GLOBAL HEALTH ACTION

2023, VOL. 16, 2206684

https://doi.org/10.1080/16549716.2023.2206684

REVIEW ARTICLE

Role of actor networks in primary health care implementation in low- and

middle-income countries: a scoping review
Dominic Dormenyo Gadeka a, Patricia Akweongoa, Eleanor Whyleb, Genevieve Cecilia Aryeeteya,
Justice Moses Ahetoc and Lucy Gilsonb,d
a
Department of Health Policy, Planning and Management, University of Ghana School of Public Health, Legon-Accra, Ghana; bDivision of
Health Policy and Systems, School of Public Health and Family Medicine, University of Cape Town, Cape Town, South Africa;
c
Department of Biostatistics, University of Ghana School of Public Health, Legon-Accra, Ghana; dDepartment of Global Health and
Development, London School of Hygiene and Tropical Medicine, London, UK

ABSTRACT ARTICLE HISTORY

Background: Primary health care (PHC) improvement is often undermined by implementa­ Received 2 February 2023
tion gaps in low- and middle-income countries (LMICs). The influence that actor networks Accepted 21 April 2023
might have on the implementation has received little attention up to this point. RESPONSIBLE EDITOR
Objective: This study sought to offer insights about actor networks and how they support Jennifer Stewart Williams
PHC implementation in LMICs.
Methods: We reviewed primary studies that utilised social network analysis (SNA) to deter­ KEYWORDS
mine actor networks and their influence on aspects of PHC in LMICs following the five-stage Policy implementation;
primary health care; actor
scoping review methodological framework by Arksey and O’Malley. Narrative synthesis was
networks; social network
applied to describe the included studies and the results. analysis; health policy
Results: Thirteen primary studies were found eligible for this review. Ten network types were analysis
identified from the included papers across different contexts and actors: professional advice
networks, peer networks, support/supervisory networks, friendship networks, referral net­
works, community health committee (CHC) networks, inter-sectoral collaboration networks,
partnership networks, communications networks, and inter-organisational network. The net­
works were found to support PHC implementation at patient/household or community-level,
health facility-level and multi-partner networks that work across levels. The study demon­
strates that: (1) patient/household or community-level networks promote early health-
seeking, continuity of care and inclusiveness by enabling network members (actors) the
support that ensures access to PHC services, (2) health facility-level networks enable colla­
boration among PHC staff and also ensure the building of social capital that enhances
accountability and access to community health services, and (3) multi-partner networks
that work across levels promote implementation by facilitating information and resource
sharing, high professional trust and effective communication among actors.
Conclusion: This body of literature reviewed suggests that, actor networks exist across
different levels and that they make a difference in PHC implementation. Social Network
Analysis may be a useful approach to health policy analysis (HPA) on implementation.

Introduction quality Universal Health Coverage (UHC) and the

Primary health care (PHC) improvement is often health-related Sustainable Development Goals
undermined by implementation gaps in low- and (SDGs) [4–7]. However, PHC implementation in
middle-income countries (LMICs). Pursuant to the LMICs remains poor [4]. Efforts to address the gap
Alma Ata declaration on PHC, which enjoined often focus on new ways of delivering services or
nations to make health care accessible, affordable, extending coverage [8,9].
and situated in the cultural context of the people Bottom-up theory suggests that networks play an
[1], LMICs adopted bottom-up approaches towards important role in policy implementation [10,11]. This
achieving the PHC goal [2,3]. Experience since the body of theory emphasises the discretionary power
declaration has shown PHC as an important strategy and critical influence of ‘policy implementors’ over
in improving both population health and in making policy implementation, even as top-down implemen­
healthcare systems more effective, responsive and tation is the norm of practice in many settings [12].
efficient [4]. Evidence further shows PHC as a cost- For instance, bottom-up theorists Hanf and collea­
effective strategy and an essential tool for achieving gues studied the goals of actors in an intervention

CONTACT Dominic Dormenyo Gadeka dgadeka@gmail.com; dgadeka@st.ug.edu.gh Department of Health Policy Planning and Management,
University of Ghana School of Public Health, P. O. Box LG13, Legon, Accra, Ghana
© 2023 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the
posting of the Accepted Manuscript in a repository by the author(s) or with their consent.
2 D. D. GADEKA ET AL.

