Journal of Cancer Policy 16 (2018) 26–32
Contents lists available at ScienceDirect
Journal of Cancer Policy
journal homepage: www.elsevier.com/locate/jcpo
Development of paediatric oncology shared-care networks in low-middle T
income countries
Elizabeth Burnsa, Meg Collingtona, Tim Edena,b,⁎, Piera Frecceroa, Lorna Rennerc,
Vivian Paintsild, Mae Dolendoe, Afiqul islamf, Aye Aye Khaingg, Jon Rossera
aWorld Child Cancer UK, London, UK
bAcademic Unit of Paediatric and Adolescent Oncology, University of Manchester, UK
c Korle Bu Teaching Hospital, Accra, Ghana
d Komfo Anokye Teaching Hospital, Kumasi, Ghana
e Southern Philippines Medical Centre, Davao, Philippines
f Bangabandhu Sheikh Mujib Medical University, Dhaka, Bangladesh
g Yangon Children’s Hospital, Yangon, Myanmar
A R T I C L E I N F O A B S T R A C T
Keywords: In order to reach out and offer optimum access for children with cancer across each country the development of
Optimising childhood cancer care networks of paediatric oncology units rather than just a single centre in one city has been the practice in high
Shared-care networks-in developing countries income countries over the last 30–40 years and is now being considered by developing countries. At a workshop
bringing delegates from 4 countries in Africa and Asia to share their concepts, ideas and experiences of devel-
oping such networks and how to move forward, it was recommended that there should be a Hub (referral
centre), shared-care centre hospitals, a real emphasis on good communication, ability to train staff within the
network and a focus on ensuring that each hospital and country’s government must be supportive for it to be
successful.
1. Introduction reduce travel times to hospital, financial burdens on families and rapid
access at the time of any acute illness to a team with knowledge of the
In 2014 World Child Cancer (WCC) was awarded UK Government patient and the treatment required.
(Department for International Development) grants to fund the devel- The budget for this work included an allowance at the end of the
opment of paediatric oncology shared-care centre networks in Ghana programme for writing up the findings as a resource guide or ‘blueprint
and Bangladesh [1]. The objective of the grant was to work towards the plan’, to share lessons learned and to be made available to others at-
achievement of MDGs 1, 4 and 6 in Bangladesh and Ghana by im- tempting similar work in low and middle-income countries. There is a
proving diagnosis, treatment and care for over 4700 children with relative dearth of such literature on network development in low-
cancer. This was measured by a number of indicators including the middle income countries.
number of children diagnosed with cancer across the shared-care net-
works in Bangladesh and Ghana and the average% increase in survival 1.1. Key definitions
rates for easily treatable child cancers across the shared-care networks
in Bangladesh and Ghana. The 3-year grants enabled improvements A discussion was generated around the most appropriate key defi-
both in access to services and the quality of service provision across nitions to use in the development of paediatric oncology networks, and
each country. Improvement in childhood cancer awareness, earlier di- minimum criteria agreed for each to function well. The definitions
agnoses, total care and cancer treatment was achieved across the de- agreed by the group are summarised in Table 1, along with what par-
veloping networks. Such a shared-care centre network comprises a ticipants felt are the minimum criteria. This work was done individually
number of hospitals working together albeit at often different levels of by participants, and then the results discussed and agreed upon as a
provision of cancer care for children. The aim is to coordinate care not group.
in just one centre in each country but to enable patients to access at The terms that were under discussion have been used widely within
least some of their treatment closer to their homes and in consequence this DfID-funded programme and in other programmes supported by
⁎ Corresponding author at: 5 South Gillsland Road, Edinburgh EH105DE, UK.
E-mail address: tim.eden@edentob.co.uk (T. Eden).
https://doi.org/10.1016/j.jcpo.2018.03.003
Received 6 February 2018; Received in revised form 20 March 2018; Accepted 21 March 2018
Available online 30 March 2018
2213-5383/ © 2018 Elsevier Ltd. All rights reserved.
E. Burns et al. Journal of Cancer Policy 16 (2018) 26–32
Table 1
Agreed shared-care network definitions.
