In a recent blog post,⁽²⁾ Devi Sridhar draws attention to the fragmentation that has accompanied the interpretation and implementation of the Millennium Development Goals (MDGs). This “siloed”⁽³⁾ approach has also characterised interventions in global health over the past few decades. Even seemingly broader attempts at strengthening health systems as a whole, especially in developing countries, have tended to focus on at most a few sub-components, ignoring potential system-level effects. This paper argues that this approach has led to a lag in progress regarding population health outcomes. Contrasting these siloed approaches with systems thinking, the discourse ends with recommendations for addressing population health challenges that leverage an understanding of the complexity of population health challenges.

The insufficiency of the siloed approach

Since 2000, the MDGs have served as a rallying point for discourse on global health, with a focus on pressing challenges in the developing world, such as malaria, HIV/AIDS, tuberculosis and maternal and child health. With two years to go to the 2015 MDG deadline, tremendous progress has been made on these issues, with declines in global malaria deaths, dramatic increases in access to antiretroviral therapy, plunges in the incidence rates of tuberculosis and a decrease in the mortality rate of children under the age of five.⁽⁴⁾ There have, however, been lags in related interventions because design did not take into cognisance the varied responses of some agents within the health system, the complexity of the causation of various diseases and could not anticipate some of the negative effects that were subsequently observed. In addition, the opportunity to leverage potential synergies between action towards reducing the prevalence of infectious diseases and other local public health priorities was not maximised.

The persistence of the polio endemic in Nigeria is a case in point: in spite of massive investments⁽⁵⁾ in polio eradication campaigns, Sarah Boseley⁽⁶⁾ points out that currently only 47% of Nigerian children receive the routine triple vaccine that provides protection against polio. In response to these gaps, an intervention that provided financial incentives to volunteers who gave children oral protective drops, was designed. This intervention fell short of expectations for several unforeseen reasons: several fixed-salaried clinic workers deserted health facilities to monitor these interventions for extra remuneration; children received multiple immunisations, which made parents less willing to cooperate; and parents often lost the record of immunisation and as a consequence it is unknown what the true extent of coverage was.

Ill-health and healthcare are complex issues. Even seemingly well understood conditions like diarrhoea are inextricably linked to social determinants like housing, water and sanitation and health system performance. In the same way, interlinked causes have been identified for seemingly unrelated diseases.⁽⁷⁾ For example, the greater majority of death and disability due to multiple chronic diseases have been attributed to three major behavioural risk factors: tobacco use; lack of physical activity; and unhealthy diets.⁽⁸⁾ At the societal level, human interaction is instrumental to the transmission of infectious diseases such as influenza. In light of the foregoing, health cannot be considered a stand-alone phenomenon.⁽⁹⁾ However, approaches to improve population health are often narrow, targeted and do not seem to take these linkages into consideration.

Health workers are important stakeholders within health care delivery and their aggregate performance is an important determinant of progress made on population health outcomes. In the last few centuries, the paradigm of science has been one of reductionism: studying the component parts of systems to ensure deeper understanding and predictability.⁽¹⁰⁾ Is the health worker trained in today’s institutions prepared to confront complex health challenges? In many countries, the education of health professionals has failed to overcome dysfunctional and inequitable health systems because of curricula rigidities, professional silos, static teaching methods, insufficient adaptation to local contexts and commercialism in the professions.⁽¹¹⁾ Indeed, scientific silos, or compartmentalised knowledge, have the potential to impede understanding of the complex inter-relationships among determinants of ill-health.⁽¹²⁾

These limitations extend to decision-making in health systems. As Rifat Atun explains in a recent article on innovation in health systems,⁽¹³⁾ decision-making in health systems is characterised by ‘detail complexity’, referring to a reduction in the amount of information used, the tendency to simplify mental cause-effect maps and limiting to a number of static options when making decisions.⁽¹⁴⁾ Unfortunately, health challenges do not limit their complexity to the capacity of the human brain for comprehension and static solutions cannot remedy dynamic problems. The world has witnessed massive progress on some fronts as a result of targeted and narrow solutions that were effective in varied settings, such as smallpox eradication by vaccination. However, the majority of the recalcitrant health challenges facing populations will demand much more comprehensive efforts. In order to design solutions that optimise the ability of health systems to deliver on results, there is a need to leverage an understanding of systemic determinants and systemic effects.⁽¹⁵⁾

