 5). For nurses, this outcome measured nurses' opinions on the strategy and its utility. Five (8%), three (55%), and two (3%) studies addressed adoption, sustainability, and appropriateness respectively. Sustainability captured nurses' ability to sustain knowledge acquired during a training for a defined period of time. Appropriateness measured nurses' perceptions of the cultural relevance of educational materials produced for a training or the benefit of training to the nursing profession. Cost, measured in only two (3%) studies, captured the amount of funds provided from the development sector to ensure that implementation strategies were carried out. Zero (0%) studies measured fidelity to implementation strategies.

Quality appraisal
All studies (n = 61, 100%) included within this review met the pre-determined or global quality criteria. Only one study, which was a multi-modal study, had a score under the cut-off criteria for its quantitative section. The qualitative score for this study met criteria, and thus was included in the review, which adhered to the predetermined criteria for multi-model studies. Fifty-five studies (90%) were quantitative or multi-modal studies. The most common MAStARI score was 5 (n = 19, 31%). The structure of the MAStaRI scales favors experimental designs; five articles with MAStARI scores of 9 or higher were randomized controlled trials. Three (4.9%) studies were purely qualitative, both examining the feasibility of an EBP implementation strategy. Three (4.9%) studies were mixed methods, and both achieved MMAT rating of 50% or higher.