Term Definition
Shared-Care Centre A facility with a relationship to the hub centre that is able to administer maintenance chemotherapy treatment, has a basic level of diagnostics and is able
to conduct follow-up of patients. Co-manages patient care and follow-up in conjunction with hub centre.
Shared-Care Network A network of paediatric cancer wards made up of a hub centre and shared-care centres.
Hub Centre A tertiary facility coordinating the activities of linked shared-care centres. Equipped with advanced facilities able to treat and diagnose childhood cancer
and train multi-disciplinary teams. Supportive care may also be given.
Referral Pathway Defined process for referral of paediatric cancer patients between two centres.
WCC. It is beneficial to discuss them in order to gain clarity and con- • Challenges, successes and possible “ways forward” summaries from
sensus moving forward, as some are used synonymously across WCC each country
work. The terms discussed were; hub centre, satellite centre, shared- • Discussion around the ‘ideal’ shared-care network model
care centre network, shared-care centre, referral centre and referral • Agreement on the most appropriate definitions and terms pertaining
pathway. to shared-care networks
The key findings from the discussion are detailed below. • Defining minimum criteria for the agreed terms and examining
these through the lens of the 6 defined WHO building blocks
• Participants agreed that a ‘hub centre’ would always be a ‘referral • Exploration of key steps towards building a shared-care network
centre’, and so using the term hub automatically implies that pa- • A reflection on lessons learned and recommendations for planning
tients will be referred there. Shared-care centres, once at a later this type of programme
stage of development, could also become referral centres so this
term could become confusing if used in isolation. The term ‘hub All participants in the workshop are co-authors of this paper. The
centre’ was therefore agreed to be used, and not ‘referral centre’. lead clinicians from Myanmar and Bangladesh joined the conference
• It was generally agreed that the terms ‘satellite centre’ and ‘shared- on-line and confirmed the outcomes of the discussions.
care’ centre can be defined as the same type of facility. The group
felt that the term ‘shared-care’ is preferred, as this negates any in-
dication of seniority or hierarchy between sites, which could poli- 2.1. Country perspectives
tically cause conflict. It was agreed that WCC would use ‘shared-
care’ centre in any future programme work. 2.1.1. Challenges to developing a shared-care network
• Learning from the information above, the term ‘shared-care net- It was clear from the discussion that the challenges in developing a
work’ is preferred to ‘satellite network’. service for paediatric oncology in a low-middle income country were
• ‘Referral pathway’ was agreed on as the defined process for referral similar across the countries represented. Participants stated that in-
between two centres. creasing patient numbers, with a lack of adequate facilities and/or
trained staff were exerting an increasing burden on the existing team. In
The agreed terms following this exercise were shared-care network, all cases the oncology services were not fully funded or supported by
hub centre, shared-care centre and referral pathway. their governments. The number of new patients can increase faster than
shared-care centres can be developed. Logistical issues such as trans-
porting patients between one centre and another, presented a major
2. Methods
challenge needing a considerable increase in funding which was diffi-
cult to obtain. Communication between hospitals was noted to be a
In order to collect data from various stakeholders to inform this
common area of concern with developing teams often lacking the
document, WCC organised a 2-day workshop in Dubai. Stakeholders
knowledge, technology and time to transfer information. All the dele-
from the two current DfID-funded programmes were invited, along with
gates stressed that strong communication links are essential for the
paediatric oncologists supported by WCC undertaking similar work in
development of networks. In some countries security issues or civil
other countries where networks were in development. Attendees were
conflict can hamper attempts to establish networks (Fig. 1).
included from Ghana, Bangladesh, Myanmar and the Philippines.
The format of the workshop comprised the following areas:
Fig. 1. The Challenges of developing a shared-care network.
27
E. Burns et al. Journal of Cancer Policy 16 (2018) 26–32
Fig. 2. Successes of Shared-Care Network Development.
