Multi-System Problems Need Multi-System Solutions
Multi-System Problems Need Multi-System Solutions
Communities are interdependent networks of people, ideas and things connected by their relationships. The vitality of the network and its component parts depend on the number, quality and distribution of relationships. As complex, self-organizing collections of intelligent people and organizations in a place and time, communities are diverse and unpredictable – and it follows that they will respond in equally diverse and unpredictable ways to various inputs, including interventions intended to solve problems.
Yet despite this, we continue to approach complex community problems as if they were academic questions to be solved by breaking them down into their component parts. But while compartmentalizing seems like the natural course to take when tackling complex community problems, treating social issues as if they were independent when they are actually part of a bigger picture often results in solutions that fail – worsening, multiplying or leaving unchanged the problem we set out to solve.
To explore the potential negative impact of over-simplifying and over-compartmentalizing social problems, consider how we have traditionally approached healthcare and education, two of the most important inputs of human value in the United States.
In economics, human value, or human capital, is calculated by combining expected years lived between the ages 20 and 64 – adjusted for health – with years of education completed (0-18) – adjusted for learning. The key elements – health and education – form the composite metric used in a study by the Institute for Health Metrics and Evaluation at the University of Washington to rank and measure the growth of human capital in 195 countries during the period from 1990 to 2016.
Published by the medical journal Lancet in 2018, the study showed that growth of human capital in the United States has fallen compared to other countries, dropping from 6th to 27th place. This change reflects the health and education improvements of other nations – and our relative stagnation.
Yet we have been far from complacent during this period. We have made major investments in financial and process control for healthcare and implemented financial penalties and rigid teacher accountability in our public education system. In both cases, our response has been ineffective.
Why? Because we have approached problems in healthcare and education as ailments of health and education, with interventions occurring only (or at least primarily) within those respective domains.
The bigger picture.
Only recently have we begun to recognize that these problems – and therefore their solutions – are not so simple. A complex set of community-based determinants significantly influences health and educational attainment. And while the healthcare and educational systems have a definite interest in supporting solutions, the problems they face are too big and diverse for either to solve alone.
Core community services in addition to, and outside the domain and expertise of, healthcare and education are necessary to demonstrably improve these social determinants – and therefore the overall health of our communities. In short, we must recognize that complex, multi-system problems require complex, multi-system interventions.
How do we achieve this? Not by putting social workers in clinics or community outreach coordinators in schools. These approaches represent an inadequate attempt by healthcare and education to internally mirror services already offered within the community, minus the range and specialized expertise needed to significantly improve core services and build capacity.
Instead, we must seek a solution that:
Creates efficient integration and outcome-based accountability for community services.
Organizes around individuals or neighborhoods.
Connects non-traditional providers into the treatment team.
Enables multiple projects to efficiently share services, reducing redundancy and growing capacity and effectiveness.
Community-based coordination networks.
Community-based coordination networks integrate the specialized expertise of multiple service providers organized by shared purpose to deliver specific, sponsor-determined desired outcomes. Internally organized by network science, what is important at this level is how such networks interact to carry out community interventions.
Interventions to address specific persistent problems could be initiated from any sector. For example, an intervention initiated by healthcare would be patient-centered from a clinically important subset; for instance, a hospital might initiate an intervention to reduce return visits to the emergency room among homeless individuals lacking the stability to pursue follow-up care for acute injuries.
A community network, operating in much the same way as a specialty provider network, would serve the intent of the initiative sponsor and be reimbursed based on outcomes – creating value that would result in a net gain for the sponsor. Internally, the network would delegate sub-outcomes to a range of service providers, together supporting the purpose of the initiative.
Most sub-outcomes would not be unique to each initiative. For example, a sub-outcome in our rehospitalization example might include having the patients apply for unemployment insurance and Medicaid. It is likely that the same sub-outcomes would be applicable for in an initiative sponsored by a school district to improve second-grade reading test scores among children in schools with higher enrollment in the free and reduced-price meals program. As, networks will begin to develop specific, sustainable expertise applicable across initiatives, and the integration of these networks through systems that support evolutionary learning grow value for the community.
Finding a way forward.
We will not solve homelessness simply by building more houses. Eliminating hunger will require more than providing food boxes. Effective health care is more than treating illness.
These complex community problems persist as a result of multiple, interdependent systems adapting to resist change. Therefore, to deliver desired change, we must develop multi-system responses that recognize and address social determinants; the intertwined roots of the problem. This is the second of four principles necessary to improve community health, and it will require community services to coordinate and collaborate at a level few have done before.
In my next article – the third in a series of four – I will explore the importance of applying adaptive responses to adaptive systems
Measuring human capital: a systematic analysis of 195 countries and territories, 1990–2016. Lancet 2018; 392: 1217–34 Published Online September 24, 2018 http://dx.doi.org/10.1016/ S0140-6736(18)31941-X Institute for Health Metrics and Evaluation, University of Washington, Seattle, WA, USA (Prof S S Lim PhD, R L Updike BA, A S Kaldjian MSc, R M Barber BS, K Cowling PhD, H York BA, J Friedman MPH, R Xu BS, J L Whisnant MPH, H J Taylor BA, A T Leever BS, Y Roman MLIS, M F Bryant MPH, J Dieleman PhD, Prof E Gakidou PhD, Prof C J L Murray DPhil); and David Geffen School of Medicine at University of California Los Angeles, Los Angeles, CA, USA