Discussion
Infectious diseases contribute significantly to patient and healthcare worker morbidity and mortality in SSA. Nurses who engage in direct patient care are at increased risk of nosocomial and other infections [22,105]. To address these HAI challenges, many IPC EBPs have been deployed at the healthcare facility level. Yet, obstacles within LMICs make it difficult to sustain these practices. Implementation science provides an opportunity to identify and execute strategies that may better sustain EBPs over time [106]. Based on the results of this review, it is clear that implementation of IPC EBPs in SSA requires additional research, practice, and funding. More research from an explicit implementation sciencespecific perspective is needed for the following reasons: (1) very few of the studies in this review were able to produce sustainable outcomes, a potential limitation of the strategies implemented. Yet, more research is needed to know if this is true or not. (2) All studies in this review, except for one, did not identify as an implementation study. Thus, the strategies were not measured using implementation science definitions for strategies. Therefore, increased implementation science research into which, what combination, and in what specific context implementation strategies should be used in SSA is important to advance the science and to fully reap the potential of IPC protocols to save the lives of patients and healthcare workers in vulnerable regions. Despite increased risk of contracting infectious diseases, nurses were underrepresented in the studies in this review. The majority of studies did not exclusively focus on nurses; they focused on non-nurse participants, including physicians, patients, auxiliary hospital staff, or some other aspect of the healthcare system, and nurses were included in the samples as generic healthcare workers. In studies where nurses were the target population, the numbers of nurses included in the study was often not reported or too low to draw generalizable conclusions. Given this finding, a shift is needed in global health research to properly determine what IPC EBP practices are best suited for nurse implementation at health facilities in SSA, especially as the tasks of nurses are different from other healthcare workers. Whereas there can be some lessons learned from other healthcare cohorts, the scope of practice for nurses, as it relates to IPC, is vastly different from other healthcare workers. While many different types of health care workers (including laboratory technicians and water and sanitation staff) are at increased risk of acquiring HAIs, this study focused on nurses for the following reasons: (a) nurses have unique needs (they spend the most amount of time with patients than any other health worker and operate in highly unstandardized and variable circumstances); (b) nurses are by far the largest cadre of the workers in SSA (even though their needs often take second place or are lumped to those of physicians). As infectious diseases evolve and become more prevalent, EBPS for IPC will become an increasingly important part of nurses' scope of practice. Thus, additional and better-quality research that targets the unique IPC EBP challenges of the different healthcare worker cohorts, especially nurses and their involvement in IPC efforts, are needed.
Standard and transmission-based precautions were underemphasized in this review; most studies focused on administrative precautions. Standard precautions represent standard care. Therefore, an IPC study without a standard precautions component overlooked a critical component of IPC. Transmission-based precautions are typically undertaken when a patient is already diseased and to prevent further disease. Very few studies focused on this specific set of IPC EBPs, despite the fact that so many diseases in SSA, like TB, require them. Based on the results of this review, transmission-based and standard precautions are not sufficiently studied within LMICs, leaving a significant gap in the IPC literature.
The implementation strategies identified in this review lacked diversity. Of the seven implementation strategies identified, the predominant method was nurse education, specifically training on an IPC protocol. Johnson et al. [107] found similar results in a systematic review assessing implementation strategies for cervical cancer in SSA. Rowe et al. [108] argue that training in conjunction with other strategies (i.e., group problem solving, management techniques, supervision, and sound infrastructures) could improve healthcare worker practices in LMICs. Implementation science, along with other disciplines, acknowledges that access to information alone is not sufficient to produce long-term change [109]. Education only strategies are not likely to produce sustainable IPC knowledge or practice change among nurses. In addition to a lack of diversity in implementation strategies, there were notable gaps: policy and finance strategies, which play an important role in overall implementation, were rarely utilized. Unfortunately, the degree to which these strategies influence each other and other strategies (i.e., quality management) is relatively unknown given gaps in implementation science literature.
Among the studies in this review, it was identified that implementation strategies for nurses compared to nonnurses are vastly different. Education and quality management strategies are predominately used for nurses, and planning and restructure strategies are most often used for non-nurses. Nurses are trained and undergo regular monitoring, whereas non-nurses, especially patients, benefit from restructured clinical services and planned healthcare activities. Additional scholarship is needed to affirm if these strategies are indeed the best for each population.
Studies in this review that utilized multi-faceted implementation strategies produced better outcomes (such as feasibility, acceptability, etc.) compared to their singlestrategy counterparts; a finding consistent with the current implementation science literature [110]. A comprehensive implementation approach, addressing potential challenges at each level of a healthcare system, yields stronger and more sustainable outcomes. Training, an educational strategy, and real-time audit/feedback, a quality management strategy, produced better outcomes when used together than alone [54]. More specifically, hand hygiene rates increased (8.0% to 21%, p < .001) when multiple strategies (quality management, education, planning) were used to promote this EBP, whereas hand hygiene rates declined (68.9% to 36.8%, p < .001) when only one strategy was used (quality management) [32,33]. Multi-faceted strategies also generally focus on more than one stakeholder. Many of the studies in this review that used multi-faceted strategies targeted patients and healthcare workers. Single strategies focused on patient outcomes, perpetuating the underrepresentation of nurses, and overlooking nurse outcomes. Restructure is an excellent example of single strategy utilization. Most restructure strategies involved taskshifting nurse duties; however, this strategy is geared toward patient care and the impact on patient care, not the impact on the nurse practice. To advance implementation research within the global health context, multifaceted strategies need to be promoted and better metrics need to be developed to capture outcomes related to multi-faceted approaches [111].
Many studies only reported on implementation strategy outcomes, not EBP outcomes. Two possible explanations for this finding are as follows: (1) most EBPs are already accepted as effective practices that do not require additional validation, and (2) measuring EBPs is challenging in a LMIC. Collecting data on an implemented strategy is fairly easy. Collecting and verifying EBP data is a significant undertaking in resourceconstrained settings. However, implementation science research conducted in SSA must focus on some of the lesser reported outcomes, like cost or sustainability. Although these outcomes are more challenging to measure and require more time and funding to capture, implementation science studies that do not report on these outcomes have missed an opportunity to inform healthcare practices and policies at a macro-level. Implementation outcomes measured for nurses compared to non-nurses are distinct. More EBP outcomes were measured for non-nurses, and more strategy outcomes were measured for nurses. The results of this review suggest that the relationship between EBP outcomes and non-nurses, specifically patients, may be related to how an EBP is defined. EBPs are "the conscientious, explicit and judicious use of current best evidence in making decisions about the care of the individual patient." [112]. Patients, not nurses, are at the center of EBPs, which may explain the observed relationship between EBP outcomes and non-nurses. Educating patients is at the heart of the nursing profession; however, nurses need access to specific IPC EBPs information that is unique to them. Within global health, a common reality is that there are too few nurses to provide patient care. Investing in nurses and their safety must start to be a priority of the global community. If sources of inertia around nurse investment are not addressed within LMICs, then entire global communities will continue to suffer poor health as a result.
Studies with similar implementation strategies often used similar outcomes to measure impact. For example, the implementation of education interventions was often measured using an indicator of penetration (e.g., a training is implemented and researchers measured number of nurses trained). Similarly, planning strategies measured stakeholder response to an intervention. These outcomes were often not informative. A potential reason for this limited outcome data is related to the funding mechanisms of global health research and the types of outcomes that donors prioritize. The development sector is primarily interested in penetration outcomes (i.e., counts or proportions of nurses trained). In LMICs, purchasing resources for a project or building a new health facility is relatively achievable. It is much harder to ensure that resources and expertise are being used appropriately, health facilities have an adequate number of staff, and that overall operations of these facilities are sustainable. The "collect and measure what is easiest" notion is not going to improve the promotion of IPC EPBs in LMICs. Different outcome-strategy combinations need to be researched, allowing for increased growth of the implementation science discipline and increased IPC innovation in LMICs.
This review is not without limitations. The overall quality of the studies in this review was highly variable. Many studies were inadequately designed to capture outcomes of interest, biased, or were neither generalizable nor reproducible. Many studies only reported descriptive statistics. Many studies poorly described their methodologies. Additionally, most of the studies in this review were only just above the cut-off criteria for inclusion.

