Article Text

Download PDFPDF

Treating diarrhoeal disease in children under five: the global picture
  1. Carla Chan Unger1,
  2. Shumona Sharmin Salam1,
  3. Md Shafiqul Alam Sarker1,
  4. Robert Black2,
  5. Alejandro Cravioto3,
  6. Shams El Arifeen1
  1. 1Centre for Child and Adolescent Health, International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR,B), Dhaka, Bangladesh
  2. 2Centre for Global Health Faculty, Johns Hopkins Centre for Global Health, Baltimore, Maryland, USA
  3. 3International Vaccine Institute, Seoul, Republic of Korea
  1. Correspondence to Carla Chan Unger, Centre for Child and Adolescent Health, International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR,B), 68 Shaheed Tajuddin Ahmed Sharani, Mohakhali, GPO Box 128, Dhaka 1212, Bangladesh; carla.chan.unger{at}


Rates of childhood mortality due to diarrhoea remain unacceptably high and call for renewed global focus and commitment. Affordable, simple and effective diarrhoeal treatments have already been available for many years, yet a shift in international health priorities has seen coverage of recommended treatments slow to a near-standstill since 1995. This article reviews coverage of recommended childhood diarrhoeal treatments (low-osmolarity oral rehydration solution (ORS) and zinc), globally and regionally, and provides an overview of the major barriers to wide-scale coverage. It is argued that to ensure smooth supply and equitable distribution of ORS and zinc, adequate financing, relevant policy changes, strong public, private and non-government organisation (NGO) collaboration, local manufacturing of pharmaceuticals, mass media awareness and campaigning, in conjunction with strong government support, are necessary for successful treatment scale-up.

  • Comm Child Health
  • General Paediatrics
  • Infectious Diseases
  • Epidemiology

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


After pneumonia, diarrhoea remains the second leading cause of death due to infection globally for children aged under 5 years,1 with an estimated 700 000 children younger than five dying of diarrhoea annually.2 Childhood diarrhoea is most prevalent in low-income countries where pathogen transmission is largely caused by poor sanitation and hygiene, contamination of drinking water, and where overall health and nutritional status is low.3 Consequently, particular regions of the world have a higher incidence of diarrhoeal disease. South Asia and Sub-Saharan Africa alone account for over 80% of all childhood deaths due to diarrhoea.4 The majority of deaths (>60%) are clustered within just 10 countries—India, Nigeria, Democratic Republic of Congo, Pakistan, Ethiopia, Kenya, Uganda, Niger, Bangladesh and Tanzania (table 1).4 These are the 10 priority countries for childhood diarrhoea, as well as for pneumonia.

Table 1

Overview of treatment coverage for the 10 countries with the highest number of child deaths due to diarrhoea4 ,24 ,46 ,76

Unfortunately, a shift in international health priorities toward other global emergencies, particularly HIV and malaria, has seen progress in the coverage of diarrhoeal treatments slow to a near-standstill in recent decades.4–6 Trend data suggest that usage of available treatments has virtually stagnated since 19957 despite the unparalleled success of oral rehydration solution (ORS) in reducing child mortality rates during the 1970s and 1980s.8 Pioneered in 1968 by the International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR,B), ORS has been heralded as the most important medical advance of the 20th Century7–9 and is estimated to have prevented some 40 million child deaths since its invention.4 Simple, affordable, evidence-based treatment interventions are available—and have been for over 40 years—yet rates of child mortality due to diarrhoea remain unacceptably high and call for renewed global focus and commitment.

This article reviews current childhood diarrhoeal treatment coverage globally and regionally, providing an overview of the major barriers to wide-scale coverage, which include financial, manufacturing, policy and demand-side constraints. Bangladesh, a country with comparatively high treatment coverage for diarrhoea, is used as a case study in this review (box 1) to illustrate how some of the obstacles can be successfully overcome at the national level, and to identify challenges that remain in the promotion and distribution of recommended treatments. While acknowledging the importance of prevention alongside treatment in reducing rates of child mortality, the scope of this review has been limited to treatment.