delivery process, and concluded that policy imple­ In accordance with the standard approach to con­
mentation through networks (such as networks of ducting scoping reviews, a quality appraisal was not
frontline service providers) are more successful than performed. However, the Preferred Reporting Item
implementation carried out through a top-down for Systematic Reviews and Meta-Analyses extension
approach [10]. Additionally, Elmore [13] emphasises for Scoping Reviews (PRISMA-ScR) [17] criteria was
that bottom-up implementors possess widely dis­ used to guide the conduct and reporting of the
persed informal power that are critical for problem- review. Ethical approval or patient consent was not
solving and which promote success or otherwise of required.
a policy implementation. However, the influence that
actor networks might have on implementation has
Identifying the research question
received little attention up to this point.
Bottom-up theory further argues that in order to The specific scoping review question is ‘What role
strengthen implementation, the goals, strategies, activ­ do actor networks play in PHC implementation
ities, and contacts of actors involved in the implemen­ and how do networks add to our understanding
tation process must be understood. Furthermore, of implementation processes in low-and-middle-
earlier work that mapped existing body of health policy income countries’?
implementation research in lower-income settings
specifically highlights the need to consider the nature
and role of actor networks in understanding why and Identifying relevant studies
how implementation varies across policy types [14]. We attempted to identify all relevant scholarly pri­
However, implementation research in LMICs is scarce, mary research studies conducted in LMICs and
and actor-network-focused research is rare. reported in the English language and with no date
Understanding how networks contribute to policy limits. We did not include unpublished studies in
implementation, particularly, PHC in LMICs may order to ensure quality of the findings of the review.
help to strengthen its implementation. Strengthening The date limit was left open to enable the collection
PHC implementation requires working through many of all relevant articles to ensure richness of data to
sets of actors, but it is not clear whether, which and answer the research question. Electronic searches of
how networks among implementors support imple­ public health, social science, and medical peer-
mentation. In recent years, Social Network Analysis reviewed journals were carried out in PubMed,
(SNA) has begun to be applied in health policy analysis Sociological Abstracts, Social Science Research
(HPA) work. Such analysis can deepen understanding Network (SSRN), and PsychINFO databases using
of who forms networks and how networks function. logical operator-based combinations of key terms.
These insights could be useful in understanding how The initial search used a broad search strategy invol­
networks influence policy change, including PHC ving free-text terms, synonyms and subject headings
implementation. relating to actor networks, social networks and PHC.
This review offers insights about actor networks The search strategy consisted of the main terms, actor
and how they support PHC implementation in networks, social networks, social network analysis,
LMICs. It further provides evidence on the usefulness while the subject headings consisted of role of social
of SNA in HPA in understanding how networks networks in PHC, role of actor networks in PHC
support policy implementation and research gaps. implementation and influence of social networks on
PHC in low- and middle-income countries. Finally,
a general search combining the terms ‘social network’
Methods
or ‘social network analysis’ or ‘actor networks’ and
The study adopted a scoping review approach to ‘primary health care’ and ‘community health work­
understand the roles actor networks play in PHC ers,’ ‘community health committees,’ ‘health promo­
implementation, specifically in LMICs and to con­ tion,’ ‘disease prevention,’ ‘rehabilitation,’ ‘palliative
sider the implications for policy and practice. Our care,’ ‘health financing’ ‘health service provision’
study therefore focused on all sources of information ‘health workers’ health information” ‘medicines and
from full range of literature available on actor net­ technology’ ‘health systems governance and leader­
works in PHC implementation in LMICs. The study ship’ or ‘low- and middle-income country,’ was com­
followed the five-stage scoping review methodological pleted. We did not use the term ‘implementation’ in
framework by Arksey and O’Malley [15] as updated our search strategy because most authors who utilised
by Levac, Colquhoun, and O’Brien [16]: (1) identify­ SNA as an approach already focused on implementa­
ing the research question (2) identifying relevant tion. Reference lists of retrieved primary articles were
studies (3) study selection (4) charting the data and checked for further potentially relevant studies that
(5) collating, summarising and reporting the results. may have been missed by the electronic search. The
 GLOBAL HEALTH ACTION 3

website of the International Network for Social The networks were then classified into types based on
Network Analysis (www.insna.org), including linked the definitions drawn from the reviewed papers for
sites and contents were hand-searched. We further each of the identified networks. Through an inductive
hand-searched the journals Social Networks and process, we extracted, coded and categorised evidence
Implementation Science to identify additional studies. from the included articles to identify the roles played
by the different network types. To understand how
the networks support PHC implementation, we
Study selection
further categorised the evidence from the included
To be eligible for the review, primary studies had to articles based on the three inter-related and synergis­
report on the relationship between networks (actor tic components of PHC: (1) individual, households/
networks or social networks) and any aspect of PHC families and community empowerment for increased
implementation in LMICs. The study selection was social participation and enhanced self-care and self-
an iterative process. It involved searching the litera­ reliance in health (2) comprehensive integrated
ture, refining the search strategy and reviewing arti­ health services that embrace primary care as public
cles for inclusion [16]. To be included, studies had to health goods and (3) multi-sectoral policies and
report use of SNA in the design of the study, for actions to address wider determinants of health [7].
example, ‘net-map’, social network mapping, assess­
ment of network structure and properties. Limiting
Results
the study to papers using SNA ensured that studies
formally considered networks as the central elements In this section, we will first present the results on the
of their analysis. We included quantitative, qualitative extent, range, and nature of research activity on actor
and mixed-method studies. All studies that met the networks in PHC implementation in LMICs, then
inclusion criteria were exported to Mendeley, with discuss the various types of actor networks identified
duplicates identified and removed. in the literature and then the different roles played by
Studies were excluded if social networks were the different network types and how they support
mentioned but the type of analysis was only descrip­ PHC implementation in the following sections.
tive, that is, without the determination of the role or Figure 1 presents a flow diagram of the selection
influence of the social networks on PHC. Also, stu­ process including the total studies identified, the
dies were excluded if the type of analysis was not excluded studies, reasons for exclusion and the final
reported although the study involved a social net­ included studies. A total of 923 potentially relevant
work. Additionally, social network studies published studies were identified from searches of electronic
in languages other than English language were databases and review article references after the
excluded. Furthermore, where studies reported the removal of duplicates. After title and abstract screen­
same data in multiple papers, only one paper was ing, 836 were excluded because the studies only men­
included to avoid duplication. tioned social network analysis without applying it to
their work. The remaining 87 full-text articles were
retrieved and assessed for eligibility. We noted that
Charting the data
most of the studies identified were descriptive in
We developed a data charting form in Microsoft nature and do not systematically examine the rela­
Excel. The data extracted included study title, study tionships that exist between networks and PHC.
objective, authors, year of publication, the country, Based on this, we excluded 74 out of the 87 articles
study design/methods, study population, networks for only providing a descriptive view of SNA, thus
reported and key findings that related to the review without the determination of the role or influence of
question. The first author performed the data chart­ social networks on PHC. The remaining 13 studies
ing and was reviewed and doubled checked by all the were considered eligible for this review.
other authors. We resolved all disagreements through
discussions until reaching consensus. We discussed
Extent, range, and nature of research activity on
the results and continuously updated the data chart­
actor networks in PHC implementation in LMICs
ing form in an iterative process.
Table 1 presents the full list of the included studies
in this review. The different papers examined dif­
Collating, summarising, and reporting the results
ferent networks, with each paper primarily exam­
We applied a narrative synthesis to present the results ining one type of network. Together, the articles
of the included studies following recommendation by reported on PHC implementation using SNA from
Levac et al. [16]. We collated and compared defini­ nine different countries. Most studies were
tions of network(s) provided by the authors in each reported from Tanzania (n = 3) [18–20], followed
paper to observe any commonalities or differences. by South Africa (n = 2) [21,22] and India (n = 2)
4 D. D. GADEKA ET AL.