Decoding complexity: Characterising systems thinking for health

The World Health Organization defines a health system as consisting of all organisations, people and actions whose primary intent is to promote, restore or maintain health.⁽¹⁶⁾ In thinking about systems, it is common for people to invoke the “machine” metaphor.⁽¹⁷⁾ Mechanical systems possess characteristics defined by sub-components that interact in a predictable manner. The machine metaphor inspires a certain approach to studying systems performance. This usually involves decomposing the system into component elements, designing a solution for each element and subsequently recomposing the system.⁽¹⁸⁾ Success in mechanical systems also depends on some entity having authority and resources to design the system in this manner.⁽¹⁹⁾ Health systems, however, defy that sort of control, and often the characteristics of the whole system are more than the sum of its component parts. It is therefore argued that understanding health system performance would be best facilitated by applying the metaphor of the system as a living organism: the complex adaptive system (CAS).⁽²⁰⁾

Figure 1: Understanding the systems thinking approach ⁽²¹⁾

The CAS is defined by certain characteristics.⁽²²⁾ Firstly, it is non-linear and dynamic and therefore does not inherently reach fixed equilibrium points. Rather, there is constant and discontinuous change in a CAS.⁽²³⁾ Secondly, such a system is composed of independent agents whose behaviour is based on physical, psychological or social rules rather than the demands of system dynamics. Thirdly, as the needs of agents are not homogenous, conflict of goals and behaviours is likely. Fourthly, agents are intelligent and able to learn and change behaviour with experience so that overall system behaviour changes with time. Fifthly, adaptation and learning among multiple interacting agents leads to self-organisation.  Therefore, system behaviour patterns tend to emerge as either innovations or negative consequences. Depending on the relationship between agents, their behaviour feeds back negatively or positively on other agents. These feedback loops can, depending on the context, cause different and counterintuitive impacts in similar systems. Understanding that these systems are embedded in larger systems additionally raises the need for analysis of relationships across levels of systems in addition to those within the system itself.⁽²⁴⁾ Finally, a CAS obviously has no single point(s) of control. System behaviours are often unpredictable and are more easily influenced than controlled.

Health systems are thus considered open systems with interlinked components that interact in the context within which the health system is situated.⁽²⁵⁾ Systems thinking encourages the deliberate, continuous and comprehensive application of an understanding of a CAS to population health challenges.⁽²⁶⁾ Systems thinking (as illustrated in the figure above) applies this understanding of the characteristics of complex adaptive systems to challenges and the design of potential solutions. However, as with many lofty theories, it is pertinent to understand how this method can be applied in practice and if indeed, there is still a place for siloed approaches within health systems today.

Moving forward on systems thinking agenda

In order to provide a roadmap for action on systems thinking for health, strategies will be discussed under three overarching themes recommended in a recent collaborative effort of health system research experts.⁽²⁷⁾ First and foremost, systems thinking will require collaboration across disciplines, sectors and organisations. Stakeholders must not only reach beyond their area of expertise, but also collaborate with colleagues with different experience, knowledge and goals. Stakeholder perspectives on the health system could vary, so it is important to manage and coordinate partnerships, and ensure ownership of policies and interventions by relevant stakeholders, without compromising objectivity or the needs of the system. Failing this, policy resistance can occur in which seemingly obvious solutions fail or worsen the situations they were designed to address.⁽²⁸⁾ In Tanzania, for example, enrolment rates for community health insurance funds were less than 10% after a decade of implementation. These low rates were attributed to reluctance among district managers to implement because of perceived difficulty, a rushed introduction and their anticipation of a lack of financial sustainability. This frustration is believed to mirror their lack of ownership of the scheme.⁽²⁹⁾ The need for multi-stakeholder collaboration makes reform in education essential. Graduating health students should have a grasp of the determinants of population health and skills to negotiate interdisciplinary practice, research, and advocacy, irrespective of their area of specialisation.⁽³⁰⁾ It is also important that theoretical learning be extended to practical experiences in community settings so that students get a first-hand sense of the complexity of interactions within health systems in communities.⁽³¹⁾