2.1.2. Successes achieved in the development of shared-centre networks opportunities within the network (Fig. 3).
The most significant benefit reported was the availability of training
opportunities within the networks. As the most essential element for 2.2. The ideal shared-care network
development, it was seen as a key activity to fund. This is helping to
build the capacity of sites to provide higher quality care for all patients. 2.2.1. Hub centre
Shared-care centres were also felt to be a sustainable solution, com- All participants defined a ‘hub’ site at the centre of the network. This
prising home-grown teams of dedicated and empowered staff. The is described as a centre of excellence with dedicated space and staff,
sustainable aspect must come through both local government and where training takes place for the multi-disciplinary team involved in
hospital administrative support, as well as being shown to have a paediatric oncology care across the network.
multiplier effect through the support of local donors and agencies.
Regional solutions, bringing people together and sharing lessons 2.2.2. Shared-care centres
learned were also mentioned as strong elements of building a network. The shared-care centre hospitals are linked with the hub centre by a
As well as the positive developments inside the system, an improvement two-way referral and communication process. They will have at least a
had been seen in care delivered to patients, numbers of patients diag- paediatrician with an interest in oncology and have the ability to per-
nosed and the number of children surviving cancer. Enabling children form basic investigations and/or deliver less complicated treatments to
to access healthcare closer to their homes has reduced treatment refusal patients. Staff from the sites may spend time at the hub receiving spe-
and subsequent abandonment within developing networks (Fig. 2). cialist training or attending workshops. The sites should be strategically
located and accessible to patients in a defined population area away
2.1.3. Lessons learned and recommendations on the way forward for others from the Hub Centre. As they develop they may establish their own
planning such developments mini shared-care centres to further improve accessibility, lighten the
One of the strongest recommendations was to ensure that shared- patient load at their centre and deliver some care closer to patients’
care centres and staff were selected strategically and carefully to ensure homes.
the best chance of success. Collaborative working and good commu- Under-pinning these depictions were strong themes of resourcing
nication are essential elements and should be emphasised at the be- and support, through funding from NGO’s or Governments, and support
ginning, as should the support of the hospital administration. The group from Health Service administration and policy makers.
felt that the best way of working in a network was through sharing and
using the same treatment protocols, developing two-way referral sys- 2.3. WHO building blocks for health system strengthening
tems between centres and sharing successes and any failures. It was also
advised that having a strategic and feasible development plan at the The World Health Organisation describes health systems in terms of
outset was crucial. This must comprise a timeline of actions, planned 6 core components or “building blocks”. These are (i) service delivery
measurable outputs and outcomes, reporting procedures and a budget. (ii) health workforce (iii) health information systems (iv) access to es-
The idea of sustainable development was stressed, through step-by-step sential medicines (v) financing (vi) leadership and governance.
stages for the project, funding support and “train-the-trainer” Participants were earlier asked to consider the criteria necessary for
Fig. 3. Lessons learned from shared-care network development.
28
E. Burns et al. Journal of Cancer Policy 16 (2018) 26–32
Fig. 4. (a) WHO building blocks: service delivery, health workforce, information systems. (b) WHO building blocks: access to essential medicines, financing,
leadership & governance.
a shared-care network to function well (key definitions). Building on this measurable outcomes and financial support are needed for devel-
work to list the minimum criteria for a functioning network, the par- opment into a centre of excellence. Support would ideally be
ticipants rearranged the listed criteria under the most appropriate WHO available for patients and families, to include accommodation,
building block. They were then each given a different stakeholder treatment costs, food and transport.
scenario and asked to re-do the criteria exercise with a ‘different hat • Strong communication networks must link the hub with any shared-
on’. This enabled us to build up a picture of the key elements which are care centres developed, to maintain sustainable relationships and
required for a functioning shared-care network, organised under the develop two-way referral pathways.
WHO framework. Participants also worked together to define the re- • Shared-care centres: need to have an interested doctor, access to
sources needed and the key actors under each block, considering the some basic training, some ward space for oncology patients, the
criteria defined. support of the hospital administration and a basic multi-disciplinary
A summary of the building blocks exercise is included in Fig. 4a and team. Patient data needs to be stored, at least on paper, and there
b, along with resources and key actors identified. must be a relationship with a hub centre.