Box 1

Case study Fighting childhood diarrhoeal disease in Bangladesh: lessons learned from the national scale-up of oral rehydration solution (ORS) and zinc62

Bangladesh has made great headway in tackling childhood diarrhoeal disease. Today, only 2% of deaths of under-5-year-olds are caused by diarrhoea, down from 20% in 1993.37 This reduction in child mortality is attributable, in part, to the comparatively high level of coverage of ORS and zinc. Close to 80% of children now receive ORS to treat their diarrhoea, while 34% receive ORS (or a homemade fluid) together with zinc supplementation.

Bangladesh enjoys strong government and NGO (non-government organisation) collaboration, which has been critical in the scale-up of diarrhoeal treatment coverage. Two major scale-up initiatives are particularly noteworthy. Over a 10-year period (1980–1990), BRAC, a large home-grown NGO, rolled out an extensive door-to-door Community Health Worker (CHW) Programme that taught over 12 million mothers across the country how to prepare a homemade form of oral rehydration therapy.63 This effort was backed by a government-led mass media campaign and strong promotion and distribution of prepackaged ORS. Similarly, a partnership between the International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR,B) and the Ministry of Health and Family Welfare has been instrumental in kick-starting the world's first national zinc scale-up campaign for children under 5 years of age, which used an innovative social marketing approach.54 By the end of the 3-year campaign in 2009, nearly 90% of urban and 70% of rural caregivers were aware of zinc treatment and its benefits for childhood diarrhoea. This project demonstrated that zinc supplementation, when coupled with appropriate education programmes, can effectively increase rates of ORS use and lower the use of unnecessary antibiotics.49 ,54 ,67

In Bangladesh, the local presence of ICDDR,B has been fundamental in increasing diarrhoeal treatment coverage and reducing childhood mortality. Renowned for its discovery and development of ORS in the 1960s,79 ICDDR,B, based in Dhaka, is the first ever recipient of the Gates Award for Global Health68 and remains an international leader in diarrhoeal disease research. With two specialised hospitals providing population-based surveillance data on diarrhoeal diseases and enteric infections, the Centre is at the forefront of laboratory-based and applied research into diarrhoeal treatment interventions. Studies conducted at the Centre have also helped to build an evidence-base for the efficacy of zinc supplementation17 ,6972 and rotavirus vaccine.7375 Furthermore, ICDDR,B is influential within Bangladesh as it provides technical input to national-level policymaking on diarrhoeal-related issues, enhancing the content of debates and discussion and advising on implementation. Over the years, the Bangladeshi government has successfully pushed through a series of policy changes conducive to the scale-up of diarrhoeal treatments. ORS and zinc have gained over-the-counter approval, appear on the essential medicines list, and have been incorporated into integrated management of childhood illnesses (IMCI) guidelines. Since diarrhoeal management was first subsumed under IMCI in 2002, training on IMCI guidelines has been rolled out to public facilities in 85% of the country's districts, and to a lesser extent incorporated into home-based care. Recognising the need to expand activities to include informal health providers, a Basic

Health Worker package was adapted at the end of 2011 specifically to train CHWs in diarrhoea and pneumonia case management. Bangladesh's 5-year plan (2011–2016) to expand facility-based and community-based IMCI coverage prioritises high-mortality populations and districts. Even more recently, the government of Bangladesh finalised (although it has not yet endorsed) a national scale-up plan for diarrhoeal treatment under the auspices of the Diarrhoea and Pneumonia Working Group.

Bangladesh also benefits from a large and competitive private sector pharmaceutical industry and has gone through a transition from being a drug-importing to a drug-exporting country, along with a move away from the promotion of homemade fluids toward the use of standardised, prepackaged ORS. Amendments to the national drug policy now require all companies to adhere strictly to Good Manufacturing Practice (GMP) guidelines, thereby ensuring quality regulation and control. Bangladesh locally manufactures over 18 000 registered allopathic products, including around 95 brands of zinc tablets and syrups, and a limited number of ORS products. In order to maintain a reliable supply of ORS in public sector facilities, the government's own Essential Drug Company Limited (EDCL) manufactures and distributes ORS, but has yet to start producing any form of zinc therapy.