Conclusion
This review provides evidence that additional research from an implementation science-specific perspective is needed to promote IPC protocols for nurses in SSA. While many of the studies included in this review did not frame their methods or outcomes as implementationfocused, they indeed were. Implementation science should be fostered and promoted in global health. While many IPC EBPs are well-known, the best strategies to successfully implement them remains undetermined for LMICs. Implementation science has a unique opportunity to expand scholarship around best implementation practices in SSA. For example, a better understanding of which strategies are best suited for specific contexts or phenomena in SSA, like an infectious disease epidemic, to promote IPC protocols is needed. In short, additional implementation science research could greatly contribute to the limited knowledge base around which strategies should be utilized at the onset and during an infectious disease epidemic to promote IPC protocols and keep nurses from becoming patients during these emergency situations. Furthermore, implementation science-specific research could also alter the way scholarship in SSA is prioritized and funded, especially as it pertains to development mechanisms. Given the region's reliance on foreign assistance to cover healthcare expenditures [113], implementation science, with an emphasis on cost and policy, could identify investments for SSA and its healthcare workers.
Prevention is a critical component of infection control. The cost of IPC inaction is large. For the countries most affected by Ebola in 2014 (Sierra Leone, Liberia, and Guinea), economic modeling projections for postepidemic healthcare strengthening initiatives estimate the cost of reconstruction to be US$877 million in 2018 [114]. These funds are necessary for infrastructure and service delivery changes, including IPC measures. With a stronger health system able to withstand a future epidemic, the loss of life, particularly of nurses, during the 2014 outbreak may have been avoided. With emerging infections currently challenging the global community, like the EVD epidemic in the DRC and Marburg virus identified in Sierra Leone for the first time [115], neglecting IPC prevention strategies, especially for nurses, is unaffordable.

Supplementary information
Supplementary information accompanies this paper at https://doi.org/10. 1186/s13012-019-0958-3. Authors' contributions AEBC, KAR, JES, and AB made substantial contributions to this systematic review. AEBC and KAR completed the database search, screened all studies for review inclusion, and appraised all studies for quality using the JBI framework. A third reviewer (JES) also appraised a random sample of 20% (n = 20) of the total articles in the review. Data extraction was conducted by the primary reviewer (AEBC) for all articles. All authors contributed to the writing and revising of the final manuscript. All authors read and approved the final manuscript.

Funding
At the time of this review, AEBC was supported by the Hillman Scholars Program in Nursing Innovation Predoctoral Fellowship; KAR was supported by the Robert Wood Johnson Foundation Predoctoral Fellowship.
Availability of data and materials All data generated, produced, or analyzed for this study are included in this published article (including within supplementary files).
Ethics approval and consent to participate Not applicable.