The stakeholder scenarios used for the exercise were; doctor at a • The over-arching principle of sustainability needs to be emphasised
shared-care hospital, patient or parent, nurse at hub hospital and a data through the availability of training within the system (preferably an
manager at a hub hospital. in-country fellowship programme) and funding. Sharing, learning
and replicating within the system are important elements.
2.4. Key steps to building a shared-care network
2.5. Reflection
Fig. 5 visually depicts the combined thoughts of the group on the
key steps necessary for developing shared-care networks in LMIC. This At the close of the workshop, participants were asked some reflec-
work was done in two groups which each had either an African or an tion questions to draw out any conclusions or pertinent observations
Asian regional focus. A comparison at the end of the exercise showed which had not been captured already, or that the group wished to
that each group had described similar steps. The diagram contains some emphasise. The questions and findings are summarised below.
elements described by individuals as if there were no constraints on Would you do anything differently if starting this work again?
resources and consequently may take time to implement.
The combined steps are more appropriately shown as diagrams, as - There should be an MOU between hub and shared-care centres, to
they are not linear in nature. The elements agreed on are summarised strengthen the understanding of the teams and of hospital man-
below. agement. It should define responsibilities and timeframes. It is also
much easier to achieve conformity and guidelines if hospital man-
• Hub centre: requires a committed doctor with interest/training in agement are involved from the start of the programme.
paediatric oncology, dedicated bed space, a multi-disciplinary team, - There should be oversight from the hub on the activities at shared-
training must be available, patient data accurately recorded, drugs care centres, especially regarding treatment protocols.
available and research opportunities accessible. A health partner- - Communication links need to be stronger, and visits to the shared-
ship with an external developed centre is beneficial. A tangible plan, care centres in person are important.
29
E. Burns et al. Journal of Cancer Policy 16 (2018) 26–32
Fig. 4. (continued)
- A Project Committee or Steering Group is important for decision- - A platform for exchanging information and for funding.
making processes and should comprise a range of stakeholders. This - Organising training and workshops, which are essential for em-
avoids the situation of all decisions being left with the defined powering healthcare staff.
project lead. - ‘Movers and shakers’ of the network development as would not have
- Outcome measures should be developed for each of the health previously thought of linking up with the other centres.
system building blocks. These should include quality of life mea- - Helped put childhood cancer on the political map/scene through
sures, where in the short-term the biggest impact will be seen. advocacy and meetings with policy makers, and kept hospital
management thinking about childhood cancer.
Would you revise the way forward as previously presented to the - Ongoing capacity building, data strengthening.
group? - Development of fundraising skills for parent groups.
- Have not been intrusive, have included stakeholders right from
- Prioritisation on talking and collaborating more within the network. outset and the needs assessment, been supportive, allow in-country
- Development of a network using the structures and systems of the stakeholders to work on programmes themselves and present WCC
WHO building blocks framework. with budget for their needs.
- Would put emphasis on revising treatment guidelines; simplifying - Created an enabling environment for childhood cancer management
these and putting them in a public space for staff to see them clearly. to improve.
- Implement more regular refresher training for staff at shared-care - Enriched the experience of child with cancer and the journey of the
centres to build confidence, and while there needs to be a focus on family.
early warning signs & symptoms topics such as palliative care should - Give the opportunity to share and learn with other projects so they
also be included. don’t go through the same mistakes and can shorten journey to
- Mentoring for centres could be better to dispel feelings of isolation. improved care for paediatric cancer.
Are you taking away any new ideas from this workshop? How could WCC help in the future?
- Encouraging simple research in audit form to enable quality im- - Make more of an impact with capacity building – one week work-
provements. shop rather than a 2-day course.
- WHO building blocks are a useful planning tool and will help with - Strengthen twinning partners. Partners should be proud of the
programme management and strategies. partnership, boast about the achievements and be able to give time.