Alongside this undeniable progress, some challenges still remain in the scale-up of ORS and zinc. ORS coverage in Bangladesh, albeit high, has hovered around 78% for the past 5 years. Further improvement in ORS coverage requires an understanding of exactly who constitutes the remaining 22% and how best to reach these under-served populations. Zinc coverage in Bangladesh has almost doubled during this same period, yet in spite of laudable gains in usage, actual use is still low, with only one in three children receiving zinc for the treatment of diarrhoea. One constraint to coverage is the lack of appropriate communication to target community-level behavioural change. With the cessation of ICDDR,B's zinc project, awareness campaigning pushing for increased uptake of zinc faded and use began to decline. More effort is needed to revive social marketing campaigns targeting, in particular, schools and government-run health facilities.

Another key constraint to coverage is the relatively high cost of a zinc treatment course. Recent data demonstrate that cost is adversely affecting usage and demand, with the poorest populations at greatest disadvantage.33 The private sector on its own has not been able to guarantee a steady and affordable zinc supply sufficient to meet needs and, as such, the government of Bangladesh is contemplating start-up of production through EDCL as one way to boost zinc supply in the public sector and provide it to service users free of charge. Unicef and Micronutrient Initiative are currently the main providers of zinc to the public sector, but greater involvement of donors is required to fill the funding gap and ensure more consistent public-sector supply. ORS, in comparison, does not face the same set of challenges as zinc. Although predominately provided through the private sector, ORS is a cheaper product than zinc and, therefore, usage rates are more or less the same across income levels.

Coverage of recommended treatments

Severe dehydration is the principle cause of death from acute diarrhoea due to the large loss of water and salt through excretion.7 As such, death is almost entirely preventable if dehydration is appropriately treated.10 ORS contains glucose and electrolytes that work to replenish lost fluids by enhancing the absorption of water back into the body. Since the early inception of ORS, there have been several important developments in diarrhoeal treatment. Compared with the original ORS formula, a reduced-osmolarity ORS formula has been shown to be clinically beneficial,11 ,12 decreasing stool output by 20% and vomiting by 30%.13 The use of ORS becomes even more efficacious when used in conjunction with zinc supplementation.14 ,15 While ORS helps in rehydration, zinc helps in reducing frequency of loose motion and duration of the diarrhoeal attack. Numerous studies have demonstrated that a supplementary 10–14-day treatment course of zinc significantly reduces the severity and duration of diarrhoeal episodes in children16 ,17 as well as a reduction in treatment failure or deaths in diarrhoeal episodes of longer duration (>14 days).18 Clinical trials of other micronutrients and vitamins for the treatment of diarrhoea have not generated sufficient evidence of their efficacy.19–22 Recent findings on the use of probiotics as an adjunct therapy look promising, though remain inconclusive.23–25

In 2004, WHO and Unicef issued a joint statement with the latest diarrhoea treatment recommendations for low-income countries, promoting a switch to low-osmolarity ORS and introduction of zinc as an adjunct therapy.26 Also promoted (albeit half-heartedly) are ‘recommended homemade fluids’ to prevent rehydration where ORS is not available.27 There is, however, lack of concrete evidence showing that homemade fluids can save lives, with most sugar-salt home fluid programmes having been abandoned due to variability in quality, ingredients and concentrations when prepared in community-based settings.10 ,28 ,29

The joint statement further recommends, alongside the administration of ORS and zinc, an increase of fluids in general, continued food intake and breastfeeding to help maintain nutritional status and fight infection.26 Use of antibiotics is recommended by WHO and Unicef only for selective cases of bacterial diarrhoea, especially dysentery or severe shigellosis.26 Randomised control trials from developing countries have shown antibiotic treatment of dysentery infections to reduce rates of clinical failure by 63–82%.30

The widespread introduction of these recommended treatments has been extremely slow when compared with global need. Only 34% of children with diarrhoea in high-burden countries receive either standard or low-osmolarity ORS4 and fewer than 5% receive zinc.31 Although the majority of children (71%) in low-income countries continue to be fed during periods of illness, nearly a third of children receive much less or no food, and only around one-quarter of children with diarrhoea drink more fluids of any type.4