Records identified through Additional records

searching multiple databases identified through

Identification
(n= 1566) other sources (n= 19)

Records after duplicates removed (n=

923)
Screening

Records screened (n= 923)

Records excluded (n= 836)
Eligibility

Full-text articles were

Full-text articles to be assessed for
excluded (n=74)
eligibility (n = 87)
Analysis was only descriptive:55
Type of analysis not reported: 8
Text of article in languages other
Included

than English language: 11

Studies included in synthesis (n = 13)

Figure 1. Flow diagram of study selection process.

[23,24]. The articles were largely recent with pub­ delivery [27]; organisational infrastructure for ser­
lication dates ranging from 2014 to 2020. The vice delivery [29] and communication between dis­
highest number of publications were seen in 2019 ease programmes and district managers [22].
[21,23–26] and 2020 [18,19,27,28]. This indicates
the recent adoption of SNA in PHC research in
LMICs. Most of the studies (n = 11) applied cross-
Actor networks identified to support PHC
sectional design [18,19,21–23,25–30]. All the stu­
implementation in LMICs
dies applied a quantitative social network data col­
lection method, while six of these applied a mixed Ten types of actor networks across different con­
methods approach [22,23,25–28]. The qualitative texts and actors were identified in the pool of
approaches used ranged from in-depth interviews, papers we reviewed (Table 2). The network types
focus group discussions, observation and document include professional advice networks [28], peer
review. One article reports on the pre-test post-test networks [19,20], support/supervisory networks
social network analysis approach [24], while [21], friendship networks [30], referral networks
another was conducted as part of a cluster- [18,24], community health committee (CHC) net­
randomised HIV prevention trial [20]. Study popu­ works [26], inter-sectoral collaboration networks
lations were diverse, ranging from community [23,25], partnership networks [27], communica­
members [19,20,24,26,30] to district health man­ tions networks [22] and inter-organisational net­
agers [22,29] and actors beyond the district level works [29]. The definitions for each of the
[25,27] who influence PHC implementation. There network types are provided in Table 2. We also
were considerable variations in the areas of PHC provided the descriptions drawn from the
addressed in each set of articles. These include reviewed papers for each of the identified net­
supervisory relationships of community health works. Different terms were used to denote the
workers in PHC [21]; health behaviours and self- meanings of the different network types based on
reported health [30] patient referrals [18]; inter- the different contexts and the network goal. For
sectoral collaboration for people-centred mental instance, friendship [30] and peer networks [19]
health care [25]; community health committees in were reported to represent two different network
rural and urban settings [26]; antenatal care utili­ types by the different authors based on the goal of
sation [19]; HIV risk behaviour and normative the network (Table 2). Similarly, referral networks,
beliefs [20]; co-ordination between women’s self- which were examined in two of the reviewed arti­
help groups and local health systems [24]; profes­ cles [18,24] were reported with different meanings
sional advice for PHC workers [28]; vaccine based on the different contexts.
 GLOBAL HEALTH ACTION 5

Table 1. Overview of the included papers.

Authors (Year) Study title Country Method/Study design Study Population
Assegaai & Schneider The supervisory relationships of community South Africa Cross-sectional study Community Health Workers (CHWs),
(2019) health workers in primary health care: design with ward-based outreach team (WBOT),
social network analysis of ward- based quantitative social PHC facility managers and local
outreach teams in Ngaka Modiri Molema network analysis area managers
District, South Africa (SNA) approach
Chami, Ahnert, Voors, Social Network Analysis Predicts Health Liberia Cross-sectional study Households in highly remote, post-
& Kontoleon, Behaviours and Self-Reported Health in design with conflict villages in Liberia
(2014) African Villages quantitative
household survey SNA
approach
Francetic, Tediosi, A network analysis of patient referrals in Tanzania Cross-sectional study Public health facilities in the districts
& Kuwawenaruwa two district health systems in Tanzania design with
(2020) quantitative facility-
based SNA approach
Hall et al. (2019) Intersectoral collaboration for Timor-Leste Cross-sectional study Key stakeholders from government,
people-centred mental health care in design with a mixed NGO, civil society and other
Timor-Leste: a mixed-methods study method SNA organisations working in national
using qualitative and social network approach mental health and social care
analysis
Hoe et al. (2019) Using social network analysis to plan, India Cross-sectional study Key informants from organisations
promote and monitor intersectoral design with mixed working on MCH and/or WASH in
collaboration for health in rural India method SNA Kachhauna, Uttar Pradesh, India.
approach
Karuga et al. (2019) “It’s like these CHCs don’t exist, are they Kenya Cross-sectional study Community discussants and health
featured anywhere?”: Social network design with a mixed professionals in a rural area and an
analysis of community health method SNA urban slum
committees in a rural and urban setting approach
in Kenya
Mukong & Burns Social networks and antenatal care Tanzania Cross-sectional study Women aged 15–49 years.
(2020) utilisation in Tanzania design with
quantitative SNA
approach
Mulawa et al. (2016) Evidence of social network influence on Tanzania Cluster-randomised HIV Urban social networks locally referred
multiple HIV risk behaviors and prevention trial with to as “camps”.
normative beliefs among young quantitative SNA
Tanzanian men approach
Ruducha et al., (2019) Measuring coordination between women’s India Pretest, post-test design Self-help groups (SHG) members,
self-help groups and local health systems with quantitative SNA village-level and block-level
in rural India: A social network analysis approach government health workers, and
other key members of the
community
Sabot et al., (2020) Professional advice for primary healthcare Ethiopia Observational, cross- Staff at primary healthcare units
workers in Ethiopia: A social network sectional study design
analysis with quantitative SNA
approach
Soi et al., (2020) Global health systems partnerships: Mozambique Cross-sectional study Frontline Ministry of Health workers,
A mixed method analysis of design with a mixed- Ministry of Education staff and
Mozambique’s HPV vaccine delivery method SNA supporting partner organisation
network actors approach members
Ssengooba, Kawooya, Application of social network analysis in the Uganda Cross-sectional study District Health Offices (DHOs) and
Namakula, & assessment of organisation infrastructure design with a mixed service provider organisations
Fustukian, (2017) for service delivery: A three district case method SNA (SPOs)
study from post-conflict northern approach
Uganda
Kawonga, Blaauw, & Exploring the use of social network analysis South Africa Cross-sectional study Health managers located at
Fonn, (2015) to measure communication between design with subnational level
disease programme and district quantitative and
managers at sub-national level in South qualitative SNA
Africa approach