Secondly, as it is understood that contexts continuously change, there must be ongoing, iterative learning. Stakeholders need to continuously adapt, learn and apply new knowledge to population health challenges. Presently, many developing countries are not able to set research or skill development priorities and allocate resources accordingly because resources for capacity building are still driven largely by international sources.⁽³²⁾ It is, however, imperative that countries build within-country capacity to apply a systems analytic perspective.⁽³³⁾ In terms of research methodology, the randomised controlled trial (RCT) is considered the gold standard in medical and public health research. However, RCTs cannot adequately explore complex health system challenges as they involve control of the variables that may be essential to understanding system relationships.⁽³⁴⁾ This provides justification for multi-method research,⁽³⁵⁾ which may include: qualitative health research to elucidate behaviours, perception and culture;⁽³⁶⁾ and systems dynamics modelling which elucidates feedback loops in dynamic systems and will aid evaluation of complex health system interventions and health policy.⁽³⁷⁾

To tie the foregoing points on collaboration and the need for iterative learning, systems thinking justifies multidisciplinary research. There is, however, little institutional incentive to undertake these collaborative, multi-disciplinary research projects.⁽³⁸⁾ The Initiative on the Study and Implementation of Systems (ISIS) of the United States of America and the United Kingdom government’s Foresight programme both involved collaborations that leveraged systems thinking approaches to enhance the understanding of, and design solutions for, population health challenges. ISIS, on the one hand, mobilised multiple stakeholders to study the factors influencing tobacco use and provide evidence to inform decision-making for strategies to deal with these factors.⁽³⁹⁾ The Foresight program, on the other hand, explored the systems view around diabetes and obesity, drawing advice from multiple disciplines.⁽⁴⁰⁾ Both projects illustrate, through their respective successes, the potential benefits involved in multi-disciplinary research on complex health challenges. This is not to suggest that unidisciplinary, reductionist science is completely irrelevant. However, there is an increased need for science that is transdisciplinary, translational and network-centric, as most disease causation is multifactorial, dynamic and nonlinear.⁽⁴¹⁾

In terms of policy and practice, systems thinking will involve consideration of alternative resulting scenarios that take into account interactions within the system and surrounding context.⁽⁴²⁾ It is essential, therefore, that a continuum is established between knowledge generation and translation. In several countries in Sub-Saharan Africa, there is poor data availability and quality to meet the need for information on health indices, sub-system characteristics and contextual factors.⁽⁴³⁾ Information flows play a critical role in driving change in systems, thus there is need for feedback pathways for data and evidence that guides decisions.⁽⁴⁴⁾

Finally, systems thinking calls for transformational leadership: leaders that challenge prevailing paradigms, sacrifice personal and organisational interests for systemic benefit, enhance inter-organisational collaboration,⁽⁴⁵⁾ and advocate for change. In addition, leadership can and should be “distributed” throughout an organisation over time. Systems thinking also advocates for policies based on widely-accepted simple rules. These will facilitate dissemination to front-line practice implementation and rally stakeholders around a common goal of improving the health system as a whole.⁽⁴⁶⁾ Reynolds showed ⁽⁴⁷⁾ that complex flocking, herding and schooling behaviour in animals could emerge from having each animal, such as each fish in a school, apply three simple rules: avoid collisions; match speeds with your neighbours; and move toward the centre of the mass of your neighbours. No central controller or director is needed and each animal can simply apply the rules locally. The behaviour of the system emerges from the interactions and this behaviour is successful in avoiding predators.⁽⁴⁸⁾ Leadership should strive not to control, but to create conditions in which the system can positively evolve naturally over time. Thus, leadership should set forth a vision and simple rules with minimal specifications. In addition, instead of control, appropriate behaviour can be encouraged using incentives. The ultimate aim is to create a wide space in which natural creativity can emerge from local actions within the system.⁽⁴⁹⁾ Because the health system is composed of adaptable parts embedded in a unique context, changes can stimulate other unexpected changes. Thus, rather than agonising over details, the goal is to generate a ‘good enough plan’, observe what happens and include modifications in an evolutionary fashion.⁽⁵⁰⁾

It is important to note that in spite of the immense potential within the systems thinking approach to revolutionise population health, not all interventions require a systems thinking approach. For example, changing microscopy guidelines in a local hospital may not require extensive multidisciplinary collaboration or consideration of alternative scenarios. Systems thinking does not necessarily have to completely replace reductionist approaches, but can complement them.