Regionally, Sub-Saharan Africa (30%) and South Asia (33%) have the lowest rates of usage of ORS, despite being the two subregions with the highest need for treatment.4 While there are indications that Sub-Saharan Africa is showing a slight improvement overall,4 rates of ORS use vary tremendously from country to country (table 1), reflecting inter-regional variations in the availability and acceptance of treatments.6 For instance, particularly low use of ORS is observed throughout the Sahel region of West Africa (<20%) while higher levels of use are typically found across countries within southern Africa (50–70%).6 Wide disparities in coverage are similarly observed among South Asian countries: India's ORS coverage rate (26%) is low32 compared with Bangladesh (78%),33 which has the highest rate of ORS coverage of any low-income country. The limited information that is available on the prevalence of zinc supplementation suggests that few countries appear to exceed 6% coverage, with the exception of El Salvador (12%), Democratic People's Republic of Korea (19%)4 and Bangladesh (34%).33

Launched in April 2013 by WHO and Unicef, a new integrated Global Action Plan for Pneumonia and Diarrhoea provides a framework for national governments to plan and implement approaches for prevention and control of the two diseases.34 Goals to be met by year 2025 include reducing mortality from diarrhoea in children under 5 years to fewer than 1 in 1000 live births, and ensuring 90% coverage of appropriate case management at national/regional level with at least 80% access in every district. The Global Action Plan recommends that governments and their partners develop country-level strategies and work plans. Enormous variation in the coverage of treatments between countries underscores the importance of targeted and tailored efforts for a sustained and equitable scale-up of new formulation ORS and zinc. Barriers to coverage are setting-specific, and formative research is required to increase understanding of which local obstacles need to be overcome to ensure that life-saving treatments reach the populations most in need.

Financing shortfalls

Following the successes of earlier decades, the 1990s saw a shift away from vertical diarrhoea-control programmes towards the integrated management of childhood illnesses (IMCI), along with a lessening of dedicated funding for diarrhoea treatment.7 ,35 IMCI is a broad strategy encompassing prevention and treatment interventions for a range of conditions including malaria, malnutrition, acute respiratory infection, immunisation and diarrhoea. With the incorporation of diarrhoea treatment into IMCI, diarrhoea management activities have typically narrowed from countrywide programmes to implementation in small geographical areas.7 ,36

Simultaneously, other diseases, such as HIV and malaria (which result in fewer deaths in children under five compared with diarrhoea), have disproportionately captured the attention of the international donor community, pushing diarrhoeal diseases further down the list of public health priorities, particularly in sub-Saharan Africa.6 ,37 ,38 Between 2007 and 2011, HIV/AIDS received US$6500 million39 and malaria US$1300 million,40 compared with zinc scale-up, which received a total of US$15 million (a figure compiled using limited available data from eight countries).41 ,42 Information on funding for diarrhoeal disease control is not available as a stand-alone line item from any source, and this lack of financial data in itself reflects a low priority for diarrhoea control in recent years.43 It is estimated that dedicated financing for diarrhoea from WHO, United States Agency for International Development (USAID) and Unicef has been reduced to around 5–10% of what it was in the mid-1980s.43

In an effort to regain momentum, a Unicef-led Essential Medicines (Diarrhoea and Pneumonia) Working Group was established in 2011 with its main goal of achieving between 60% and 80% pneumonia and diarrhoea treatment coverage by securing funds to support implementation of treatment scale-up plans for each of the 10 countries with the highest number of child deaths due to pneumonia and diarrhoea.44 The Working Group's current estimates of national funding gaps for pneumonia and diarrhoea (combined) total US$965 million for the 10 priority countries (table 1).45 In the absence of adequate financing, the two leading causes of deaths among young children will prevail largely untreated.

Investment in diarrhoeal treatment is one of the most economical means for donors to achieve a rapid decrease in child mortality. A full course of ORS and zinc costs as little as US$0.50 and associated distribution and marketing costs are comparatively low.46 Reduced-osmolarity ORS is 15–20% cheaper to produce than its predecessor, and upgrading from the original formula does not necessitate any additional funding.14 Zinc production also offers an especially high return on investment for donors;47 leading global economists at the 2008 Copenhagen Consensus-ranked zinc supplementation as the most cost-effective intervention for advancing human development.48

Recent evaluations of zinc treatment scale-up programmes in Indonesia, India and Tanzania, however, have found the cost of zinc to be an obstacle for the poorest population groups.49 The cost of a treatment course varies widely from country to country, underlining the problem of relying on the private sector alone to ensure equitable access to treatment for all users. A zinc course in Nepal, for example, costs US$0.19–US$0.52 contrasting with Indonesia where prices range from between US$0.66 and US$3.50.49 ,50 Governments have an important role to play in balancing affordability and profitability considerations, and must work in collaboration with the manufacturing sector to increase product availability at reduced prices. Moreover, many countries face difficulties with securing funds for initial zinc start-up activities, such as the procurement of zinc, dissemination of guidelines and training of healthcare providers.14 ,48 A boost in global financing for diarrhoeal treatment is imperative if governments are to cover start-up costs and the wide distribution of ORS and zinc, either free of charge or at a nominal subsidised fee.