How networks support PHC implementation in that support health care providers) [18,21,24,28]; the
LMICs review also revealed multi-partner networks that work
across levels [22,23,25,27,29].
We identified that although the reviewed articles
focused on specific networks, they also provide evi­
dence on the wider context within which the
networks were based. In line with the three inter- Patient/household/community-level networks
related and synergistic components of PHC [7], the The study demonstrated that patient/household or com­
review showed (Tables 3 and 4) that PHC-linked net­ munity-level networks promote early health-seeking,
works are engaged at the patient/household/commu­ continuity of care and inclusiveness by enabling network
nity level [19,20,26,30], health facility-level (networks members (actors) the support that ensures access to PHC
6 D. D. GADEKA ET AL.

Table 2. Actor networks and their definitions.

Network type Network definition Network description from reviewed articles
Friendship Close individuals who comfortably help each other on a given Close friends who do not live in the same household but are
network issue comfortable to either turn to each other for advice, interest-
free loan, or ask for help with harvest without paying [30].
Referral network Relations between two or more health facilities involved in the Networks where each referral represents a directional tie
sharing of patients from lower to higher levels of care between a pair of health facilities [22].
Inter-sectoral Sets of organisations from different sectors which interact to Any planning, information and resource sharing for a common
collaboration achieve a particular objective purpose between organisations from different sectors and/or
network across thematic areas [24,29].
Partnership Sets of actors or agencies that formally or informally engage for A collaboration with the mission of accomplishing a common
network a common purpose goal either contractually or non-contractually [27].
Community Group of individuals who serve as liaisons between community A group of community-based committees who provide and
health members and frontline health workers in the provision of received health-related information and are able to
committee health services to the community participate in the exchange of information and decision-
networks making on community health services [26].
Support/ Two or more actors who provide support to one another based A set of relationships embedded in the wider context of social
supervisory on hierarchies of the health system and professional relationships and hierarchies within the
networks health system [20].
Peer networks Group or individuals with similar sociodemographic Semiformal groups who are either friends, acquaintances, or
categorisation and who get along either formally, informally people who get along and/or socialise regularly [19,20].
or both and interact on a given issue
Inter-organi Network formed by two or more organisations with the purpose Relational architecture or networks among organisations to
sational of achieving a common goal strengthening the health workforce in post-conflict northern
network Uganda or any linkage [support/activity/engagement]
between the respondent organisation and another external
organisation for strengthening health workforce [29.]
Communications Communication relationship that facilitates action or decision- Communications involving managers through one-on-one task-
networks making among individuals or organisations related communication or communication through co-
participating in management committees [22].
Professional Individuals or group of healthcare professionals who share Healthcare professionals, either having provided or sought
advice advice on a common issue of interest advice for a common objective or policy of interest [28].
networks

services (Table 4). The synthesis showed that patient/ Health facility-level networks
household or community-level networks vary consider­
In relation to health facility-level networks, the review
ably in their specific application and level of support for
showed that these networks enable collaboration
PHC (Tables 3 and 4). Four patient/household or com­
among PHC staff and also ensure the building of
munity-level network types were identified in the
social capital that enhances accountability and access
reviewed articles [19,20,26,30]. Two were peer networks
among health care users [19,20] while the rest were to community health services. The reviewed articles
friendship [30] and community health committee reported four facility-level networks located at the
(CHC) networks [26]. With regard to the peer networks, district level of care (Tables 3 and 4). These include
it was noted that pregnant women leveraged their con­ professional advice networks [28], support/supervi­
tacts and sources of information externalities regarding sory networks [21] and patient referral networks
antenatal care utilisation to increase their probability of [18,24]. In the study by Sabot et al., [28]) where
early antenatal check-up and antenatal completion [19]. existing professional advice networks among health­
Similarly, young Tanzanian men used their peer net­ care workers in PHC units were examined, it was
works to support individuals’ engagement in preventative found that professional advice networks provide sup­
health in the prevention of multiple HIV risk behaviours port for improvements in health provider practices
and normative beliefs [20]. than when health care workers are only given training
In relation to friendship networks, it was noted that (example, in-service training) without the involve­
they enable individuals at the community levels the ability ment of their networks. These networks were further
to create opportunities to take preventive health actions seen to provide support for healthcare professionals
and thereby promote effective uptake of health interven­ because they share expert knowledge on PHC issues
tions in rural poor villages [30]. Similarly, in Karuga et al. and other resources by leveraging the capabilities of
[26] community health committee networks were seen to network members.
enhance community participation in health services. Similarly, it was observed that support/supervisory
These networks (community health committee net­ networks that draw on sympathetic cadres among PHC
works) were further observed to use their level of influ­ facility staff for support among ward-based outreach
ence and resourcefulness in serving as back-ups to teams (WBOTs) in a rural South African district pro­
government in addressing PHC challenges at the com­ mote teamwork, problem-solving and participatory
munity levels. The commonality among these networks is decision-making, and strengthen interpersonal rela­
that they support information sharing at the local/com­ tions among PHC providers [21]. The WBOTs are the
munity level, enabling easy reach to communities. teams responsible for the provision of preventive and
 GLOBAL HEALTH ACTION 7

Table 3. Findings on networks among implementors.

