COVID-19 has upended a decade of easy certainties in the understanding of health systems, with a lot of attention now being paid to how to recover from the pandemic and prepare for the next one.
Doing this requires new ideas, as too often efforts to build health systems have been constrained by mental models that are driven by isomorphic mimicry, the habit of “mimic[king] other governments’ successes, replicating processes, systems, and even products of the ‘best practice’ examples.”
To break free of this, we suggest that those committed to improving health outcomes could learn something from a field that has been thinking about building for a long time: architecture.
In a seminal 1965 article, Reyner Banham called out the trap of mimicry in the context of the habit in American architecture of emulating European monumentalism.
He contrasted the shallowness of this approach with the benefits of grounding architecture in local norms and values.
At a moment when the global health community is starting to grapple with the need to decolonize its work, this call to move past imported archetypes and focus on the richness and importance of local traditions has particular value.
But Banham went further than this and proposed a more radical reckoning with the status quo driven by the insight that it is easy to look at a system and fail to recognize that some elements are central to its functioning while others are masquerading as essential but in fact are superfluous.
He contrasted the vital role of the plumbing, electricity, and heating with the walls of a house, which he mused could be removed without impeding the functioning of a home.
Reimagining the system without these unnecessary components opens up a host of new possibilities for how the system can operate.
The present set-up of health services hide the fact that some elements are detracting rather than enhancing the system’s ability to improve people’s lives
Current arrangements—whether for a house or a health system—have a strong hold on our thinking, to the extent that it can be difficult not to be fooled into thinking that they represent an immutable logic.
But just as Banham calls out the fact that walls can obscure what truly makes a house a home, so too can the present set-up of health services hide the fact that some elements are detracting rather than enhancing the system’s ability to improve people’s lives.
Stripping away parts of an overloaded edifice can free up the rest of the system to perform better.
This paper seeks to provoke some reimagining by taking seriously the idea that a house without walls may be better than one with them—that the act of erasure may open up some critical perspectives that challenge the status quo and lead to new ways of improving health outcomes.
These four provocations are not a step-by-step prescription for reform but rather are intended to stimulate new ways of envisioning health systems, whether on the part of a minister of health, a district administrator, or a donor representative.
As highlighted in the boxes below, a number of leaders have already put these ideas into practice—and delivered impressive outcomes as a result.
Many countries suffer from what could be called the tyranny of the health facility model.
A huge weight is placed on these facilities (particularly primary care facilities), driven by the implicit assumption that a single structure is the most effective way of meeting the multiple health needs of vast swathes of the population.
They are expected to deliver an increasingly unmanageable set of services—preventive, promotive, outpatient, and, increasingly, chronic care to address the shifting burden of disease.
The reality is far from this.
Health facility surveys regularly paint a grim picture: most do not have the staffing (either quantity or quality) or the supplies to offer quality care. One recent study of 10 African countries found an average 30% absence rate among healthcare providers, with only 42% of facilities stocked with a set of four basic medicines for common conditions.
To address persistent service delivery challenges, we need to dissolve the walls of health facilities—to move from a mental model fixated on a single point of care responsible for the full range of services to a model in which facilities are seamlessly connected both with the communities in which they are situated and with specialist providers located remotely.
We need to move from a model fixated on a single point of care to a model in which facilities are seamlessly connected with communities and specialist providers
Community health workers (CHWs) are central to this vision, not as isolated volunteers roaming from house-to-house but as formal, compensated parts of the system, connected with other providers through supportive supervision and mobile applications that ensure that their work is guided by national protocols and that data flow into national systems.
Technological advances allow other ways of rethinking the facility model.
Healthcare personnel no longer need to rely on the expertise of the staff physically present; instead, they can send diagnostic imaging to specialists located remotely or benefit from second opinions from providers looking at the same electronic medical records from a different city.
Meanwhile, patients increasingly have new options available to them, whether in the form of a telemedicine consultation that connects them with a nurse or doctor from the comfort of their own home or a text-based system that provides information about key topics and helps individuals understand when visiting a facility is necessary.
Brazil has been a pioneer in thinking differently about the delivery of primary health care.
It goes beyond traditional models of care by using “Family Health Teams” that are each composed of physician, nurse, nurse technician, and four to six full-time community health agents to provide comprehensive and continuous community-based PHC to a defined group of patients.
This approach has delivered impressive gains on a range of health areas, from infant mortality to cardiovascular disease.
The process of planning service delivery has become increasingly more sophisticated in recent decades, with data-driven and results-oriented approaches leading to a blizzard of ever-more-complex tools, processes, and software that are intended to optimally design service delivery systems.
This impressive edifice obscures some cracks in the foundations of these efforts.
Most health services are still designed in an archaic manner: top-down, implicitly resting on a build-it-and-they-will-come mentality.
A strong supply-side bias dominates health systems planning and financing.
The focus is almost always on public sector service delivery, ignoring the fact that in many countries a large proportion of people access services from the private sector.
Planners may concoct grand visions for health reforms but too often little attention is paid to how people actually behave when they are not well.
It is little surprise that health services that are designed without understanding the factors that shape the demand for these services often underperform.