Policy barriers

There are a number of government policies and actions considered crucial in scaling up treatment successfully. These include the incorporation of zinc and low-osmolarity ORS into national diarrhoea guidelines and IMCI programmes, their classification as over-the-counter (OTC) products, and incorporation of zinc into essential medicines lists.49 Adapting policies to include recommended treatments can be complex, leading to long delays or even termination of the process.51 Indonesia, for example, has pushed through almost all relevant policy changes needed for the wide-scale distribution of zinc,49 except for inclusion of zinc on the national essential medicine list. As the government wants to first conduct its own clinical safety trials in-country, this has limited the use of zinc in public hospitals and health centres and, consequently, coverage in Indonesia remains low. Country ownership of clinical data has been shown to be critical in building confidence among policymakers in the safety of zinc, yet adequate financing to carry out these safety studies is often not available.49 Today, zinc is included on only 15% of national essential medicines lists.52

Many countries have successfully changed their policies to incorporate low-osmolarity ORS and zinc since the release of the 2004 joint WHO/Unicef statement.27 ,31 ,53 There remains, however, a gap between policy change and programme implementation. Fewer than 30 countries have initiated pilot programmes for zinc treatment, and less than 10 countries have attempted nationwide zinc scale-up initiatives, with varying degrees of success.49 ,54–56 Momentum generated by the Diarrhoea and Pneumonia Working Group has led to the first-ever development of national-level scale-up plans for the 10 highest-burden countries.7 To date, approximately half these plans have been endorsed and the other half remain in progress (table 1).

The rate of usage can also hinge on whether sale of ORS and zinc are restricted by licensed pharmacies or receive OTC status, allowing the product to be sold and administered by informal health providers beyond the limited pharmacy network.49 ,57 As opposed to targeting a select group of children for distribution, a strategy of ‘flooding the market’ with ORS and zinc can be pursued safely.46 ORS is often registered as an OTC product, whereas zinc, despite having a strong safety profile, typically requires a prescription before it can be purchased.58 Among the 10 highest-burden countries, OTC approval for the sale of zinc still needs to be achieved in Ethiopia, Democratic Republic of Congo, Uganda and Niger (table 1).45

Manufacturing constraints

ORS and zinc are considered low-profit products compared with more lucrative antibiotics, antimalarial and antiretroviral medications.31 ,53 Relative to Africa—where the production of HIV and malaria drugs take precedence—Asia has a more competitive private sector market for the production of ORS.6 This competition has decreased dependency on donor funds and led to the availability of a wide assortment of product brands in health facilities and local stores. Broadly speaking, pharmaceutical production in Africa has been crippled by diminished flows of direct foreign investment, by a ‘scaling problem’ whereby small country size restrains economies of scale,59 and by an influx of imported counterfeit drugs originating mainly from India and China.60 ,61

Local manufacturing of zinc supplementation is still confined to a small handful of low-income countries6 whose products may or may not meet WHO's Good Manufacturing Practice (GMP) standards.51 In many countries, governments and manufacturers are caught in a catch-22 situation, whereby, companies are unwilling to produce zinc until there have been changes in national policy, while many governments are hesitant to commit to policy change without a guarantee that there will be a locally manufactured product.6 ,51 It is, therefore, critical that countries receive additional financial and technical support to enable them to set up local production of quality standard products and to adapt their national policies accordingly.