Identified actor
networks Authors (Year) Study objective Key findings
Professional Sabot et al., (2020) To compare professional advice networks of healthcare The study established that primary health care unit
advice workers in eight primary healthcare units across four staff involved in the delivery of maternal and
networks regions of Ethiopia newborn health services have informal advice
networks outside of supervisory structures. These
advice networks of the healthcare workers in PHC
units were observed to provide better change in
health provider practices compared to when
health care workers we-re only given training
(example, in-service training) without the
involvement of their networks.
Peer networks Mukong & Burns, To examine the impact of information externalities Peer networks were seen to increase the probability
(2020) generated through network membership on of early antenatal check-up and antenatal
antenatal care utilisation in Tanzania completion. The study further found that relying
on strong peer network increases the probability
of antenatal completion by 23% and early
antenatal initiation by 48%.
Mulawa et al., The study sought to evaluate the effectiveness of It was observed that peer networks of young men
(2016) a camp-randomised Microfinance and health explained between 5.78 and 7.17% of variance in
leadership intervention on sexually transmitted men’s normative beliefs and between 1.93 and
infections, gender-based violence and HIV risk 15.79% of variance in HIV risk behaviors
behaviors
Support/ Assegaai & To identify critical actors and patterns of relationships Support/supervisory networks were found to enable
supervisory Schneider, in the supervision of ward-based outreach teams the WBOTs in drawing on sympathetic cadres
networks (2019) (WBOT) in a rural South African district identified among the PHC facility staff for support.
Additionally, supervisory relationships within
networks (teams) were observed to function
better than those between teams and the rest of
the PHC system
Friendship Chami, Ahnert, To predict health outcomes and to explain the The findings show that friendship network helps in
networks Voors, & susceptibility of a household to partake in health determining the number of ill family members,
Kontoleon, interventions in impoverished and under-researched whereas the closeness of the members of the
(2014) developing country contexts friendship network determines the individuals’
engagement in preventative health. Also was
found to influence the individuals’ self-reported
health status. The study further shows that the
closeness between members in a friendship
network predicts susceptibility to, instead of
influence over, good health behaviours
Referral Francetic, Tediosi, & To explore patient referral networks in two rural It was shown that referral networks facilitate the
networks Kuwawenaruwa, districts in Tanzania, Kilolo and sharing of patients among health facilities
(2020) Msalala
Ruducha et al., To assess how the health coordination and emergency It was observed that the referral networks expanded
(2019) referral networks between women’s self-help groups relationships at the village level. It was
(SHGs) and local health systems have changed over established that the new relationships between
the course of a 2-year learning phase of the Uttar the traditionally under-represented communities
Pradesh Community Mobilisation Project, and local government could serve as vehicle for
India. building social capital that could lead to a more
accountable and accessible community health
delivery system
Community Karuga et al., To explore the structure of a rural and an urban CHC networks served as back-ups to government in
Health (2019) Community Health Committee (CHC) network and addressing primary health care challenges by
Committee to analyze how health-related information flowed in supporting Chiefs to implement government
(CHC) these networks directives, such as compulsory child immunisation
networks and discouraging banned cultural practices.
Inter-sectoral Hall et al., (2019) To investigate intersectoral Intersectoral collaboration networks facilitate
collaboration collaboration for people-centred mental health care in information and resource sharing among
networks Timor-Leste, a South-East Asian country in the organisations working within the health and
process of strengthening its mental health system social (disability and violence support) sectors in
Timor-Leste even in resource restricted settings. It
further revealed that contrary to the assumption
that mental health services and system
strengthening are led by the Ministry of Health,
intersectoral collaboration networks resulted in
split in stewardship for mental health between
subnetworks in the health and social sectors
Hoe et al., (2019) To understand intersectoral collaboration between the Intersectoral collaboration networks enabled growth
organisations working on maternal & child health of the network of organisations working on the
(MCH) and water & sanitation (WASH) before and MCH and WASH. Additionally, intersectoral
immediately after the implementation of the HCL collaboration networks were able to enhance
Foundation (HCLF)-funded HCL Samuday Project organisations’ states in serving as gatekeepers of
(2015–2017) in a rural block of Uttar Pradesh state, information and also enhanced their ability to
India play a coordinator role
(Continued )
8 D. D. GADEKA ET AL.

Table 3. (Continued).
Identified actor
networks Authors (Year) Study objective Key findings
Partnership Soi et al., (2020) To examine Mozambique’s Gavi driven partnership Partnership networks enabled high professional
networks network which delivered human papillomavirus trust. Also, the composition and practices of the
(HPV) vaccine during the demonstration phase partnership resulted in complementary
relationship that engendered a favorable
collaborative environment in which the best that
each institution had to offer was leveraged
Communication Kawonga, Blaauw, To explore the use of social network analysis (SNA) to Communication networks between the actors
networks & Fonn, (2015) measure communication between disease involved enhance the flow of information among
programme and district managers at sub-national provincial programme managers. The study
level in South Africa further shows that communication networks
facilitate complex and varied interactions,
including both collaborative as well as siloed
communication amongst programme and district
actors
Inter-organi Ssengooba, To assess the inter-organisation infrastructure that Inter-organisational networks enabled organisations
sational Kawooya, supports the provision of selected health services in to exhibit a broad range of functional roles in
networks Namakula, & the reconstruction phase after conflict in northern supporting the provision of the services. It further
Fustukian, Uganda shows that inter-organisational networks present
(2017) a better opportunity for organisations to leverage
for faster communication and resource flow to
boost the delivery of health services

Table 4. Summary of how networks support implementation.