Efforts to improve service delivery should start with deep engagement with those who use the services
The increasingly common habit of bringing a more diverse set of stakeholders into planning processes is a step in the right direction, but the shift must be more profound.
The real experts on service delivery are the mother who has to shuttle between multiple providers in search of a qualified practitioner who can correctly diagnose her daughter’s illness and the person living with HIV who has deep insight into what it will take for him to manage daily antiretroviral therapy and the implications of that for how services are delivered.
Planning needs to escape the walls of ministries of health to engage with these experts by drawing on the approaches of human-centered design.
Instead of prospective changes being developed in insular discussions among technocrats and workshopped in capital cities with elite audiences of donors, NGO leaders, and academics, efforts to improve service delivery should start with deep engagement with those who use the services.
The rapid growth of machine learning and other forms of big data analytics means that it is now possible to complement qualitative research with quantitative understandings of behavior drawn from data sources such as mobile phones and internet search terms.
The insights from both qualitative and quantitative research should lead to inform the rapid development of prototypes that can quickly be tested with end-users.
The NGO Medic Mobile starting working with the Kenyan Ministry of Health and the NGO Kilifi Kids in 2010 on a technology-enabled toolkit to improve the coordination and quality of antenatal care and immunizations delivered by CHWs.
The process engaged CHWs and nurses in an iterative design process that created a system that combined SMS messages to CHWs, data collection via SMS, and a data dashboard for supervisors.
Repeated engagement with users was used to improve the system over time.
More recently, Medic Mobile has been combining these user insights with analysis of usage data from their SMS systems, supported by data science, to further refine a community health toolkit.
Intersectoral collaboration is often discussed but rarely delivered.
Initiatives to build partnerships between health and other sectors tend to focus on the “comparative advantages” of different sectors, which often boils down to the most visible characteristics of each: the education sector offering access to school children, agriculture offering the possibility of improving nutrition, etc.
The challenge has been that these efforts regularly fall victim to competing priorities and limited incentives to work together.
These mental models obscure a more powerful way that the sectors can connect, if they are willing to think beyond their classic roles.
The rapid fall in cost of collecting data is resulting in huge amounts of data being collected by multiple agencies, often on the same people.
Technological advances mean that it is far easier than ever before to draw connections between disparate data sources but progress has been slow in integrating more diverse data sources that capture the social and economic determinants of health that play crucial roles in health outcomes.
These new sources of data are critical because too often the health sector is stuck in the past: considerable investments have been made over the past two decades that have resulted in important improvements in routine data collection, but the types of data gathered have generally been narrow—mostly basic epidemiology and intervention coverage—and overly focused on understanding progress retrospectively, often in the context of the needs of funders.
To cope with an increasingly unpredictable world, we need health systems that can be more adaptive, integrative, and predictive
Although everyone is aware that past performance is no guarantee of future returns, the large investments of time and money in gathering retrospective data have focused attention on these rather than on efforts to integrate an understanding of future trends into the planning of health systems.
Large parts of the world are in the middle of a massive demographic transition and an epochal shift from a burden of disease driven by communicable conditions to one where non-communicable diseases predominate.
Meanwhile, disruptive threats such as climate change, pandemics, and antimicrobial resistance, and macrotrends such as urbanization all have significant consequences for health.
These factors are widely acknowledged but it is rare that countries or donors meaningfully integrate the implications of these longer-term changes into today’s plans.
To cope with an increasingly unpredictable world, we need health systems that can look beyond the walls of the sector and the confines of the present, and be more adaptive, integrative, and predictive.
The Cardiff Violence Prevention Programme in Cardiff, UK, arose from the insight that more than half of the injuries from interpersonal violence seen the emergency department were not being reported by the local police.
A new intersectoral partnership combined data from the emergency department with police intelligence to inform targeted policing and other strategies, leading to impressive reductions in violent injury. This resulted in the approach serving as a model in other settings.
The transition from the Millennium Development Goals to the Sustainable Development Goals was monumental in many ways.
For the health sector, the focus shifted from a concentration on infectious diseases to a much broader agenda.
What did not evolve was the global health architecture that is supposed to be supporting the achievement of these goals: the move from the MDG era to the SDG era brought no changes at all in the arrangement of organizations.
This stasis impacts service delivery because in many low- and middle-income countries, international agencies and donors play key roles in supporting service delivery, but they are doing so using an architecture that is no longer fit-for-purpose and so creates inefficiencies.
The international community owes it to low- and middle-income countries to be bolder
Does the world still need three UN agencies that focus primarily on health (plus a fourth with large health programs focused on children)?
Or three financing mechanisms that are oriented around specific diseases, but none with the lead responsibility for primary health care?
Most of the work of reimagining service delivery must happen at national and sub-national level, but COVID-19 has put on full display just how broken the international system is and how empty its promises of equity are.
The international community owes it to low- and middle-income countries to be bolder in reconceptualizing the global health architecture so it can live up to its ideals.
Mergers are a standard practice the corporate world and a less frequent but still regular occurrence among government departments and non-governmental organizations.
The idea occasionally emerges around the global health architecture, but inertia and risk aversion have reigned supreme: there do not appear to be any examples of large multilateral organizations working on health merging (or even seriously exploring merging), despite the massive shifts in the health landscape in recent decades.