Tanzania is the first African country to manufacture zinc treatment for diarrhoea, having initiated production in 2007.49 Since overcoming difficulties with securing start-up funds, a local Tanzanian company has finally received GMP accreditation and, as of 2011, was poised to supply zinc products to other countries across the continent49 although this has not yet come to fruition.42 By comparison, Asian countries with a pre-existing pharmaceutical industry have found local production of zinc products to be swift and feasible.6 With strong backing from the ministry of health and professional medical associations, India effectively used donor funds to catalyse local production of zinc.49 By the end of a 5-year USAID project in 2010, 30 Indian companies were manufacturing and/or distributing zinc for sale in the commercial market.

Low awareness and demand

Well-executed diarrhoea control interventions have demonstrated the importance of generating awareness and demand for ORS and zinc through social marketing campaigns as well as face-to-face contact.63 In the 1980s Egypt, for example, government used televised advertisements to promote ORS and trained nearly 14 000 practicing nurses and 10 000 physicians on the importance of using ORS in the management of diarrhoea.64 As testament to the success of the programme, awareness of ORS among caregivers increased to over 90% and the use of ORS rose quickly to 60%. Egyptian civil registration data shows that diarrhoea mortality fell 82% in infants and 62% in children between 1982 and 1987. Actual use of zinc, however, can lag behind awareness.54 ,65 As new treatment innovations take time for providers and caregivers to integrate into their diarrhoeal management decision making, there is a need for longer-term promotion strategies with a concerted focus on eliciting behavioural change.54

A major challenge to behavioural change is the entrenched perception that ORS and zinc are not ‘medicines’ as they do not provide immediate relief of diarrhoeal symptoms, resulting in a tendency for providers to prescribe and caregivers to demand ineffective and potentially harmful antidiarrhoeal medicines, antibiotics and local remedies.7 ,50 ,66 Providers even continue to prescribe alternative therapies in the face of specific training on new diarrhoeal treatment guidelines.50 An evaluation of a USAID zinc scale-up intervention in Benin, for example, found that 84% of public clinics whose health personnel had received training still recommended an antibiotic or antidiarrhoeal in addition to prescribing zinc.50 Increasing the retail profit margin on ORS and zinc product sales is one way of incentivising providers to change their prescribing behaviours and, in turn, encourage caregivers to comply with recommended ORS and zinc-treatment regimes.

A greater focus and more resources need to be allocated to strategies that improve provider outreach, especially with respect to community case management of childhood diarrhoea. A WHO global review of IMCI implementation (2003) found that ORS usage and fluid intake declined significantly in a number of countries following discontinuation of dedicated diarrhoea control programmes when there was not an adequate alternative provided through community-IMCI activities.36 WHO attributed this to the poor working conditions of community health workers and a high turnover of personnel. Further research is needed to ascertain how IMCI can be harnessed more effectively in future to generate greater awareness and behavioural change among providers and caregivers.


In order to ensure a smooth supply and equitable distribution of ORS and zinc, relevant policy changes, strong public, private and non-government organisation (NGO) collaboration, local manufacturing of treatments and mass media awareness campaigning, in conjunction government support, are required to achieve the successful scale-up of diarrhoeal treatment coverage. Basic IMCI implemented through community healthcare workers shows promise if targeted for high-need populations and hard-to-reach areas, but greater attention must be paid to staff training and retention. Adequate financing coupled with appropriate budgeting is imperative for any scale-up intervention to succeed.

Lessons learned from the successes of national-level diarrhoea treatment scale-up programmes in countries like Bangladesh should be used to promote such intervention efforts in other high-burden countries. At the same time, it is important to recognise that each country has its own distinct set of issues and obstacles that require an individualised approach to treatment scale-up. Formative research to gather locally contextualised information prior to programme design and implementation is a fundamental first step needed to develop and improve on existing interventions that seek to expand diarrhoeal treatment for children under 5 years of age.


Our gratitude to Olivier Fontaine, Shamim Ahmad Qazi and Yolande Robertson for providing clarification on information and resources where needed.


View Abstract


  • Contributors CCU: Designed, wrote and reviewed manuscript. SSS, SEA: Designed, reviewed and revised manuscript. MSAS, RB and AC: Reviewed and revised the draft manuscript. CCU, SEA, SSS conceptualised and designed the paper. CCU carried out the literature review and wrote the first draft. CCU, SEA, SSS, MSAS, RB and AC reviewed and revised the subsequent drafts.

  • Competing interests None.

  • Provenance and peer review Commissioned; externally peer reviewed.