Networks Authors (Year) How networks support PHC implementation
Patient/household/ Mukong & Burns, (2020); Mulawa et al., (2016); Chami, Increase the probability of early antenatal check-up and
community-level Ahnert, Voors, & Kontoleon, (2014); Karuga et al., (2019) antenatal completion
networks Ensures alignment in good behavioral practices at the
community level
Enables the engagement of community members in
preventative health.
Help in predicting the susceptibility of individuals to good
health behaviours
Enables local implementers such as Chiefs to implement
government directives
Health facility-level Sabot et al., (2020); Ruducha et al., (2019); Enable the provision of better change in health provider
networks Francetic, Tediosi, & Kuwawenaruwa, (2020); Assegaai & practices
Schneider, (2019) Helps health workers in drawing on sympathetic cadres
identified among the PHC facility staff for support
Facilitate the sharing of patients among health facilities
Helps in building social capital that could lead to a more
accountable and accessible community health delivery
system
Multi-partner networks Hall et al., (2019); Hoe et al., (2019); Soi et al., (2020); Facilitate information and resource sharing among
that work across Kawonga, Blaauw, & Fonn, (2015); Ssengooba, Kawooya, organisations
levels Namakula, & Fustukian, (2017) Enabled growth of the network of organisations serving as
gatekeepers of information
Enabled high professional trust among partners/stakeholders
Facilitate complex and varied interactions among stakeholders
Enable organisations to leverage for faster communication and
resource flow to boost the delivery of health services

promotive services at the community and household and accessible community health delivery system
levels within a municipal ward. One of the reviewed Although the health facility-level networks (Table 4)
articles [18] that explored the influence of patient refer­ were different, it is clear that they work towards
ral networks on policy implementation showed that a common goal of promoting/improving PHC
referral networks influence the treatment of childhood implementation.
illness and non-communicable diseases by facilitating
the sharing of patients among PHC facilities and higher
levels of care as well as support patients from the com­ Multi-partner networks that work across levels
munity access to PHC facilities. Referral networks The review further demonstrated that multi-partner
between women’s self-help groups (SHGs) and local networks that work across levels promote implemen­
health systems in India [24] were also noted to help in tation by facilitating information and resource shar­
expanding relationships at village level, building the ing, high professional trust and effective
social capital that could lead to a more accountable communication among actors (Tables 3 and 4).
 GLOBAL HEALTH ACTION 9

Five of the reviewed articles [22,23,25,27,29] actor networks and how they support PHC imple­
reported on multi-partner networks, involving mul­ mentation in LMICs. It further provides evidence on
tiple partners working together to achieve a common the usefulness of SNA in HPA in understanding how
goal (Tables 3 and 4). They included intersectoral networks support policy implementation and
collaboration networks [23,25], partnership networks research gaps. In this regard, the review makes two
[27], interorganisational networks [29] and commu­ main contributions to the field of health policy imple­
nication networks [22]. The different names of the mentation and the value of SNA in health policy
networks connote the objectives of the networks. implementation work.
Two of the articles reported on intersectoral colla­ These are against the background that as it is
boration networks [23,25]. In one of the articles that common with scoping reviews, the literature search
reported on intersectoral collaborations between was not as methodologically rigorous as that for
organisations working on maternal and child health a systematic review. However, we ensured that only
(MCH) and water and sanitation (WASH) in rural published peer-reviewed articles were included in the
India [23], it was observed that when networks study. Also, the PRISMA-ScR [17] criteria were fol­
involve multiple partners, they enable growth of the lowed to guide the conduct and reporting of the
organisations involved and also provide these orga­ review. We could not compare our findings to evi­
nisations the opportunity to serve as gatekeepers of dence in high-income countries as actor network
information. Similarly, with regard to mental health studies in those countries were not PHC-focused
care, intersectoral collaboration networks were and generally descriptive in nature [31].
found to facilitate information and resource sharing First, the review provides evidence that, actor net­
among the involved organisations as means to works more likely play a role in PHC implementation
strengthening people-centred care in Timor-Leste in all LMICs. However, we could only obtain evidence
[25]. for some countries. In these countries, actor networks
A partnership network was examined in one of the play a role in PHC implementation across different
reviewed papers [27]. The specific focus was levels of the health system. The review shows that
Mozambique’s Gavi-driven partnership network, networks support PHC implementation, offering
which delivered human papillomavirus (HPV) vaccine empirical support for the role of networks highlighted
during the demonstration phase to test a model for HPV in bottom-up policy implementation theory [10,11].
vaccine delivery to girls aged between 9 and 13 years. Our review also shows that networks support imple­
The findings showed that this partnership network mentation at patient/household or community-level,
enabled high professional trust and also a favourable health facility-level and through multi-partner net­
environment for leveraging partners’ capabilities. works that work across levels. In other words,
In terms of inter-organisational networks, patient/household or community-level networks
Ssengooba et al. [29] assessed organisations that sup­ could drive individual, households/families and com­
port the provision of (1) HIV treatment, (2) maternal munity empowerment for increased social participa­
delivery services and (3) workforce strengthening. tion and enhanced self-care and self-reliance in health.
The study showed that the inter-organisational net­ Community empowerment and participation are
work enabled organisations to exhibit a broad range known beneficial strategies for implementation of
of functional roles in supporting the provision of the health services [32]. Moreover, effective community
services. It was further noted that inter-organisational engagement is shown to positively impact on social
networks present opportunities to leverage faster capital that ultimately improves health status and
communication and resource flow to boost the deliv­ reduces health inequalities [24]. For instance, patient/
ery of health services. household or community-level networks could use
Communication networks were specifically exam­ their level of influence to help local implementers
ined by Kawonga et al. [22] who sought to under­ adapt PHC policy to local circumstances in ways that
stand the extent to which the health programme and ensures broad policy and performance gains. Similarly,
district managers communicate in South Africa. The health facility-level networks could ensure comprehen­
study revealed that communication networks sive integrated PHC services, while multi-partner-level
enhance the flow of information among the pro­ networks would enhance multi-sectoral policies and
gramme managers. It also facilitates complex and actions to address wider determinants of health.
varied interactions among programme and district Existing evidence shows that integrated health services
managers. are key to achieving universal coverage [32,33] while
multi-sectoral approaches to health has been identified
as key for implementing health promotion and the
Discussion
prevention and control of diseases [33]. The evidence
This scoping review of papers reporting on the use of on SNA suggests that networks play an important role
SNA in PHC implementation offers insights about in these processes.
10 D. D. GADEKA ET AL.