Conclusion

Julio Frenk, a former Secretary of Health in Mexico who pioneered nationwide health system reform, recently remarked: “There are four revolutions currently underway that will transform health and health systems. These are the revolutions in: a) life sciences; b) information and communications technology; c) social justice and equity; and d) systems thinking to transcend complexity.”⁽⁵¹⁾

As countries continue to battle familiar health challenges such as infectious diseases, they are confronted by new problems such as aging populations, increases in the incidence of chronic disease and health effects of climate change, among other. Even though there will occasionally be a place for siloed approaches in addressing these challenges, it is imperative to also move forward on these complex issues leveraging a systems thinking perspective. This will require multi-stakeholder collaboration and could stimulate innovative, locally-adapted solutions guided by simple rules. Such an approach will enable progress - not only on population health, but through action on the determinants of health, advancement on development as a whole.

Written by Adanna Chukwuma ⁽¹⁾

NOTES:

(1) Contact Adanna Chukwuma through Consultancy Africa Intelligence's Public Health Unit public.health@consultancyafrica.com). This CAI discussion paper was developed with the assistance of Tsholofelo Thomas and was edited by Liezl Stretton.
(2) Sridhar, D., ‘Moving beyond MDG Silos’, September 2010, http://www.globalhealthpolicy.net.
(3) ‘Non-communicable diseases (NCDs): Central to the post-2015 development framework’, The NCD Alliance, 2012, http://www.idf.org.
(4) ‘The Millennium Development Goals report’, United Nations, 2012, http://www.undp.org.
(5) Cooke, G. J. and Tahir, F., ‘Polio in Nigeria: The race to eradication’, The CSIS Global Health Report Center, 2012, http://csis.org.
(6) Boseley, S., ‘Polio vaccination in Nigeria – a series of unfortunate events’, The Guardian, December 2012, http://www.guardian.co.uk.
(7) ‘Draft resolution submitted by the President of the General Assembly: Political declaration of the High-level Meeting of the General Assembly on the Prevention and Control of Non-communicable Diseases’, 2011, United Nations General Assembly, Sixty-sixth session, Agenda item 117.
(8) ‘Preventing chronic diseases: A vital investment’, World Health Organization Global Report, 2005, http://www.who.int.
(9) Ibid.
(10) Plsek, P., 2001. “Redesigning health care with insights from the science of complex adaptive systems”, in Institute of Medicine. Crossing the quality chasm: A new health system for the 21st Century. The National Academies Press, http://www.nap.edu.
(11) Frenk, J., et al., 2010. Health professionals for a new century: Transforming education to strengthen health systems in an interdependent world. The Lancet, 376, pp. 1923–58.
(12) Satterfield, D. and De Bruyn, L.M., 2005. The malignment of metaphor: Silos revisited – repositories and sanctuaries for these times. American Journal of Preventative Medicine, 29, pp. 240–241.
(13) Atun, R., 2012. Health systems, systems thinking and innovation. Health Policy and Planning, 27, pp. iv4-iv8.
(14) Sengupta, K. and Abdelhamid, T. K., 1993. Alternative conceptions of feedback in dynamic decision environments: An experimental investigation. Management Science, 39, pp. 411–28.
(15) Savigny, D. and Adam, T. (eds.), 2009. Systems thinking for health systems strengthening. Alliance for Health Policy and Systems Research, WHO.
(16) ‘Everybody's business: Strengthening health systems to improve health outcomes: WHO's framework for action’, World Health Organization, 2007, http://www.who.int.
(17) Morgan, G., 1997. Images of organization  (2nd edition). Sage: Thousand Oaks, CA.