- Sharing with the centres in the network how they can become sus- Some of the twinning partners have not involved hospital manage-
tainable and emphasise local support rather than waiting for support ment as much as they should have, the involvement is more at
to come from above. clinician level. If management had signed the MOU they may have
- Having ‘celebration of life’ parties to raise awareness and provide had more ownership.
psychological support. - Regional training workshops (with other countries like Malawi,
Cameroon) would be helpful to build capacity. The workshop should
What is the role of World Child Cancer in this work? last for 1 week and focus on just one topic like Leukaemia,
30
E. Burns et al. Journal of Cancer Policy 16 (2018) 26–32
Fig. 5. Key Steps to Building a Paediatric Oncology Shared-Care Network.
Retinoblastoma or Oncology Nursing. It would deal with the issue 3. Discussion
from basics, genetics, treatment, rehabilitation.
- Shared-care centre get-together learning workshops – have not There is such a disparity in survival for children with cancer in low-
brought all the centres together before. income countries compared with those in high-income countries [2–5]
that as paediatricians start to see an increase in children presenting
31
E. Burns et al. Journal of Cancer Policy 16 (2018) 26–32
with cancer at their hospitals they frequently seek help from those in A resource manual will be developed using the information and
established paediatric cancer centres elsewhere. Development of in- diagrams in this report, to be made available to the paediatric oncology
ternational health partnerships has been a good way to provide advice; community for those wishing to develop shared-care programmes.
share expertise gained from experiences of successes and failures over
the last few decades in high income countries; provide support for Conflict of interest
training, and enable technology transfer to help those starting up cancer
care services to overcome the challenges facing them [6–10]. It is es- None of the authors have any conflict of interest.
sential that the ownership of any development is in the hands of those
in each country trying to develop a cancer service and not by the Acknowledgements
supporting high–income partner. Usually with hospital to hospital
partnerships, the initial project is always to create an effective single We would like to thank our partners from Ghana, Bangladesh, the
hub centre with adequate space, staffing, equipment, access to medi- Philippines and Myanmar for participating in this work.
cines and a training programme. However, if all children in a country
who develop a malignancy are to be diagnosed and treated in an ap- References
propriate and timely way a single unit, most often in the country’s
capital, will not be able to reach out to all children who need help. [1] E. Burns, T. Eden, J. Rosser, World child cancer: supporting partnership models in
Creation of shared-care networks is how high-income countries have paediatric oncology, Cancer Control (2016) 22–26.
[2] L.M. McGregor, M.L. Metzger, R. Sanders, V.M. Santana, Paediatric cancer in the
developed their services over the last few decades with significant new millennium: dramatic progress, new challenges, Oncology 21 (7) (2007)
success and is a concept well worth sharing. What each shared-centre 809–820.
can provide for patients depends on adequate staffing, training and [3] K. Pritchard-Jones, R. Pieters, G.H. Reaman, L. Hjorth, P. Downie, G. Calaminus,
capacity. Some hospitals can provide basic investigations, blood tests, et al., Sustaining innovation and improvement in the treatment of childhood cancer:lessons from high income countries, Lancet Oncol. 14 (3) (2013) e95–e103, http://
and sanctuary of first call when a child is ill at home whilst others are dx.doi.org/10.1016/s1470-2045(13)70010-x.
able to provide some or, in due course as they develop, all of treatment [4] I. Magrath, E. Steliarova-Foucher, S. Epelman, R.C. Ribeiro, M. Harif, C.-K. Li, et al.,
after initial diagnostics at a hub centre. All of this has led to develop- Pediatric cancer in low-income and middle-income countries, Lancet Oncol. 14 (3)
(2013) e104–116, http://dx.doi.org/10.1016/S1470-2045(13)70008-1.
ment of “hub and spoke” networks with an essential need for good [5] K. Pritchard-Jones, R. Sullivan, Children with cancer: driving the global agenda,
communication and a collective team approach. There is a relative Lancet Oncol. 14 (March) (2013) 189–191.