The review also reveals commonalities across the The review also provides a foundation for such
networks in terms of the way they support imple­ future primary, empirical research. To the best of our
mentation. For example, the exchange of information knowledge, this is the first scoping review on the role of
among actors or implementers was seen across multi­ actor networks in PHC implementation in LMICs.
ple networks. The review also shows that differences Given the very limited number of empirical studies
in networks tend to be driven by the context, goal identified, there is considerable scope for more research
and activities of a network. For instance, friendship on the role of actor networks in PHC implementation,
network, tends to be driven by the context, goal and the paper therefore adds to the literature that docu­
activities of close friends who do not live in same ments actor networks and their role in the implemen­
household but are comfortable to either turn to each tation of PHC. From the review, all networks were
other for advice, interest-free loan, or ask for help identified and defined within a particular context, how­
with harvest without paying [30], while peer net­ ever, there is limited understanding of the influence of
works, were driven by activities of people getting the context on the networks and how the networks are
along or socialising regularly in a fixed location in managed in order to influence implementation.
predicting the susceptibility of individuals to good
health behaviours [20,30]. This supports the argu­ Conclusion
ment of bottom-up theorists that the actors involved
in policy implementation are influenced by the con­ This body of literature we reviewed suggests that, actor
text, goals and activities of a policy. Therefore, in networks exist across different levels and that they make
order to strengthen implementation, these compo­ a difference in PHC implementation. We argue that
nents must be understood [12]. SNA could be a useful approach for health policy ana­
Bottom-up theory emphasises the discretionary lysts looking at implementation. However, because of
power and critical influence of ‘policy implementors’ the limited body of evidence considered in this review,
over policy implementation [12]. Our review shows and the narrow range of country contexts covered in
the use of discretionary power either during the for­ the included studies, we suggest it will be useful to
mation of a network or among the actors in the further explore the use of SNA as part of HPA approach
network. For instance, in the formation of in implementation in a wider range of contexts.
a partnership network for HPV vaccine delivery,
actors used their discretionary power to determine Acknowledgments
who they would partner with [27]. Additionally, for
professional advice networks, health providers used This paper has been partially funded and supported through
the Health Policy Analysis Fellowship programme, by the
their discretion to determine whether to seek or to
Alliance for Health Policy and Systems Research, World
give advice on the provision of antenatal, childbirth, Health Organisation, Switzerland, and in addition, the first
postnatal and newborn care [26]. The consequences author received mentorship and peer support as a Fellow of
of these practices of discretionary power among the 2020 cohort. We are grateful to Zubin Shroff for the
actors within the networks could contribute to the critical comments and reviews of the manuscript.
commonalities or the differences observed but also
demonstrate the significance for understanding net­ Author contributions
works in mitigating the implementation gap.
Our review further shows that SNA could be Dominic Dormenyo Gadeka was responsible for the con­
ceptualisation and design of the study, analysis and inter­
a useful tool in HPA, particularly in providing under­
pretation of the data, and writing and revising the
standing of how networks influence policy imple­ manuscript. Lucy Gilson and Eleanor Whyle contributed
mentation. In the reviewed papers, SNA was applied to the validation of the research design, review and editing
within different contexts and across different actors. analysis and interpretation of the data. Patricia Akweongo,
SNA provided descriptions of the set of actors and Genevieve Cecilia Aryeetey and Justice Moses Aheto
members of actor networks, and characterised the reviewed the paper and provided relevant feedback. All
authors have read and approved the final version of the
relationships between the actors and the influence manuscript.
of these relationships. It was applied within several
research designs and either as a stand-alone method
or in combination with other relevant methods, Disclosure statement
implying the flexibility of SNA as a methodology. In No potential conflict of interest was reported by the
the reviewed studies, SNA provided understanding of author(s).
actors, context, content and processes of policy
implementation, suggesting that this analytic
approach offers value in generating policy relevant Funding information
theoretical insights about networks and their influ­ This paper has been partially funded and supported
ences on policy implementation. through the Health Policy Analysis Fellowship programme,
 GLOBAL HEALTH ACTION 11