(18) Rouse, W.B., 2008. Health care as a complex adaptive system: Implications for design and management. The Bridge, pp. 17-25.
(19) Ibid.
(20) Begun, J. W., Zimmerman, B. and Dooley, K., 2003. “Health care organizations as complex adaptive systems”, in Mick, S.M. and Wyttenbach, M. (eds.). Advances in health care organization theory. Jossey-Bass: San Francisco.
(21) Savigny, D. and Adam, T. (eds.), 2009. Systems thinking for health systems strengthening. Alliance for Health Policy and Systems Research, WHO. Modified from: Richmond, B., 2000. The "thinking" in systems thinking: Seven essential skills. Pegasus Communications: Waltham, MA.
(22) Rouse, W.B., 2000. Managing complexity: Disease control as a complex adaptive system. Systems Management, 2, pp. 143–165.
(23) Ibid.
(24) Ibid.
(25) Atun, R.A., et al., 2006. Implementing complex health innovations — primary health care reforms in Estonia: Multimethod evaluation. Health Policy, 79, pp. 79-91.
(26) Best, A., et al., ‘Greater than the sum: Systems thinking in tobacco control’, National Cancer Institute, US Department of Health and Human Services, National Institutes of Health, 2007.
(27) Swanson, R.C., et al., 2012. Rethinking health systems strengthening: Key systems thinking tools and strategies for transformational change. Health Policy and Planning, 27, pp. iv54–iv61.
(28) Meadows, D., 2008. Thinking in systems: A primer. Sustainability Institute: White River, VT.
(29) Lipsky, M., 1980. Street-level bureaucracy: Dilemmas of the individual in public services. Russell Sage Foundation: New York.
(30) Ibid.
(31) Ibid.
(32) White, F., 2002. Capacity-building for health research in developing countries: A manager's approach. Pan American Journal of Public Health, 12, pp. 165-172.
(33) Ibid.
(34) Mabry, P.L., et al., 2010. Systems science: A revolution in public health policy research. American Journal of Public Health, 100, pp. 1161–1163.
(35) Mills, A., 2012. Health policy and systems research: Defining the terrain; identifying the methods. Health Policy and Planning, 27, pp. 1–7.
(36) Atun, R., et al., 2005. Analysis of how the health systems context shapes responses to the control of human immunodeficiency virus: Case-studies from the Russian Federation. Bulletin of the World Health Organization, 83, pp. 730–738.
(37) Atun, R., et al, 2007. High coverage with HAART is required to substantially reduce the number of deaths from tuberculosis: System dynamics simulation. International Journal of STD & AIDS, 18, pp. 267–73.
(38) Ibid.
(39) Leischow, S.J., et al., 2008. Systems thinking to improve the public's health. American Journal of Preventive Medicine, 35(2 Suppl), pp. S196-S203.
(40) Butland, B., et al., ‘Foresight: Tackling obesities: Future choices’, Government Office for Science, 2007, http://www.bis.gov.uk.
(41) Barabasi, A.L., 2007. Network medicine — from obesity to the “diseasome”. New England Journal of Medicine, 357, pp. 404 –407.
(42) Ibid.
(43) Oomman, N., Bernstein, M. and Rosenzweig, S., ‘Seizing the opportunity on AIDS and health systems’, Centre for Global Development, 2008; Health Metrics Network, 2008. Framework and standards for country health information systems (2nd edition). World Health Organization: Geneva.
(44) Ibid.
(45) Best, A. and Holmes, B., 2010. Systems thinking, knowledge and action: Towards better models and methods. Evidence & Policy: A Journal of Research, Debate and Practice, 6, pp. 145–59.
(46) Ibid.
(47) Reynolds, C.W., 1987. Flocks, herds, and schools: A distributed behavioral model. Computer Graphics, 21, pp. 25–34.
(48) Ibid.
(49) Ibid.
(50) Ibid.
(51) Frenk, J., ‘Acknowledging the past, committing to the future’, Address delivered to introduce the new dean, Harvard School of Public Health, Boston, Massachusetts, 5 September 2008, http://www.hsph.harvard.edu.