dearth of papers describing the optimal way to create such a network [6] J. Hopkins, E. Burns, T. Eden, International twinning partnerships: an effective
method of improving diagnosis, treatment and care for children with cancer in low-
within low-middle income countries [7–9,11]. middle income countries, J. Cancer Policy (1-2) (2013) e8–e19, http://dx.doi.org/
Clearly the major challenges are to obtain hospital administration 10.1016/j.jcpo.2013.06.001.
support and even more importantly, governmental support and finan- [7] R.C. Ribeiro, C.-H. Pui, Saving the children: improving childhood cancer treatment
in developing countries, New Engl. J. Med. 352 (2005) 2158–2160.
cing for such developments. Without universal health coverage, the cost [8] S.C. Howard, M. Pedrosa, M. Lins, A. Pedrosa, C.-H. Pui, R.C. Ribeiro, et al.,
of treating a child with cancer falls on the families [12,13] but service Establishment of a paediatric oncology program and outcomes of childhood lym-
delivery must be the duty of health providers and governments. Ribeiro phoblastic leukaemia in a resource–poor area, J. Am. Med. Assoc. 291 (2004)
2471–2475.
et al. reported that 5-year survival for children in 10 low to middle- [9] G. Masera, F. Baez, A. Biondi, F. Cavilli, V. Conter, A. Flores, et al., North-South
income countries was directly proportional to per capita annual total twinning in paediatric haematology-oncology: the La Mascota Programme:
health expenditure, per capita gross domestic product, per capita gross Nicaragua, Lancet 352 (1998) 1923–1926.
national income, number of physicians and nurses per 1000 population [10] L.A. Renner, C. Segbefia, E. Johnson, E. Burns, V. Sharma, S. Kerr, et al., Challengesand success in a twinning partnership in Ghana, Oncol. News 8 (2) (2013) 20–22.
and annual governmental health–care expenditure per capita [14]. As [11] S. Gupta, R. Rivera-Luna, R.C. Ribeiro, S.C. Howard, Paediatric Oncology as the
countries improve economically then it is essential that the case is made next Global Health priority: the need for National Childhood Cancer Strategies in
to health care providers and governments that they do need to develop low-middle income countries, PLoS Med. 11 (6) (2014) e1001656.
[12] A. Islam, A. Akhter, T. Eden, Cost of treatment for children with acute lympho-
paediatric services including for childhood malignancies. Childhood blastic leukemia in Bangladesh, J. Cancer Policy 6 (2015) 37–43.
cancer has increased significantly worldwide since the 1980s especially [13] S. Mostert, C.M. Sitaresmi, S. Gundy, S. Sutaryo, A.J.P. Veerman, Influence of socio-
following progressive reduction of communicable diseases in the last economic status on childhood acute lymphoblastic leukemia treatment in
Indonesia, Pediatrics (2006) e1600–e1606.
15–20 years [15,16] and it is predicted to become more prevalent in [14] R.C. Ribeiro, E. Steliarova-Foucher, I. Magrath, J. Lemerle, T. Eden, C. Forget, et al.,
LMICs just as it did in high income ones from the 1960s onwards. Baseline status of paediatric oncology care in ten low- income and mid-income
The workshop was a successful way of bringing together stake- countries receiving My Child Matters support: a descriptive study, Lancet Oncol. 9
(2008) 721–729.
holders involved with the shared-care centre development programmes [15] L. Liu, H.L. Johnson, S. Cousens, J. Perin, S. Scott, J.E. Lawn, et al., Global, regional
and other colleagues who had already created a network to share their and national causes of child mortality: an updated systematic analysis for 2010 with
experiences in developing such a network. It was a useful mechanism to time trends since 2000, Lancet 379 (2012) 2151–2161.
[16] E. Steliarova-Foucher, M. Colombet, L.A.G. Ries, F. Moreno, A. Dolya, F. Bray, et al.,
develop coherent steps to potentially help others in the future trying to International incidence of childhood cancer 2001–10: a population-based registry
undertake this type of work, and for informing the next stages of World study, Lancet Oncol. (April) (2017), http://dx.doi.org/10.1016/S1470-2045(17)
Child Cancer programmes. 30186-9.
32