by the Alliance for Health Policy and Systems Research, primary healthcare services to persons with physical
World Health Organisation, Switzerland. disabilities in rural Ghana. Prim Health Care Res Dev.
2019;20:20.
[10] Hanf K, Hjern B, Porter D. Local networks of manpower
Paper context training in the federal republic of Germany and Sweden.
In: Hanf K Scharpf FW, editors. Interorganizational
Primary health care improvement is often undermined by policy making: limits to coordination and central con­
implementation gaps in low- and middle-income countries. trol. London: Sage Publication; 1978.
Limited attention has been paid to the role actor networks [11] Hjern B, Porter DO. Implementation structures: a new
could play over the implementation. This body of literature unit of administrative analysis. Organ Stud.
reviewed suggests that, actor networks exist across different 1981;2:211–227.
levels and that they make a difference in primary health [12] Barrett SM. Implementation studies: time for
care implementation. Social Network Analysis may be a a revival? Personal reflections on 20 years of imple­
useful approach to health policy analysis on mentation studies. Public Adm. 2004;82:249–262.
implementation. [13] Elmore RE. Organizational models of social program
implementation. Public Policy. 1978;26:185–22.
[14] Erasmus E, Orgill M, Schneider H, Gilson L. Mapping
ORCID the existing body of health policy implementation
Dominic Dormenyo Gadeka http://orcid.org/0000-0002- research in lower income settings: what is covered
1583-4572 and what are the gaps? Heal Policy Plan. 2014;29:
iii35–50.
[15] Arksey H, O’Malley L. Scoping studies: towards
References a methodological framework. Int J Soc Res Methodol
Theory Pract. 2005;8:19–32.
[1] World Health Organization. Primary health care. [16] Levac D, Colquhoun H, O’Brien KK. Scoping studies:
Report of the international conference on primary advancing the methodology. Implement Sci.
health care, Alma-Ata, USSR. World Heal Organ; 2010;5:1–9.
1978. [17] Tricco AC, Lillie E, Zarin W, O’Brien KK,
[2] Adongo PB, Tapsoba P, Phillips JF, Tabong PTN, Colquhoun H, Levac D, et al. PRISMA extension for
Stone A, Kuffour E, et al. The role of scoping reviews (PRISMA-ScR): checklist and
community-based health planning and services strat­ explanation. Ann Intern Med. 2018;169:467–473.
egy in involving males in the provision of family [18] Francetic I, Tediosi F, Kuwawenaruwa A. A network
planning services: a qualitative study in Southern analysis of patient referrals in two district health sys­
Ghana. Reprod Health. 2013;10:1–15. tems in Tanzania. Health Policy Plan. 2020;36:1–14.
[3] Christopher J, May AL, Lewin S. Review of the impact [19] Mukong AK, Burns J. Social networks and antenatal
of community health workers delivering curative care utilisation in Tanzania. Sci African [Internet].
interventions against malaria, pneumonia and diar­ 2020;9:e00535.
rhoea on child mortality and morbidity in. Hum [20] Mulawa M, Yamanis TJ, Hill LM, Balvanz P,
Resour Health. 2011;27:1–11. Kajula LJ, Maman S. Evidence of social network influ­
[4] Bitton A, Fifield J, Ratcliffe H, Karlage A, Wang H, ence on multiple HIV risk behaviors and normative
Veillard JH, et al. Primary healthcare system perfor­ beliefs among young Tanzanian men. Soc Sci Med.
mance in low-income and middle-income countries: 2016 Mar 1;153:35–43.
a scoping review of the evidence from 2010 to 2017. [21] Assegaai T, Schneider H. The supervisory relation­
BMJ Glob Heal. 2019;4:e001551. ships of community health workers in primary health
[5] Bitton A, Ratcliffe HL, Veillard JH, Kress DH, care: social network analysis of ward-based outreach
Barkley S, Kimball M, et al. Primary health care as teams in Ngaka Modiri Molema District, South Africa.
a foundation for strengthening health systems in low- BMJ Glob Heal. 2019;4:1–9.
and middle-income countries. J Gen Intern Med. [22] Kawonga M, Blaauw D, Fonn S. Exploring the use of
2017;32:566–571. social network analysis to measure communication
[6] Stenberg K, Hanssen O, Bertram M, Brindley C, between disease programme and district managers at
Meshreky A, Barkley S, et al. Guide posts for invest­ sub-national level in South Africa. Soc Sci Med.
ment in primary health care and projected resource 2015;135:1–14.
needs in 67 low-income and middle-income countries: [23] Hoe C, Adhikari B, Glandon D, Das A, Kaur N,
a modelling study. Lancet Glob Health [Internet]. Gupta S. Using social network analysis to plan, pro­
2019;7:e1500–10. mote and monitor intersectoral collaboration for
[7] World Health Organization. Primary health care. health in rural India. PLoS ONE. 2019;14:1–12.
World Heal Organ [Internet]; 2019. Available from: [24] Ruducha J, Hariharan D, Potter J, Ahmad D,
https://www.who.int/news-room/fact-sheets/detail/pri Kumar S, Mohanan PS, et al. Measuring coordination
mary-health-care between women’s self-help groups and local health
[8] Awoonor-Williams J, Tindana P, Dalinjong P, systems in rural India: a social network analysis.
Nartey H, Akazili J. Does the operations of the BMJ Open. 2019;9:1–13.
National Health Insurance Scheme (NHIS) in Ghana [25] Hall T, Kakuma R, Palmer L, Minas H, Martins J,
align with the goals of primary health care? Armstrong G. Intersectoral collaboration for
Perspectives of key stakeholders in northern Ghana. people-centred mental health care in timor-leste: a
BMC Int Health Hum Rights [Internet]. 2016;16:1–11. mixed-methods study using qualitative and social net­
[9] Dassah E, Aldersey HM, McColl MA, Davison C. work analysis. Int J Ment Health Syst [Internet].
Healthcare providers’ perspectives of providing 2019;13:1–13.
12 D. D. GADEKA ET AL.

[26] Karuga RN, Kok M, Mbindyo P, Hilverda F, Otiso L, northern Uganda. Health Policy Plan.
Kavoo D, et al. “It’s like these CHCs don’t exist, are 2017;32:1193–1202.
they featured anywhere?”: social network analysis of [30] Chami GF, Ahnert SE, Voors MJ, Kontoleon AA.
community health committees in a rural and urban Social network analysis predicts health behaviours
setting in Kenya. PLoS ONE. 2019;14:1–19. and self-reported health in African villages. PLoS
[27] Soi C, Shearer J, Chilundo B, Muchanga V, ONE. 2014;9:e103500.
Matsinhe L, Gimbel S, et al. Global health systems [31] Chambers D, Wilson P, Thompson C, Harden M.
partnerships: a mixed methods analysis of Social network analysis in healthcare settings:
Mozambique’s HPV vaccine delivery network actors. a systematic scoping review. PLoS ONE. 2012;7:
BMC Public Health. 2020;20:1–13. e41911.
[28] Sabot K, Blanchet K, Berhanu D, Spicer N, [32] Marston C, Hinton R, Kean S, Baral S, Ahuja A,
Schellenberg J. Professional advice for primary health­ Costello A, et al. Community participation for
care workers in Ethiopia: a social network analysis. transformative action on women’s, children’s and
BMC Health Serv Res. 2020;20:1–16. adolescents’ health. Bull World Health Organ.
[29] Ssengooba F, Kawooya V, Namakula J, Fustukian S. 2016;94:376–382.
Application of social network analysis in the assess­ [33] Narain J. Integrating services for noncommunicable dis­
ment of organization infrastructure for service deliv­ eases prevention and control: use of primary health care
ery: a three district case study from post-conflict approach. Indian J Community Med. 2011;36:67–71.